From mailnull@www1.ietf.org  Mon Feb  3 06:48:20 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA03781
	for <sigtran-archive@odin.ietf.org>; Mon, 3 Feb 2003 06:48:19 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h13BrGg14438
	for sigtran-archive@odin.ietf.org; Mon, 3 Feb 2003 06:53:16 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h13Br5J14412;
	Mon, 3 Feb 2003 06:53:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h13Bo2J14233
	for <sigtran@optimus.ietf.org>; Mon, 3 Feb 2003 06:50:02 -0500
Received: from CNRI.Reston.VA.US (localhost [127.0.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA03487;
	Mon, 3 Feb 2003 06:44:35 -0500 (EST)
Message-Id: <200302031144.GAA03487@ietf.org>
Mime-Version: 1.0
Content-Type: Multipart/Mixed; Boundary="NextPart"
To: IETF-Announce: ;
Cc: sigtran@ietf.org
From: Internet-Drafts@ietf.org
Reply-to: Internet-Drafts@ietf.org
Date: Mon, 03 Feb 2003 06:44:35 -0500
Subject: [Sigtran] I-D ACTION:draft-ietf-sigtran-security-01.txt
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

--NextPart

A New Internet-Draft is available from the on-line Internet-Drafts directories.
This draft is a work item of the Signaling Transport Working Group of the IETF.

	Title		: Security Considerations for SIGTRAN Protocols
	Author(s)	: J. Loughney et al.
	Filename	: draft-ietf-sigtran-security-01.txt
	Pages		: 13
	Date		: 2003-1-31
	
This documents discusses how TLS and IPSec can be used to secure the
communication for SIGTRAN protocols.  The support of IPSec is
mandatory for all nodes running SIGTRAN protocols and the support of
TLS is optional.

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-sigtran-security-01.txt

To remove yourself from the IETF Announcement list, send a message to 
ietf-announce-request with the word unsubscribe in the body of the message.

Internet-Drafts are also available by anonymous FTP. Login with the username
"anonymous" and a password of your e-mail address. After logging in,
type "cd internet-drafts" and then
	"get draft-ietf-sigtran-security-01.txt".

A list of Internet-Drafts directories can be found in
http://www.ietf.org/shadow.html 
or ftp://ftp.ietf.org/ietf/1shadow-sites.txt


Internet-Drafts can also be obtained by e-mail.

Send a message to:
	mailserv@ietf.org.
In the body type:
	"FILE /internet-drafts/draft-ietf-sigtran-security-01.txt".
	
NOTE:	The mail server at ietf.org can return the document in
	MIME-encoded form by using the "mpack" utility.  To use this
	feature, insert the command "ENCODING mime" before the "FILE"
	command.  To decode the response(s), you will need "munpack" or
	a MIME-compliant mail reader.  Different MIME-compliant mail readers
	exhibit different behavior, especially when dealing with
	"multipart" MIME messages (i.e. documents which have been split
	up into multiple messages), so check your local documentation on
	how to manipulate these messages.
		
		
Below is the data which will enable a MIME compliant mail reader
implementation to automatically retrieve the ASCII version of the
Internet-Draft.

--NextPart
Content-Type: Multipart/Alternative; Boundary="OtherAccess"

--OtherAccess
Content-Type: Message/External-body;
	access-type="mail-server";
	server="mailserv@ietf.org"

Content-Type: text/plain
Content-ID:	<2003-1-31143013.I-D@ietf.org>

ENCODING mime
FILE /internet-drafts/draft-ietf-sigtran-security-01.txt

--OtherAccess
Content-Type: Message/External-body;
	name="draft-ietf-sigtran-security-01.txt";
	site="ftp.ietf.org";
	access-type="anon-ftp";
	directory="internet-drafts"

Content-Type: text/plain
Content-ID:	<2003-1-31143013.I-D@ietf.org>

--OtherAccess--

--NextPart--


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb  3 11:11:47 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA11657
	for <sigtran-archive@odin.ietf.org>; Mon, 3 Feb 2003 11:11:47 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h13GGoS00743
	for sigtran-archive@odin.ietf.org; Mon, 3 Feb 2003 11:16:50 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h13GG7J00715;
	Mon, 3 Feb 2003 11:16:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h13GEeJ00646
	for <sigtran@optimus.ietf.org>; Mon, 3 Feb 2003 11:14:40 -0500
Received: from pt.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA11609
	for <sigtran@ietf.org>; Mon, 3 Feb 2003 11:09:05 -0500 (EST)
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id LAA14175
	for <sigtran@ietf.org>; Mon, 3 Feb 2003 11:09:15 -0500
Message-ID: <3E3E94EA.2080805@pt.com>
Date: Mon, 03 Feb 2003 11:12:26 -0500
From: Andrew Booth <abooth@pt.com>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.1) Gecko/20020826
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: sigtran@ietf.org
Subject: [sigtran] draft-ietf-sigtran-security-01.txt
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Hi All,

I'm not overly familiar with IPSec or TLS, but here are some questions 
anyway.

1) Section 6. TLS Usage says
 "A SIGTRAN node that initiates a SCTP association to another SIGTRAN
  node acts as a TLS client according to [3], and a SIGTRAN node that
  accepts a connection acts as a TLS server."

I'm under the impression that an SCTP Endpoint can act as both a client 
and server simultaneously while trying to establish an association.  Can 
a TLS endpoint act a both a client and server also?  Would it just mean 
more certificate exchanges?

2) line 424
  "It is recommended that SIGTRAN peers use the same security mechanism
  (IPSec or TLS) across all its sessions with other SIGTRAN peers."

  This recommendation seems to encourage the use of IPSec, since peers 
are required to support IPSec while TLS is optional.  For example, 
suppose that a server interacts with implementations from several 
vendors, where additional vendors could be added later, then there is no 
guarantee that the new implementations would support TLS.   So the 
server would be "recommended" to use IPSec.  All of this is technically 
fine, it just seems to stack the odds against general TLS support.

Comments?
Andrew



Internet-Drafts@ietf.org wrote:

> A New Internet-Draft is available from the on-line Internet-Drafts 
> directories.
> This draft is a work item of the Signaling Transport Working Group of 
> the IETF.
>
>     Title        : Security Considerations for SIGTRAN Protocols
>     Author(s)    : J. Loughney et al.
>     Filename    : draft-ietf-sigtran-security-01.txt
>     Pages        : 13
>     Date        : 2003-1-31
>     
> This documents discusses how TLS and IPSec can be used to secure the
> communication for SIGTRAN protocols.  The support of IPSec is
> mandatory for all nodes running SIGTRAN protocols and the support of
> TLS is optional.
>
> A URL for this Internet-Draft is:
> http://www.ietf.org/internet-drafts/draft-ietf-sigtran-security-01.txt
>
> To remove yourself from the IETF Announcement list, send a message to 
> ietf-announce-request with the word unsubscribe in the body of the 
> message.
>
> Internet-Drafts are also available by anonymous FTP. Login with the 
> username
> "anonymous" and a password of your e-mail address. After logging in,
> type "cd internet-drafts" and then
>     "get draft-ietf-sigtran-security-01.txt".
>
> A list of Internet-Drafts directories can be found in
> http://www.ietf.org/shadow.html or 
> ftp://ftp.ietf.org/ietf/1shadow-sites.txt
>
>
> Internet-Drafts can also be obtained by e-mail.
>
> Send a message to:
>     mailserv@ietf.org.
> In the body type:
>     "FILE /internet-drafts/draft-ietf-sigtran-security-01.txt".
>     
> NOTE:    The mail server at ietf.org can return the document in
>     MIME-encoded form by using the "mpack" utility.  To use this
>     feature, insert the command "ENCODING mime" before the "FILE"
>     command.  To decode the response(s), you will need "munpack" or
>     a MIME-compliant mail reader.  Different MIME-compliant mail readers
>     exhibit different behavior, especially when dealing with
>     "multipart" MIME messages (i.e. documents which have been split
>     up into multiple messages), so check your local documentation on
>     how to manipulate these messages.
>        
>        
> Below is the data which will enable a MIME compliant mail reader
> implementation to automatically retrieve the ASCII version of the
> Internet-Draft.
>  
>

-- 
Andrew Booth, Software Developer
Signaling Systems Group, Performance Technologies
613-237-4344

Visit www.pt.com/mailing.html to subscribe to our Packets & Signals eLetter.


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb  3 16:10:22 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA22002
	for <sigtran-archive@odin.ietf.org>; Mon, 3 Feb 2003 16:10:21 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h13LFUP26251
	for sigtran-archive@odin.ietf.org; Mon, 3 Feb 2003 16:15:30 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h13LEjJ26192;
	Mon, 3 Feb 2003 16:14:46 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h13LDMJ26146
	for <sigtran@optimus.ietf.org>; Mon, 3 Feb 2003 16:13:22 -0500
Received: from ilsa.franken.de (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA21907
	for <sigtran@ietf.org>; Mon, 3 Feb 2003 16:07:41 -0500 (EST)
Received: from lurchi.franken.de (pD958FE1C.dip.t-dialin.net [217.88.254.28])
	by ilsa.franken.de (Postfix) with ESMTP
	id 2760724648; Mon,  3 Feb 2003 22:09:59 +0100 (CET)
Date: Mon, 3 Feb 2003 22:06:49 +0100
Subject: Re: [sigtran] draft-ietf-sigtran-security-01.txt
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v551)
Cc: sigtran@ietf.org
To: Andrew Booth <abooth@pt.com>
From: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>
In-Reply-To: <3E3E94EA.2080805@pt.com>
Message-Id: <69424B8C-37BB-11D7-9521-0030654C1AB6@lurchi.franken.de>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.551)
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Hi Andrew,

see my comments below.

Best regards
Michael

On Monday, Feb 3, 2003, at 17:12 Europe/Berlin, Andrew Booth wrote:

> Hi All,
>
> I'm not overly familiar with IPSec or TLS, but here are some questions 
> anyway.
>
> 1) Section 6. TLS Usage says
> "A SIGTRAN node that initiates a SCTP association to another SIGTRAN
>  node acts as a TLS client according to [3], and a SIGTRAN node that
>  accepts a connection acts as a TLS server."
>
> I'm under the impression that an SCTP Endpoint can act as both a 
> client and server simultaneously while trying to establish an 
> association.  Can a TLS endpoint act a both a client and server also?  
> Would it just mean more certificate exchanges?
You are correct, SCTP supports the peer-to-peer model, whereas TLS uses 
the client-server model.
>
> 2) line 424
>  "It is recommended that SIGTRAN peers use the same security mechanism
>  (IPSec or TLS) across all its sessions with other SIGTRAN peers."
>
>  This recommendation seems to encourage the use of IPSec, since peers 
> are required to support IPSec while TLS is optional.  For example, 
> suppose that a server interacts with implementations from several 
> vendors, where additional vendors could be added later, then there is 
> no guarantee that the new implementations would support TLS.   So the 
> server would be "recommended" to use IPSec.  All of this is 
> technically fine, it just seems to stack the odds against general TLS 
> support.
We had to make a choice and selected IPSec. So if you are in a 
multivendor environment the only assumption
you can make is, if you only buy equipment which is compliant to this 
'RFC', it supports IPSec. Even without
the recommendation that you use only one technology. So if wou provide 
the environment, you will focus on IPSec.
But I think the operators will make a choice and request that from the 
vendors. And if an operator wants to
use TLS he can use it, if he want to use IPSec he can do it.
>
> Comments?
> Andrew
>
>
>
> Internet-Drafts@ietf.org wrote:
>
>> A New Internet-Draft is available from the on-line Internet-Drafts 
>> directories.
>> This draft is a work item of the Signaling Transport Working Group of 
>> the IETF.
>>
>>     Title        : Security Considerations for SIGTRAN Protocols
>>     Author(s)    : J. Loughney et al.
>>     Filename    : draft-ietf-sigtran-security-01.txt
>>     Pages        : 13
>>     Date        : 2003-1-31
>>     This documents discusses how TLS and IPSec can be used to secure 
>> the
>> communication for SIGTRAN protocols.  The support of IPSec is
>> mandatory for all nodes running SIGTRAN protocols and the support of
>> TLS is optional.
>>
>> A URL for this Internet-Draft is:
>> http://www.ietf.org/internet-drafts/draft-ietf-sigtran-security-01.txt
>>
>> To remove yourself from the IETF Announcement list, send a message to 
>> ietf-announce-request with the word unsubscribe in the body of the 
>> message.
>>
>> Internet-Drafts are also available by anonymous FTP. Login with the 
>> username
>> "anonymous" and a password of your e-mail address. After logging in,
>> type "cd internet-drafts" and then
>>     "get draft-ietf-sigtran-security-01.txt".
>>
>> A list of Internet-Drafts directories can be found in
>> http://www.ietf.org/shadow.html or 
>> ftp://ftp.ietf.org/ietf/1shadow-sites.txt
>>
>>
>> Internet-Drafts can also be obtained by e-mail.
>>
>> Send a message to:
>>     mailserv@ietf.org.
>> In the body type:
>>     "FILE /internet-drafts/draft-ietf-sigtran-security-01.txt".
>>     NOTE:    The mail server at ietf.org can return the document in
>>     MIME-encoded form by using the "mpack" utility.  To use this
>>     feature, insert the command "ENCODING mime" before the "FILE"
>>     command.  To decode the response(s), you will need "munpack" or
>>     a MIME-compliant mail reader.  Different MIME-compliant mail 
>> readers
>>     exhibit different behavior, especially when dealing with
>>     "multipart" MIME messages (i.e. documents which have been split
>>     up into multiple messages), so check your local documentation on
>>     how to manipulate these messages.
>>               Below is the data which will enable a MIME compliant 
>> mail reader
>> implementation to automatically retrieve the ASCII version of the
>> Internet-Draft.
>>
>
> -- 
> Andrew Booth, Software Developer
> Signaling Systems Group, Performance Technologies
> 613-237-4344
>
> Visit www.pt.com/mailing.html to subscribe to our Packets & Signals 
> eLetter.
>
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>
>
--
Michael.Tuexen@micmac.franken.de

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb  3 17:20:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA23733
	for <sigtran-archive@odin.ietf.org>; Mon, 3 Feb 2003 17:20:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h13MQ8530826
	for sigtran-archive@odin.ietf.org; Mon, 3 Feb 2003 17:26:08 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h13MPeJ30790;
	Mon, 3 Feb 2003 17:25:41 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h0V9kuJ14340
	for <sigtran@optimus.ietf.org>; Fri, 31 Jan 2003 04:46:56 -0500
Received: from beamer.mchh.siemens.de (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA02206
	for <sigtran@ietf.org>; Fri, 31 Jan 2003 04:42:57 -0500 (EST)
Received: from blues.mchh.siemens.de (mail2.mchh.siemens.de [194.138.158.227])
	by beamer.mchh.siemens.de (8.9.3/8.9.3) with ESMTP id KAA28344;
	Fri, 31 Jan 2003 10:46:30 +0100 (MET)
Received: from mchh273e.demchh201e.icn.siemens.de ([139.21.200.83])
	by blues.mchh.siemens.de (8.9.1/8.9.1) with ESMTP id KAA20223;
	Fri, 31 Jan 2003 10:46:08 +0100 (MET)
Received: by MCHH273E with Internet Mail Service (5.5.2656.59)
	id <D13GJB72>; Fri, 31 Jan 2003 10:45:41 +0100
Message-ID: <3D72C18DF140D311A54F0008C7DBAC8D02917431@MCHH252E>
From: Gaiser Franz <franz.gaiser@siemens.com>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>,
        Tom George
	 <Tom.George@alcatel.com>
Cc: sigtran@ietf.org
Subject: AW: [Sigtran] M2PA LPO during alignment/proving (was: M2PA versio
	n 7 changes)
Date: Fri, 31 Jan 2003 10:45:39 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2656.59)
Content-Type: text/plain;
	charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h0V9kuJ14341
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Hi All,

two additions to the Link State Transition Diagram below: 

1.) Figure 7. M2PA Link State Transition Diagram:
Missing the handling of receiving LS PO in state ALIGNED READY: Situation: After Proving the local side is in state ALIGNED READY and sends LSR; the peer side is in state ALIGNED NOT READY experiencing Processor Outage and starts sending LS PO. To which state the local M2PA is moving ? Transition to PROCESSOR OUTAGE ? If so, the state PROCESSOR OUTAGE handles both local and remote processor outage and then there must be an arrow from PO to IN SERVICE in case of receiving LS PO Recovered (and there is no local PO situation).

2.) Figure 7. M2PA Link State Transition Diagram:
Missing the handling of receiving User data in state ALIGNED NOT READY. Transition to PROCESSOR OUTAGE ? What is to do with this message then (to be discarded) ?


regards, franz 

-----Ursprüngliche Nachricht-----
Von: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
Gesendet: Donnerstag, 30. Januar 2003 01:07
An: Tom George
Cc: sigtran@ietf.org
Betreff: Re: [Sigtran] M2PA LPO during alignment/proving (was: M2PA
version 7 changes)


Tom,

The problem is the LSC doesn't really have a proving state (only IAC
has proving state).  LSC always moves from INITIAL ALIGNMENT to either
ALIGNED READY or ALIGNED NOT READY.  You might change the diagram as
follows:

 From: 

  |                       |                            |                
  |                       | MTP3 Start                 |                
  |                       V                            |                
  |                 +-----------+                      |                
  |                 |  INITIAL  |                      |                
  +<----------------| ALIGNMENT |--------------------->+                
  |      MTP3 Stop  +-----------+     SCTP Comm Error  |                
  |   OR T2 Expiry        |  ^     OR SCTP Comm Lost   |                
  |   OR Receive LS OOS   |  |                         |                
  |                       |  +-------------------------|--------+      
  |                       |              LPO/LPR       |        |      
  |                       |                            |        |
  |                       | Receive LS Align           |        |
  |                       |                            |        |
  |                       V                            |        |
  |                 +-----------+        LPO/LPR       |        |
  |                 |  PROVING  |<---------------------|--------+
  |<----------------|           |--------------------->+        |
  |      MTP3 Stop  +-----------+     SCTP Comm Error  |        |
  |   OR T3 Expiry        |        OR SCTP Comm Lost   |        |
  |   OR Receive LS OOS   |                            |        |      
  |                       | T4 Expiry                  |        |      
  |                       | Send LS Ready              |        |      
  |                       V                            |        V      
  |                 +-----------+        LPO/LPR       |  +-----------+
  |                 |  ALIGNED  |<---------------------|->|  ALIGNED  |
  +<----------------|   READY   |--------------------->+  | NOT READY |
  |      MTP3 Stop  +-----------+     SCTP Comm Error     +-----------+
  |   OR T1 Expiry        |        OR SCTP Comm Lost         |       |
  |   OR Receive LS OOS   |                                  |   T1  | 
  |   OR Receive LS Align | Receive LS Ready      Receive    | Expiry|
  |                       | OR Receive User Data    LS Ready | OR Rcv|
  |                       |                       OR Receive |    LS |
  |                       |                         LS PO    |  Align|
  |                       |                                  |       |

 To:

  |                       |                            |               
  |                       | MTP3 Start                 |               
  |                       V                            |               
  |                 +-----------+        LPO/LPR       |  +-----------+
  |                 |  INITIAL  |<---------------------|->|  INITIAL  |
  +<----------------| ALIGNMENT |--------------------->+  | ALIGNMENT |
  |      MTP3 Stop  +-----------+     SCTP Comm Error  |  +-----------+
  |   OR T2 Expiry        |        OR SCTP Comm Lost   |        |      
  |   OR Receive LS OOS   |                            |        |      
  |                       | Receive                    |   Recv |      
  |                       | LS Align                   |     LS |      
  |                       |                            |  Align |      
  |                       V                            |        V      
  |                 +-----------+        LPO/LPR       |  +-----------+
  |                 |  PROVING  |<---------------------|->|  PROVING  |
  |<----------------|           |--------------------->+<-|           |
  |      MTP3 Stop  +-----------+     SCTP Comm Error  |  +-----------+
  |   OR T3 Expiry        |        OR SCTP Comm Lost   |        |      
  |   OR Receive LS OOS   |                            | T4 Exp |      
  |                       | T4 Expiry                  |   Send |      
  |                       | Send LS Ready              |  LS PO |      
  |                       V                            |        V      
  |                 +-----------+        LPO/LPR       |  +-----------+
  |                 |  ALIGNED  |<---------------------|->|  ALIGNED  |
  +<----------------|   READY   |--------------------->+  | NOT READY |
  |      MTP3 Stop  +-----------+     SCTP Comm Error     +-----------+
  |   OR T1 Expiry        |        OR SCTP Comm Lost         |       | 
  |   OR Receive LS OOS   |                                  |   T1  | 
  |   OR Receive LS Align | Receive LS Ready      Receive    | Expiry| 
  |                       | OR Receive User Data    LS Ready | OR Rcv| 
  |                       |                       OR Receive |    LS | 
  |                       |                         LS PO    |  Align| 
  |                       |                                  |       | 
  |                       |                                  |       | 
  |                       V                                  V       | 
  |                 +-----------+        LPO/LPR       +-----------+ | 
  |                 |    IN     |<-------------------->| PROCESSOR | | 
  |                 |  SERVICE  |                      |   OUTAGE  | | 
  |                 +-----------+                      +-----------+ | 

This would indicates that the LSC moves from INITIAL ALIGNMENT to
ALIGNED READY or ALIGNED NOT READY regardless of PROVING.  That is what the
line from INITIAL alignment to ALIGNED NOT READY was intended to indicate
in shorthand.

I suppose one could draw it as:

  |                       | MTP3 Start                 |               
  |                       |                            |               
  |                       |   +-------+                |               
  |                       |   |       | LPO/LPR        |               
  |                       V   V       |                |               
  |                 +-----------+     |                |               
  |                 |  INITIAL  |<----+                |               
  +<----------------| ALIGNMENT |--------------------->+               
  |      MTP3 Stop  +-----------+     SCTP Comm Error  |               
  |   OR T2 Expiry        |        OR SCTP Comm Lost   |               
  |   OR Receive LS OOS   |                            |               
  |                       | Receive                    |               
  |                       | LS Align                   |               
  |                       |                            |               
  |                       |   +-------+                |               
  |                       |   |       | LPO/LPR        |               
  |                       V   V       |                |               
  |                 +-----------+     |                |               
  |                 |  PROVING  |<----+                |               
  |<----------------|           |--------------------->+               
  |      MTP3 Stop  +-----------+     SCTP Comm Error  |               
  |   OR T3 Expiry        | |      OR SCTP Comm Lost   |               
  |   OR Receive LS OOS   | |                          |               
  |                       | +--------------------------|--------+      
  |                       |                            |   T4   |      
  |                       | T4 Expiry                  | Expiry |      
  |                       | Send LS Ready              | Send   |      
  |                       V                            | LS PO  V      
  |                 +-----------+        LPO/LPR       |  +-----------+
  |                 |  ALIGNED  |<---------------------|->|  ALIGNED  |
  +<----------------|   READY   |--------------------->+  | NOT READY |
  |      MTP3 Stop  +-----------+     SCTP Comm Error     +-----------+
  |   OR T1 Expiry        |        OR SCTP Comm Lost         |       | 
  |   OR Receive LS OOS   |                                  |   T1  | 
  |   OR Receive LS Align | Receive LS Ready      Receive    | Expiry| 
  |                       | OR Receive User Data    LS Ready | OR Rcv| 
  |                       |                       OR Receive |    LS | 
  |                       |                         LS PO    |  Align| 
  |                       |                                  |       | 
  |                       |                                  |       | 
  |                       V                                  V       | 
  |                 +-----------+        LPO/LPR       +-----------+ | 
  |                 |    IN     |<-------------------->| PROCESSOR | | 
  |                 |  SERVICE  |                      |   OUTAGE  | | 
  |                 +-----------+                      +-----------+ | 

The effect is the same.  The diagram is certainly not intended to be 100%
complete, it is an overview.  The text should be consulted.

--brian

Tom George wrote:                                   (Wed, 29 Jan 2003 17:18:23)
> IMHO if there is a local PO while in INITIAL ALIGNMENT, M2PA should
> continue sending/receiving LS Alignment, i.e., remain in INITIAL
> ALIGNMENT. 
> 
> After the LS Alignment is sent and received, M2PA would move to PROVING
> and send/receive LS Proving messages. 
> 
> When T4 expires, it seems to me that M2PA would go ahead and
> send/receive the LS Ready messages. That's the ALIGNED READY state. 
> 
> In short, it looks to me that if there is a local PO, M2PA should
> continue with alignment and proving and should not go into ALIGNED NOT
> READY until it is completed with the ALIGNED READY state.
> 
> Or if M2PA does go to ALIGNED NOT READY earlier than that, it should
> continue doing what it would have done in INITIAL ALIGNMENT, PROVING,
> and ALIGNED READY. Which looks like a confusing way to explain it. 
> 
> tomG.
> 
> Gaiser Franz wrote:
> > 
> > Dear All,
> > 
> > some remarks to the new draft version 07:
> > 
> > 1.)  Figure 7. M2PA Link State Transition Diagram:
> > Why is there an arrow from state INITIAL ALIGNMENT to state 
> > ALIGNED NOT READY  ? If the local side is in processor outage, 
> > the PROVING period should be passed before moving to 
> > ALIGNED NOT READY state.
>   [...]
> 
> -- 
> * Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
> * 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
> * Alcatel USA, Inc.; 1000 Coit Rd, MS PB7-SWPRO; Plano TX 75075-5813  *
> * I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
> * to advertise or represent its position,  operations, or activities. *
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb  3 17:50:28 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA24393
	for <sigtran-archive@odin.ietf.org>; Mon, 3 Feb 2003 17:50:28 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h13Mtdv32736
	for sigtran-archive@odin.ietf.org; Mon, 3 Feb 2003 17:55:39 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h13MtJJ32725;
	Mon, 3 Feb 2003 17:55:19 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h13MsiJ32689
	for <sigtran@optimus.ietf.org>; Mon, 3 Feb 2003 17:54:44 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA24343
	for <sigtran@ietf.org>; Mon, 3 Feb 2003 17:49:02 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h13MqceW021467;
	Mon, 3 Feb 2003 16:52:38 -0600 (CST)
Message-ID: <3E3EF2B6.2534BDFB@alcatel.com>
Date: Mon, 03 Feb 2003 16:52:38 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA Flush Buffers 
References: <OF058D8FAB.B6526DCE-ONC2256CBE.001D41FE@LocalDomain>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Tolga,

The interface between SCTP and its upper layer does not provide for a
way to flush the SCTP buffers. So the approach we took in M2PA was to
discard the messages waiting to be sent or received. 

If you have some implementation-dependent means for flushing SCTP
buffers and you can use it to meet the M2PA requirements, then that
would be, well, an implementation matter.

tomG.


TOLGA ASVEREN wrote:
> 
> Tom,
> 
> I mean, during flushing, what to do with SCTP buffers?
> 
>    Tolga
> --- Tom George <Tom.George@alcatel.com> wrote:
> > Tolga,
> >
> > I am not sure what you are asking. Could you be more
> > specific?
> >
> > tomG.
> >
> > TOLGA ASVEREN wrote:
> > >
> > > Tom,
> > > Is there also a decision about what to do while
> > flushing buffers?
> > >
> > >    Tolga

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb  4 00:52:06 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA01328
	for <sigtran-archive@odin.ietf.org>; Tue, 4 Feb 2003 00:52:06 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h145vPr20946
	for sigtran-archive@odin.ietf.org; Tue, 4 Feb 2003 00:57:25 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h145uhJ20915;
	Tue, 4 Feb 2003 00:56:43 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h145tiJ20889
	for <sigtran@optimus.ietf.org>; Tue, 4 Feb 2003 00:55:44 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA01286
	for <sigtran@ietf.org>; Tue, 4 Feb 2003 00:49:49 -0500 (EST)
Subject: Re: [Sigtran] M2PA Flush Buffers
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF2B181B04.7281C66B-ONC2256CC3.001DA5D7@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Tue, 4 Feb 2003 07:49:58 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/04/2003
 07:50:07 AM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Tom,

4.2.4 says:
"The Flush Buffers and Continue commands allow M2PA to resume normal
operations (i.e., transmission of messages to SCTP and receiving
messages from SCTP) after a processor outage (local and/or remote)
ceases.

If M2PA receives a Flush Buffers command from MTP3, M2PA:

   (a) SHALL NOT transmit any messages to SCTP that are currently
       waiting to be transmitted to SCTP. These messages SHALL be
       discarded.

   (b) SHALL discard all messages currently waiting to be passed
       to MTP3."
So M2PA will stop sending messages to SCTP but continue to receive messages
from SCTP and will discard them(?) What about scenarios, where a SCTP
message, which needs to be discarded is ready to be delivered to M2PA
(because of SCTP retransmission) after Continue?

     Tolga


Tolga,

The interface between SCTP and its upper layer does not provide for a
way to flush the SCTP buffers. So the approach we took in M2PA was to
discard the messages waiting to be sent or received.

If you have some implementation-dependent means for flushing SCTP
buffers and you can use it to meet the M2PA requirements, then that
would be, well, an implementation matter.

tomG.


TOLGA ASVEREN wrote:
>
> Tom,
>
> I mean, during flushing, what to do with SCTP buffers?
>
>    Tolga
> --- Tom George <Tom.George@alcatel.com> wrote:
> > Tolga,
> >
> > I am not sure what you are asking. Could you be more
> > specific?
> >
> > tomG.
> >
> > TOLGA ASVEREN wrote:
> > >
> > > Tom,
> > > Is there also a decision about what to do while
> > flushing buffers?
> > >
> > >    Tolga

--
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb  4 01:04:43 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA01611
	for <sigtran-archive@odin.ietf.org>; Tue, 4 Feb 2003 01:04:43 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h146A1I22156
	for sigtran-archive@odin.ietf.org; Tue, 4 Feb 2003 01:10:01 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1469MJ22111;
	Tue, 4 Feb 2003 01:09:24 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1468KJ22064
	for <sigtran@optimus.ietf.org>; Tue, 4 Feb 2003 01:08:20 -0500
Received: from mgw-x1.nokia.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA01584
	for <sigtran@ietf.org>; Tue, 4 Feb 2003 01:02:30 -0500 (EST)
From: john.loughney@nokia.com
Received: from esvir01nok.ntc.nokia.com (esvir01nokt.ntc.nokia.com [172.21.143.33])
	by mgw-x1.nokia.com (Switch-2.2.1/Switch-2.2.0) with ESMTP id h1464x203654
	for <sigtran@ietf.org>; Tue, 4 Feb 2003 08:04:59 +0200 (EET)
Received: from esebh002.NOE.Nokia.com (unverified) by esvir01nok.ntc.nokia.com
 (Content Technologies SMTPRS 4.2.5) with ESMTP id <T6034478c25ac158f21083@esvir01nok.ntc.nokia.com>;
 Tue, 4 Feb 2003 08:06:04 +0200
Received: from esebe009.NOE.Nokia.com ([172.21.138.41]) by esebh002.NOE.Nokia.com with Microsoft SMTPSVC(5.0.2195.6139);
	 Tue, 4 Feb 2003 08:06:04 +0200
Received: from esebe022.NOE.Nokia.com ([172.21.138.113]) by esebe009.NOE.Nokia.com with Microsoft SMTPSVC(5.0.2195.6139);
	 Tue, 4 Feb 2003 08:06:04 +0200
X-MimeOLE: Produced By Microsoft Exchange V6.0.6375.0
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: RE: [sigtran] draft-ietf-sigtran-security-01.txt
Date: Tue, 4 Feb 2003 08:06:03 +0200
Message-ID: <A16A3EE4D4CA124FADC7987B1AC89FE440EC39@esebe022.ntc.nokia.com>
Thread-Topic: [sigtran] draft-ietf-sigtran-security-01.txt
Thread-Index: AcLLn+8NPTzLkWVCRqCO2AMw2au5lgAc0i+w
To: <abooth@pt.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 04 Feb 2003 06:06:04.0661 (UTC) FILETIME=[802C0A50:01C2CC13]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h1468KJ22065
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Hi Andrew,

> 2) line 424
>   "It is recommended that SIGTRAN peers use the same security mechanism
>   (IPSec or TLS) across all its sessions with other SIGTRAN peers."
> 
>   This recommendation seems to encourage the use of IPSec, since peers 
> are required to support IPSec while TLS is optional.  For example, 
> suppose that a server interacts with implementations from several 
> vendors, where additional vendors could be added later, then there is no 
> guarantee that the new implementations would support TLS.   So the 
> server would be "recommended" to use IPSec.  All of this is technically 
> fine, it just seems to stack the odds against general TLS support.

The opinion of the security area was that a single security solution
needs to be mandatory, to ensure interoperability.  In addition, mixing
TLS on some links and IPsec on others leads to known possible security holes.
Now, in deployments, TLS or IPsec could be chosen, depending upon the
service providers needs.

br,
John
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb  4 01:39:38 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA02322
	for <sigtran-archive@odin.ietf.org>; Tue, 4 Feb 2003 01:39:38 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h146iwO23976
	for sigtran-archive@odin.ietf.org; Tue, 4 Feb 2003 01:44:58 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h146iHJ23951;
	Tue, 4 Feb 2003 01:44:17 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h146h2J23924
	for <sigtran@optimus.ietf.org>; Tue, 4 Feb 2003 01:43:02 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA02274
	for <sigtran@ietf.org>; Tue, 4 Feb 2003 01:37:11 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:YlXscFg1RxOD7gGRRQdmPok3/dOzDy+Y@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h146elD00752;
	Mon, 3 Feb 2003 23:40:47 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h146elX24661;
	Mon, 3 Feb 2003 23:40:47 -0700
Date: Mon, 3 Feb 2003 23:40:46 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA Flush Buffers
Message-ID: <20030203234046.A23003@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OF2B181B04.7281C66B-ONC2256CC3.001DA5D7@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF2B181B04.7281C66B-ONC2256CC3.001DA5D7@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Tue, Feb 04, 2003 at 07:49:58AM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Tolga,

Flush buffers is only used after recovery from local processor outage.
The SS7 peer should not be sending any messages and cannot expect
acknolwedgement for any messages that are outstanding when it receives
the remote processor outage indication.

Flush buffers is only performed on recovery from local processor outage
after MTP Level 3 T1 timeout or remote changeover (which normally takes
about a second or two).

Also, changeover orders have been sent to the other side before buffer
flushing occurs and the other side should have already placed the link
out of service.

So, in any case, the only messages are messages caught in local transmit
and retransmit buffers.  Any message which SCTP happens to send or receive
late will merely interfere with the new alignment process and will be
discarded anyway.

That is why I asked several times: what does it matter?  It doesn't.

--brian


TOLGA ASVEREN wrote:                           (Tue, 04 Feb 2003 07:49:58)
> 
> Tom,
> 
> 4.2.4 says:
> "The Flush Buffers and Continue commands allow M2PA to resume normal
> operations (i.e., transmission of messages to SCTP and receiving
> messages from SCTP) after a processor outage (local and/or remote)
> ceases.
> 
> If M2PA receives a Flush Buffers command from MTP3, M2PA:
> 
>    (a) SHALL NOT transmit any messages to SCTP that are currently
>        waiting to be transmitted to SCTP. These messages SHALL be
>        discarded.
> 
>    (b) SHALL discard all messages currently waiting to be passed
>        to MTP3."
> So M2PA will stop sending messages to SCTP but continue to receive messages
> from SCTP and will discard them(?) What about scenarios, where a SCTP
> message, which needs to be discarded is ready to be delivered to M2PA
> (because of SCTP retransmission) after Continue?
> 
>      Tolga
> 
> 
> Tolga,
> 
> The interface between SCTP and its upper layer does not provide for a
> way to flush the SCTP buffers. So the approach we took in M2PA was to
> discard the messages waiting to be sent or received.
> 
> If you have some implementation-dependent means for flushing SCTP
> buffers and you can use it to meet the M2PA requirements, then that
> would be, well, an implementation matter.
> 
> tomG.
> 
> 
> TOLGA ASVEREN wrote:
> >
> > Tom,
> >
> > I mean, during flushing, what to do with SCTP buffers?
> >
> >    Tolga
> > --- Tom George <Tom.George@alcatel.com> wrote:
> > > Tolga,
> > >
> > > I am not sure what you are asking. Could you be more
> > > specific?
> > >
> > > tomG.
> > >
> > > TOLGA ASVEREN wrote:
> > > >
> > > > Tom,
> > > > Is there also a decision about what to do while
> > > flushing buffers?
> > > >
> > > >    Tolga
> 
> --
> * Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
> * 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
> * Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
> * I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
> * to advertise or represent its position,  operations, or activities. *
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 
> 
> 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb  4 02:47:27 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA13168
	for <sigtran-archive@odin.ietf.org>; Tue, 4 Feb 2003 02:47:27 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h147qmW05659
	for sigtran-archive@odin.ietf.org; Tue, 4 Feb 2003 02:52:49 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h147qaJ05645;
	Tue, 4 Feb 2003 02:52:36 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h147oNJ05557
	for <sigtran@optimus.ietf.org>; Tue, 4 Feb 2003 02:50:23 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA13123
	for <sigtran@ietf.org>; Tue, 4 Feb 2003 02:44:29 -0500 (EST)
Subject: Re: [Sigtran] M2PA Flush Buffers
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OFA7237DA0.04EFF34A-ONC2256CC3.00280EC6@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Tue, 4 Feb 2003 09:44:38 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/04/2003
 09:44:42 AM
MIME-Version: 1.0
Content-type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h147oNJ05558
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit


Brian,
--- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
> Tolga,
>
> Flush buffers is only used after recovery from local
> processor outage.
> The SS7 peer should not be sending any messages and
> cannot expect
> acknolwedgement for any messages that are
> outstanding when it receives
> the remote processor outage indication.
>
> Flush buffers is only performed on recovery from
> local processor outage
> after MTP Level 3 T1 timeout or remote changeover
> (which normally takes
> about a second or two).
>
> Also, changeover orders have been sent to the other
> side before buffer
> flushing occurs and the other side should have
> already placed the link
> out of service.
>
> So, in any case, the only messages are messages
> caught in local transmit
> and retransmit buffers.  Any message which SCTP
> happens to send or receive
> late will merely interfere with the new alignment
> process and will be
> discarded anyway.
[TOLGA] If it happens to be delivered after the alignment process, still I
think there may the possibility of a problem but I wouldn´t think this as a
very likely scenario.
BTW, somehow related with this topic, I think the following sentence needs
to be changed:
4.1.3
"Link alignment takes place after the association is established." (please
don´t say that it does not say that link alignment does not happen at any
other case.)

And one question, when M2PA receives LS(Processor Outage), what does it do
with User Data messages arriving afterwards from that peer? Discarding them
or considering this as end of processor outage and processing them? If the
former is true, what happens if User Data sent before LS(Processor Outage)
arrives after it? It seems a message which shouldn´t be dropped will be. If
the latter is true, again when LS(Processor Outage)arrives before User Data
sent earlier, that User Data will be interpreted mistekenly as end of
processor outage.
>
> That is why I asked several times: what does it
> matter?  It doesn't.
>
> --brian
>
>
> TOLGA ASVEREN wrote:                           (Tue,
> 04 Feb 2003 07:49:58)
> >
> > Tom,
> >
> > 4.2.4 says:
> > "The Flush Buffers and Continue commands allow
> M2PA to resume normal
> > operations (i.e., transmission of messages to SCTP
> and receiving
> > messages from SCTP) after a processor outage
> (local and/or remote)
> > ceases.
> >
> > If M2PA receives a Flush Buffers command from
> MTP3, M2PA:
> >
> >    (a) SHALL NOT transmit any messages to SCTP
> that are currently
> >        waiting to be transmitted to SCTP. These
> messages SHALL be
> >        discarded.
> >
> >    (b) SHALL discard all messages currently
> waiting to be passed
> >        to MTP3."
> > So M2PA will stop sending messages to SCTP but
> continue to receive messages
> > from SCTP and will discard them(?) What about
> scenarios, where a SCTP
> > message, which needs to be discarded is ready to
> be delivered to M2PA
> > (because of SCTP retransmission) after Continue?
> >
> >      Tolga
> >
> >
> > Tolga,
> >
> > The interface between SCTP and its upper layer
> does not provide for a
> > way to flush the SCTP buffers. So the approach we
> took in M2PA was to
> > discard the messages waiting to be sent or
> received.
> >
> > If you have some implementation-dependent means
> for flushing SCTP
> > buffers and you can use it to meet the M2PA
> requirements, then that
> > would be, well, an implementation matter.
> >
> > tomG.
> >
> >
> > TOLGA ASVEREN wrote:
> > >
> > > Tom,
> > >
> > > I mean, during flushing, what to do with SCTP
> buffers?

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb  4 03:05:31 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA13524
	for <sigtran-archive@odin.ietf.org>; Tue, 4 Feb 2003 03:05:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h148AqY07047
	for sigtran-archive@odin.ietf.org; Tue, 4 Feb 2003 03:10:52 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h148ABJ07011;
	Tue, 4 Feb 2003 03:10:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1489vJ06971
	for <sigtran@optimus.ietf.org>; Tue, 4 Feb 2003 03:09:57 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA13506
	for <sigtran@ietf.org>; Tue, 4 Feb 2003 03:04:04 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:ZDfzsr5Je6ihLoXl+CKrh6Dk/MY/fzw2@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1487eD01943;
	Tue, 4 Feb 2003 01:07:40 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1487ex02796;
	Tue, 4 Feb 2003 01:07:40 -0700
Date: Tue, 4 Feb 2003 01:07:40 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA Flush Buffers
Message-ID: <20030204010740.A1629@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OFA7237DA0.04EFF34A-ONC2256CC3.00280EC6@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OFA7237DA0.04EFF34A-ONC2256CC3.00280EC6@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Tue, Feb 04, 2003 at 09:44:38AM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: 8bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

TOLGA,

TOLGA ASVEREN wrote:                                                                 (Tue, 04 Feb 2003 09:44:38)
> [TOLGA] If it happens to be delivered after the alignment process, still I
> think there may the possibility of a problem but I wouldn´t think this as a
> very likely scenario.

2 out of 3 rule for BSN should catch it anyway.

> BTW, somehow related with this topic, I think the following sentence needs
> to be changed:
> 4.1.3
> "Link alignment takes place after the association is established." (please
> don´t say that it does not say that link alignment does not happen at any
> other case.)

But it doesn't way it doesn't happen at any other time...  Would you say that
link alignment takes place before the association is established?  (If not,
then it must always be after.)

> 
> And one question, when M2PA receives LS(Processor Outage), what does it do
> with User Data messages arriving afterwards from that peer? Discarding them
> or considering this as end of processor outage and processing them? If the
> former is true, what happens if User Data sent before LS(Processor Outage)
> arrives after it? It seems a message which shouldn´t be dropped will be. If
> the latter is true, again when LS(Processor Outage)arrives before User Data
> sent earlier, that User Data will be interpreted mistekenly as end of
> processor outage.

IMO LS should be sent in same stream as DATA.

In that case DATA received after LS PO is cause to place the link out of
service.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb  4 12:12:39 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA28957
	for <sigtran-archive@odin.ietf.org>; Tue, 4 Feb 2003 12:12:39 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h14HID610360
	for sigtran-archive@odin.ietf.org; Tue, 4 Feb 2003 12:18:13 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h14HHuJ10340;
	Tue, 4 Feb 2003 12:17:57 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h14HGoJ10262
	for <sigtran@optimus.ietf.org>; Tue, 4 Feb 2003 12:16:50 -0500
Received: from w2ksjexg01.ciena.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA28893
	for <sigtran@ietf.org>; Tue, 4 Feb 2003 12:10:46 -0500 (EST)
Received: from wntcsdexg01.csd.ciena.com ([10.34.31.31]) by w2ksjexg01.ciena.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2653.13)
	id 1H7HJ7KB; Tue, 4 Feb 2003 09:14:08 -0800
Received: by webdev-llnt.oni.com with Internet Mail Service (5.5.2653.19)
	id <Y5W9MN6D>; Tue, 4 Feb 2003 09:14:22 -0800
Message-ID: <2135200C183FD5119588009027DE572302836C4D@webdev-owa.oni.com>
From: "Ong, Lyndon" <LyOng@ciena.com>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Date: Tue, 4 Feb 2003 09:14:16 -0800 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Sigtran] security framework Last Call
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Hi Folks,

Now that the security framework is out, I would like to start up WG Last
Call
for this as well.  Let's start Last Call on Wednesday, February 5th, ending
on
Wednesday, February 19th.  Please put "Security Comment" in the subject
line of any comment emails.  

The current draft is available at
http://ietf.org/internet-drafts/draft-ietf-sigtran-security-01.txt.

BTW, Last Call on M2PA ends tomorrow, so please get any additional comments
in by then.

Thanks,

L. Ong
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb  4 15:38:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA03738
	for <sigtran-archive@odin.ietf.org>; Tue, 4 Feb 2003 15:38:56 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h14KiYG22270
	for sigtran-archive@odin.ietf.org; Tue, 4 Feb 2003 15:44:34 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h14KiLJ22260;
	Tue, 4 Feb 2003 15:44:21 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h14KhlJ22226
	for <sigtran@optimus.ietf.org>; Tue, 4 Feb 2003 15:43:47 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA03681
	for <sigtran@ietf.org>; Tue, 4 Feb 2003 15:37:39 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h14KfFeW003383;
	Tue, 4 Feb 2003 14:41:16 -0600 (CST)
Message-ID: <3E40256B.31DC2107@alcatel.com>
Date: Tue, 04 Feb 2003 14:41:15 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA Flush Buffers
References: <OFA7237DA0.04EFF34A-ONC2256CC3.00280EC6@LocalDomain>
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Tolga,

This is one place where M2PA works differently from MTP2. In MTP2,
receipt of a MSU (or FISU) indicates the end of a remote processor
outage. There is no LSSU that indicates the end of remote processor
outage.

I think this would not work well in M2PA, since User Data messages might
get out of order with the Link Status messages. So a User Data message
received during remote processor outage does not indicate the end of the
RPO. It has to be interpreted as having been sent before the RPO
condition began. 

In M2PA, the LS Processor Recovered message is there to indicate the end
of RPO.

That may not answer your question, but maybe it gets us a little closer. 

tomG.


TOLGA ASVEREN wrote:
  [...] 
> And one question, when M2PA receives LS(Processor Outage), what does it do
> with User Data messages arriving afterwards from that peer? Discarding them
> or considering this as end of processor outage and processing them? If the
> former is true, what happens if User Data sent before LS(Processor Outage)
> arrives after it? It seems a message which shouldn´t be dropped will be. If
> the latter is true, again when LS(Processor Outage)arrives before User Data
> sent earlier, that User Data will be interpreted mistekenly as end of
> processor outage.
  [...]

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb  4 15:47:47 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA03950
	for <sigtran-archive@odin.ietf.org>; Tue, 4 Feb 2003 15:47:47 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h14KrO422581
	for sigtran-archive@odin.ietf.org; Tue, 4 Feb 2003 15:53:24 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h14KrBJ22560;
	Tue, 4 Feb 2003 15:53:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h14Kq5J22531
	for <sigtran@optimus.ietf.org>; Tue, 4 Feb 2003 15:52:05 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA03935
	for <sigtran@ietf.org>; Tue, 4 Feb 2003 15:45:56 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h14KnXeW004874;
	Tue, 4 Feb 2003 14:49:33 -0600 (CST)
Message-ID: <3E40275D.DA05F75A@alcatel.com>
Date: Tue, 04 Feb 2003 14:49:33 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA Flush Buffers
References: <OFA7237DA0.04EFF34A-ONC2256CC3.00280EC6@LocalDomain> <20030204010740.A1629@openss7.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

I think we had this dicussion at the M2PA Interop and probably talked
about it on the list long ago. 8-)

Putting the LS and Data messages on the same stream does make it easy to
decide in which order they were sent.

The thing I have never liked about putting them on the same stream is
that a LS message could get stuck in a queue behind many User Data
messages. 

MTP2 does not have this problem. It puts LSSUs ahead of MSUs in its
transmission buffer. 

If SCTP had an option to put an outgoing message at the head of the
queue, it would be OK to send LS and User Data on the same stream.

tomG.

"Brian F. G. Bidulock" wrote:
  [...]
> IMO LS should be sent in same stream as DATA.
> 
> In that case DATA received after LS PO is cause to place the link out of
> service.
> 
> --brian
  [...]

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb  4 16:28:53 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA04808
	for <sigtran-archive@odin.ietf.org>; Tue, 4 Feb 2003 16:28:53 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h14LYV724766
	for sigtran-archive@odin.ietf.org; Tue, 4 Feb 2003 16:34:31 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h14LYKJ24710;
	Tue, 4 Feb 2003 16:34:20 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h14LRaJ24470
	for <sigtran@optimus.ietf.org>; Tue, 4 Feb 2003 16:27:36 -0500
Received: from w2ksjexg01.ciena.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA04671
	for <sigtran@ietf.org>; Tue, 4 Feb 2003 16:21:27 -0500 (EST)
Received: from wntcsdexg01.csd.ciena.com ([10.34.31.31]) by w2ksjexg01.ciena.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2653.13)
	id 1H7HJ98F; Tue, 4 Feb 2003 13:24:49 -0800
Received: by webdev-llnt.oni.com with Internet Mail Service (5.5.2653.19)
	id <Y5W9M3JL>; Tue, 4 Feb 2003 13:25:03 -0800
Message-ID: <2135200C183FD5119588009027DE572302836C55@webdev-owa.oni.com>
From: "Ong, Lyndon" <LyOng@ciena.com>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Date: Tue, 4 Feb 2003 13:25:01 -0800 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Sigtran] Security Last Call starting today
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Hi Folks,

To align things with the IESG schedule, I'm going to move up Last Call
on the security framework 1 day, so let's start Last Call today (2/4) 
and end it on 2/18.

Thanks,

L. Ong
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb  4 16:48:56 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA05416
	for <sigtran-archive@odin.ietf.org>; Tue, 4 Feb 2003 16:48:55 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h14LsY026462
	for sigtran-archive@odin.ietf.org; Tue, 4 Feb 2003 16:54:34 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h14LsMJ26432;
	Tue, 4 Feb 2003 16:54:22 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h14Lr5J26389
	for <sigtran@optimus.ietf.org>; Tue, 4 Feb 2003 16:53:05 -0500
Received: from chuckie.dgms.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA05351
	for <sigtran@ietf.org>; Tue, 4 Feb 2003 16:46:55 -0500 (EST)
Received: from pckhusidp (localhost [127.0.0.1])
	by chuckie.dgms.com (8.9.3/8.9.3) with SMTP id QAA13017
	for <sigtran@ietf.org>; Tue, 4 Feb 2003 16:50:32 -0500 (EST)
From: "Anatoly Khusid" <anatoly.khusid@ulticom.com>
To: <sigtran@ietf.org>
Date: Tue, 4 Feb 2003 16:48:47 -0500
Message-ID: <009d01c2cc97$32aa8840$5a2119ac@pckhusidp>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook CWS, Build 9.0.2416 (9.0.2911.0)
x-mimeole: Produced By Microsoft MimeOLE V6.00.2600.0000
Importance: Normal
In-Reply-To: <3E40275D.DA05F75A@alcatel.com>
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] M2UA RFC 3331 Question
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Hi,

I am not clear on one item in the M2UA RFC:

[Page 11] (Section 1.5.4.1):

"SCTP Stream '0' SHOULD NOT be used for MTP2 User Adaptation (MAUP)
messages (see Section 3) since stream '0' SHOULD only be used for ASP
Management (ASPM) messages (see Section 4.3.3)."

I assume ASPM acronym combines  ASPSM(Application Server Process State
Maintenance) and ASPTM(Application Server Process Traffic Maintenance),
which means all ASPTM messages must be sent on SCTP stream '0'

Now, if you look at [Page 60] (Section 4.2.1) :
"All MGMT messages, except BEAT and BEAT Ack, SHOULD be sent with
sequenced delivery to ensure ordering.  All MGMT messages, with the
exception of ASPTM, BEAT and BEAT Ack messages, SHOULD be sent on SCTP
stream '0'.  All ASPTM messages SHOULD be sent on the stream which
normally carries the data traffic to which the message applies.  BEAT
and BEAT Ack messages MAY be sent using out-of-order delivery, and
MAY be sent on any stream."


This section says that ALL ASPTM messages should be sent on the stream which
normally carries the data traffic to which message applies.

Am I missing something, but I am confused which SCTP stream should be used
for ASPTM messages.

I appreciate if somebody could clear this out for me.

Thanks a lot!

Anatoly

 -----Original Message-----
From: 	sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]  On Behalf Of
Tom George
Sent:	Tuesday, February 04, 2003 3:50 PM
Cc:	sigtran@ietf.org
Subject:	Re: [Sigtran] M2PA Flush Buffers

I think we had this dicussion at the M2PA Interop and probably talked
about it on the list long ago. 8-)

Putting the LS and Data messages on the same stream does make it easy to
decide in which order they were sent.

The thing I have never liked about putting them on the same stream is
that a LS message could get stuck in a queue behind many User Data
messages.

MTP2 does not have this problem. It puts LSSUs ahead of MSUs in its
transmission buffer.

If SCTP had an option to put an outgoing message at the head of the
queue, it would be OK to send LS and User Data on the same stream.

tomG.

"Brian F. G. Bidulock" wrote:
  [...]
> IMO LS should be sent in same stream as DATA.
>
> In that case DATA received after LS PO is cause to place the link out of
> service.
>
> --brian
  [...]

--
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb  4 17:02:45 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA05944
	for <sigtran-archive@odin.ietf.org>; Tue, 4 Feb 2003 17:02:45 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h14M8O228301
	for sigtran-archive@odin.ietf.org; Tue, 4 Feb 2003 17:08:24 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h14M8AJ28272;
	Tue, 4 Feb 2003 17:08:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h14M7OJ27984
	for <sigtran@optimus.ietf.org>; Tue, 4 Feb 2003 17:07:24 -0500
Received: from w2ksjexg01.ciena.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA05905
	for <sigtran@ietf.org>; Tue, 4 Feb 2003 17:01:14 -0500 (EST)
Received: from wntcsdexg01.csd.ciena.com ([10.34.31.31]) by w2ksjexg01.ciena.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2653.13)
	id 1H7HJ0H5; Tue, 4 Feb 2003 14:04:37 -0800
Received: by webdev-llnt.oni.com with Internet Mail Service (5.5.2653.19)
	id <Y5W9M3LF>; Tue, 4 Feb 2003 14:04:51 -0800
Message-ID: <2135200C183FD5119588009027DE572302836C57@webdev-owa.oni.com>
From: "Ong, Lyndon" <LyOng@ciena.com>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Date: Tue, 4 Feb 2003 14:04:50 -0800 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Sigtran] Security Last Call
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Hi Folks,

Sorry if this was not made clear - the security framework document is 
intended to be standards track.

Cheers,

L. Ong
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb  4 18:13:20 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA07552
	for <sigtran-archive@odin.ietf.org>; Tue, 4 Feb 2003 18:13:20 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h14NJ1B32627
	for sigtran-archive@odin.ietf.org; Tue, 4 Feb 2003 18:19:01 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h14NIhJ32613;
	Tue, 4 Feb 2003 18:18:43 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h14NHsJ32589
	for <sigtran@optimus.ietf.org>; Tue, 4 Feb 2003 18:17:54 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA07542
	for <sigtran@ietf.org>; Tue, 4 Feb 2003 18:11:42 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h14NFJqe024518;
	Tue, 4 Feb 2003 17:15:19 -0600 (CST)
Message-ID: <3E404987.7329ED4@alcatel.com>
Date: Tue, 04 Feb 2003 17:15:19 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA LPO during alignment/proving (was: M2PA version 7 
 changes)
References: <3D72C18DF140D311A54F0008C7DBAC8D02917425@MCHH252E> <3E38613F.FF645442@alcatel.com> <20030129170659.B13613@openss7.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Brian,

I like the 2nd version (below). I will update the draft accordingly.

And when you say
> The effect is the same.  The diagram is certainly not intended to be 100%
> complete, it is an overview.  The text should be consulted.

I agree whole heartedly. I am running out of blank spaces on the
diagram. 8-)

tomG.

"Brian F. G. Bidulock" wrote:
> 
> Tom,
> 
> The problem is the LSC doesn't really have a proving state (only IAC
> has proving state).  LSC always moves from INITIAL ALIGNMENT to either
> ALIGNED READY or ALIGNED NOT READY.  You might change the diagram as
> follows:
> 
>  From:
  [...]
>  To:
  [...]
>   |                       | MTP3 Start                 |
>   |                       |                            |
>   |                       |   +-------+                |
>   |                       |   |       | LPO/LPR        |
>   |                       V   V       |                |
>   |                 +-----------+     |                |
>   |                 |  INITIAL  |<----+                |
>   +<----------------| ALIGNMENT |--------------------->+
>   |      MTP3 Stop  +-----------+     SCTP Comm Error  |
>   |   OR T2 Expiry        |        OR SCTP Comm Lost   |
>   |   OR Receive LS OOS   |                            |
>   |                       | Receive                    |
>   |                       | LS Align                   |
>   |                       |                            |
>   |                       |   +-------+                |
>   |                       |   |       | LPO/LPR        |
>   |                       V   V       |                |
>   |                 +-----------+     |                |
>   |                 |  PROVING  |<----+                |
>   |<----------------|           |--------------------->+
>   |      MTP3 Stop  +-----------+     SCTP Comm Error  |
>   |   OR T3 Expiry        | |      OR SCTP Comm Lost   |
>   |   OR Receive LS OOS   | |                          |
>   |                       | +--------------------------|--------+
>   |                       |                            |   T4   |
>   |                       | T4 Expiry                  | Expiry |
>   |                       | Send LS Ready              | Send   |
>   |                       V                            | LS PO  V
>   |                 +-----------+        LPO/LPR       |  +-----------+
>   |                 |  ALIGNED  |<---------------------|->|  ALIGNED  |
>   +<----------------|   READY   |--------------------->+  | NOT READY |
>   |      MTP3 Stop  +-----------+     SCTP Comm Error     +-----------+
>   |   OR T1 Expiry        |        OR SCTP Comm Lost         |       |
>   |   OR Receive LS OOS   |                                  |   T1  |
>   |   OR Receive LS Align | Receive LS Ready      Receive    | Expiry|
>   |                       | OR Receive User Data    LS Ready | OR Rcv|
>   |                       |                       OR Receive |    LS |
>   |                       |                         LS PO    |  Align|
>   |                       |                                  |       |
>   |                       |                                  |       |
>   |                       V                                  V       |
>   |                 +-----------+        LPO/LPR       +-----------+ |
>   |                 |    IN     |<-------------------->| PROCESSOR | |
>   |                 |  SERVICE  |                      |   OUTAGE  | |
>   |                 +-----------+                      +-----------+ |
> 
> The effect is the same.  The diagram is certainly not intended to be 100%
> complete, it is an overview.  The text should be consulted.
> 
> --brian

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb  4 18:28:50 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA07902
	for <sigtran-archive@odin.ietf.org>; Tue, 4 Feb 2003 18:28:50 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h14NYV600708
	for sigtran-archive@odin.ietf.org; Tue, 4 Feb 2003 18:34:31 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h14NYCJ00697;
	Tue, 4 Feb 2003 18:34:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h14NXnJ00664
	for <sigtran@optimus.ietf.org>; Tue, 4 Feb 2003 18:33:49 -0500
Received: from pt.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA07869
	for <sigtran@ietf.org>; Tue, 4 Feb 2003 18:27:36 -0500 (EST)
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id SAA00697;
	Tue, 4 Feb 2003 18:27:39 -0500
Message-ID: <3E404D32.1090200@pt.com>
Date: Tue, 04 Feb 2003 18:30:58 -0500
From: Andrew Booth <abooth@pt.com>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.1) Gecko/20020826
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Tom George <Tom.George@alcatel.com>
CC: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA Flush Buffers
References: <OFA7237DA0.04EFF34A-ONC2256CC3.00280EC6@LocalDomain> <20030204010740.A1629@openss7.org> <3E40275D.DA05F75A@alcatel.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Hi Tom,

I'm not an M2PA expert, so I'm not sure how important this issue is.  In 
case it makes things easier, here are some thoughts on quick solutions.

Problem statement:
==============
It looks like there are a few cases in M2PA where it would be nice to 
know the relative order in which a LS and a User Data message were sent. 
 Also, it would be nice if LS messages were delivered "as quickly as 
possible".

Solution 1:
========
If the issue is to decide where in the User Data sequence an LS message 
fits, why not use the FSN field to indicate it?  Since the FSN field in 
a LS message is no longer used for acknowledgements, you could just fill 
the FSN of the LS message with the FSN of the last transmitted User Data 
message.  This would let you have the logical ordering you get on a wire 
while reducing head of line blocking for LS messages.  There is still 
the possiblity that User Data sent after a LS message could arrive at 
the receiver first.

Solution 2:
========
Alternatively, LS messages could get sent on both streams and we could 
use the reserved space in the M2PA header to eliminate duplicates.  This 
approach involves slightly more complicated procedures than the first 
proposal, but would ensure that an LS message arrives before any data 
that is sent chronologically after it.

In either case, the procedures would be very restricted in scope and 
would allow statements like "If a User Data is received that was sent 
after the PO message then...".  Is this worth anything?

Thoughts?
Andrew

Tom George wrote:

>I think we had this dicussion at the M2PA Interop and probably talked
>about it on the list long ago. 8-)
>
>Putting the LS and Data messages on the same stream does make it easy to
>decide in which order they were sent.
>
>The thing I have never liked about putting them on the same stream is
>that a LS message could get stuck in a queue behind many User Data
>messages. 
>
>MTP2 does not have this problem. It puts LSSUs ahead of MSUs in its
>transmission buffer. 
>
>If SCTP had an option to put an outgoing message at the head of the
>queue, it would be OK to send LS and User Data on the same stream.
>
>tomG.
>
>"Brian F. G. Bidulock" wrote:
>  [...]
>  
>
>>IMO LS should be sent in same stream as DATA.
>>
>>In that case DATA received after LS PO is cause to place the link out of
>>service.
>>
>>--brian
>>    
>>
>  [...]
>
>  
>

-- 
Andrew Booth, Software Developer
Signaling Systems Group, Performance Technologies
613-237-4344

Visit www.pt.com/mailing.html to subscribe to our Packets & Signals eLetter.


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb  4 18:56:46 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA08269
	for <sigtran-archive@odin.ietf.org>; Tue, 4 Feb 2003 18:56:45 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1502S402225
	for sigtran-archive@odin.ietf.org; Tue, 4 Feb 2003 19:02:28 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1502DJ02214;
	Tue, 4 Feb 2003 19:02:13 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1501RJ02188
	for <sigtran@optimus.ietf.org>; Tue, 4 Feb 2003 19:01:27 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA08236
	for <sigtran@ietf.org>; Tue, 4 Feb 2003 18:55:13 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h14NwpeW002779;
	Tue, 4 Feb 2003 17:58:51 -0600 (CST)
Message-ID: <3E4053BB.B55AE42D@alcatel.com>
Date: Tue, 04 Feb 2003 17:58:51 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA Flush Buffers
References: <OFA7237DA0.04EFF34A-ONC2256CC3.00280EC6@LocalDomain> <20030204010740.A1629@openss7.org> <3E40275D.DA05F75A@alcatel.com> <3E404D32.1090200@pt.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Andrew,

Thanks for the comments. I will gladly defer to those with more SS7
smarts, but I think there is no case where the receiver has to know the
order in which LS and User Data messages were sent. MTP2 itself has
LSSUs jumping to the head of the line when they are transmitted.

So what do we do with User Data messages that are received after a LS
PO? We can't just copy MTP2, since the same situation does not occur. In
MTP2, receiving an MSU ends the RPO condition. 

The current M2PA does not say exactly what to do, which I would
interpret (if I may "interpret" my own writing) that the receiver should
go ahead and process the User Data messages. The presumption would be
that they are valid messages sent before the RPO occurred. 

Does anyone see a problem with this?

tomG.
 

Andrew Booth wrote:
> 
> Hi Tom,
> 
> I'm not an M2PA expert, so I'm not sure how important this issue is.  In
> case it makes things easier, here are some thoughts on quick solutions.
> 
> Problem statement:
> ==============
> It looks like there are a few cases in M2PA where it would be nice to
> know the relative order in which a LS and a User Data message were sent.
>  Also, it would be nice if LS messages were delivered "as quickly as
> possible".
> 
> Solution 1:
> ========
> If the issue is to decide where in the User Data sequence an LS message
> fits, why not use the FSN field to indicate it?  Since the FSN field in
> a LS message is no longer used for acknowledgements, you could just fill
> the FSN of the LS message with the FSN of the last transmitted User Data
> message.  This would let you have the logical ordering you get on a wire
> while reducing head of line blocking for LS messages.  There is still
> the possiblity that User Data sent after a LS message could arrive at
> the receiver first.
> 
> Solution 2:
> ========
> Alternatively, LS messages could get sent on both streams and we could
> use the reserved space in the M2PA header to eliminate duplicates.  This
> approach involves slightly more complicated procedures than the first
> proposal, but would ensure that an LS message arrives before any data
> that is sent chronologically after it.
> 
> In either case, the procedures would be very restricted in scope and
> would allow statements like "If a User Data is received that was sent
> after the PO message then...".  Is this worth anything?
> 
> Thoughts?
> Andrew
> 
> Tom George wrote:
> 
> >I think we had this dicussion at the M2PA Interop and probably talked
> >about it on the list long ago. 8-)
> >
> >Putting the LS and Data messages on the same stream does make it easy to
> >decide in which order they were sent.
> >
> >The thing I have never liked about putting them on the same stream is
> >that a LS message could get stuck in a queue behind many User Data
> >messages.
> >
> >MTP2 does not have this problem. It puts LSSUs ahead of MSUs in its
> >transmission buffer.
> >
> >If SCTP had an option to put an outgoing message at the head of the
> >queue, it would be OK to send LS and User Data on the same stream.
> >
> >tomG.
> >
> >"Brian F. G. Bidulock" wrote:
> >  [...]
> >
> >
> >>IMO LS should be sent in same stream as DATA.
> >>
> >>In that case DATA received after LS PO is cause to place the link out of
> >>service.
> >>
> >>--brian
> >>
> >>
> >  [...]
> >
> >
> >
> 
> --
> Andrew Booth, Software Developer
> Signaling Systems Group, Performance Technologies
> 613-237-4344
> 
> Visit www.pt.com/mailing.html to subscribe to our Packets & Signals eLetter.

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb  4 21:24:23 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA11507
	for <sigtran-archive@odin.ietf.org>; Tue, 4 Feb 2003 21:24:23 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h152U7q09683
	for sigtran-archive@odin.ietf.org; Tue, 4 Feb 2003 21:30:07 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h152TqJ09664;
	Tue, 4 Feb 2003 21:29:52 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h152SAJ09622
	for <sigtran@optimus.ietf.org>; Tue, 4 Feb 2003 21:28:10 -0500
Received: from rtp-msg-core-1.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA11467
	for <sigtran@ietf.org>; Tue, 4 Feb 2003 21:21:55 -0500 (EST)
Received: from cia.cisco.com (localhost [127.0.0.1])
	by rtp-msg-core-1.cisco.com (8.12.2/8.12.6) with ESMTP id h152QO90029882;
	Tue, 4 Feb 2003 21:26:24 -0500 (EST)
Received: from kmorneau-w2k.cisco.com (rtp-vpn2-467.cisco.com [10.82.241.211])
	by cia.cisco.com (Mirapoint)
	with ESMTP id ABO43687;
	Tue, 4 Feb 2003 21:15:29 -0500 (EST)
Message-Id: <4.3.2.7.2.20030204212139.048f5f80@cia.cisco.com>
X-Sender: kmorneau@cia.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Tue, 04 Feb 2003 21:25:28 -0500
To: "Anatoly Khusid" <anatoly.khusid@ulticom.com>, <sigtran@ietf.org>
From: Ken Morneault <kmorneau@cisco.com>
Subject: Re: [Sigtran] M2UA RFC 3331 Question
In-Reply-To: <009d01c2cc97$32aa8840$5a2119ac@pckhusidp>
References: <3E40275D.DA05F75A@alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>



Anatoly,

Pls see comments below.

Regards,

Ken

At 04:48 PM 2/4/2003 -0500, Anatoly Khusid wrote:
>Hi,
>
>I am not clear on one item in the M2UA RFC:
>
>[Page 11] (Section 1.5.4.1):
>
>"SCTP Stream '0' SHOULD NOT be used for MTP2 User Adaptation (MAUP)
>messages (see Section 3) since stream '0' SHOULD only be used for ASP
>Management (ASPM) messages (see Section 4.3.3)."

The "SHOULD" only applies to ASPSM messages.  

>I assume ASPM acronym combines  ASPSM(Application Server Process State
>Maintenance) and ASPTM(Application Server Process Traffic Maintenance),
>which means all ASPTM messages must be sent on SCTP stream '0'
>
>Now, if you look at [Page 60] (Section 4.2.1) :
>"All MGMT messages, except BEAT and BEAT Ack, SHOULD be sent with
>sequenced delivery to ensure ordering.  All MGMT messages, with the
>exception of ASPTM, BEAT and BEAT Ack messages, SHOULD be sent on SCTP
>stream '0'.  All ASPTM messages SHOULD be sent on the stream which
>normally carries the data traffic to which the message applies.  BEAT
>and BEAT Ack messages MAY be sent using out-of-order delivery, and
>MAY be sent on any stream."
>
>This section says that ALL ASPTM messages should be sent on the stream which
>normally carries the data traffic to which message applies.

Yes that is correct.  I agree that this information should only be listed
in one place in the RFC.  Will clarify in an implementors guide.

>Am I missing something, but I am confused which SCTP stream should be used
>for ASPTM messages.
>
>I appreciate if somebody could clear this out for me.
>
>Thanks a lot!
>
>Anatoly
>
> -----Original Message-----
>From:   sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]  On Behalf Of
>Tom George
>Sent:   Tuesday, February 04, 2003 3:50 PM
>Cc:     sigtran@ietf.org
>Subject:        Re: [Sigtran] M2PA Flush Buffers
>
>I think we had this dicussion at the M2PA Interop and probably talked
>about it on the list long ago. 8-)
>
>Putting the LS and Data messages on the same stream does make it easy to
>decide in which order they were sent.
>
>The thing I have never liked about putting them on the same stream is
>that a LS message could get stuck in a queue behind many User Data
>messages.
>
>MTP2 does not have this problem. It puts LSSUs ahead of MSUs in its
>transmission buffer.
>
>If SCTP had an option to put an outgoing message at the head of the
>queue, it would be OK to send LS and User Data on the same stream.
>
>tomG.
>
>"Brian F. G. Bidulock" wrote:
>  [...]
>> IMO LS should be sent in same stream as DATA.
>>
>> In that case DATA received after LS PO is cause to place the link out of
>> service.
>>
>> --brian
>  [...]
>
>--
>* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
>* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
>* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
>* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
>* to advertise or represent its position,  operations, or activities. *
>_______________________________________________
>Sigtran mailing list
>Sigtran@ietf.org
>https://www1.ietf.org/mailman/listinfo/sigtran
>
>_______________________________________________
>Sigtran mailing list
>Sigtran@ietf.org
>https://www1.ietf.org/mailman/listinfo/sigtran 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb  5 01:28:09 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA17544
	for <sigtran-archive@odin.ietf.org>; Wed, 5 Feb 2003 01:28:09 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h156Xxu21438
	for sigtran-archive@odin.ietf.org; Wed, 5 Feb 2003 01:33:59 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h156XiJ21429;
	Wed, 5 Feb 2003 01:33:46 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h156WMJ21390
	for <sigtran@optimus.ietf.org>; Wed, 5 Feb 2003 01:32:22 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA17517
	for <sigtran@ietf.org>; Wed, 5 Feb 2003 01:25:58 -0500 (EST)
Subject: Re: [Sigtran] M2PA Flush Buffers
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF8F72E3AF.9EAE793B-ONC2256CC4.001CEA23@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Wed, 5 Feb 2003 08:25:56 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/05/2003
 08:26:15 AM
MIME-Version: 1.0
Content-type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h156WMJ21391
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit


Tom,
--- Tom George <Tom.George@alcatel.com> wrote:
> Andrew,
>
> Thanks for the comments. I will gladly defer to
> those with more SS7
> smarts, but I think there is no case where the
> receiver has to know the
> order in which LS and User Data messages were sent.
> MTP2 itself has
> LSSUs jumping to the head of the line when they are
> transmitted.
>
> So what do we do with User Data messages that are
> received after a LS
> PO? We can't just copy MTP2, since the same
> situation does not occur. In
> MTP2, receiving an MSU ends the RPO condition.
>
> The current M2PA does not say exactly what to do,
> which I would
> interpret (if I may "interpret" my own writing) that
> the receiver should
> go ahead and process the User Data messages. The
> presumption would be
> that they are valid messages sent before the RPO
> occurred.
[TOLGA]IMO it is better to mention this in the document. I have some more
questions though:
I assume even after receiving LS(PO) from the peer, M2PA continues to
acknowledge received messages(?)
Similarliy I assume during local processor outage, based on received
acknowledgements buffers are updated(?) IMO, it might be good to clarify
this in the text(if this is the case)
What about T7 on the peer experiencing local processor outage? Will it be
stopped or not? If the above is true, I would think it shouldn´t be
stopped.
In 4.1.4 there is a sentence
"Alternatively, MTP3 may perform data retrieval as part of a changeover
procedure."
Are the described procedures in that sentence and flushing buffers really
alternatives to each other?
>
> Does anyone see a problem with this?
>
> tomG.
>
>
> Andrew Booth wrote:
> >
> > Hi Tom,
> >
> > I'm not an M2PA expert, so I'm not sure how
> important this issue is.  In
> > case it makes things easier, here are some
> thoughts on quick solutions.
> >
> > Problem statement:
> > ==============
> > It looks like there are a few cases in M2PA where
> it would be nice to
> > know the relative order in which a LS and a User
> Data message were sent.
> >  Also, it would be nice if LS messages were
> delivered "as quickly as
> > possible".
> >
> > Solution 1:
> > ========
> > If the issue is to decide where in the User Data
> sequence an LS message
> > fits, why not use the FSN field to indicate it?
> Since the FSN field in
> > a LS message is no longer used for
> acknowledgements, you could just fill
> > the FSN of the LS message with the FSN of the last
> transmitted User Data
> > message.  This would let you have the logical
> ordering you get on a wire
> > while reducing head of line blocking for LS
> messages.  There is still
> > the possiblity that User Data sent after a LS
> message could arrive at
> > the receiver first.
> >
> > Solution 2:
> > ========
> > Alternatively, LS messages could get sent on both
> streams and we could
> > use the reserved space in the M2PA header to
> eliminate duplicates.  This
> > approach involves slightly more complicated
> procedures than the first
> > proposal, but would ensure that an LS message
> arrives before any data
> > that is sent chronologically after it.
> >
> > In either case, the procedures would be very
> restricted in scope and
> > would allow statements like "If a User Data is
> received that was sent
> > after the PO message then...".  Is this worth
> anything?
> >
> > Thoughts?
> > Andrew
> >
> > Tom George wrote:
> >
> > >I think we had this dicussion at the M2PA Interop
> and probably talked
> > >about it on the list long ago. 8-)
> > >
> > >Putting the LS and Data messages on the same
> stream does make it easy to
> > >decide in which order they were sent.
> > >
> > >The thing I have never liked about putting them
> on the same stream is
> > >that a LS message could get stuck in a queue
> behind many User Data
> > >messages.
> > >
> > >MTP2 does not have this problem. It puts LSSUs
> ahead of MSUs in its
> > >transmission buffer.
> > >
> > >If SCTP had an option to put an outgoing message
> at the head of the
> > >queue, it would be OK to send LS and User Data on
> the same stream.
> > >
> > >tomG.
> > >
> > >"Brian F. G. Bidulock" wrote:
> > >  [...]
> > >
> > >
> > >>IMO LS should be sent in same stream as DATA.
> > >>
> > >>In that case DATA received after LS PO is cause
> to place the link out of
> > >>service.
> > >>
> > >>--brian

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb  5 03:02:06 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA13759
	for <sigtran-archive@odin.ietf.org>; Wed, 5 Feb 2003 03:02:06 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1587uw05374
	for sigtran-archive@odin.ietf.org; Wed, 5 Feb 2003 03:07:56 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1587gJ05331;
	Wed, 5 Feb 2003 03:07:43 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15865J04629
	for <sigtran@optimus.ietf.org>; Wed, 5 Feb 2003 03:06:05 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA13684
	for <sigtran@ietf.org>; Wed, 5 Feb 2003 02:59:43 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:QEOOSCh0lJUy6tcdrwCvGcItLOz6lkyL@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1583HD25265;
	Wed, 5 Feb 2003 01:03:17 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1583FZ21385;
	Wed, 5 Feb 2003 01:03:15 -0700
Date: Wed, 5 Feb 2003 01:03:15 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA Flush Buffers
Message-ID: <20030205010315.A12910@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OF8F72E3AF.9EAE793B-ONC2256CC4.001CEA23@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF8F72E3AF.9EAE793B-ONC2256CC4.001CEA23@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Wed, Feb 05, 2003 at 08:25:56AM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: 8bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

TOLGA,

TOLGA ASVEREN wrote:                                                                 (Wed, 05 Feb 2003 08:25:56)
> 
> Tom,
> --- Tom George <Tom.George@alcatel.com> wrote:
> > Andrew,
> >
> > Thanks for the comments. I will gladly defer to
> > those with more SS7
> > smarts, but I think there is no case where the
> > receiver has to know the
> > order in which LS and User Data messages were sent.
> > MTP2 itself has
> > LSSUs jumping to the head of the line when they are
> > transmitted.
> >
> > So what do we do with User Data messages that are
> > received after a LS
> > PO? We can't just copy MTP2, since the same
> > situation does not occur. In
> > MTP2, receiving an MSU ends the RPO condition.
> >
> > The current M2PA does not say exactly what to do,
> > which I would
> > interpret (if I may "interpret" my own writing) that
> > the receiver should
> > go ahead and process the User Data messages. The
> > presumption would be
> > that they are valid messages sent before the RPO
> > occurred.
> [TOLGA]IMO it is better to mention this in the document. I have some more
> questions though:
> I assume even after receiving LS(PO) from the peer, M2PA continues to
> acknowledge received messages(?)
> Similarliy I assume during local processor outage, based on received
> acknowledgements buffers are updated(?) IMO, it might be good to clarify
> this in the text(if this is the case)
> What about T7 on the peer experiencing local processor outage? Will it be
> stopped or not? If the above is true, I would think it shouldn´t be
> stopped.
> In 4.1.4 there is a sentence
> "Alternatively, MTP3 may perform data retrieval as part of a changeover
> procedure."
> Are the described procedures in that sentence and flushing buffers really
> alternatives to each other?

To avoid mis-sequencing of messages, neither end should send or
acknowledge messages during LPO.

The receiver of RPO must not acknowledge any messages after the
receipt of RPO to remain consistent with Q.703 and interactions
with Q.704 (time-controlled changeover procedure).  The MTP
Level 3 suffering LPO must know that when LPO begins that
message transmission has ceased so that the time-controlled
changeover procedure time duration is correctly referenced
against the last message transmitted on the blocked link.

To remain consistent with ANSI T1.111.3, both the sender and
the receiver of RPO should re-synchronize sequence numbers
(acknowlege received messages awaiting ack) and cease
transmission of DATA.

Again, just as the side experiencing LPO must reference its
time-controlled changeover at Level 3 against the last message
sent on the blocked link, so must the side detecting RPO
reference its time-controlled changeover procedure against the
last message transmitted peer in RPO.

The peer sending RPO must stop its T7 timer as it cannot expect
an acknowledgement for any messages during RPO (see test case
3.4.1 in draft-bidulock-sigtran-m2pa-test-02.txt).  Both sides
also stop T6 and cease any busy condition (sends BUSY-ENDED if
BUSY was sent previously).

To summarize:

- any data received by an M2PA peer experiencing LPO after an RPO
  has been sent MUST be discarded and MUST NOT be acknowledged.

- the M2PA peer experiencing LPO MUST NOT send any data messages
  between sending PROCESSOR-OUTAGE and PROCESSOR-OUTAGE-ENDED

- the M2PA peer sending PROCESSOR-OUTAGE SHOULD resynchornize
  sequence numbers by acknowledging all outstanding received
  data.

- any data received by an M2PA after receiving PROCESSOR-OUTAGE
  and before receiving PROCESSOR-OUTAGE-ENDED from the peer
  MUST be discarded and MUST NOT be acknowledged.

- the M2PA peer receiving PROCESSOR-OUTAGE MUST NOT send any
  data until receiving PROCESSOR-OUTAGE-ENDED.

- the M2PA peer receiving PROCESSOR-OUTAGE SHOULD resynchronize
  sequence numbers by acknowledging all outstanding data
  received before receiving PROCESSOR-OUTAGE.

This is completely consistent with Q.703/T1.111.3 and has the
benefit that harmful interactions with MTP Level 3 will not be
caused by introducing unexpected behavior (i.e. sending messages
for seconds after MTP Level 3 expects that the link has blocked).

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb  5 05:37:08 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA16529
	for <sigtran-archive@odin.ietf.org>; Wed, 5 Feb 2003 05:37:08 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h15Ah3713208
	for sigtran-archive@odin.ietf.org; Wed, 5 Feb 2003 05:43:03 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15AgdJ13168;
	Wed, 5 Feb 2003 05:42:40 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15Ae8J13078
	for <sigtran@optimus.ietf.org>; Wed, 5 Feb 2003 05:40:08 -0500
Received: from web12803.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id FAA16457
	for <sigtran@ietf.org>; Wed, 5 Feb 2003 05:33:42 -0500 (EST)
Message-ID: <20030205103720.32742.qmail@web12803.mail.yahoo.com>
Received: from [217.31.225.14] by web12803.mail.yahoo.com via HTTP; Wed, 05 Feb 2003 02:37:20 PST
Date: Wed, 5 Feb 2003 02:37:20 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] M2PA Flush Buffers
To: sigtran@ietf.org
In-Reply-To: <20030205010315.A12910@openss7.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Brian,
> To summarize:
> 
> - any data received by an M2PA peer experiencing LPO
> after an RPO
>   has been sent MUST be discarded and MUST NOT be
> acknowledged.
> 
> - the M2PA peer experiencing LPO MUST NOT send any
> data messages
>   between sending PROCESSOR-OUTAGE and
> PROCESSOR-OUTAGE-ENDED
> 
> - the M2PA peer sending PROCESSOR-OUTAGE SHOULD
> resynchornize
>   sequence numbers by acknowledging all outstanding
> received
>   data.
> 
> - any data received by an M2PA after receiving
> PROCESSOR-OUTAGE
>   and before receiving PROCESSOR-OUTAGE-ENDED from
> the peer
>   MUST be discarded and MUST NOT be acknowledged.
[TOLGA] In that case, what happens with messages sent
before LS(PO)and arrive after LS(PO)? Similarly what
will happen with messages, which are sent after LS(PE)
but arrive before it?
> 
> - the M2PA peer receiving PROCESSOR-OUTAGE MUST NOT
> send any
>   data until receiving PROCESSOR-OUTAGE-ENDED.
> 
> - the M2PA peer receiving PROCESSOR-OUTAGE SHOULD
> resynchronize
>   sequence numbers by acknowledging all outstanding
> data
>   received before receiving PROCESSOR-OUTAGE.
> 
> This is completely consistent with Q.703/T1.111.3
> and has the
> benefit that harmful interactions with MTP Level 3
> will not be
> caused by introducing unexpected behavior (i.e.
> sending messages
> for seconds after MTP Level 3 expects that the link
> has blocked).


__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb  5 09:02:05 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA25375
	for <sigtran-archive@odin.ietf.org>; Wed, 5 Feb 2003 09:02:04 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h15E82027242
	for sigtran-archive@odin.ietf.org; Wed, 5 Feb 2003 09:08:02 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15E7mJ27232;
	Wed, 5 Feb 2003 09:07:48 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15E6TJ26552
	for <sigtran@optimus.ietf.org>; Wed, 5 Feb 2003 09:06:29 -0500
Received: from chuckie.dgms.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA25299
	for <sigtran@ietf.org>; Wed, 5 Feb 2003 09:00:00 -0500 (EST)
Received: from pckhusidp (localhost [127.0.0.1])
	by chuckie.dgms.com (8.9.3/8.9.3) with SMTP id JAA10165
	for <sigtran@ietf.org>; Wed, 5 Feb 2003 09:03:36 -0500 (EST)
From: "Anatoly Khusid" <anatoly.khusid@ulticom.com>
To: <sigtran@ietf.org>
Subject: RE: [Sigtran] M2UA RFC 3331 Question
Date: Wed, 5 Feb 2003 09:01:47 -0500
Message-ID: <00a101c2cd1f$1f9abcb0$5a2119ac@pckhusidp>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook CWS, Build 9.0.2416 (9.0.2911.0)
x-mimeole: Produced By Microsoft MimeOLE V6.00.2600.0000
Importance: Normal
In-Reply-To: <4.3.2.7.2.20030204212139.048f5f80@cia.cisco.com>
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Ken,

Thanks for the clarification.
Now, it seems confusing to me that all APSTM messages are sent on the
individual streams.

For example, lets look at the ASP Active(ASPAC) message:
First case,  ASPAC has no optional Interface Identifier ( the message is for
all provisioned Interface Identifies within the AS ).
What SCTP stream should be used for this message?
Second case, ASPAC has a list or a range of optional Interface Identifies,
what SCTP stream should be used for this message?

I assume that in both cases ASP only sends ONE ASPAC message ( will not send
multiple ASPAC messages on different SCTP streams ).


Thanks

Anatoly

 -----Original Message-----
From: 	sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]  On Behalf Of
Ken Morneault
Sent:	Tuesday, February 04, 2003 9:25 PM
To:	Anatoly Khusid; sigtran@ietf.org
Subject:	Re: [Sigtran] M2UA RFC 3331 Question



Anatoly,

Pls see comments below.

Regards,

Ken

At 04:48 PM 2/4/2003 -0500, Anatoly Khusid wrote:
>Hi,
>
>I am not clear on one item in the M2UA RFC:
>
>[Page 11] (Section 1.5.4.1):
>
>"SCTP Stream '0' SHOULD NOT be used for MTP2 User Adaptation (MAUP)
>messages (see Section 3) since stream '0' SHOULD only be used for ASP
>Management (ASPM) messages (see Section 4.3.3)."

The "SHOULD" only applies to ASPSM messages.

>I assume ASPM acronym combines  ASPSM(Application Server Process State
>Maintenance) and ASPTM(Application Server Process Traffic Maintenance),
>which means all ASPTM messages must be sent on SCTP stream '0'
>
>Now, if you look at [Page 60] (Section 4.2.1) :
>"All MGMT messages, except BEAT and BEAT Ack, SHOULD be sent with
>sequenced delivery to ensure ordering.  All MGMT messages, with the
>exception of ASPTM, BEAT and BEAT Ack messages, SHOULD be sent on SCTP
>stream '0'.  All ASPTM messages SHOULD be sent on the stream which
>normally carries the data traffic to which the message applies.  BEAT
>and BEAT Ack messages MAY be sent using out-of-order delivery, and
>MAY be sent on any stream."
>
>This section says that ALL ASPTM messages should be sent on the stream
which
>normally carries the data traffic to which message applies.

Yes that is correct.  I agree that this information should only be listed
in one place in the RFC.  Will clarify in an implementors guide.

>Am I missing something, but I am confused which SCTP stream should be used
>for ASPTM messages.
>
>I appreciate if somebody could clear this out for me.
>
>Thanks a lot!
>
>Anatoly
>
> -----Original Message-----
>From:   sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]  On Behalf
Of
>Tom George
>Sent:   Tuesday, February 04, 2003 3:50 PM
>Cc:     sigtran@ietf.org
>Subject:        Re: [Sigtran] M2PA Flush Buffers
>
>I think we had this dicussion at the M2PA Interop and probably talked
>about it on the list long ago. 8-)
>
>Putting the LS and Data messages on the same stream does make it easy to
>decide in which order they were sent.
>
>The thing I have never liked about putting them on the same stream is
>that a LS message could get stuck in a queue behind many User Data
>messages.
>
>MTP2 does not have this problem. It puts LSSUs ahead of MSUs in its
>transmission buffer.
>
>If SCTP had an option to put an outgoing message at the head of the
>queue, it would be OK to send LS and User Data on the same stream.
>
>tomG.
>
>"Brian F. G. Bidulock" wrote:
>  [...]
>> IMO LS should be sent in same stream as DATA.
>>
>> In that case DATA received after LS PO is cause to place the link out of
>> service.
>>
>> --brian
>  [...]
>
>--
>* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
>* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
>* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
>* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
>* to advertise or represent its position,  operations, or activities. *
>_______________________________________________
>Sigtran mailing list
>Sigtran@ietf.org
>https://www1.ietf.org/mailman/listinfo/sigtran
>
>_______________________________________________
>Sigtran mailing list
>Sigtran@ietf.org
>https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb  5 09:19:28 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA26161
	for <sigtran-archive@odin.ietf.org>; Wed, 5 Feb 2003 09:19:28 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h15EPQv28063
	for sigtran-archive@odin.ietf.org; Wed, 5 Feb 2003 09:25:26 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15EPCJ28051;
	Wed, 5 Feb 2003 09:25:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15EOBJ28006
	for <sigtran@optimus.ietf.org>; Wed, 5 Feb 2003 09:24:11 -0500
Received: from rtp-msg-core-1.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA26124
	for <sigtran@ietf.org>; Wed, 5 Feb 2003 09:17:41 -0500 (EST)
Received: from cia.cisco.com (localhost [127.0.0.1])
	by rtp-msg-core-1.cisco.com (8.12.2/8.12.6) with ESMTP id h15EMBb8026939;
	Wed, 5 Feb 2003 09:22:11 -0500 (EST)
Received: from kmorneau-w2k.cisco.com (hrn2-dhcp-161-44-86-125.cisco.com [161.44.86.125])
	by cia.cisco.com (Mirapoint)
	with ESMTP id ABO46316;
	Wed, 5 Feb 2003 09:11:18 -0500 (EST)
Message-Id: <4.3.2.7.2.20030205091918.04652ca8@cia.cisco.com>
X-Sender: kmorneau@cia.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Wed, 05 Feb 2003 09:21:15 -0500
To: "Anatoly Khusid" <anatoly.khusid@ulticom.com>, <sigtran@ietf.org>
From: Ken Morneault <kmorneau@cisco.com>
Subject: RE: [Sigtran] M2UA RFC 3331 Question
In-Reply-To: <00a101c2cd1f$1f9abcb0$5a2119ac@pckhusidp>
References: <4.3.2.7.2.20030204212139.048f5f80@cia.cisco.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>



Anatoly,

Yes, in the cases you describe, only one ASPAC message would
be sent and it would not matter which stream it was sent on.

Regards,

Ken

At 09:01 AM 2/5/2003 -0500, Anatoly Khusid wrote:
>Ken,
>
>Thanks for the clarification.
>Now, it seems confusing to me that all APSTM messages are sent on the
>individual streams.
>
>For example, lets look at the ASP Active(ASPAC) message:
>First case,  ASPAC has no optional Interface Identifier ( the message is for
>all provisioned Interface Identifies within the AS ).
>What SCTP stream should be used for this message?
>Second case, ASPAC has a list or a range of optional Interface Identifies,
>what SCTP stream should be used for this message?
>
>I assume that in both cases ASP only sends ONE ASPAC message ( will not send
>multiple ASPAC messages on different SCTP streams ).
>
>
>Thanks
>
>Anatoly
>
> -----Original Message-----
>From:   sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]  On Behalf Of
>Ken Morneault
>Sent:   Tuesday, February 04, 2003 9:25 PM
>To:     Anatoly Khusid; sigtran@ietf.org
>Subject:        Re: [Sigtran] M2UA RFC 3331 Question
>
>
>
>Anatoly,
>
>Pls see comments below.
>
>Regards,
>
>Ken
>
>At 04:48 PM 2/4/2003 -0500, Anatoly Khusid wrote:
>>Hi,
>>
>>I am not clear on one item in the M2UA RFC:
>>
>>[Page 11] (Section 1.5.4.1):
>>
>>"SCTP Stream '0' SHOULD NOT be used for MTP2 User Adaptation (MAUP)
>>messages (see Section 3) since stream '0' SHOULD only be used for ASP
>>Management (ASPM) messages (see Section 4.3.3)."
>
>The "SHOULD" only applies to ASPSM messages.
>
>>I assume ASPM acronym combines  ASPSM(Application Server Process State
>>Maintenance) and ASPTM(Application Server Process Traffic Maintenance),
>>which means all ASPTM messages must be sent on SCTP stream '0'
>>
>>Now, if you look at [Page 60] (Section 4.2.1) :
>>"All MGMT messages, except BEAT and BEAT Ack, SHOULD be sent with
>>sequenced delivery to ensure ordering.  All MGMT messages, with the
>>exception of ASPTM, BEAT and BEAT Ack messages, SHOULD be sent on SCTP
>>stream '0'.  All ASPTM messages SHOULD be sent on the stream which
>>normally carries the data traffic to which the message applies.  BEAT
>>and BEAT Ack messages MAY be sent using out-of-order delivery, and
>>MAY be sent on any stream."
>>
>>This section says that ALL ASPTM messages should be sent on the stream
>which
>>normally carries the data traffic to which message applies.
>
>Yes that is correct.  I agree that this information should only be listed
>in one place in the RFC.  Will clarify in an implementors guide.
>
>>Am I missing something, but I am confused which SCTP stream should be used
>>for ASPTM messages.
>>
>>I appreciate if somebody could clear this out for me.
>>
>>Thanks a lot!
>>
>>Anatoly
>>
>> -----Original Message-----
>>From:   sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]  On Behalf
>Of
>>Tom George
>>Sent:   Tuesday, February 04, 2003 3:50 PM
>>Cc:     sigtran@ietf.org
>>Subject:        Re: [Sigtran] M2PA Flush Buffers
>>
>>I think we had this dicussion at the M2PA Interop and probably talked
>>about it on the list long ago. 8-)
>>
>>Putting the LS and Data messages on the same stream does make it easy to
>>decide in which order they were sent.
>>
>>The thing I have never liked about putting them on the same stream is
>>that a LS message could get stuck in a queue behind many User Data
>>messages.
>>
>>MTP2 does not have this problem. It puts LSSUs ahead of MSUs in its
>>transmission buffer.
>>
>>If SCTP had an option to put an outgoing message at the head of the
>>queue, it would be OK to send LS and User Data on the same stream.
>>
>>tomG.
>>
>>"Brian F. G. Bidulock" wrote:
>>  [...]
>>> IMO LS should be sent in same stream as DATA.
>>>
>>> In that case DATA received after LS PO is cause to place the link out of
>>> service.
>>>
>>> --brian
>>  [...]
>>
>>--
>>* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
>>* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
>>* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
>>* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
>>* to advertise or represent its position,  operations, or activities. *
>>_______________________________________________
>>Sigtran mailing list
>>Sigtran@ietf.org
>>https://www1.ietf.org/mailman/listinfo/sigtran
>>
>>_______________________________________________
>>Sigtran mailing list
>>Sigtran@ietf.org
>>https://www1.ietf.org/mailman/listinfo/sigtran
>
>_______________________________________________
>Sigtran mailing list
>Sigtran@ietf.org
>https://www1.ietf.org/mailman/listinfo/sigtran
>
>_______________________________________________
>Sigtran mailing list
>Sigtran@ietf.org
>https://www1.ietf.org/mailman/listinfo/sigtran 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb  5 09:43:37 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA26996
	for <sigtran-archive@odin.ietf.org>; Wed, 5 Feb 2003 09:43:37 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h15EnZs29755
	for sigtran-archive@odin.ietf.org; Wed, 5 Feb 2003 09:49:35 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15EnFJ29742;
	Wed, 5 Feb 2003 09:49:17 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15Em4J29706
	for <sigtran@optimus.ietf.org>; Wed, 5 Feb 2003 09:48:04 -0500
Received: from chuckie.dgms.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA26931
	for <sigtran@ietf.org>; Wed, 5 Feb 2003 09:41:33 -0500 (EST)
Received: from pckhusidp (localhost [127.0.0.1])
	by chuckie.dgms.com (8.9.3/8.9.3) with SMTP id JAA22573;
	Wed, 5 Feb 2003 09:45:00 -0500 (EST)
From: "Anatoly Khusid" <anatoly.khusid@ulticom.com>
To: "'Ken Morneault'" <kmorneau@cisco.com>, <sigtran@ietf.org>
Subject: RE: [Sigtran] M2UA RFC 3331 Question
Date: Wed, 5 Feb 2003 09:43:10 -0500
Message-ID: <00a201c2cd24$e81e93f0$5a2119ac@pckhusidp>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook CWS, Build 9.0.2416 (9.0.2911.0)
x-mimeole: Produced By Microsoft MimeOLE V6.00.2600.0000
Importance: Normal
In-Reply-To: <4.3.2.7.2.20030205091918.04652ca8@cia.cisco.com>
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Ken,

Thanks.  As I can see there are only 4 ASPTM messages used by M2UA:
ASP Active
ASP Active ACK
ASP Inactive
ASP Inactive ACK

I don't really understand when it is necessary to use specific stream vs.
*any* stream to send ASPTM messages.

Lets say SGP receives ASP Active on SCTP (stream 4, association 1), is
necessary to send ASP Active ACK on the same stream (stream 4, association
1) or any stream?  Does the same logic apply to ASP Inactive / ASP Inactive
ACK.


Thanks

Anatoly

 -----Original Message-----
From: 	Ken Morneault [mailto:kmorneau@cisco.com]
Sent:	Wednesday, February 05, 2003 9:21 AM
To:	Anatoly Khusid; sigtran@ietf.org
Subject:	RE: [Sigtran] M2UA RFC 3331 Question



Anatoly,

Yes, in the cases you describe, only one ASPAC message would
be sent and it would not matter which stream it was sent on.

Regards,

Ken

At 09:01 AM 2/5/2003 -0500, Anatoly Khusid wrote:
>Ken,
>
>Thanks for the clarification.
>Now, it seems confusing to me that all APSTM messages are sent on the
>individual streams.
>
>For example, lets look at the ASP Active(ASPAC) message:
>First case,  ASPAC has no optional Interface Identifier ( the message is
for
>all provisioned Interface Identifies within the AS ).
>What SCTP stream should be used for this message?
>Second case, ASPAC has a list or a range of optional Interface Identifies,
>what SCTP stream should be used for this message?
>
>I assume that in both cases ASP only sends ONE ASPAC message ( will not
send
>multiple ASPAC messages on different SCTP streams ).
>
>
>Thanks
>
>Anatoly
>
> -----Original Message-----
>From:   sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]  On Behalf
Of
>Ken Morneault
>Sent:   Tuesday, February 04, 2003 9:25 PM
>To:     Anatoly Khusid; sigtran@ietf.org
>Subject:        Re: [Sigtran] M2UA RFC 3331 Question
>
>
>
>Anatoly,
>
>Pls see comments below.
>
>Regards,
>
>Ken
>
>At 04:48 PM 2/4/2003 -0500, Anatoly Khusid wrote:
>>Hi,
>>
>>I am not clear on one item in the M2UA RFC:
>>
>>[Page 11] (Section 1.5.4.1):
>>
>>"SCTP Stream '0' SHOULD NOT be used for MTP2 User Adaptation (MAUP)
>>messages (see Section 3) since stream '0' SHOULD only be used for ASP
>>Management (ASPM) messages (see Section 4.3.3)."
>
>The "SHOULD" only applies to ASPSM messages.
>
>>I assume ASPM acronym combines  ASPSM(Application Server Process State
>>Maintenance) and ASPTM(Application Server Process Traffic Maintenance),
>>which means all ASPTM messages must be sent on SCTP stream '0'
>>
>>Now, if you look at [Page 60] (Section 4.2.1) :
>>"All MGMT messages, except BEAT and BEAT Ack, SHOULD be sent with
>>sequenced delivery to ensure ordering.  All MGMT messages, with the
>>exception of ASPTM, BEAT and BEAT Ack messages, SHOULD be sent on SCTP
>>stream '0'.  All ASPTM messages SHOULD be sent on the stream which
>>normally carries the data traffic to which the message applies.  BEAT
>>and BEAT Ack messages MAY be sent using out-of-order delivery, and
>>MAY be sent on any stream."
>>
>>This section says that ALL ASPTM messages should be sent on the stream
>which
>>normally carries the data traffic to which message applies.
>
>Yes that is correct.  I agree that this information should only be listed
>in one place in the RFC.  Will clarify in an implementors guide.
>
>>Am I missing something, but I am confused which SCTP stream should be used
>>for ASPTM messages.
>>
>>I appreciate if somebody could clear this out for me.
>>
>>Thanks a lot!
>>
>>Anatoly
>>
>> -----Original Message-----
>>From:   sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]  On Behalf
>Of
>>Tom George
>>Sent:   Tuesday, February 04, 2003 3:50 PM
>>Cc:     sigtran@ietf.org
>>Subject:        Re: [Sigtran] M2PA Flush Buffers
>>
>>I think we had this dicussion at the M2PA Interop and probably talked
>>about it on the list long ago. 8-)
>>
>>Putting the LS and Data messages on the same stream does make it easy to
>>decide in which order they were sent.
>>
>>The thing I have never liked about putting them on the same stream is
>>that a LS message could get stuck in a queue behind many User Data
>>messages.
>>
>>MTP2 does not have this problem. It puts LSSUs ahead of MSUs in its
>>transmission buffer.
>>
>>If SCTP had an option to put an outgoing message at the head of the
>>queue, it would be OK to send LS and User Data on the same stream.
>>
>>tomG.
>>
>>"Brian F. G. Bidulock" wrote:
>>  [...]
>>> IMO LS should be sent in same stream as DATA.
>>>
>>> In that case DATA received after LS PO is cause to place the link out of
>>> service.
>>>
>>> --brian
>>  [...]
>>
>>--
>>* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
>>* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
>>* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
>>* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
>>* to advertise or represent its position,  operations, or activities. *
>>_______________________________________________
>>Sigtran mailing list
>>Sigtran@ietf.org
>>https://www1.ietf.org/mailman/listinfo/sigtran
>>
>>_______________________________________________
>>Sigtran mailing list
>>Sigtran@ietf.org
>>https://www1.ietf.org/mailman/listinfo/sigtran
>
>_______________________________________________
>Sigtran mailing list
>Sigtran@ietf.org
>https://www1.ietf.org/mailman/listinfo/sigtran
>
>_______________________________________________
>Sigtran mailing list
>Sigtran@ietf.org
>https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb  5 09:52:39 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA27224
	for <sigtran-archive@odin.ietf.org>; Wed, 5 Feb 2003 09:52:39 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h15EwcU30088
	for sigtran-archive@odin.ietf.org; Wed, 5 Feb 2003 09:58:38 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15EwLJ30055;
	Wed, 5 Feb 2003 09:58:21 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15EvxJ30023
	for <sigtran@optimus.ietf.org>; Wed, 5 Feb 2003 09:57:59 -0500
Received: from rtp-msg-core-1.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA27189
	for <sigtran@ietf.org>; Wed, 5 Feb 2003 09:51:23 -0500 (EST)
Received: from cia.cisco.com (localhost [127.0.0.1])
	by rtp-msg-core-1.cisco.com (8.12.2/8.12.6) with ESMTP id h15EtrOw002072;
	Wed, 5 Feb 2003 09:55:54 -0500 (EST)
Received: from kmorneau-w2k.cisco.com (hrn2-dhcp-161-44-86-125.cisco.com [161.44.86.125])
	by cia.cisco.com (Mirapoint)
	with ESMTP id ABO46707;
	Wed, 5 Feb 2003 09:44:59 -0500 (EST)
Message-Id: <4.3.2.7.2.20030205095154.046696a0@cia.cisco.com>
X-Sender: kmorneau@cia.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Wed, 05 Feb 2003 09:54:58 -0500
To: "Anatoly Khusid" <anatoly.khusid@ulticom.com>, <sigtran@ietf.org>
From: Ken Morneault <kmorneau@cisco.com>
Subject: RE: [Sigtran] M2UA RFC 3331 Question
In-Reply-To: <00a201c2cd24$e81e93f0$5a2119ac@pckhusidp>
References: <4.3.2.7.2.20030205091918.04652ca8@cia.cisco.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>



Anatoly,

If you want all of the traffic related to a particular Interface Identifier (including
ASPTM messages) to be ordered, then these messages must be sent on the same 
stream.  

The SGP does not need to respond to the ASPAC or ASPIA on the
same stream as they were received, just as it does not have to select
 the same stream the ASP selects to send Data messages associated 
with an Interface Identifier.

Regards,

Ken

At 09:43 AM 2/5/2003 -0500, Anatoly Khusid wrote:
>Ken,
>
>Thanks.  As I can see there are only 4 ASPTM messages used by M2UA:
>ASP Active
>ASP Active ACK
>ASP Inactive
>ASP Inactive ACK
>
>I don't really understand when it is necessary to use specific stream vs.
>*any* stream to send ASPTM messages.
>
>Lets say SGP receives ASP Active on SCTP (stream 4, association 1), is
>necessary to send ASP Active ACK on the same stream (stream 4, association
>1) or any stream?  Does the same logic apply to ASP Inactive / ASP Inactive
>ACK.
>
>
>Thanks
>
>Anatoly
>
> -----Original Message-----
>From:   Ken Morneault [mailto:kmorneau@cisco.com]
>Sent:   Wednesday, February 05, 2003 9:21 AM
>To:     Anatoly Khusid; sigtran@ietf.org
>Subject:        RE: [Sigtran] M2UA RFC 3331 Question
>
>
>
>Anatoly,
>
>Yes, in the cases you describe, only one ASPAC message would
>be sent and it would not matter which stream it was sent on.
>
>Regards,
>
>Ken
>
>At 09:01 AM 2/5/2003 -0500, Anatoly Khusid wrote:
>>Ken,
>>
>>Thanks for the clarification.
>>Now, it seems confusing to me that all APSTM messages are sent on the
>>individual streams.
>>
>>For example, lets look at the ASP Active(ASPAC) message:
>>First case,  ASPAC has no optional Interface Identifier ( the message is
>for
>>all provisioned Interface Identifies within the AS ).
>>What SCTP stream should be used for this message?
>>Second case, ASPAC has a list or a range of optional Interface Identifies,
>>what SCTP stream should be used for this message?
>>
>>I assume that in both cases ASP only sends ONE ASPAC message ( will not
>send
>>multiple ASPAC messages on different SCTP streams ).
>>
>>
>>Thanks
>>
>>Anatoly
>>
>> -----Original Message-----
>>From:   sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]  On Behalf
>Of
>>Ken Morneault
>>Sent:   Tuesday, February 04, 2003 9:25 PM
>>To:     Anatoly Khusid; sigtran@ietf.org
>>Subject:        Re: [Sigtran] M2UA RFC 3331 Question
>>
>>
>>
>>Anatoly,
>>
>>Pls see comments below.
>>
>>Regards,
>>
>>Ken
>>
>>At 04:48 PM 2/4/2003 -0500, Anatoly Khusid wrote:
>>>Hi,
>>>
>>>I am not clear on one item in the M2UA RFC:
>>>
>>>[Page 11] (Section 1.5.4.1):
>>>
>>>"SCTP Stream '0' SHOULD NOT be used for MTP2 User Adaptation (MAUP)
>>>messages (see Section 3) since stream '0' SHOULD only be used for ASP
>>>Management (ASPM) messages (see Section 4.3.3)."
>>
>>The "SHOULD" only applies to ASPSM messages.
>>
>>>I assume ASPM acronym combines  ASPSM(Application Server Process State
>>>Maintenance) and ASPTM(Application Server Process Traffic Maintenance),
>>>which means all ASPTM messages must be sent on SCTP stream '0'
>>>
>>>Now, if you look at [Page 60] (Section 4.2.1) :
>>>"All MGMT messages, except BEAT and BEAT Ack, SHOULD be sent with
>>>sequenced delivery to ensure ordering.  All MGMT messages, with the
>>>exception of ASPTM, BEAT and BEAT Ack messages, SHOULD be sent on SCTP
>>>stream '0'.  All ASPTM messages SHOULD be sent on the stream which
>>>normally carries the data traffic to which the message applies.  BEAT
>>>and BEAT Ack messages MAY be sent using out-of-order delivery, and
>>>MAY be sent on any stream."
>>>
>>>This section says that ALL ASPTM messages should be sent on the stream
>>which
>>>normally carries the data traffic to which message applies.
>>
>>Yes that is correct.  I agree that this information should only be listed
>>in one place in the RFC.  Will clarify in an implementors guide.
>>
>>>Am I missing something, but I am confused which SCTP stream should be used
>>>for ASPTM messages.
>>>
>>>I appreciate if somebody could clear this out for me.
>>>
>>>Thanks a lot!
>>>
>>>Anatoly
>>>
>>> -----Original Message-----
>>>From:   sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]  On Behalf
>>Of
>>>Tom George
>>>Sent:   Tuesday, February 04, 2003 3:50 PM
>>>Cc:     sigtran@ietf.org
>>>Subject:        Re: [Sigtran] M2PA Flush Buffers
>>>
>>>I think we had this dicussion at the M2PA Interop and probably talked
>>>about it on the list long ago. 8-)
>>>
>>>Putting the LS and Data messages on the same stream does make it easy to
>>>decide in which order they were sent.
>>>
>>>The thing I have never liked about putting them on the same stream is
>>>that a LS message could get stuck in a queue behind many User Data
>>>messages.
>>>
>>>MTP2 does not have this problem. It puts LSSUs ahead of MSUs in its
>>>transmission buffer.
>>>
>>>If SCTP had an option to put an outgoing message at the head of the
>>>queue, it would be OK to send LS and User Data on the same stream.
>>>
>>>tomG.
>>>
>>>"Brian F. G. Bidulock" wrote:
>>>  [...]
>>>> IMO LS should be sent in same stream as DATA.
>>>>
>>>> In that case DATA received after LS PO is cause to place the link out of
>>>> service.
>>>>
>>>> --brian
>>>  [...]
>>>
>>>--
>>>* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
>>>* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
>>>* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
>>>* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
>>>* to advertise or represent its position,  operations, or activities. *
>>>_______________________________________________
>>>Sigtran mailing list
>>>Sigtran@ietf.org
>>>https://www1.ietf.org/mailman/listinfo/sigtran
>>>
>>>_______________________________________________
>>>Sigtran mailing list
>>>Sigtran@ietf.org
>>>https://www1.ietf.org/mailman/listinfo/sigtran
>>
>>_______________________________________________
>>Sigtran mailing list
>>Sigtran@ietf.org
>>https://www1.ietf.org/mailman/listinfo/sigtran
>>
>>_______________________________________________
>>Sigtran mailing list
>>Sigtran@ietf.org
>>https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb  5 10:52:56 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA29181
	for <sigtran-archive@odin.ietf.org>; Wed, 5 Feb 2003 10:52:56 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h15Fwuv01629
	for sigtran-archive@odin.ietf.org; Wed, 5 Feb 2003 10:58:56 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15FwbJ01604;
	Wed, 5 Feb 2003 10:58:38 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15FvTJ01565
	for <sigtran@optimus.ietf.org>; Wed, 5 Feb 2003 10:57:29 -0500
Received: from domino2.nc.tekelec.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA29100
	for <sigtran@ietf.org>; Wed, 5 Feb 2003 10:50:57 -0500 (EST)
From: jeffrey.craig@tekelec.com
Received: from domino4.nc.tekelec.com ([198.89.40.9])
          by domino2.nc.tekelec.com (Lotus Domino Release 5.0.10)
          with ESMTP id 2003020510482163:19115 ;
          Wed, 5 Feb 2003 10:48:21 -0500 
Subject: Re: [Sigtran] M2PA Flush Buffers
To: bidulock@openss7.org
Cc: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.7  March 21, 2001
Message-ID: <OFF9E41E19.04BD4D8C-ON85256CC4.004ED0B7@nc.tekelec.com>
Date: Wed, 5 Feb 2003 10:50:21 -0500
MIME-Version: 1.0
X-MIMETrack: Serialize by Router on Domino4/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/05/2003 10:50:22 AM,
	Itemize by SMTP Server on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/05/2003 10:48:21 AM,
	Serialize by Router on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/05/2003 10:48:24 AM,
	Serialize complete at 02/05/2003 10:48:24 AM
Content-type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h15FvTJ01566
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit


Hello Brian,

I agree with most of what you have said. It appears, however, that only
the "clear buffers" case for exiting processor outage is being addressed.
The more challenging case involves the "resume" (recovery prior to MTP3
T1 expiry and execution of timed-controlled changeover).

The following characteristics of M2PA v7 make support of processor outage
problematic:
1) M2PA maintains sequence numbers but does not retransmit, unlike MTP2.
2) Reception order of user data/ACKs vs. LSPO/LSPOE cannot be ensured
due to the use of multiple SCTP steams.

Perhaps I am missing something, but it seems to me that your statement

"- any data received by an M2PA peer experiencing LPO after an RPO
  has been sent MUST be discarded and MUST NOT be acknowledged."

guarantees that the M2PA link will go out of service for the "resume"
case of exiting PO. My understanding is that for this case, the
retransmit queue is not cleared and M2PA cannot retransmit. No re-synching
of sequence numbers should be necessary for this case.

Having said this, I think that while M2PA v7 clarity can be improved,
there is no issue with the draft's description of PO procedures. The
draft currently does not require that the receiver discard messages
during PO.

A solution for these issues is that upon PO onset (either local or remote)
M2PA begins buffering received user data without acknowledgement. This
can be accomplished by not receiving from SCTP stream 1, or by buffering
user data at the M2PA layer. Because the received messages are buffered,
it does not matter about the order of receipt of data/ACK and LSPO/LSPOE.
As discussed on the list previously, M2PA must continue to receive and
process messages on SCTP stream 0 (link status) during PO.

In the case of "resume" after PO, M2PA should simply unbuffer the
received user data messages, validate them, acknowledge receipt, send them
to L3, and begin transmitting as normal. No retrieval or rerouting occurs,
and no messages are lost.

In the case of "clear buffers" after PO, M2PA should flush the all
outbound and inbound queues (possibly including SCTP receive buffer),
and then re-synch the sequence numbers with the peer.

Regards,

Jeff






"Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 02/05/2003
03:03:15 AM

Please respond to bidulock@openss7.org

Sent by:    sigtran-admin@ietf.org


To:    TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
cc:    sigtran@ietf.org
Subject:    Re: [Sigtran] M2PA Flush Buffers


TOLGA,

TOLGA ASVEREN wrote:
(Wed, 05 Feb 2003 08:25:56)
>
> Tom,
> --- Tom George <Tom.George@alcatel.com> wrote:
> > Andrew,
> >
> > Thanks for the comments. I will gladly defer to
> > those with more SS7
> > smarts, but I think there is no case where the
> > receiver has to know the
> > order in which LS and User Data messages were sent.
> > MTP2 itself has
> > LSSUs jumping to the head of the line when they are
> > transmitted.
> >
> > So what do we do with User Data messages that are
> > received after a LS
> > PO? We can't just copy MTP2, since the same
> > situation does not occur. In
> > MTP2, receiving an MSU ends the RPO condition.
> >
> > The current M2PA does not say exactly what to do,
> > which I would
> > interpret (if I may "interpret" my own writing) that
> > the receiver should
> > go ahead and process the User Data messages. The
> > presumption would be
> > that they are valid messages sent before the RPO
> > occurred.
> [TOLGA]IMO it is better to mention this in the document. I have some more
> questions though:
> I assume even after receiving LS(PO) from the peer, M2PA continues to
> acknowledge received messages(?)
> Similarliy I assume during local processor outage, based on received
> acknowledgements buffers are updated(?) IMO, it might be good to clarify
> this in the text(if this is the case)
> What about T7 on the peer experiencing local processor outage? Will it be
> stopped or not? If the above is true, I would think it shouldn´t be
> stopped.
> In 4.1.4 there is a sentence
> "Alternatively, MTP3 may perform data retrieval as part of a changeover
> procedure."
> Are the described procedures in that sentence and flushing buffers really
> alternatives to each other?

To avoid mis-sequencing of messages, neither end should send or
acknowledge messages during LPO.

The receiver of RPO must not acknowledge any messages after the
receipt of RPO to remain consistent with Q.703 and interactions
with Q.704 (time-controlled changeover procedure).  The MTP
Level 3 suffering LPO must know that when LPO begins that
message transmission has ceased so that the time-controlled
changeover procedure time duration is correctly referenced
against the last message transmitted on the blocked link.

To remain consistent with ANSI T1.111.3, both the sender and
the receiver of RPO should re-synchronize sequence numbers
(acknowlege received messages awaiting ack) and cease
transmission of DATA.

Again, just as the side experiencing LPO must reference its
time-controlled changeover at Level 3 against the last message
sent on the blocked link, so must the side detecting RPO
reference its time-controlled changeover procedure against the
last message transmitted peer in RPO.

The peer sending RPO must stop its T7 timer as it cannot expect
an acknowledgement for any messages during RPO (see test case
3.4.1 in draft-bidulock-sigtran-m2pa-test-02.txt).  Both sides
also stop T6 and cease any busy condition (sends BUSY-ENDED if
BUSY was sent previously).

To summarize:

- any data received by an M2PA peer experiencing LPO after an RPO
  has been sent MUST be discarded and MUST NOT be acknowledged.

- the M2PA peer experiencing LPO MUST NOT send any data messages
  between sending PROCESSOR-OUTAGE and PROCESSOR-OUTAGE-ENDED

- the M2PA peer sending PROCESSOR-OUTAGE SHOULD resynchornize
  sequence numbers by acknowledging all outstanding received
  data.

- any data received by an M2PA after receiving PROCESSOR-OUTAGE
  and before receiving PROCESSOR-OUTAGE-ENDED from the peer
  MUST be discarded and MUST NOT be acknowledged.

- the M2PA peer receiving PROCESSOR-OUTAGE MUST NOT send any
  data until receiving PROCESSOR-OUTAGE-ENDED.

- the M2PA peer receiving PROCESSOR-OUTAGE SHOULD resynchronize
  sequence numbers by acknowledging all outstanding data
  received before receiving PROCESSOR-OUTAGE.

This is completely consistent with Q.703/T1.111.3 and has the
benefit that harmful interactions with MTP Level 3 will not be
caused by introducing unexpected behavior (i.e. sending messages
for seconds after MTP Level 3 expects that the link has blocked).

--brian

--
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb  5 12:26:50 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA02683
	for <sigtran-archive@odin.ietf.org>; Wed, 5 Feb 2003 12:26:50 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h15HWrB08668
	for sigtran-archive@odin.ietf.org; Wed, 5 Feb 2003 12:32:53 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15HWeJ08656;
	Wed, 5 Feb 2003 12:32:40 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15HVWJ08614
	for <sigtran@optimus.ietf.org>; Wed, 5 Feb 2003 12:31:33 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA02612
	for <sigtran@ietf.org>; Wed, 5 Feb 2003 12:24:58 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h15HSaqe027445;
	Wed, 5 Feb 2003 11:28:36 -0600 (CST)
Message-ID: <3E4149C3.FC2A24B2@alcatel.com>
Date: Wed, 05 Feb 2003 11:28:35 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA LPO during alignment/proving (was: M2PA version 7 
 changes)
References: <3D72C18DF140D311A54F0008C7DBAC8D02917431@MCHH252E>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Franz,

Comments below...

Gaiser Franz wrote:
> 
> Hi All,
> 
> two additions to the Link State Transition Diagram below:
> 
> 1.) Figure 7. M2PA Link State Transition Diagram:
> Missing the handling of receiving LS PO in state ALIGNED READY: Situation: After Proving the local side is in state ALIGNED READY and sends LSR; the peer side is in state ALIGNED NOT READY experiencing Processor Outage and starts sending LS PO. To which state the local M2PA is moving ? Transition to PROCESSOR OUTAGE ? If so, the state PROCESSOR OUTAGE handles both local and remote processor outage and then there must be an arrow from PO to IN SERVICE in case of receiving LS PO Recovered (and there is no local PO situation).

If M2PA is in ALIGNED READY and receives LS PO, it should go to ALIGNED
NOT READY. When it gets LS Ready from the other end, then it would go to
PROCESSOR OUTAGE. 

I updated the diagram accordingly. See below.

> 
> 2.) Figure 7. M2PA Link State Transition Diagram:
> Missing the handling of receiving User data in state ALIGNED NOT READY. Transition to PROCESSOR OUTAGE ? What is to do with this message then (to be discarded) ?

We are in the middle of discussing this in another thread. It looks like
User Data should not be sent/received when in ALIGNED NOT READY. If it
can happen, it would be an error case that I would prefer not to show in
the diagram.

> 
> regards, franz

Updated piece of the diagram:

  |                       |   +-----+                  |                
  |                       |   |     | LPO/LPR          |                
  |                       V   V     |                  |                
  |                 +-----------+   |                  |
  |                 |  PROVING  |---+                  |
  |<----------------|           |--------------------->+
  |      MTP3 Stop  +-----------+     SCTP Comm Error  |
  |   OR T3 Expiry        |  |     OR SCTP Comm Lost   |
  |   OR Receive LS OOS   |  |                         |
  |                       |  +-------------------------|--------+
  |                       |     T4 Expiry              |        |      
  |                       |     Send LS PO             |        |      
  |                       |                            |        |      
  |                       | T4 Expiry                  |        |      
  |                       | Send LS Ready              |        |      
  |                       |                            |        |      
  |                       |                            |        |      
  |                       V           SCTP Comm Error  |        V      
  |                 +-----------+  OR SCTP Comm Lost   |  +-----------+
  |                 |  ALIGNED  |--------------------->+  |  ALIGNED  |
  +<----------------|   READY   |<----------------------->| NOT READY |
  |      MTP3 Stop  +-----------+     LPO/LPR             +-----------+
  |   OR T1 Expiry        |        OR Receive LS PO/PR       |       |
  |   OR Receive LS OOS   |                                  |   T1  | 
  |   OR Receive LS Align | Receive LS Ready      Receive    | Expiry|
  |                       | OR Receive User Data    LS Ready | OR Rcv|
  |                       |                                  |    LS |
  |                       |                                  |  Align|
  |                       |                                  |       |
  |                       V                                  V       |
  |                 +-----------+                      +-----------+ |
  |                 |    IN     |<-------------------->| PROCESSOR | |
  |                 |  SERVICE  |     LPO/LPR          |   OUTAGE  | |
  |                 +-----------+  OR Receive LS PO/PR +-----------+ |
  |                       |                                  |       |
  |                       |                                  |       |



-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb  5 14:13:55 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA07048
	for <sigtran-archive@odin.ietf.org>; Wed, 5 Feb 2003 14:13:55 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h15JK1F15724
	for sigtran-archive@odin.ietf.org; Wed, 5 Feb 2003 14:20:01 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15JJbJ15679;
	Wed, 5 Feb 2003 14:19:39 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15JIPJ15636
	for <sigtran@optimus.ietf.org>; Wed, 5 Feb 2003 14:18:25 -0500
Received: from domino2.nc.tekelec.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA06990
	for <sigtran@ietf.org>; Wed, 5 Feb 2003 14:11:48 -0500 (EST)
From: jeffrey.craig@tekelec.com
Received: from domino4.nc.tekelec.com ([198.89.40.9])
          by domino2.nc.tekelec.com (Lotus Domino Release 5.0.10)
          with ESMTP id 2003020514091349:19844 ;
          Wed, 5 Feb 2003 14:09:13 -0500 
Subject: Re: [Sigtran] M2PA LPO during alignment/proving (was: M2PA version 7
 changes)
To: Tom George <Tom.George@alcatel.com>
Cc: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.7  March 21, 2001
Message-ID: <OF5DFAA082.4616B342-ON85256CC4.00649086@nc.tekelec.com>
Date: Wed, 5 Feb 2003 14:11:13 -0500
MIME-Version: 1.0
X-MIMETrack: Serialize by Router on Domino4/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/05/2003 02:11:14 PM,
	Itemize by SMTP Server on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/05/2003 02:09:13 PM,
	Serialize by Router on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/05/2003 02:09:16 PM,
	Serialize complete at 02/05/2003 02:09:16 PM
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hello Tom,

>If M2PA is in ALIGNED READY and receives LS PO, it should go to ALIGNED
>NOT READY. When it gets LS Ready from the other end, then it would go to
>PROCESSOR OUTAGE.

I think you meant 'Local Processor Outage Request' (LPO) primitive from
MTP3, rather than 'LS PO' in the preceding paragraph.

I recommend that the M2PA state machine be split into two figures, one
that addresses the 'normal' alignment case, and another that addresses
the 'processor outage' case. I am not sure that the Q.703 SDL can be
accurately reproduced in a single text-style figure.

The M2PA state machine is missing transitions and has incorrectly labeled
transitions, relative to Q.703:

1) LSPO received during ALIGNED_READY  causes a transition to
PROCESSOR_OUTAGE (Q.703, figure 8, sheet 7 of 14).

2) LSPO or LSR received during ALIGNED_NOT_READY causes a transition
to PROCESSOR_OUTAGE (Q.703, figure 8, sheet 9 of 14).

3) LSPOE received during PROCESSOR_OUTAGE causes M2PA to notify
MTP3 of end of remote processor outage (no state transition).
(Q.703, figure 8, sheet 12 of 14).

4) In the absence of both local or remote processor outage,
a 'continue' primitive or 'flush buffers' primitive from MTP3
during PROCESSOR_OUTAGE causes a transition to IN_SERVICE
(Q.703, figure 8, sheet 13 of 14). Reception of LPR during
PROCESSOR_OUTAGE does not cause a transition, but rather
prompts M2PA to signal MTP3 (Q.703, figure 8, sheet 13 of 14).

5) A 'Local Processor Outage Request' (LPO) primtive received from MTP3
during PROCESSOR_OUTAGE results in no state transition, though M2PA
records the LPO state.

----

After all of this....
Perhaps M2PA, just like SAAL/ATM, should abandon support of this
obsolete (processor outage) procedure, and instead, fail the link
when it receives the LPO primitive from MTP3. Many, if not most,
modern MTP3 implementations never generate this primitive.

Why does M2PA need processor outage? SAAL/ATM does just fine
without it.

Regards,

Jeff






Tom George <Tom.George@alcatel.com>@ietf.org on 02/05/2003 12:28:35 PM

Sent by:    sigtran-admin@ietf.org


To:
cc:    sigtran@ietf.org
Subject:    Re: [Sigtran] M2PA LPO during alignment/proving (was: M2PA
       version 7  changes)


Franz,

Comments below...

Gaiser Franz wrote:
>
> Hi All,
>
> two additions to the Link State Transition Diagram below:
>
> 1.) Figure 7. M2PA Link State Transition Diagram:
> Missing the handling of receiving LS PO in state ALIGNED READY:
Situation: After Proving the local side is in state ALIGNED READY and sends
LSR; the peer side is in state ALIGNED NOT READY experiencing Processor
Outage and starts sending LS PO. To which state the local M2PA is moving ?
Transition to PROCESSOR OUTAGE ? If so, the state PROCESSOR OUTAGE handles
both local and remote processor outage and then there must be an arrow from
PO to IN SERVICE in case of receiving LS PO Recovered (and there is no
local PO situation).

If M2PA is in ALIGNED READY and receives LS PO, it should go to ALIGNED
NOT READY. When it gets LS Ready from the other end, then it would go to
PROCESSOR OUTAGE.

I updated the diagram accordingly. See below.

>
> 2.) Figure 7. M2PA Link State Transition Diagram:
> Missing the handling of receiving User data in state ALIGNED NOT READY.
Transition to PROCESSOR OUTAGE ? What is to do with this message then (to
be discarded) ?

We are in the middle of discussing this in another thread. It looks like
User Data should not be sent/received when in ALIGNED NOT READY. If it
can happen, it would be an error case that I would prefer not to show in
the diagram.

>
> regards, franz

Updated piece of the diagram:

  |                       |   +-----+                  |
  |                       |   |     | LPO/LPR          |
  |                       V   V     |                  |
  |                 +-----------+   |                  |
  |                 |  PROVING  |---+                  |
  |<----------------|           |--------------------->+
  |      MTP3 Stop  +-----------+     SCTP Comm Error  |
  |   OR T3 Expiry        |  |     OR SCTP Comm Lost   |
  |   OR Receive LS OOS   |  |                         |
  |                       |  +-------------------------|--------+
  |                       |     T4 Expiry              |        |
  |                       |     Send LS PO             |        |
  |                       |                            |        |
  |                       | T4 Expiry                  |        |
  |                       | Send LS Ready              |        |
  |                       |                            |        |
  |                       |                            |        |
  |                       V           SCTP Comm Error  |        V
  |                 +-----------+  OR SCTP Comm Lost   |  +-----------+
  |                 |  ALIGNED  |--------------------->+  |  ALIGNED  |
  +<----------------|   READY   |<----------------------->| NOT READY |
  |      MTP3 Stop  +-----------+     LPO/LPR             +-----------+
  |   OR T1 Expiry        |        OR Receive LS PO/PR       |       |
  |   OR Receive LS OOS   |                                  |   T1  |
  |   OR Receive LS Align | Receive LS Ready      Receive    | Expiry|
  |                       | OR Receive User Data    LS Ready | OR Rcv|
  |                       |                                  |    LS |
  |                       |                                  |  Align|
  |                       |                                  |       |
  |                       V                                  V       |
  |                 +-----------+                      +-----------+ |
  |                 |    IN     |<-------------------->| PROCESSOR | |
  |                 |  SERVICE  |     LPO/LPR          |   OUTAGE  | |
  |                 +-----------+  OR Receive LS PO/PR +-----------+ |
  |                       |                                  |       |
  |                       |                                  |       |



--
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb  5 15:22:35 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA09355
	for <sigtran-archive@odin.ietf.org>; Wed, 5 Feb 2003 15:22:35 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h15KSfW20443
	for sigtran-archive@odin.ietf.org; Wed, 5 Feb 2003 15:28:41 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15KSMJ20426;
	Wed, 5 Feb 2003 15:28:22 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15KRGJ20393
	for <sigtran@optimus.ietf.org>; Wed, 5 Feb 2003 15:27:16 -0500
Received: from web12805.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id PAA09306
	for <sigtran@ietf.org>; Wed, 5 Feb 2003 15:20:39 -0500 (EST)
Message-ID: <20030205202416.67967.qmail@web12805.mail.yahoo.com>
Received: from [62.29.80.165] by web12805.mail.yahoo.com via HTTP; Wed, 05 Feb 2003 12:24:16 PST
Date: Wed, 5 Feb 2003 12:24:16 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] M2PA Flush Buffers
To: sigtran@ietf.org
In-Reply-To: <OFF9E41E19.04BD4D8C-ON85256CC4.004ED0B7@nc.tekelec.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Jeff,
--- jeffrey.craig@tekelec.com wrote:
> 
> Hello Brian,
> 
> I agree with most of what you have said. It appears,
> however, that only
> the "clear buffers" case for exiting processor
> outage is being addressed.
> The more challenging case involves the "resume"
> (recovery prior to MTP3
> T1 expiry and execution of timed-controlled
> changeover).
> 
> The following characteristics of M2PA v7 make
> support of processor outage
> problematic:
> 1) M2PA maintains sequence numbers but does not
> retransmit, unlike MTP2.
> 2) Reception order of user data/ACKs vs. LSPO/LSPOE
> cannot be ensured
> due to the use of multiple SCTP steams.
> 
> Perhaps I am missing something, but it seems to me
> that your statement
> 
> "- any data received by an M2PA peer experiencing
> LPO after an RPO
>   has been sent MUST be discarded and MUST NOT be
> acknowledged."
> 
> guarantees that the M2PA link will go out of service
> for the "resume"
> case of exiting PO. My understanding is that for
> this case, the
> retransmit queue is not cleared and M2PA cannot
> retransmit. No re-synching
> of sequence numbers should be necessary for this
> case.
> 
> Having said this, I think that while M2PA v7 clarity
> can be improved,
> there is no issue with the draft's description of PO
> procedures. The
> draft currently does not require that the receiver
> discard messages
> during PO.
> 
> A solution for these issues is that upon PO onset
> (either local or remote)
> M2PA begins buffering received user data without
> acknowledgement. This
> can be accomplished by not receiving from SCTP
> stream 1, or by buffering
> user data at the M2PA layer. Because the received
> messages are buffered,
> it does not matter about the order of receipt of
> data/ACK and LSPO/LSPOE.
> As discussed on the list previously, M2PA must
> continue to receive and
> process messages on SCTP stream 0 (link status)
> during PO.
[TOLGA] Tom's interpretation of text sounds a bit
better to me, i.e. processing and acknowledging user
data messages even during remote processor outage.
This might help for the case where a message
experiences retransmission and arrives after LS(PO).
If the acknowledgement is delayed till LS(POE) is
received, T7 might expire. Afterall I think it is not
a bad assumption to think that any data message
received between LS(PO) and LS(POE) is in fact either
sent before LS(PO) or after LS(POE). 
IMO it is better if M2PA is more clear about the
procedure to be followed during processor outage as we
can see during this thread, different people might
have different interpretations of the current text.
> 
> In the case of "resume" after PO, M2PA should simply
> unbuffer the
> received user data messages, validate them,
> acknowledge receipt, send them
> to L3, and begin transmitting as normal. No
> retrieval or rerouting occurs,
> and no messages are lost.
> 
> In the case of "clear buffers" after PO, M2PA should
> flush the all
> outbound and inbound queues (possibly including SCTP
> receive buffer),
> and then re-synch the sequence numbers with the
> peer.
> 
> Regards,
> 
> Jeff
> 
> 
> 
> 
> 
> 
> "Brian F. G. Bidulock"
> <bidulock@openss7.org>@ietf.org on 02/05/2003
> 03:03:15 AM
> 
> Please respond to bidulock@openss7.org
> 
> Sent by:    sigtran-admin@ietf.org
> 
> 
> To:    TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
> cc:    sigtran@ietf.org
> Subject:    Re: [Sigtran] M2PA Flush Buffers
> 
> 
> TOLGA,
> 
> TOLGA ASVEREN wrote:
> (Wed, 05 Feb 2003 08:25:56)
> >
> > Tom,
> > --- Tom George <Tom.George@alcatel.com> wrote:
> > > Andrew,
> > >
> > > Thanks for the comments. I will gladly defer to
> > > those with more SS7
> > > smarts, but I think there is no case where the
> > > receiver has to know the
> > > order in which LS and User Data messages were
> sent.
> > > MTP2 itself has
> > > LSSUs jumping to the head of the line when they
> are
> > > transmitted.
> > >
> > > So what do we do with User Data messages that
> are
> > > received after a LS
> > > PO? We can't just copy MTP2, since the same
> > > situation does not occur. In
> > > MTP2, receiving an MSU ends the RPO condition.
> > >
> > > The current M2PA does not say exactly what to
> do,
> > > which I would
> > > interpret (if I may "interpret" my own writing)
> that
> > > the receiver should
> > > go ahead and process the User Data messages. The
> > > presumption would be
> > > that they are valid messages sent before the RPO
> > > occurred.
> > [TOLGA]IMO it is better to mention this in the
> document. I have some more
> > questions though:
> > I assume even after receiving LS(PO) from the
> peer, M2PA continues to
> > acknowledge received messages(?)
> > Similarliy I assume during local processor outage,
> based on received
> > acknowledgements buffers are updated(?) IMO, it
> might be good to clarify
> > this in the text(if this is the case)
> > What about T7 on the peer experiencing local
> processor outage? Will it be
> > stopped or not? If the above is true, I would
> think it shouldn´t be
> > stopped.
> > In 4.1.4 there is a sentence
> > "Alternatively, MTP3 may perform data retrieval as
> part of a changeover
> > procedure."
> > Are the described procedures in that sentence and
> flushing buffers really
> > alternatives to each other?
> 
> To avoid mis-sequencing of messages, neither end
> should send or
> acknowledge messages during LPO.
> 
> The receiver of RPO must not acknowledge any
> messages after the
> receipt of RPO to remain consistent with Q.703 and
> interactions
> with Q.704 (time-controlled changeover procedure). 
> The MTP
> Level 3 suffering LPO must know that when LPO begins
> that
> message transmission has ceased so that the
> time-controlled
> changeover procedure time duration is correctly
> referenced
> against the last message transmitted on the blocked
> link.
> 
> To remain consistent with ANSI T1.111.3, both the
> sender and
> the receiver of RPO should re-synchronize sequence
> numbers
> (acknowlege received messages awaiting ack) and
> cease
> transmission of DATA.
> 
> Again, just as the side experiencing LPO must
> reference its
> time-controlled changeover at Level 3 against the
> last message
> sent on the blocked link, so must the side detecting
> RPO
> reference its time-controlled changeover procedure
> against the
> last message transmitted peer in RPO.
> 
> The peer sending RPO must stop its T7 timer as it
> cannot expect
> an acknowledgement for any messages during RPO (see
> test case
> 
=== message truncated ===


__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb  5 15:51:23 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA10199
	for <sigtran-archive@odin.ietf.org>; Wed, 5 Feb 2003 15:51:23 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h15KvUc22378
	for sigtran-archive@odin.ietf.org; Wed, 5 Feb 2003 15:57:30 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15KvHJ22370;
	Wed, 5 Feb 2003 15:57:17 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15KuwJ22348
	for <sigtran@optimus.ietf.org>; Wed, 5 Feb 2003 15:56:58 -0500
Received: from pt.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA10157
	for <sigtran@ietf.org>; Wed, 5 Feb 2003 15:50:20 -0500 (EST)
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id PAA17246
	for <sigtran@ietf.org>; Wed, 5 Feb 2003 15:50:18 -0500
Message-ID: <3E4179D6.8080502@pt.com>
Date: Wed, 05 Feb 2003 15:53:42 -0500
From: Andrew Booth <abooth@pt.com>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.1) Gecko/20020826
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: sigtran@ietf.org
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] M2PA 7 questions
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Hi All,

I was looking over M2PA draft 7 and I have a few questions.


3. M2PA Link State Control
SCTP Restart is not mentioned in the state transistion diagram for SCTP 
associations in Figure 8.  Is this an oversight, or intentional?  There 
don't seem to be procedures for SCTP Restart mentioned elsewhere either.


Timer T7
========
In section 4.2.1 it sounds like a per message timer, or even an interval 
e.g.
   Timer T7 provides an indication of excessive delay of
   acknowledgement. If the following conditions are true:

   (a) There is at least one message in the M2PA retransmit buffer.

   (b) The remote M2PA is not in a Busy condition (i.e., the local
       timer T6 is not running).

   (c) There is a message in the M2PA retransmit buffer that has not
       received an acknowledgement in the span of T7 since its last
       transmission.

   then M2PA SHOULD take the link out of service.

The interpretation of T7 as an interval (or just a condition on 
messages) is reinforced by the fact there are no references to starting T7.

On the other hand in section 4.1.4 there is a reference to stopping T7 
during processor outage.  In 4.1.5 there is the comment
   "The peer M2PA SHALL NOT fail the link due to expiration of timer T7
   excessive delay of acknowledgement (see section 4.2.1 Sending and
   receiving messages) while its T6 timer is running, i.e., while the
   other end is Busy."
There is no comment on how to handle T7 after Busy clears.  Is it 
appropriate to immediately check the conditions from 4.2.1?

Timer T6
========
T6 is stopped during LPO, there is no comment to restart it at the end.  
Conclude that LPO ends the Busy state.  Is this correct?

4.1.5
In MTP2 a receiver that is experiencing congestion stops acknowledging 
messages until Busy clears.  In M2PA neither the sender nor receiver 
seems to do anything differently than when one end is not congested 
(other than start T6 and refuse to fail the link because of T7).  Is 
this appropriate?


Thanks for any comments,
Andrew

-- 
Andrew Booth, Software Developer
Signaling Systems Group, Performance Technologies
613-237-4344

Visit www.pt.com/mailing.html to subscribe to our Packets & Signals eLetter.


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb  5 16:26:42 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA11446
	for <sigtran-archive@odin.ietf.org>; Wed, 5 Feb 2003 16:26:42 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h15LWnL24402
	for sigtran-archive@odin.ietf.org; Wed, 5 Feb 2003 16:32:49 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15LWSJ24386;
	Wed, 5 Feb 2003 16:32:29 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15LV0J24323
	for <sigtran@optimus.ietf.org>; Wed, 5 Feb 2003 16:31:00 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA11316
	for <sigtran@ietf.org>; Wed, 5 Feb 2003 16:24:21 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:ZK56h/AQmPaA/5LsnikY+aM51zwLXZ85@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h15LRwD04962;
	Wed, 5 Feb 2003 14:27:58 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h15LRw529281;
	Wed, 5 Feb 2003 14:27:58 -0700
Date: Wed, 5 Feb 2003 14:27:58 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: jeffrey.craig@tekelec.com
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA Flush Buffers
Message-ID: <20030205142758.B27731@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: jeffrey.craig@tekelec.com, sigtran@ietf.org
References: <OFF9E41E19.04BD4D8C-ON85256CC4.004ED0B7@nc.tekelec.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OFF9E41E19.04BD4D8C-ON85256CC4.004ED0B7@nc.tekelec.com>; from jeffrey.craig@tekelec.com on Wed, Feb 05, 2003 at 10:50:21AM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: 8bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

jeffrey.craig,

jeffrey.craig@tekelec.com wrote:                                             (Wed, 05 Feb 2003 10:50:21)
> 
> Hello Brian,
> 
> I agree with most of what you have said. It appears, however, that only
> the "clear buffers" case for exiting processor outage is being addressed.
> The more challenging case involves the "resume" (recovery prior to MTP3
> T1 expiry and execution of timed-controlled changeover).
> 
> The following characteristics of M2PA v7 make support of processor outage
> problematic:
> 1) M2PA maintains sequence numbers but does not retransmit, unlike MTP2.
> 2) Reception order of user data/ACKs vs. LSPO/LSPOE cannot be ensured
> due to the use of multiple SCTP steams.
> 
> Perhaps I am missing something, but it seems to me that your statement
> 
> "- any data received by an M2PA peer experiencing LPO after an RPO
>   has been sent MUST be discarded and MUST NOT be acknowledged."
> 
> guarantees that the M2PA link will go out of service for the "resume"
> case of exiting PO. My understanding is that for this case, the
> retransmit queue is not cleared and M2PA cannot retransmit. No re-synching
> of sequence numbers should be necessary for this case.
> 
> Having said this, I think that while M2PA v7 clarity can be improved,
> there is no issue with the draft's description of PO procedures. The
> draft currently does not require that the receiver discard messages
> during PO.
> 
> A solution for these issues is that upon PO onset (either local or remote)
> M2PA begins buffering received user data without acknowledgement. This

Yes, of course.  I'm sorry, I don't know what I was thinking saying "dicard."
Yes, buffer and do not acknowledge.

 To summarize:
 
 - any data received by an M2PA peer experiencing LPO after an RPO
   has been sent MUST be buffered and MUST NOT be acknowledged.
 
 - the M2PA peer experiencing LPO MUST NOT send any data messages
   between sending PROCESSOR-OUTAGE and PROCESSOR-OUTAGE-ENDED
 
 - the M2PA peer sending PROCESSOR-OUTAGE SHOULD resynchornize
   sequence numbers by acknowledging all outstanding received
   data.
 
 - any data received by an M2PA after receiving PROCESSOR-OUTAGE
   and before receiving PROCESSOR-OUTAGE-ENDED from the peer
   MUST be buffered and MUST NOT be acknowledged.
 
 - the M2PA peer receiving PROCESSOR-OUTAGE MUST NOT send any
   data until receiving PROCESSOR-OUTAGE-ENDED.
 
 - the M2PA peer receiving PROCESSOR-OUTAGE SHOULD resynchronize
   sequence numbers by acknowledging all outstanding data
   received before receiving PROCESSOR-OUTAGE.

> can be accomplished by not receiving from SCTP stream 1, or by buffering
> user data at the M2PA layer. Because the received messages are buffered,
> it does not matter about the order of receipt of data/ACK and LSPO/LSPOE.
> As discussed on the list previously, M2PA must continue to receive and
> process messages on SCTP stream 0 (link status) during PO.
> 
> In the case of "resume" after PO, M2PA should simply unbuffer the
> received user data messages, validate them, acknowledge receipt, send them
> to L3, and begin transmitting as normal. No retrieval or rerouting occurs,
> and no messages are lost.
> 
> In the case of "clear buffers" after PO, M2PA should flush the all
> outbound and inbound queues (possibly including SCTP receive buffer),
> and then re-synch the sequence numbers with the peer.
> 
> Regards,
> 
> Jeff
> 
> 
> 
> 
> 
> 
> "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 02/05/2003
> 03:03:15 AM
> 
> Please respond to bidulock@openss7.org
> 
> Sent by:    sigtran-admin@ietf.org
> 
> 
> To:    TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
> cc:    sigtran@ietf.org
> Subject:    Re: [Sigtran] M2PA Flush Buffers
> 
> 
> TOLGA,
> 
> TOLGA ASVEREN wrote:
> (Wed, 05 Feb 2003 08:25:56)
> >
> > Tom,
> > --- Tom George <Tom.George@alcatel.com> wrote:
> > > Andrew,
> > >
> > > Thanks for the comments. I will gladly defer to
> > > those with more SS7
> > > smarts, but I think there is no case where the
> > > receiver has to know the
> > > order in which LS and User Data messages were sent.
> > > MTP2 itself has
> > > LSSUs jumping to the head of the line when they are
> > > transmitted.
> > >
> > > So what do we do with User Data messages that are
> > > received after a LS
> > > PO? We can't just copy MTP2, since the same
> > > situation does not occur. In
> > > MTP2, receiving an MSU ends the RPO condition.
> > >
> > > The current M2PA does not say exactly what to do,
> > > which I would
> > > interpret (if I may "interpret" my own writing) that
> > > the receiver should
> > > go ahead and process the User Data messages. The
> > > presumption would be
> > > that they are valid messages sent before the RPO
> > > occurred.
> > [TOLGA]IMO it is better to mention this in the document. I have some more
> > questions though:
> > I assume even after receiving LS(PO) from the peer, M2PA continues to
> > acknowledge received messages(?)
> > Similarliy I assume during local processor outage, based on received
> > acknowledgements buffers are updated(?) IMO, it might be good to clarify
> > this in the text(if this is the case)
> > What about T7 on the peer experiencing local processor outage? Will it be
> > stopped or not? If the above is true, I would think it shouldn´t be
> > stopped.
> > In 4.1.4 there is a sentence
> > "Alternatively, MTP3 may perform data retrieval as part of a changeover
> > procedure."
> > Are the described procedures in that sentence and flushing buffers really
> > alternatives to each other?
> 
> To avoid mis-sequencing of messages, neither end should send or
> acknowledge messages during LPO.
> 
> The receiver of RPO must not acknowledge any messages after the
> receipt of RPO to remain consistent with Q.703 and interactions
> with Q.704 (time-controlled changeover procedure).  The MTP
> Level 3 suffering LPO must know that when LPO begins that
> message transmission has ceased so that the time-controlled
> changeover procedure time duration is correctly referenced
> against the last message transmitted on the blocked link.
> 
> To remain consistent with ANSI T1.111.3, both the sender and
> the receiver of RPO should re-synchronize sequence numbers
> (acknowlege received messages awaiting ack) and cease
> transmission of DATA.
> 
> Again, just as the side experiencing LPO must reference its
> time-controlled changeover at Level 3 against the last message
> sent on the blocked link, so must the side detecting RPO
> reference its time-controlled changeover procedure against the
> last message transmitted peer in RPO.
> 
> The peer sending RPO must stop its T7 timer as it cannot expect
> an acknowledgement for any messages during RPO (see test case
> 3.4.1 in draft-bidulock-sigtran-m2pa-test-02.txt).  Both sides
> also stop T6 and cease any busy condition (sends BUSY-ENDED if
> BUSY was sent previously).
> 
> To summarize:
> 
> - any data received by an M2PA peer experiencing LPO after an RPO
>   has been sent MUST be discarded and MUST NOT be acknowledged.
> 
> - the M2PA peer experiencing LPO MUST NOT send any data messages
>   between sending PROCESSOR-OUTAGE and PROCESSOR-OUTAGE-ENDED
> 
> - the M2PA peer sending PROCESSOR-OUTAGE SHOULD resynchornize
>   sequence numbers by acknowledging all outstanding received
>   data.
> 
> - any data received by an M2PA after receiving PROCESSOR-OUTAGE
>   and before receiving PROCESSOR-OUTAGE-ENDED from the peer
>   MUST be discarded and MUST NOT be acknowledged.
> 
> - the M2PA peer receiving PROCESSOR-OUTAGE MUST NOT send any
>   data until receiving PROCESSOR-OUTAGE-ENDED.
> 
> - the M2PA peer receiving PROCESSOR-OUTAGE SHOULD resynchronize
>   sequence numbers by acknowledging all outstanding data
>   received before receiving PROCESSOR-OUTAGE.
> 
> This is completely consistent with Q.703/T1.111.3 and has the
> benefit that harmful interactions with MTP Level 3 will not be
> caused by introducing unexpected behavior (i.e. sending messages
> for seconds after MTP Level 3 expects that the link has blocked).
> 
> --brian
> 
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 
> 
> 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb  5 16:38:20 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA11913
	for <sigtran-archive@odin.ietf.org>; Wed, 5 Feb 2003 16:38:20 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h15LiRF25475
	for sigtran-archive@odin.ietf.org; Wed, 5 Feb 2003 16:44:27 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15LiBJ25463;
	Wed, 5 Feb 2003 16:44:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15LhNJ25416
	for <sigtran@optimus.ietf.org>; Wed, 5 Feb 2003 16:43:23 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA11866
	for <sigtran@ietf.org>; Wed, 5 Feb 2003 16:36:44 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:Eh8h1+E4p4Xk8q0aAAPSwnhZlY4Ey+ty@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h15LeMD05217;
	Wed, 5 Feb 2003 14:40:22 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h15LeM830882;
	Wed, 5 Feb 2003 14:40:22 -0700
Date: Wed, 5 Feb 2003 14:40:22 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tolga Asveren <tolga_asveren@yahoo.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA Flush Buffers
Message-ID: <20030205144022.C27731@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030205010315.A12910@openss7.org> <20030205103720.32742.qmail@web12803.mail.yahoo.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20030205103720.32742.qmail@web12803.mail.yahoo.com>; from tolga_asveren@yahoo.com on Wed, Feb 05, 2003 at 02:37:20AM -0800
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Tolga,

Tolga Asveren wrote:                                                                 (Wed, 05 Feb 2003 02:37:20)
> Brian,
> > To summarize:
> > 
> > - any data received by an M2PA peer experiencing LPO
> > after an RPO
> >   has been sent MUST be discarded and MUST NOT be
> > acknowledged.
> > 
> > - the M2PA peer experiencing LPO MUST NOT send any
> > data messages
> >   between sending PROCESSOR-OUTAGE and
> > PROCESSOR-OUTAGE-ENDED
> > 
> > - the M2PA peer sending PROCESSOR-OUTAGE SHOULD
> > resynchornize
> >   sequence numbers by acknowledging all outstanding
> > received
> >   data.
> > 
> > - any data received by an M2PA after receiving
> > PROCESSOR-OUTAGE
> >   and before receiving PROCESSOR-OUTAGE-ENDED from
> > the peer
> >   MUST be discarded and MUST NOT be acknowledged.
> [TOLGA] In that case, what happens with messages sent
> before LS(PO)and arrive after LS(PO)? Similarly what
> will happen with messages, which are sent after LS(PE)
> but arrive before it?

Will Jeff's correction ("buffer rather than discard",
doh!):

They are treated according to order of arrival not
transmission.  Messages which were actually sent before
LS PO will not be acknowledged until after LS PE.
Messages sent after LS PE that arrive bfore LS PE will
not be acknowledged until the LS PE is actulally
received.  That's what you get by putting LS PO/PE on
stream "0".  I would prefer that the implemnetation be
allowed to decide which stream to place LS on.


> > 
> > - the M2PA peer receiving PROCESSOR-OUTAGE MUST NOT
> > send any
> >   data until receiving PROCESSOR-OUTAGE-ENDED.
> > 
> > - the M2PA peer receiving PROCESSOR-OUTAGE SHOULD
> > resynchronize
> >   sequence numbers by acknowledging all outstanding
> > data
> >   received before receiving PROCESSOR-OUTAGE.
> > 
> > This is completely consistent with Q.703/T1.111.3
> > and has the
> > benefit that harmful interactions with MTP Level 3
> > will not be
> > caused by introducing unexpected behavior (i.e.
> > sending messages
> > for seconds after MTP Level 3 expects that the link
> > has blocked).
> 
> 
> __________________________________________________
> Do you Yahoo!?
> Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
> http://mailplus.yahoo.com
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb  5 18:22:25 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA15160
	for <sigtran-archive@odin.ietf.org>; Wed, 5 Feb 2003 18:22:25 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h15NSaH05670
	for sigtran-archive@odin.ietf.org; Wed, 5 Feb 2003 18:28:36 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15NSHp05654;
	Wed, 5 Feb 2003 18:28:18 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h15NQLp05609
	for <sigtran@optimus.ietf.org>; Wed, 5 Feb 2003 18:26:21 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA15070
	for <sigtran@ietf.org>; Wed, 5 Feb 2003 18:19:39 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h15NNHqe027859;
	Wed, 5 Feb 2003 17:23:17 -0600 (CST)
Message-ID: <3E419CE5.99B40D2@alcatel.com>
Date: Wed, 05 Feb 2003 17:23:17 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA 7 questions
References: <3E4179D6.8080502@pt.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Andrew,

Thanks for the comments. Replies below...

Andrew Booth wrote:
> 
> Hi All,
> 
> I was looking over M2PA draft 7 and I have a few questions.
> 
> 3. M2PA Link State Control
> SCTP Restart is not mentioned in the state transistion diagram for SCTP
> associations in Figure 8.  Is this an oversight, or intentional?  There
> don't seem to be procedures for SCTP Restart mentioned elsewhere either.

I don't see how SCTP restart would affect M2PA. 


> 
> Timer T7
> ========
> In section 4.2.1 it sounds like a per message timer, or even an interval

It is intended to be a per-message timer. 


> e.g.
>    Timer T7 provides an indication of excessive delay of
>    acknowledgement. If the following conditions are true:
> 
>    (a) There is at least one message in the M2PA retransmit buffer.
> 
>    (b) The remote M2PA is not in a Busy condition (i.e., the local
>        timer T6 is not running).
> 
>    (c) There is a message in the M2PA retransmit buffer that has not
>        received an acknowledgement in the span of T7 since its last
>        transmission.
> 
>    then M2PA SHOULD take the link out of service.
> 
> The interpretation of T7 as an interval (or just a condition on
> messages) is reinforced by the fact there are no references to starting T7.
> 
> On the other hand in section 4.1.4 there is a reference to stopping T7
> during processor outage.  In 4.1.5 there is the comment
>    "The peer M2PA SHALL NOT fail the link due to expiration of timer T7
>    excessive delay of acknowledgement (see section 4.2.1 Sending and
>    receiving messages) while its T6 timer is running, i.e., while the
>    other end is Busy."
> There is no comment on how to handle T7 after Busy clears.  Is it
> appropriate to immediately check the conditions from 4.2.1?

I see the problem. After Busy clears, you may have a some unacked
messages that have been waiting for longer than T7. It doesn't seem like
you would want to fail the link for that. So it looks like T7 should be
restarted when Busy clears.


> 
> Timer T6
> ========
> T6 is stopped during LPO, there is no comment to restart it at the end.
> Conclude that LPO ends the Busy state.  Is this correct?

T6 is timing the other end's Busy condition, so it would seem that a
local PO should not end the Busy state. So if there is an LPR, I would
think M2PA should restart the T6 timer if the other end is still Busy.


> 
> 4.1.5
> In MTP2 a receiver that is experiencing congestion stops acknowledging
> messages until Busy clears.  In M2PA neither the sender nor receiver
> seems to do anything differently than when one end is not congested
> (other than start T6 and refuse to fail the link because of T7).  Is
> this appropriate?

I think there was a discussion about this a long time ago. SCTP has its
own congestion procedures. So the only thing I was trying to do with
"M2PA flow control" was make sure SCTP was able to fix the congestion
problem within a reasonable (T6) amount of time. 



> 
> Thanks for any comments,
> Andrew
> 
> --
> Andrew Booth, Software Developer
> Signaling Systems Group, Performance Technologies
> 613-237-4344
  [...]

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb  5 20:36:00 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA17977
	for <sigtran-archive@odin.ietf.org>; Wed, 5 Feb 2003 20:35:56 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h161g9212554
	for sigtran-archive@odin.ietf.org; Wed, 5 Feb 2003 20:42:09 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h161fvp12545;
	Wed, 5 Feb 2003 20:41:57 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h161dMp12447
	for <sigtran@optimus.ietf.org>; Wed, 5 Feb 2003 20:39:22 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA17899
	for <sigtran@ietf.org>; Wed, 5 Feb 2003 20:32:33 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:7uvsd82Aj9Bvg2tbl1GQY7X7RA2xP8nR@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h161aCD08543;
	Wed, 5 Feb 2003 18:36:12 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h161aBl29780;
	Wed, 5 Feb 2003 18:36:11 -0700
Date: Wed, 5 Feb 2003 18:36:11 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tom George <Tom.George@alcatel.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA 7 questions
Message-ID: <20030205183611.A28979@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tom George <Tom.George@alcatel.com>, sigtran@ietf.org
References: <3E4179D6.8080502@pt.com> <3E419CE5.99B40D2@alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E419CE5.99B40D2@alcatel.com>; from Tom.George@alcatel.com on Wed, Feb 05, 2003 at 05:23:17PM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Tom,

Tom George wrote:                                                                      (Wed, 05 Feb 2003 17:23:17)
> Andrew,
> 
> Thanks for the comments. Replies below...
> 
> Andrew Booth wrote:
> > 
> > Hi All,
> > 
> > I was looking over M2PA draft 7 and I have a few questions.
> > 
> > 3. M2PA Link State Control
> > SCTP Restart is not mentioned in the state transistion diagram for SCTP
> > associations in Figure 8.  Is this an oversight, or intentional?  There
> > don't seem to be procedures for SCTP Restart mentioned elsewhere either.
> 
> I don't see how SCTP restart would affect M2PA. 
> 
> 
> > 
> > Timer T7
> > ========
> > In section 4.2.1 it sounds like a per message timer, or even an interval
> 
> It is intended to be a per-message timer. 
> 
> 
> > e.g.
> >    Timer T7 provides an indication of excessive delay of
> >    acknowledgement. If the following conditions are true:
> > 
> >    (a) There is at least one message in the M2PA retransmit buffer.
> > 
> >    (b) The remote M2PA is not in a Busy condition (i.e., the local
> >        timer T6 is not running).
> > 
> >    (c) There is a message in the M2PA retransmit buffer that has not
> >        received an acknowledgement in the span of T7 since its last
> >        transmission.
> > 
> >    then M2PA SHOULD take the link out of service.
> > 
> > The interpretation of T7 as an interval (or just a condition on
> > messages) is reinforced by the fact there are no references to starting T7.
> > 
> > On the other hand in section 4.1.4 there is a reference to stopping T7
> > during processor outage.  In 4.1.5 there is the comment
> >    "The peer M2PA SHALL NOT fail the link due to expiration of timer T7
> >    excessive delay of acknowledgement (see section 4.2.1 Sending and
> >    receiving messages) while its T6 timer is running, i.e., while the
> >    other end is Busy."
> > There is no comment on how to handle T7 after Busy clears.  Is it
> > appropriate to immediately check the conditions from 4.2.1?
> 
> I see the problem. After Busy clears, you may have a some unacked
> messages that have been waiting for longer than T7. It doesn't seem like
> you would want to fail the link for that. So it looks like T7 should be
> restarted when Busy clears.

T7 continues to run while busy (but not while processor outage).  This
is the same as Q.703.  T7 expiry just does not fail the link while busy,
T6 expiry does.

> 
> 
> > 
> > Timer T6
> > ========
> > T6 is stopped during LPO, there is no comment to restart it at the end.
> > Conclude that LPO ends the Busy state.  Is this correct?
> 
> T6 is timing the other end's Busy condition, so it would seem that a
> local PO should not end the Busy state. So if there is an LPR, I would
> think M2PA should restart the T6 timer if the other end is still Busy.

No. Busy is cleared during processor outage because neither side should
be sending anything during processor outage and the busy condition can
no longer exist.  When recovering from processor outage, the busy condition
cannot exist because messages were not being sent.  Both T6 and T7 are
stopped and the busy condition is cleared when entering processor outage
state.  This is consistent with Q.703.

> 
> 
> > 
> > 4.1.5
> > In MTP2 a receiver that is experiencing congestion stops acknowledging
> > messages until Busy clears.  In M2PA neither the sender nor receiver
> > seems to do anything differently than when one end is not congested
> > (other than start T6 and refuse to fail the link because of T7).  Is
> > this appropriate?
> 
> I think there was a discussion about this a long time ago. SCTP has its
> own congestion procedures. So the only thing I was trying to do with
> "M2PA flow control" was make sure SCTP was able to fix the congestion
> problem within a reasonable (T6) amount of time. 

Neither end stops sending or acknowledging messages during the busy
condition.  The busy end is just given more time (T6 instead of T7) to
acknowledge.  This is consistent with Q.703.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb  5 23:16:26 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA21786
	for <sigtran-archive@odin.ietf.org>; Wed, 5 Feb 2003 23:16:26 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h164Mg120007
	for sigtran-archive@odin.ietf.org; Wed, 5 Feb 2003 23:22:42 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h164MNp19990;
	Wed, 5 Feb 2003 23:22:23 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h164Lop19953
	for <sigtran@optimus.ietf.org>; Wed, 5 Feb 2003 23:21:51 -0500
Received: from pt.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA21761
	for <sigtran@ietf.org>; Wed, 5 Feb 2003 23:15:03 -0500 (EST)
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id XAA04096;
	Wed, 5 Feb 2003 23:14:50 -0500
Message-ID: <3E41E209.9010104@pt.com>
Date: Wed, 05 Feb 2003 23:18:17 -0500
From: Andrew Booth <abooth@pt.com>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.1) Gecko/20020826
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: bidulock@openss7.org
CC: Tom George <Tom.George@alcatel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M2PA 7 questions
References: <3E4179D6.8080502@pt.com> <3E419CE5.99B40D2@alcatel.com> <20030205183611.A28979@openss7.org>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Tom, Brian,

Thanks for the comments.  More questions below.

Brian F. G. Bidulock wrote:

>Tom,
>
>Tom George wrote:                                                                      (Wed, 05 Feb 2003 17:23:17)
>  
>
>>Andrew,
>>
>>Thanks for the comments. Replies below...
>>
>>Andrew Booth wrote:
>>    
>>
>>>Hi All,
>>>
>>>I was looking over M2PA draft 7 and I have a few questions.
>>>
>>>3. M2PA Link State Control
>>>SCTP Restart is not mentioned in the state transistion diagram for SCTP
>>>associations in Figure 8.  Is this an oversight, or intentional?  There
>>>don't seem to be procedures for SCTP Restart mentioned elsewhere either.
>>>      
>>>
>>I don't see how SCTP restart would affect M2PA. 
>>
Should SCTP RESTART be a transition event to OUT OF SERVICE in Figure 7?
I think the following scenario is possible.

If M2PA peer A detects the loss of association before M2PA peer B and A 
immediately reestablishes the association, then the SCTP implementation 
at B could see this as an association restart instead of a lost 
association.  In this case (RFC 2960, section 5.2.4):

    -  Any SCTP DATA Chunks MAY be retained (this is an implementation
         specific option).

      -  A notification of RESTART SHOULD be sent to the ULP instead of
         a "COMMUNICATION LOST" notification.
     
     All the congestion control parameters (e.g., cwnd, ssthresh)
      related to this peer MUST be reset to their initial values (see
      Section 6.2.1).

      After this the endpoint shall enter the ESTABLISHED state.


Assuming that a RESTART indication is a transition event to OOS:
Note that if the SCTP implementation does retain data chunks then there 
could still be a fair amount of stale data to retransmitted.  I don't 
think this stale data should cause big problems, but it could slow down 
the alignment after a restart.

>>
>>    
>>
>>>Timer T7
>>>========
>>>In section 4.2.1 it sounds like a per message timer, or even an interval
>>>      
>>>
>>It is intended to be a per-message timer. 
>>
>>
>>    
>>
>>>e.g.
>>>   Timer T7 provides an indication of excessive delay of
>>>   acknowledgement. If the following conditions are true:
>>>
>>>   (a) There is at least one message in the M2PA retransmit buffer.
>>>
>>>   (b) The remote M2PA is not in a Busy condition (i.e., the local
>>>       timer T6 is not running).
>>>
>>>   (c) There is a message in the M2PA retransmit buffer that has not
>>>       received an acknowledgement in the span of T7 since its last
>>>       transmission.
>>>
>>>   then M2PA SHOULD take the link out of service.
>>>
>>>The interpretation of T7 as an interval (or just a condition on
>>>messages) is reinforced by the fact there are no references to starting T7.
>>>
>>>On the other hand in section 4.1.4 there is a reference to stopping T7
>>>during processor outage.  In 4.1.5 there is the comment
>>>   "The peer M2PA SHALL NOT fail the link due to expiration of timer T7
>>>   excessive delay of acknowledgement (see section 4.2.1 Sending and
>>>   receiving messages) while its T6 timer is running, i.e., while the
>>>   other end is Busy."
>>>There is no comment on how to handle T7 after Busy clears.  Is it
>>>appropriate to immediately check the conditions from 4.2.1?
>>>      
>>>
>>I see the problem. After Busy clears, you may have a some unacked
>>messages that have been waiting for longer than T7. It doesn't seem like
>>you would want to fail the link for that. So it looks like T7 should be
>>restarted when Busy clears.
>>    
>>
>
>T7 continues to run while busy (but not while processor outage).  This
>is the same as Q.703.  T7 expiry just does not fail the link while busy,
>T6 expiry does.
>  
>
Is it then appropriate to check the condition in 4.2.1 when Busy abates, 
to see if T7 has expired for any unacknowledged messages?  If so, is it 
worth stating explicitly?

Andrew

>  
>
>>    
>>
>>>Timer T6
>>>========
>>>T6 is stopped during LPO, there is no comment to restart it at the end.
>>>Conclude that LPO ends the Busy state.  Is this correct?
>>>      
>>>
>>T6 is timing the other end's Busy condition, so it would seem that a
>>local PO should not end the Busy state. So if there is an LPR, I would
>>think M2PA should restart the T6 timer if the other end is still Busy.
>>    
>>
>
>No. Busy is cleared during processor outage because neither side should
>be sending anything during processor outage and the busy condition can
>no longer exist.  When recovering from processor outage, the busy condition
>cannot exist because messages were not being sent.  Both T6 and T7 are
>stopped and the busy condition is cleared when entering processor outage
>state.  This is consistent with Q.703.
>
>  
>
>>    
>>
>>>4.1.5
>>>In MTP2 a receiver that is experiencing congestion stops acknowledging
>>>messages until Busy clears.  In M2PA neither the sender nor receiver
>>>seems to do anything differently than when one end is not congested
>>>(other than start T6 and refuse to fail the link because of T7).  Is
>>>this appropriate?
>>>      
>>>
>>I think there was a discussion about this a long time ago. SCTP has its
>>own congestion procedures. So the only thing I was trying to do with
>>"M2PA flow control" was make sure SCTP was able to fix the congestion
>>problem within a reasonable (T6) amount of time. 
>>    
>>
>
>Neither end stops sending or acknowledging messages during the busy
>condition.  The busy end is just given more time (T6 instead of T7) to
>acknowledge.  This is consistent with Q.703.
>
>--brian
>
>  
>

-- 
Andrew Booth, Software Developer
Signaling Systems Group, Performance Technologies
613-237-4344

Visit www.pt.com/mailing.html to subscribe to our Packets & Signals eLetter.


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 00:56:29 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA24400
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 00:56:29 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1662mo24479
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 01:02:48 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1662Zp24472;
	Thu, 6 Feb 2003 01:02:36 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1661Dp24433
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 01:01:13 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA24360
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 00:54:23 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h165w2qe000401
	for <sigtran@ietf.org>; Wed, 5 Feb 2003 23:58:02 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h165vAM04235
	for <sigtran@ietf.org>; Wed, 5 Feb 2003 23:57:10 -0600 (CST)
Date: Wed, 5 Feb 2003 23:57:10 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: sigtran@ietf.org
Message-ID: <Pine.SOL.4.10.10302052218190.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Sigtran] M3UA - ASP database equalization
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hi all,

Currently RFC 3332 says that SG sends SSNM messages to all 'concerned
ASPs'(means only ACTIVE ASPs).

+----+           +---+           +-----+
|SEP | ----------|SG |-----------| ASP1|
+----+           +---+           +-----+
                   |             
                   |             +-----+
                   +-------------| ASP2|
                                 +-----+

(ASP1 is active and ASP2 is incative - override mode)

When SEP is getting restarted ,SG sends DUNA to ASP1.
But SEP status is not changed in ASP2.There are two ways to solve this
problem

1. ASP1 can equalize its database with ASP2
or
2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)

IMO, SG can send SSNM messages to ASP (either ASP is in active or inact
state)

Any comments?

-----

section 5.4.2.1

...
      - Pass the MTP-TRANSFER request primitive to the MTP3 of the
        concerned Network Appearance.

It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER request is
sent to MTP3.

But there is no primitive availabe between M3UA and MTP3.

I think it should be corrected.

----
Presently ASP is unaware of SG's MTP3 restart, 
ASP receives only DUNA for all SS7 point codes which can be
reachable via that SG.But in case of peer to peer protocol
SG and AS point codes are different, So why dont we send
DUNA with SG's point code to ASP? So that just by getting
single DUNA(with SG's PC) ASP will not route any message via that SG.
This can avoid sending multiple DUNAs to ASP.


Any comments?

-----
section 1.4.1

...
   Where a particular AS can be reached via more than one SGP, the
   corresponding Routing Keys in the SGPs should be identical.  (Note:
   It is possible for the SGP Routing Key configuration data to be
   temporarily out-of-sync during configuration updates).
 
                              +--------+
                              |        |
                 +------------+  SG 1  +--------------+
     +-------+   |  SS7 links | "STP"  |  IP network  |     ----
     |  SEP  +---+            +--------+              +---/      \
     |   or  |                    |*                      | ASPs  |
     |  STP  +---+            +--------+              +---\      /
     +-------+   |            |        |              |     ----
                 +------------+  SG 2  +--------------+
                              | "STP"  |
                              +--------+
 
                    Figure 1  Example with mated SGs

Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are located
geographically in different places.The ideas is when SG1 goes down in
CITY1(due to earthquake...) , all the traffice still can be handled by SG2
which is in CITY2.

SGP2 is mate for SGP1, On failure of SGP1, all the messages will be
routed from SGP1 to SGP2.

Does all these things mean to say that SGP1's mate 'should be' present in
SG2 and not in SG1 itself ?
 

Comments are welcome.

--Samuel




  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 00:58:59 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA24480
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 00:58:59 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1665Ih24579
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 01:05:18 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16658p24571;
	Thu, 6 Feb 2003 01:05:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1664Up24523
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 01:04:30 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA24451
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 00:57:40 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:29xRJ0appnskL6XFv8iD+/gcC6iQWNSb@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1661ID11921;
	Wed, 5 Feb 2003 23:01:18 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1661IY29324;
	Wed, 5 Feb 2003 23:01:18 -0700
Date: Wed, 5 Feb 2003 23:01:18 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Andrew Booth <abooth@pt.com>
Cc: Tom George <Tom.George@alcatel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M2PA 7 questions
Message-ID: <20030205230118.A26810@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Andrew Booth <abooth@pt.com>,
	Tom George <Tom.George@alcatel.com>, sigtran@ietf.org
References: <3E4179D6.8080502@pt.com> <3E419CE5.99B40D2@alcatel.com> <20030205183611.A28979@openss7.org> <3E41E209.9010104@pt.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E41E209.9010104@pt.com>; from abooth@pt.com on Wed, Feb 05, 2003 at 11:18:17PM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Andrew,

Andrew Booth wrote:                                                                 (Wed, 05 Feb 2003 23:18:17)
> Tom, Brian,
> 
> Thanks for the comments.  More questions below.
> 
> Brian F. G. Bidulock wrote:
> 
> >Tom,
> >
> >Tom George wrote:                                                                      (Wed, 05 Feb 2003 17:23:17)
> >  
> >
> >>Andrew,
> >>
> >>Thanks for the comments. Replies below...
> >>
> >>Andrew Booth wrote:
> >>    
> >>
> >>>Hi All,
> >>>
> >>>I was looking over M2PA draft 7 and I have a few questions.
> >>>
> >>>3. M2PA Link State Control
> >>>SCTP Restart is not mentioned in the state transistion diagram for SCTP
> >>>associations in Figure 8.  Is this an oversight, or intentional?  There
> >>>don't seem to be procedures for SCTP Restart mentioned elsewhere either.
> >>>      
> >>>
> >>I don't see how SCTP restart would affect M2PA. 
> >>
> Should SCTP RESTART be a transition event to OUT OF SERVICE in Figure 7?
> I think the following scenario is possible.
> 
> If M2PA peer A detects the loss of association before M2PA peer B and A 
> immediately reestablishes the association, then the SCTP implementation 
> at B could see this as an association restart instead of a lost 
> association.  In this case (RFC 2960, section 5.2.4):
> 
>     -  Any SCTP DATA Chunks MAY be retained (this is an implementation
>          specific option).
> 
>       -  A notification of RESTART SHOULD be sent to the ULP instead of
>          a "COMMUNICATION LOST" notification.
>      
>      All the congestion control parameters (e.g., cwnd, ssthresh)
>       related to this peer MUST be reset to their initial values (see
>       Section 6.2.1).
> 
>       After this the endpoint shall enter the ESTABLISHED state.
> 
> 
> Assuming that a RESTART indication is a transition event to OOS:
> Note that if the SCTP implementation does retain data chunks then there 
> could still be a fair amount of stale data to retransmitted.  I don't 
> think this stale data should cause big problems, but it could slow down 
> the alignment after a restart.

The peer that gets the restart indication doesn't have to do anthing:
Look at both sides:

The side getting the restart indication, if it supports the optional
procedure of RFC 2960, it will allow the M2PA to continue and deliver
a restart indication; otherwise, it will typically disconnect the ULP
and provide another init indication.  The latter fall into the existing
procedures for loss of an association and starting an association.

The side restarting either is capable of continuing where it left off
(unlikely, it has closed the association and is reopening it: there is
no reason to do so unless the M2PA has crashed and restarted).  If the
side restarting is capable of continuing, it continues where it left off.
If it is not capable of continuiing, it follows the procedures for starting
a new association and sends alignment.

(IMO the first thing that an M2PA peer should do when forming the
association is send LS OOS.  This is reflected in in the test spec
draft-bidulock-sigtran-m2pa-test-02.txt.  That way, if MTP3 has not
issued the Start signal, the other end will know the state of the link
immediately.)

So, no procedures are required outside of the normal procedures.

I agree with Tom: I don't see how SCTP restart would affect M2PA.

> 
> >>
> >>    
> >>
> >>>Timer T7
> >>>========
> >>>In section 4.2.1 it sounds like a per message timer, or even an interval
> >>>      
> >>>
> >>It is intended to be a per-message timer. 
> >>
> >>
> >>    
> >>
> >>>e.g.
> >>>   Timer T7 provides an indication of excessive delay of
> >>>   acknowledgement. If the following conditions are true:
> >>>
> >>>   (a) There is at least one message in the M2PA retransmit buffer.
> >>>
> >>>   (b) The remote M2PA is not in a Busy condition (i.e., the local
> >>>       timer T6 is not running).
> >>>
> >>>   (c) There is a message in the M2PA retransmit buffer that has not
> >>>       received an acknowledgement in the span of T7 since its last
> >>>       transmission.
> >>>
> >>>   then M2PA SHOULD take the link out of service.
> >>>
> >>>The interpretation of T7 as an interval (or just a condition on
> >>>messages) is reinforced by the fact there are no references to starting T7.
> >>>
> >>>On the other hand in section 4.1.4 there is a reference to stopping T7
> >>>during processor outage.  In 4.1.5 there is the comment
> >>>   "The peer M2PA SHALL NOT fail the link due to expiration of timer T7
> >>>   excessive delay of acknowledgement (see section 4.2.1 Sending and
> >>>   receiving messages) while its T6 timer is running, i.e., while the
> >>>   other end is Busy."
> >>>There is no comment on how to handle T7 after Busy clears.  Is it
> >>>appropriate to immediately check the conditions from 4.2.1?
> >>>      
> >>>
> >>I see the problem. After Busy clears, you may have a some unacked
> >>messages that have been waiting for longer than T7. It doesn't seem like
> >>you would want to fail the link for that. So it looks like T7 should be
> >>restarted when Busy clears.
> >>    
> >>
> >
> >T7 continues to run while busy (but not while processor outage).  This
> >is the same as Q.703.  T7 expiry just does not fail the link while busy,
> >T6 expiry does.
> >  
> >
> Is it then appropriate to check the condition in 4.2.1 when Busy abates, 
> to see if T7 has expired for any unacknowledged messages?  If so, is it 
> worth stating explicitly?

As stated in 4.2.1, if T7 has expired for any messages during the busy
condition, it does not take the link out of service.  T7 does not need
to be "checked" when the busy condition ceases: if T7 expires and there
is no busy condition the link should be taken out of service as proscribed
in section 4.2.1.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 01:05:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA24633
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 01:05:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h166CGe25466
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 01:12:16 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h166C5p25453;
	Thu, 6 Feb 2003 01:12:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h166Bop25437
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 01:11:50 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA24597
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 01:05:00 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:XMA7+FdgxiXqZaFieX2S/W5LktzTfw/F@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1668dD11997;
	Wed, 5 Feb 2003 23:08:39 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1668do30174;
	Wed, 5 Feb 2003 23:08:39 -0700
Date: Wed, 5 Feb 2003 23:08:38 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030205230838.B26810@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	sigtran@ietf.org
References: <Pine.SOL.4.10.10302052218190.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302052218190.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Wed, Feb 05, 2003 at 11:57:10PM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Wed, 05 Feb 2003 23:57:10)
> 
> Hi all,
> 
> Currently RFC 3332 says that SG sends SSNM messages to all 'concerned
> ASPs'(means only ACTIVE ASPs).
> 
> +----+           +---+           +-----+
> |SEP | ----------|SG |-----------| ASP1|
> +----+           +---+           +-----+
>                    |             
>                    |             +-----+
>                    +-------------| ASP2|
>                                  +-----+
> 
> (ASP1 is active and ASP2 is incative - override mode)
> 
> When SEP is getting restarted ,SG sends DUNA to ASP1.
> But SEP status is not changed in ASP2.There are two ways to solve this
> problem
> 
> 1. ASP1 can equalize its database with ASP2
> or
> 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> 
> IMO, SG can send SSNM messages to ASP (either ASP is in active or inact
> state)
> 
> Any comments?

The audit procedure is provided to allow any ASP to determine the
current accessibility of destinations from the SG.

The prosrcibed procedure in RFC 3332 is as follows:

  If ASP2 can know from ASP1 that the destination is inaccessible
  then nothing is required.

  If ASP2 does not know from ASP1 that destinations are inaccessible,
  it issues DAUD when it goes active and the SG tells it which
  destinations are accessible.

SSNM are NOT sent to inactive ASPs.

The reason for this is as follows:

Consider that ASP1 is active and ASP2 is down.  SG sends DUNA
to ASP1.  APS2 does not receive this because APS2 is not even
running.

ASP2 goes ASPUP and moves to the inactive state for the AS for
which the DUNA was sent.

Can the SG be expected to then send this backlog of information?

No, it is the repsonsibility of ASP2 to send DAUD to discover
the information.

This is also the case for the inactive state.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 01:29:07 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA25008
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 01:29:06 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h166ZQC26128
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 01:35:26 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h166ZCp26111;
	Thu, 6 Feb 2003 01:35:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h166YFp26068
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 01:34:15 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA24967
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 01:27:24 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h166V2qg003178;
	Thu, 6 Feb 2003 00:31:03 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h166UUM16128;
	Thu, 6 Feb 2003 00:30:30 -0600 (CST)
Date: Thu, 6 Feb 2003 00:30:30 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030205230838.B26810@openss7.org>
Message-ID: <Pine.SOL.4.10.10302060016580.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Wed, 5 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Wed, 05 Feb 2003 23:57:10)
> > 
> > Hi all,
> > 
> > Currently RFC 3332 says that SG sends SSNM messages to all 'concerned
> > ASPs'(means only ACTIVE ASPs).
> > 
> > +----+           +---+           +-----+
> > |SEP | ----------|SG |-----------| ASP1|
> > +----+           +---+           +-----+
> >                    |             
> >                    |             +-----+
> >                    +-------------| ASP2|
> >                                  +-----+
> > 
> > (ASP1 is active and ASP2 is incative - override mode)
> > 
> > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > But SEP status is not changed in ASP2.There are two ways to solve this
> > problem
> > 
> > 1. ASP1 can equalize its database with ASP2
> > or
> > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > 
> > IMO, SG can send SSNM messages to ASP (either ASP is in active or inact
> > state)
> > 
> > Any comments?
> 
> The audit procedure is provided to allow any ASP to determine the
> current accessibility of destinations from the SG.
> 
> The prosrcibed procedure in RFC 3332 is as follows:
> 
>   If ASP2 can know from ASP1 that the destination is inaccessible
>   then nothing is required.
This is out of the scope of RFC.

> 
>   If ASP2 does not know from ASP1 that destinations are inaccessible,
>   it issues DAUD when it goes active and the SG tells it which
>   destinations are accessible.
I think getting DUNA from SG is better than sending DAUD and receiving
response.If SG is able to do it ,let SG send DUNA to inactive ASPs.
IMO there is no harm.
> 
> SSNM are NOT sent to inactive ASPs.
> 
> The reason for this is as follows:
> 
> Consider that ASP1 is active and ASP2 is down.  SG sends DUNA
> to ASP1.  APS2 does not receive this because APS2 is not even
> running.
> 
> ASP2 goes ASPUP and moves to the inactive state for the AS for
> which the DUNA was sent.
> 
> Can the SG be expected to then send this backlog of information?
> 

I mean to say that when SG receives TFP from SEP and if the state
of ASP is ACTIVE or INACT then sending DUNA is good idea.

If ASP is in DOWN state then SG will not send DUNA.
IMO, practically speaking backup ASP will not be in DOWN state,
It should be in ASP INACT state.

My point is if SG is able to send DUNA , we should not put any
restriction.

--Samuel

> No, it is the repsonsibility of ASP2 to send DAUD to discover
> the information.
> 
> This is also the case for the inactive state.
> 
> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 01:41:03 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA25327
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 01:41:03 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h166lNr27211
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 01:47:23 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h166lCp27195;
	Thu, 6 Feb 2003 01:47:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h166ktp27176
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 01:46:55 -0500
Received: from hermes.fm.intel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA25297
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 01:40:03 -0500 (EST)
Received: from talaria.fm.intel.com (talaria.fm.intel.com [10.1.192.39])
	by hermes.fm.intel.com (8.11.6/8.11.6/d: outer.mc,v 1.51 2002/09/23 20:43:23 dmccart Exp $) with ESMTP id h166ena07378
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 06:40:49 GMT
Received: from fmsmsxvs041.fm.intel.com (fmsmsxvs041.fm.intel.com [132.233.42.126])
	by talaria.fm.intel.com (8.11.6/8.11.6/d: inner.mc,v 1.28 2003/01/13 19:44:39 dmccart Exp $) with SMTP id h166jVp23733
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 06:45:31 GMT
Received: from fmsmsx28.fm.intel.com ([132.233.42.28])
 by fmsmsxvs041.fm.intel.com (NAVGW 2.5.2.11) with SMTP id M2003020522413702096
 ; Wed, 05 Feb 2003 22:41:37 -0800
Received: by fmsmsx28.fm.intel.com with Internet Mail Service (5.5.2653.19)
	id <DYW2KZ39>; Wed, 5 Feb 2003 22:43:42 -0800
Message-ID: <12B638FEE763F74696D8544752E7204802C479A3@bgsmsx101.iind.intel.com>
From: "Nagar, Ashish" <ashish.nagar@intel.com>
To: "'Samuel Dur D. Jeyaseelan'" <sjeyasee@ssd.usa.alcatel.com>,
        sigtran@ietf.org
Subject: RE: [Sigtran] M3UA - ASP database equalization
Date: Wed, 5 Feb 2003 22:38:43 -0800 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

You can generate DAUD procedure at any time, so whenever ASP changes state
from Inactive to Active Just generate DAUD.

Thanks & Regards
Ashish Nagar


-----Original Message-----
From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
Sent: Thursday, February 06, 2003 11:27 AM
To: sigtran@ietf.org
Subject: [Sigtran] M3UA - ASP database equalization


Hi all,

Currently RFC 3332 says that SG sends SSNM messages to all 'concerned
ASPs'(means only ACTIVE ASPs).

+----+           +---+           +-----+
|SEP | ----------|SG |-----------| ASP1|
+----+           +---+           +-----+
                   |             
                   |             +-----+
                   +-------------| ASP2|
                                 +-----+

(ASP1 is active and ASP2 is incative - override mode)

When SEP is getting restarted ,SG sends DUNA to ASP1.
But SEP status is not changed in ASP2.There are two ways to solve this
problem

1. ASP1 can equalize its database with ASP2
or
2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)

IMO, SG can send SSNM messages to ASP (either ASP is in active or inact
state)

Any comments?

-----

section 5.4.2.1

...
      - Pass the MTP-TRANSFER request primitive to the MTP3 of the
        concerned Network Appearance.

It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER request is
sent to MTP3.

But there is no primitive availabe between M3UA and MTP3.

I think it should be corrected.

----
Presently ASP is unaware of SG's MTP3 restart, 
ASP receives only DUNA for all SS7 point codes which can be
reachable via that SG.But in case of peer to peer protocol
SG and AS point codes are different, So why dont we send
DUNA with SG's point code to ASP? So that just by getting
single DUNA(with SG's PC) ASP will not route any message via that SG.
This can avoid sending multiple DUNAs to ASP.


Any comments?

-----
section 1.4.1

...
   Where a particular AS can be reached via more than one SGP, the
   corresponding Routing Keys in the SGPs should be identical.  (Note:
   It is possible for the SGP Routing Key configuration data to be
   temporarily out-of-sync during configuration updates).
 
                              +--------+
                              |        |
                 +------------+  SG 1  +--------------+
     +-------+   |  SS7 links | "STP"  |  IP network  |     ----
     |  SEP  +---+            +--------+              +---/      \
     |   or  |                    |*                      | ASPs  |
     |  STP  +---+            +--------+              +---\      /
     +-------+   |            |        |              |     ----
                 +------------+  SG 2  +--------------+
                              | "STP"  |
                              +--------+
 
                    Figure 1  Example with mated SGs

Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are located
geographically in different places.The ideas is when SG1 goes down in
CITY1(due to earthquake...) , all the traffice still can be handled by SG2
which is in CITY2.

SGP2 is mate for SGP1, On failure of SGP1, all the messages will be
routed from SGP1 to SGP2.

Does all these things mean to say that SGP1's mate 'should be' present in
SG2 and not in SG1 itself ?
 

Comments are welcome.

--Samuel




  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 01:58:28 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA25759
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 01:58:28 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1674mj30230
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 02:04:48 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1674Zp30039;
	Thu, 6 Feb 2003 02:04:36 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1673Hp28084
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 02:03:17 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA25714
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 01:56:25 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h16702JV020230;
	Thu, 6 Feb 2003 01:00:04 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h166xsM26293;
	Thu, 6 Feb 2003 00:59:54 -0600 (CST)
Date: Thu, 6 Feb 2003 00:59:54 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Nagar, Ashish" <ashish.nagar@intel.com>
cc: sigtran@ietf.org
Subject: RE: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <12B638FEE763F74696D8544752E7204802C479A3@bgsmsx101.iind.intel.com>
Message-ID: <Pine.SOL.4.10.10302060055270.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Ashish,

My point is if SG is able to send DUNA to all inact and active ASPs,
we should not restrict it.

Sending DAUD is an alternative way ,If SG has direct way of sending
DUNA then Why do you need alternative way?

--Samuel

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Wed, 5 Feb 2003, Nagar, Ashish wrote:

> You can generate DAUD procedure at any time, so whenever ASP changes state
> from Inactive to Active Just generate DAUD.
> 
> Thanks & Regards
> Ashish Nagar
> 
> 
> -----Original Message-----
> From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> Sent: Thursday, February 06, 2003 11:27 AM
> To: sigtran@ietf.org
> Subject: [Sigtran] M3UA - ASP database equalization
> 
> 
> Hi all,
> 
> Currently RFC 3332 says that SG sends SSNM messages to all 'concerned
> ASPs'(means only ACTIVE ASPs).
> 
> +----+           +---+           +-----+
> |SEP | ----------|SG |-----------| ASP1|
> +----+           +---+           +-----+
>                    |             
>                    |             +-----+
>                    +-------------| ASP2|
>                                  +-----+
> 
> (ASP1 is active and ASP2 is incative - override mode)
> 
> When SEP is getting restarted ,SG sends DUNA to ASP1.
> But SEP status is not changed in ASP2.There are two ways to solve this
> problem
> 
> 1. ASP1 can equalize its database with ASP2
> or
> 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> 
> IMO, SG can send SSNM messages to ASP (either ASP is in active or inact
> state)
> 
> Any comments?
> 
> -----
> 
> section 5.4.2.1
> 
> ...
>       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
>         concerned Network Appearance.
> 
> It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER request is
> sent to MTP3.
> 
> But there is no primitive availabe between M3UA and MTP3.
> 
> I think it should be corrected.
> 
> ----
> Presently ASP is unaware of SG's MTP3 restart, 
> ASP receives only DUNA for all SS7 point codes which can be
> reachable via that SG.But in case of peer to peer protocol
> SG and AS point codes are different, So why dont we send
> DUNA with SG's point code to ASP? So that just by getting
> single DUNA(with SG's PC) ASP will not route any message via that SG.
> This can avoid sending multiple DUNAs to ASP.
> 
> 
> Any comments?
> 
> -----
> section 1.4.1
> 
> ...
>    Where a particular AS can be reached via more than one SGP, the
>    corresponding Routing Keys in the SGPs should be identical.  (Note:
>    It is possible for the SGP Routing Key configuration data to be
>    temporarily out-of-sync during configuration updates).
>  
>                               +--------+
>                               |        |
>                  +------------+  SG 1  +--------------+
>      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
>      |  SEP  +---+            +--------+              +---/      \
>      |   or  |                    |*                      | ASPs  |
>      |  STP  +---+            +--------+              +---\      /
>      +-------+   |            |        |              |     ----
>                  +------------+  SG 2  +--------------+
>                               | "STP"  |
>                               +--------+
>  
>                     Figure 1  Example with mated SGs
> 
> Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are located
> geographically in different places.The ideas is when SG1 goes down in
> CITY1(due to earthquake...) , all the traffice still can be handled by SG2
> which is in CITY2.
> 
> SGP2 is mate for SGP1, On failure of SGP1, all the messages will be
> routed from SGP1 to SGP2.
> 
> Does all these things mean to say that SGP1's mate 'should be' present in
> SG2 and not in SG1 itself ?
>  
> 
> Comments are welcome.
> 
> --Samuel
> 
> 
> 
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 02:16:59 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA06375
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 02:16:59 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h167NKb07170
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 02:23:20 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h167N6p07143;
	Thu, 6 Feb 2003 02:23:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h167LDp07073
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 02:21:13 -0500
Received: from hermes.hd.intel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA06326
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 02:14:20 -0500 (EST)
Received: from petasus.hd.intel.com (petasus.hd.intel.com [10.127.45.3])
	by hermes.hd.intel.com (8.11.6/8.11.6/d: outer.mc,v 1.51 2002/09/23 20:43:23 dmccart Exp $) with ESMTP id h167G9f11199
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 07:16:09 GMT
Received: from orsmsxvs040.jf.intel.com (orsmsxvs040.jf.intel.com [192.168.65.206])
	by petasus.hd.intel.com (8.11.6/8.11.6/d: inner.mc,v 1.28 2003/01/13 19:44:39 dmccart Exp $) with SMTP id h167EgE24561
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 07:14:42 GMT
Received: from orsmsx26.jf.intel.com ([192.168.65.26])
 by orsmsxvs040.jf.intel.com (NAVGW 2.5.2.11) with SMTP id M2003020523245015863
 ; Wed, 05 Feb 2003 23:24:50 -0800
Received: by orsmsx26.jf.intel.com with Internet Mail Service (5.5.2653.19)
	id <D9A3TMWY>; Wed, 5 Feb 2003 23:17:58 -0800
Message-ID: <12B638FEE763F74696D8544752E7204802C479A4@bgsmsx101.iind.intel.com>
From: "Nagar, Ashish" <ashish.nagar@intel.com>
To: "'Samuel Dur D. Jeyaseelan'" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>
Cc: sigtran@ietf.org
Subject: RE: [Sigtran] M3UA - ASP database equalization
Date: Wed, 5 Feb 2003 23:13:09 -0800 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Hi Samuel

My point is why to do any processing if ASP is in Inactive state.
Think about ASP is in Inactive state for longer period and receiving lots of
DUNA and DAVA, while what matter is the present only when ASP state changes
to ASP-ACTIVE.

Thanks & Regards
Ashish Nagar


-----Original Message-----
From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
Sent: Thursday, February 06, 2003 12:30 PM
To: Nagar, Ashish
Cc: sigtran@ietf.org
Subject: RE: [Sigtran] M3UA - ASP database equalization


Ashish,

My point is if SG is able to send DUNA to all inact and active ASPs,
we should not restrict it.

Sending DAUD is an alternative way ,If SG has direct way of sending
DUNA then Why do you need alternative way?

--Samuel

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Wed, 5 Feb 2003, Nagar, Ashish wrote:

> You can generate DAUD procedure at any time, so whenever ASP changes state
> from Inactive to Active Just generate DAUD.
> 
> Thanks & Regards
> Ashish Nagar
> 
> 
> -----Original Message-----
> From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> Sent: Thursday, February 06, 2003 11:27 AM
> To: sigtran@ietf.org
> Subject: [Sigtran] M3UA - ASP database equalization
> 
> 
> Hi all,
> 
> Currently RFC 3332 says that SG sends SSNM messages to all 'concerned
> ASPs'(means only ACTIVE ASPs).
> 
> +----+           +---+           +-----+
> |SEP | ----------|SG |-----------| ASP1|
> +----+           +---+           +-----+
>                    |             
>                    |             +-----+
>                    +-------------| ASP2|
>                                  +-----+
> 
> (ASP1 is active and ASP2 is incative - override mode)
> 
> When SEP is getting restarted ,SG sends DUNA to ASP1.
> But SEP status is not changed in ASP2.There are two ways to solve this
> problem
> 
> 1. ASP1 can equalize its database with ASP2
> or
> 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> 
> IMO, SG can send SSNM messages to ASP (either ASP is in active or inact
> state)
> 
> Any comments?
> 
> -----
> 
> section 5.4.2.1
> 
> ...
>       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
>         concerned Network Appearance.
> 
> It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER request is
> sent to MTP3.
> 
> But there is no primitive availabe between M3UA and MTP3.
> 
> I think it should be corrected.
> 
> ----
> Presently ASP is unaware of SG's MTP3 restart, 
> ASP receives only DUNA for all SS7 point codes which can be
> reachable via that SG.But in case of peer to peer protocol
> SG and AS point codes are different, So why dont we send
> DUNA with SG's point code to ASP? So that just by getting
> single DUNA(with SG's PC) ASP will not route any message via that SG.
> This can avoid sending multiple DUNAs to ASP.
> 
> 
> Any comments?
> 
> -----
> section 1.4.1
> 
> ...
>    Where a particular AS can be reached via more than one SGP, the
>    corresponding Routing Keys in the SGPs should be identical.  (Note:
>    It is possible for the SGP Routing Key configuration data to be
>    temporarily out-of-sync during configuration updates).
>  
>                               +--------+
>                               |        |
>                  +------------+  SG 1  +--------------+
>      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
>      |  SEP  +---+            +--------+              +---/      \
>      |   or  |                    |*                      | ASPs  |
>      |  STP  +---+            +--------+              +---\      /
>      +-------+   |            |        |              |     ----
>                  +------------+  SG 2  +--------------+
>                               | "STP"  |
>                               +--------+
>  
>                     Figure 1  Example with mated SGs
> 
> Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are located
> geographically in different places.The ideas is when SG1 goes down in
> CITY1(due to earthquake...) , all the traffice still can be handled by SG2
> which is in CITY2.
> 
> SGP2 is mate for SGP1, On failure of SGP1, all the messages will be
> routed from SGP1 to SGP2.
> 
> Does all these things mean to say that SGP1's mate 'should be' present in
> SG2 and not in SG1 itself ?
>  
> 
> Comments are welcome.
> 
> --Samuel
> 
> 
> 
> 
>   Alcatel USA, Inc.                  Internet:
<userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc.
*******
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 02:33:04 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA06792
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 02:33:04 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h167dPd08533
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 02:39:25 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h167dCp08522;
	Thu, 6 Feb 2003 02:39:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h167cIp08480
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 02:38:18 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA06751
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 02:31:24 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h167Z2JT023039;
	Thu, 6 Feb 2003 01:35:02 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h167YnM08979;
	Thu, 6 Feb 2003 01:34:49 -0600 (CST)
Date: Thu, 6 Feb 2003 01:34:49 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Nagar, Ashish" <ashish.nagar@intel.com>
cc: sigtran@ietf.org
Subject: RE: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <12B638FEE763F74696D8544752E7204802C479A4@bgsmsx101.iind.intel.com>
Message-ID: <Pine.SOL.4.10.10302060123090.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Ashish,

But when inactive ASP changes state to active, first of all
it does not know whether to send DAUD or not.

It will try to send traffic - but actually destination is unavailable.

The worst case is when SG triggers restart procedures this case all the
SS7 destinations are unavailable via restarted-SG.But this poor ASP
(newly come-up ASP) keeps sending traffic via restarted-SG to reach the
SS7 destination and for every message ASP receives DUNA.

Are we not doing double-work here?

--Samuel


  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Wed, 5 Feb 2003, Nagar, Ashish wrote:

> Hi Samuel
> 
> My point is why to do any processing if ASP is in Inactive state.
> Think about ASP is in Inactive state for longer period and receiving lots of
> DUNA and DAVA, while what matter is the present only when ASP state changes
> to ASP-ACTIVE.
> 
> Thanks & Regards
> Ashish Nagar
> 
> 
> -----Original Message-----
> From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> Sent: Thursday, February 06, 2003 12:30 PM
> To: Nagar, Ashish
> Cc: sigtran@ietf.org
> Subject: RE: [Sigtran] M3UA - ASP database equalization
> 
> 
> Ashish,
> 
> My point is if SG is able to send DUNA to all inact and active ASPs,
> we should not restrict it.
> 
> Sending DAUD is an alternative way ,If SG has direct way of sending
> DUNA then Why do you need alternative way?
> 
> --Samuel
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> 
> > You can generate DAUD procedure at any time, so whenever ASP changes state
> > from Inactive to Active Just generate DAUD.
> > 
> > Thanks & Regards
> > Ashish Nagar
> > 
> > 
> > -----Original Message-----
> > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> > Sent: Thursday, February 06, 2003 11:27 AM
> > To: sigtran@ietf.org
> > Subject: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > Hi all,
> > 
> > Currently RFC 3332 says that SG sends SSNM messages to all 'concerned
> > ASPs'(means only ACTIVE ASPs).
> > 
> > +----+           +---+           +-----+
> > |SEP | ----------|SG |-----------| ASP1|
> > +----+           +---+           +-----+
> >                    |             
> >                    |             +-----+
> >                    +-------------| ASP2|
> >                                  +-----+
> > 
> > (ASP1 is active and ASP2 is incative - override mode)
> > 
> > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > But SEP status is not changed in ASP2.There are two ways to solve this
> > problem
> > 
> > 1. ASP1 can equalize its database with ASP2
> > or
> > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > 
> > IMO, SG can send SSNM messages to ASP (either ASP is in active or inact
> > state)
> > 
> > Any comments?
> > 
> > -----
> > 
> > section 5.4.2.1
> > 
> > ...
> >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> >         concerned Network Appearance.
> > 
> > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER request is
> > sent to MTP3.
> > 
> > But there is no primitive availabe between M3UA and MTP3.
> > 
> > I think it should be corrected.
> > 
> > ----
> > Presently ASP is unaware of SG's MTP3 restart, 
> > ASP receives only DUNA for all SS7 point codes which can be
> > reachable via that SG.But in case of peer to peer protocol
> > SG and AS point codes are different, So why dont we send
> > DUNA with SG's point code to ASP? So that just by getting
> > single DUNA(with SG's PC) ASP will not route any message via that SG.
> > This can avoid sending multiple DUNAs to ASP.
> > 
> > 
> > Any comments?
> > 
> > -----
> > section 1.4.1
> > 
> > ...
> >    Where a particular AS can be reached via more than one SGP, the
> >    corresponding Routing Keys in the SGPs should be identical.  (Note:
> >    It is possible for the SGP Routing Key configuration data to be
> >    temporarily out-of-sync during configuration updates).
> >  
> >                               +--------+
> >                               |        |
> >                  +------------+  SG 1  +--------------+
> >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> >      |  SEP  +---+            +--------+              +---/      \
> >      |   or  |                    |*                      | ASPs  |
> >      |  STP  +---+            +--------+              +---\      /
> >      +-------+   |            |        |              |     ----
> >                  +------------+  SG 2  +--------------+
> >                               | "STP"  |
> >                               +--------+
> >  
> >                     Figure 1  Example with mated SGs
> > 
> > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are located
> > geographically in different places.The ideas is when SG1 goes down in
> > CITY1(due to earthquake...) , all the traffice still can be handled by SG2
> > which is in CITY2.
> > 
> > SGP2 is mate for SGP1, On failure of SGP1, all the messages will be
> > routed from SGP1 to SGP2.
> > 
> > Does all these things mean to say that SGP1's mate 'should be' present in
> > SG2 and not in SG1 itself ?
> >  
> > 
> > Comments are welcome.
> > 
> > --Samuel
> > 
> > 
> > 
> > 
> >   Alcatel USA, Inc.                  Internet:
> <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> *******
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > 
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 02:35:54 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA06877
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 02:35:54 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h167gFc08663
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 02:42:15 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h167g4p08656;
	Thu, 6 Feb 2003 02:42:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h167fsp08639
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 02:41:54 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA06851
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 02:35:01 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:nJdbCWS87C0X4XXW0fRM3nZ27DB0FXs+@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h167ceD13087;
	Thu, 6 Feb 2003 00:38:40 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h167ceT08484;
	Thu, 6 Feb 2003 00:38:40 -0700
Date: Thu, 6 Feb 2003 00:38:40 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206003840.A8281@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	sigtran@ietf.org
References: <20030205230838.B26810@openss7.org> <Pine.SOL.4.10.10302060016580.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302060016580.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 12:30:30AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

IMO sending SSNM to inactive ASPs is a BAD idea.

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 00:30:30)
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Wed, 5 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Wed, 05 Feb 2003 23:57:10)
> > > 
> > > Hi all,
> > > 
> > > Currently RFC 3332 says that SG sends SSNM messages to all 'concerned
> > > ASPs'(means only ACTIVE ASPs).
> > > 
> > > +----+           +---+           +-----+
> > > |SEP | ----------|SG |-----------| ASP1|
> > > +----+           +---+           +-----+
> > >                    |             
> > >                    |             +-----+
> > >                    +-------------| ASP2|
> > >                                  +-----+
> > > 
> > > (ASP1 is active and ASP2 is incative - override mode)
> > > 
> > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > But SEP status is not changed in ASP2.There are two ways to solve this
> > > problem
> > > 
> > > 1. ASP1 can equalize its database with ASP2
> > > or
> > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > 
> > > IMO, SG can send SSNM messages to ASP (either ASP is in active or inact
> > > state)
> > > 
> > > Any comments?
> > 
> > The audit procedure is provided to allow any ASP to determine the
> > current accessibility of destinations from the SG.
> > 
> > The prosrcibed procedure in RFC 3332 is as follows:
> > 
> >   If ASP2 can know from ASP1 that the destination is inaccessible
> >   then nothing is required.
> This is out of the scope of RFC.
> 
> > 
> >   If ASP2 does not know from ASP1 that destinations are inaccessible,
> >   it issues DAUD when it goes active and the SG tells it which
> >   destinations are accessible.
> I think getting DUNA from SG is better than sending DAUD and receiving
> response.If SG is able to do it ,let SG send DUNA to inactive ASPs.
> IMO there is no harm.
> > 
> > SSNM are NOT sent to inactive ASPs.
> > 
> > The reason for this is as follows:
> > 
> > Consider that ASP1 is active and ASP2 is down.  SG sends DUNA
> > to ASP1.  APS2 does not receive this because APS2 is not even
> > running.
> > 
> > ASP2 goes ASPUP and moves to the inactive state for the AS for
> > which the DUNA was sent.
> > 
> > Can the SG be expected to then send this backlog of information?
> > 
> 
> I mean to say that when SG receives TFP from SEP and if the state
> of ASP is ACTIVE or INACT then sending DUNA is good idea.
> 
> If ASP is in DOWN state then SG will not send DUNA.
> IMO, practically speaking backup ASP will not be in DOWN state,
> It should be in ASP INACT state.
> 
> My point is if SG is able to send DUNA , we should not put any
> restriction.
> 
> --Samuel
> 
> > No, it is the repsonsibility of ASP2 to send DAUD to discover
> > the information.
> > 
> > This is also the case for the inactive state.
> > 
> > --brian
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 02:38:56 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA06934
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 02:38:56 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h167jHI08823
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 02:45:17 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h167j5p08809;
	Thu, 6 Feb 2003 02:45:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h167iPp08768
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 02:44:25 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA06913
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 02:37:32 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:wF5NznDTh84qV1BWDHV1VAuaHr2Cyyrt@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h167fBD13131;
	Thu, 6 Feb 2003 00:41:11 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h167fBp08897;
	Thu, 6 Feb 2003 00:41:11 -0700
Date: Thu, 6 Feb 2003 00:41:11 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206004111.B8281@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <12B638FEE763F74696D8544752E7204802C479A4@bgsmsx101.iind.intel.com> <Pine.SOL.4.10.10302060123090.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302060123090.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 01:34:49AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 01:34:49)
> 
> Ashish,
> 
> But when inactive ASP changes state to active, first of all
> it does not know whether to send DAUD or not.

It simply sends DAUD whenever it goes active.

> 
> It will try to send traffic - but actually destination is unavailable.

No, becuase it will send DAUD.

> 
> The worst case is when SG triggers restart procedures this case all the
> SS7 destinations are unavailable via restarted-SG.But this poor ASP
> (newly come-up ASP) keeps sending traffic via restarted-SG to reach the
> SS7 destination and for every message ASP receives DUNA.

It does not send traffic until DAUD is has completed because it does
not send MTP-RESUME or MTP-RESTART-ENDS to the user until DAUD completes.

> 
> Are we not doing double-work here?


You have not considered my example where the ASP is down.

--brian

> 
> --Samuel
> 
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> 
> > Hi Samuel
> > 
> > My point is why to do any processing if ASP is in Inactive state.
> > Think about ASP is in Inactive state for longer period and receiving lots of
> > DUNA and DAVA, while what matter is the present only when ASP state changes
> > to ASP-ACTIVE.
> > 
> > Thanks & Regards
> > Ashish Nagar
> > 
> > 
> > -----Original Message-----
> > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> > Sent: Thursday, February 06, 2003 12:30 PM
> > To: Nagar, Ashish
> > Cc: sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > Ashish,
> > 
> > My point is if SG is able to send DUNA to all inact and active ASPs,
> > we should not restrict it.
> > 
> > Sending DAUD is an alternative way ,If SG has direct way of sending
> > DUNA then Why do you need alternative way?
> > 
> > --Samuel
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > 
> > > You can generate DAUD procedure at any time, so whenever ASP changes state
> > > from Inactive to Active Just generate DAUD.
> > > 
> > > Thanks & Regards
> > > Ashish Nagar
> > > 
> > > 
> > > -----Original Message-----
> > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> > > Sent: Thursday, February 06, 2003 11:27 AM
> > > To: sigtran@ietf.org
> > > Subject: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > Hi all,
> > > 
> > > Currently RFC 3332 says that SG sends SSNM messages to all 'concerned
> > > ASPs'(means only ACTIVE ASPs).
> > > 
> > > +----+           +---+           +-----+
> > > |SEP | ----------|SG |-----------| ASP1|
> > > +----+           +---+           +-----+
> > >                    |             
> > >                    |             +-----+
> > >                    +-------------| ASP2|
> > >                                  +-----+
> > > 
> > > (ASP1 is active and ASP2 is incative - override mode)
> > > 
> > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > But SEP status is not changed in ASP2.There are two ways to solve this
> > > problem
> > > 
> > > 1. ASP1 can equalize its database with ASP2
> > > or
> > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > 
> > > IMO, SG can send SSNM messages to ASP (either ASP is in active or inact
> > > state)
> > > 
> > > Any comments?
> > > 
> > > -----
> > > 
> > > section 5.4.2.1
> > > 
> > > ...
> > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> > >         concerned Network Appearance.
> > > 
> > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER request is
> > > sent to MTP3.
> > > 
> > > But there is no primitive availabe between M3UA and MTP3.
> > > 
> > > I think it should be corrected.
> > > 
> > > ----
> > > Presently ASP is unaware of SG's MTP3 restart, 
> > > ASP receives only DUNA for all SS7 point codes which can be
> > > reachable via that SG.But in case of peer to peer protocol
> > > SG and AS point codes are different, So why dont we send
> > > DUNA with SG's point code to ASP? So that just by getting
> > > single DUNA(with SG's PC) ASP will not route any message via that SG.
> > > This can avoid sending multiple DUNAs to ASP.
> > > 
> > > 
> > > Any comments?
> > > 
> > > -----
> > > section 1.4.1
> > > 
> > > ...
> > >    Where a particular AS can be reached via more than one SGP, the
> > >    corresponding Routing Keys in the SGPs should be identical.  (Note:
> > >    It is possible for the SGP Routing Key configuration data to be
> > >    temporarily out-of-sync during configuration updates).
> > >  
> > >                               +--------+
> > >                               |        |
> > >                  +------------+  SG 1  +--------------+
> > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > >      |  SEP  +---+            +--------+              +---/      \
> > >      |   or  |                    |*                      | ASPs  |
> > >      |  STP  +---+            +--------+              +---\      /
> > >      +-------+   |            |        |              |     ----
> > >                  +------------+  SG 2  +--------------+
> > >                               | "STP"  |
> > >                               +--------+
> > >  
> > >                     Figure 1  Example with mated SGs
> > > 
> > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are located
> > > geographically in different places.The ideas is when SG1 goes down in
> > > CITY1(due to earthquake...) , all the traffice still can be handled by SG2
> > > which is in CITY2.
> > > 
> > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will be
> > > routed from SGP1 to SGP2.
> > > 
> > > Does all these things mean to say that SGP1's mate 'should be' present in
> > > SG2 and not in SG1 itself ?
> > >  
> > > 
> > > Comments are welcome.
> > > 
> > > --Samuel
> > > 
> > > 
> > > 
> > > 
> > >   Alcatel USA, Inc.                  Internet:
> > <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > *******
> > > 
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> > 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 03:00:32 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA07643
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 03:00:32 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1686qF09514
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 03:06:52 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1686ap09505;
	Thu, 6 Feb 2003 03:06:37 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1685wp09474
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 03:05:58 -0500
Received: from zdemail03.zdem.compaq.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA07600
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 02:59:06 -0500 (EST)
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail03.zdem.compaq.com (Postfix) with ESMTP
	id A064162F; Thu,  6 Feb 2003 09:02:42 +0100 (CET)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.2966);
	 Thu, 6 Feb 2003 09:02:42 +0100
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Subject: RE: [Sigtran] M3UA - ASP database equalization
Date: Thu, 6 Feb 2003 09:02:41 +0100
Message-ID: <06CAA4D3753C53408D2CE6340B684442014888CA@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA - ASP database equalization
Thread-Index: AcLNsviu3zyak/DkR0ydXwBa8Ut6WAAAU5TQ
From: "Zannin, Francois" <Francois.Zannin@hp.com>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>
Cc: <sigtran@ietf.org>
X-OriginalArrivalTime: 06 Feb 2003 08:02:42.0576 (UTC) FILETIME=[2014A100:01C2CDB6]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h1685wp09475
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Samuel,

IMO, it's a good idea to inform inactive (BUT UP) ASP of the state of
the ss7 network especially in the override mode where an ASP can be
ready to recover an AS as soon as it receive an AS-Pending notification
from the SGp. In this condition, the ASP is ready to operate quickly and
minimize the risk of buffered message lost during the pending timer.

If someone does not want to receive SSNM because it is not able to
recover an AS, it just stays in an ASP-Down state.

Because SSNM reflects ss7 Point Code state that is not "SI" dependant,
it affects all AS for a network and then, unless SG that plays multiple
networks, the ASP UP is sufficient to filter the undesired SSNM.

Francois.

-----Original Message-----
From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
Sent: Thursday, February 06, 2003 8:35 AM
To: Nagar, Ashish
Cc: sigtran@ietf.org
Subject: RE: [Sigtran] M3UA - ASP database equalization



Ashish,

But when inactive ASP changes state to active, first of all
it does not know whether to send DAUD or not.

It will try to send traffic - but actually destination is unavailable.

The worst case is when SG triggers restart procedures this case all the
SS7 destinations are unavailable via restarted-SG.But this poor ASP
(newly come-up ASP) keeps sending traffic via restarted-SG to reach the
SS7 destination and for every message ASP receives DUNA.

Are we not doing double-work here?

--Samuel


  Alcatel USA, Inc.                  Internet:
<userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc.
*******

On Wed, 5 Feb 2003, Nagar, Ashish wrote:

> Hi Samuel
> 
> My point is why to do any processing if ASP is in Inactive state. 
> Think about ASP is in Inactive state for longer period and receiving 
> lots of DUNA and DAVA, while what matter is the present only when ASP 
> state changes to ASP-ACTIVE.
> 
> Thanks & Regards
> Ashish Nagar
> 
> 
> -----Original Message-----
> From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> Sent: Thursday, February 06, 2003 12:30 PM
> To: Nagar, Ashish
> Cc: sigtran@ietf.org
> Subject: RE: [Sigtran] M3UA - ASP database equalization
> 
> 
> Ashish,
> 
> My point is if SG is able to send DUNA to all inact and active ASPs, 
> we should not restrict it.
> 
> Sending DAUD is an alternative way ,If SG has direct way of sending 
> DUNA then Why do you need alternative way?
> 
> --Samuel
> 
>   Alcatel USA, Inc.                  Internet:
<userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. 
> *******
> 
> On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> 
> > You can generate DAUD procedure at any time, so whenever ASP changes

> > state from Inactive to Active Just generate DAUD.
> > 
> > Thanks & Regards
> > Ashish Nagar
> > 
> > 
> > -----Original Message-----
> > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > Sent: Thursday, February 06, 2003 11:27 AM
> > To: sigtran@ietf.org
> > Subject: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > Hi all,
> > 
> > Currently RFC 3332 says that SG sends SSNM messages to all 
> > 'concerned ASPs'(means only ACTIVE ASPs).
> > 
> > +----+           +---+           +-----+
> > |SEP | ----------|SG |-----------| ASP1|
> > +----+           +---+           +-----+
> >                    |             
> >                    |             +-----+
> >                    +-------------| ASP2|
> >                                  +-----+
> > 
> > (ASP1 is active and ASP2 is incative - override mode)
> > 
> > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > But SEP status is not changed in ASP2.There are two ways to solve 
> > this problem
> > 
> > 1. ASP1 can equalize its database with ASP2
> > or
> > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > 
> > IMO, SG can send SSNM messages to ASP (either ASP is in active or 
> > inact
> > state)
> > 
> > Any comments?
> > 
> > -----
> > 
> > section 5.4.2.1
> > 
> > ...
> >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> >         concerned Network Appearance.
> > 
> > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER request

> > is sent to MTP3.
> > 
> > But there is no primitive availabe between M3UA and MTP3.
> > 
> > I think it should be corrected.
> > 
> > ----
> > Presently ASP is unaware of SG's MTP3 restart,
> > ASP receives only DUNA for all SS7 point codes which can be
> > reachable via that SG.But in case of peer to peer protocol
> > SG and AS point codes are different, So why dont we send
> > DUNA with SG's point code to ASP? So that just by getting
> > single DUNA(with SG's PC) ASP will not route any message via that
SG.
> > This can avoid sending multiple DUNAs to ASP.
> > 
> > 
> > Any comments?
> > 
> > -----
> > section 1.4.1
> > 
> > ...
> >    Where a particular AS can be reached via more than one SGP, the
> >    corresponding Routing Keys in the SGPs should be identical.
(Note:
> >    It is possible for the SGP Routing Key configuration data to be
> >    temporarily out-of-sync during configuration updates).
> >  
> >                               +--------+
> >                               |        |
> >                  +------------+  SG 1  +--------------+
> >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> >      |  SEP  +---+            +--------+              +---/      \
> >      |   or  |                    |*                      | ASPs  |
> >      |  STP  +---+            +--------+              +---\      /
> >      +-------+   |            |        |              |     ----
> >                  +------------+  SG 2  +--------------+
> >                               | "STP"  |
> >                               +--------+
> >  
> >                     Figure 1  Example with mated SGs
> > 
> > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are 
> > located geographically in different places.The ideas is when SG1 
> > goes down in CITY1(due to earthquake...) , all the traffice still 
> > can be handled by SG2 which is in CITY2.
> > 
> > SGP2 is mate for SGP1, On failure of SGP1, all the messages will be 
> > routed from SGP1 to SGP2.
> > 
> > Does all these things mean to say that SGP1's mate 'should be' 
> > present in SG2 and not in SG1 itself ?
> >  
> > 
> > Comments are welcome.
> > 
> > --Samuel
> > 
> > 
> > 
> > 
> >   Alcatel USA, Inc.                  Internet:
> <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> *******
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > 
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 03:12:59 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA07922
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 03:12:59 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h168JJa10537
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 03:19:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168J7p10530;
	Thu, 6 Feb 2003 03:19:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168ITp10513
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 03:18:29 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA07893
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 03:11:36 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:/dGJptcYWiD147MTsTZI3I7AC/XPLMKS@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h168FED13574;
	Thu, 6 Feb 2003 01:15:14 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h168FEg13082;
	Thu, 6 Feb 2003 01:15:14 -0700
Date: Thu, 6 Feb 2003 01:15:13 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Zannin, Francois" <Francois.Zannin@hp.com>
Cc: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206011513.A12848@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Zannin, Francois" <Francois.Zannin@hp.com>,
	"Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <06CAA4D3753C53408D2CE6340B684442014888CA@vbeexc02.emea.cpqcorp.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <06CAA4D3753C53408D2CE6340B684442014888CA@vbeexc02.emea.cpqcorp.net>; from Francois.Zannin@hp.com on Thu, Feb 06, 2003 at 09:02:41AM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Zannin,,

That is contrary to the conensus of the RFC.

Consider 1 ASP active, 200,000 ASPs inactive.

The RFC cannot require an SG to send SSNM to an inactive ASP.

--brian

Zannin, Francois wrote:                                                            (Thu, 06 Feb 2003 09:02:41)
> Samuel,
> 
> IMO, it's a good idea to inform inactive (BUT UP) ASP of the state of
> the ss7 network especially in the override mode where an ASP can be
> ready to recover an AS as soon as it receive an AS-Pending notification
> from the SGp. In this condition, the ASP is ready to operate quickly and
> minimize the risk of buffered message lost during the pending timer.
> 
> If someone does not want to receive SSNM because it is not able to
> recover an AS, it just stays in an ASP-Down state.
> 
> Because SSNM reflects ss7 Point Code state that is not "SI" dependant,
> it affects all AS for a network and then, unless SG that plays multiple
> networks, the ASP UP is sufficient to filter the undesired SSNM.
> 
> Francois.
> 
> -----Original Message-----
> From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> Sent: Thursday, February 06, 2003 8:35 AM
> To: Nagar, Ashish
> Cc: sigtran@ietf.org
> Subject: RE: [Sigtran] M3UA - ASP database equalization
> 
> 
> 
> Ashish,
> 
> But when inactive ASP changes state to active, first of all
> it does not know whether to send DAUD or not.
> 
> It will try to send traffic - but actually destination is unavailable.
> 
> The worst case is when SG triggers restart procedures this case all the
> SS7 destinations are unavailable via restarted-SG.But this poor ASP
> (newly come-up ASP) keeps sending traffic via restarted-SG to reach the
> SS7 destination and for every message ASP receives DUNA.
> 
> Are we not doing double-work here?
> 
> --Samuel
> 
> 
>   Alcatel USA, Inc.                  Internet:
> <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc.
> *******
> 
> On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> 
> > Hi Samuel
> > 
> > My point is why to do any processing if ASP is in Inactive state. 
> > Think about ASP is in Inactive state for longer period and receiving 
> > lots of DUNA and DAVA, while what matter is the present only when ASP 
> > state changes to ASP-ACTIVE.
> > 
> > Thanks & Regards
> > Ashish Nagar
> > 
> > 
> > -----Original Message-----
> > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > Sent: Thursday, February 06, 2003 12:30 PM
> > To: Nagar, Ashish
> > Cc: sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > Ashish,
> > 
> > My point is if SG is able to send DUNA to all inact and active ASPs, 
> > we should not restrict it.
> > 
> > Sending DAUD is an alternative way ,If SG has direct way of sending 
> > DUNA then Why do you need alternative way?
> > 
> > --Samuel
> > 
> >   Alcatel USA, Inc.                  Internet:
> <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. 
> > *******
> > 
> > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > 
> > > You can generate DAUD procedure at any time, so whenever ASP changes
> 
> > > state from Inactive to Active Just generate DAUD.
> > > 
> > > Thanks & Regards
> > > Ashish Nagar
> > > 
> > > 
> > > -----Original Message-----
> > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > Sent: Thursday, February 06, 2003 11:27 AM
> > > To: sigtran@ietf.org
> > > Subject: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > Hi all,
> > > 
> > > Currently RFC 3332 says that SG sends SSNM messages to all 
> > > 'concerned ASPs'(means only ACTIVE ASPs).
> > > 
> > > +----+           +---+           +-----+
> > > |SEP | ----------|SG |-----------| ASP1|
> > > +----+           +---+           +-----+
> > >                    |             
> > >                    |             +-----+
> > >                    +-------------| ASP2|
> > >                                  +-----+
> > > 
> > > (ASP1 is active and ASP2 is incative - override mode)
> > > 
> > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > But SEP status is not changed in ASP2.There are two ways to solve 
> > > this problem
> > > 
> > > 1. ASP1 can equalize its database with ASP2
> > > or
> > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > 
> > > IMO, SG can send SSNM messages to ASP (either ASP is in active or 
> > > inact
> > > state)
> > > 
> > > Any comments?
> > > 
> > > -----
> > > 
> > > section 5.4.2.1
> > > 
> > > ...
> > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> > >         concerned Network Appearance.
> > > 
> > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER request
> 
> > > is sent to MTP3.
> > > 
> > > But there is no primitive availabe between M3UA and MTP3.
> > > 
> > > I think it should be corrected.
> > > 
> > > ----
> > > Presently ASP is unaware of SG's MTP3 restart,
> > > ASP receives only DUNA for all SS7 point codes which can be
> > > reachable via that SG.But in case of peer to peer protocol
> > > SG and AS point codes are different, So why dont we send
> > > DUNA with SG's point code to ASP? So that just by getting
> > > single DUNA(with SG's PC) ASP will not route any message via that
> SG.
> > > This can avoid sending multiple DUNAs to ASP.
> > > 
> > > 
> > > Any comments?
> > > 
> > > -----
> > > section 1.4.1
> > > 
> > > ...
> > >    Where a particular AS can be reached via more than one SGP, the
> > >    corresponding Routing Keys in the SGPs should be identical.
> (Note:
> > >    It is possible for the SGP Routing Key configuration data to be
> > >    temporarily out-of-sync during configuration updates).
> > >  
> > >                               +--------+
> > >                               |        |
> > >                  +------------+  SG 1  +--------------+
> > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > >      |  SEP  +---+            +--------+              +---/      \
> > >      |   or  |                    |*                      | ASPs  |
> > >      |  STP  +---+            +--------+              +---\      /
> > >      +-------+   |            |        |              |     ----
> > >                  +------------+  SG 2  +--------------+
> > >                               | "STP"  |
> > >                               +--------+
> > >  
> > >                     Figure 1  Example with mated SGs
> > > 
> > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are 
> > > located geographically in different places.The ideas is when SG1 
> > > goes down in CITY1(due to earthquake...) , all the traffice still 
> > > can be handled by SG2 which is in CITY2.
> > > 
> > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will be 
> > > routed from SGP1 to SGP2.
> > > 
> > > Does all these things mean to say that SGP1's mate 'should be' 
> > > present in SG2 and not in SG1 itself ?
> > >  
> > > 
> > > Comments are welcome.
> > > 
> > > --Samuel
> > > 
> > > 
> > > 
> > > 
> > >   Alcatel USA, Inc.                  Internet:
> > <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > *******
> > > 
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> > 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 03:16:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08018
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 03:16:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h168NIq10692
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 03:23:18 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168N6p10679;
	Thu, 6 Feb 2003 03:23:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168MLp10653
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 03:22:21 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA07996
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 03:15:29 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h168J2ec017264;
	Thu, 6 Feb 2003 02:19:06 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h168ISM24491;
	Thu, 6 Feb 2003 02:18:28 -0600 (CST)
Date: Thu, 6 Feb 2003 02:18:28 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206003840.A8281@openss7.org>
Message-ID: <Pine.SOL.4.10.10302060215470.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> IMO sending SSNM to inactive ASPs is a BAD idea.
> 
Which is bad idea sending DAUD and get the DUNA from SG
or SG itself can send DUNA to inact ASP?

IMO my idea is not BAD.It will increase the M3UA stack
efficency.

> --brian
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 00:30:30)
> > 
> > Brian,
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Wed, 5 Feb 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                                                     (Wed, 05 Feb 2003 23:57:10)
> > > > 
> > > > Hi all,
> > > > 
> > > > Currently RFC 3332 says that SG sends SSNM messages to all 'concerned
> > > > ASPs'(means only ACTIVE ASPs).
> > > > 
> > > > +----+           +---+           +-----+
> > > > |SEP | ----------|SG |-----------| ASP1|
> > > > +----+           +---+           +-----+
> > > >                    |             
> > > >                    |             +-----+
> > > >                    +-------------| ASP2|
> > > >                                  +-----+
> > > > 
> > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > 
> > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > But SEP status is not changed in ASP2.There are two ways to solve this
> > > > problem
> > > > 
> > > > 1. ASP1 can equalize its database with ASP2
> > > > or
> > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > 
> > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or inact
> > > > state)
> > > > 
> > > > Any comments?
> > > 
> > > The audit procedure is provided to allow any ASP to determine the
> > > current accessibility of destinations from the SG.
> > > 
> > > The prosrcibed procedure in RFC 3332 is as follows:
> > > 
> > >   If ASP2 can know from ASP1 that the destination is inaccessible
> > >   then nothing is required.
> > This is out of the scope of RFC.
> > 
> > > 
> > >   If ASP2 does not know from ASP1 that destinations are inaccessible,
> > >   it issues DAUD when it goes active and the SG tells it which
> > >   destinations are accessible.
> > I think getting DUNA from SG is better than sending DAUD and receiving
> > response.If SG is able to do it ,let SG send DUNA to inactive ASPs.
> > IMO there is no harm.
> > > 
> > > SSNM are NOT sent to inactive ASPs.
> > > 
> > > The reason for this is as follows:
> > > 
> > > Consider that ASP1 is active and ASP2 is down.  SG sends DUNA
> > > to ASP1.  APS2 does not receive this because APS2 is not even
> > > running.
> > > 
> > > ASP2 goes ASPUP and moves to the inactive state for the AS for
> > > which the DUNA was sent.
> > > 
> > > Can the SG be expected to then send this backlog of information?
> > > 
> > 
> > I mean to say that when SG receives TFP from SEP and if the state
> > of ASP is ACTIVE or INACT then sending DUNA is good idea.
> > 
> > If ASP is in DOWN state then SG will not send DUNA.
> > IMO, practically speaking backup ASP will not be in DOWN state,
> > It should be in ASP INACT state.
> > 
> > My point is if SG is able to send DUNA , we should not put any
> > restriction.
> > 
> > --Samuel
> > 
> > > No, it is the repsonsibility of ASP2 to send DAUD to discover
> > > the information.
> > > 
> > > This is also the case for the inactive state.
> > > 
> > > --brian
> > > 
> > > -- 
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 03:25:04 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08180
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 03:25:04 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h168VOv10992
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 03:31:24 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168VCp10971;
	Thu, 6 Feb 2003 03:31:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168Udp10911
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 03:30:39 -0500
Received: from zdemail04.zdem.compaq.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08160
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 03:23:46 -0500 (EST)
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail04.zdem.compaq.com (Postfix) with ESMTP
	id 37200DA6; Thu,  6 Feb 2003 09:27:24 +0100 (CET)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.2966);
	 Thu, 6 Feb 2003 09:27:24 +0100
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Subject: RE: [Sigtran] M3UA - ASP database equalization
Date: Thu, 6 Feb 2003 09:27:21 +0100
Message-ID: <06CAA4D3753C53408D2CE6340B68444214C40C@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA - ASP database equalization
Thread-Index: AcLNuA9gq6k8u/nbSwmUZeecsZphSAAAA3pQ
From: "Zannin, Francois" <Francois.Zannin@hp.com>
To: <bidulock@openss7.org>
Cc: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 06 Feb 2003 08:27:24.0154 (UTC) FILETIME=[932B7DA0:01C2CDB9]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h168Udp10912
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Bidulock,,,

I think your example of 200,000 ASP "UP" as backup for an AS is really
realistic for a network!

The goal is to agree on procedure that can be used for the
implementation of M3UA. There were a lot of threads discussing on this
subject and not always concluded. The role of the IG is to clarify the
RFC in order to make it usable.

This procedure similar to the ss7 network (TFP inform all the SP of the
inaccessibility). If a SG is unable to reach all the ss7 SP (200000 PC,
means 200000 ungrouped SSNM) it can use the DOWN_ACK or INACTIVE_ACK
procedure for its own isolation.

-----Original Message-----
From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
Sent: Thursday, February 06, 2003 9:15 AM
To: Zannin, Francois
Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization


Zannin,,

That is contrary to the conensus of the RFC.

Consider 1 ASP active, 200,000 ASPs inactive.

The RFC cannot require an SG to send SSNM to an inactive ASP.

--brian

Zannin, Francois wrote:
(Thu, 06 Feb 2003 09:02:41)
> Samuel,
> 
> IMO, it's a good idea to inform inactive (BUT UP) ASP of the state of 
> the ss7 network especially in the override mode where an ASP can be 
> ready to recover an AS as soon as it receive an AS-Pending 
> notification from the SGp. In this condition, the ASP is ready to 
> operate quickly and minimize the risk of buffered message lost during 
> the pending timer.
> 
> If someone does not want to receive SSNM because it is not able to 
> recover an AS, it just stays in an ASP-Down state.
> 
> Because SSNM reflects ss7 Point Code state that is not "SI" dependant,

> it affects all AS for a network and then, unless SG that plays 
> multiple networks, the ASP UP is sufficient to filter the undesired 
> SSNM.
> 
> Francois.
> 
> -----Original Message-----
> From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> Sent: Thursday, February 06, 2003 8:35 AM
> To: Nagar, Ashish
> Cc: sigtran@ietf.org
> Subject: RE: [Sigtran] M3UA - ASP database equalization
> 
> 
> 
> Ashish,
> 
> But when inactive ASP changes state to active, first of all it does 
> not know whether to send DAUD or not.
> 
> It will try to send traffic - but actually destination is unavailable.
> 
> The worst case is when SG triggers restart procedures this case all 
> the SS7 destinations are unavailable via restarted-SG.But this poor 
> ASP (newly come-up ASP) keeps sending traffic via restarted-SG to 
> reach the SS7 destination and for every message ASP receives DUNA.
> 
> Are we not doing double-work here?
> 
> --Samuel
> 
> 
>   Alcatel USA, Inc.                  Internet:
> <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc.
> *******
> 
> On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> 
> > Hi Samuel
> > 
> > My point is why to do any processing if ASP is in Inactive state.
> > Think about ASP is in Inactive state for longer period and receiving

> > lots of DUNA and DAVA, while what matter is the present only when
ASP 
> > state changes to ASP-ACTIVE.
> > 
> > Thanks & Regards
> > Ashish Nagar
> > 
> > 
> > -----Original Message-----
> > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > Sent: Thursday, February 06, 2003 12:30 PM
> > To: Nagar, Ashish
> > Cc: sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > Ashish,
> > 
> > My point is if SG is able to send DUNA to all inact and active ASPs,
> > we should not restrict it.
> > 
> > Sending DAUD is an alternative way ,If SG has direct way of sending
> > DUNA then Why do you need alternative way?
> > 
> > --Samuel
> > 
> >   Alcatel USA, Inc.                  Internet:
> <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > *******
> > 
> > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > 
> > > You can generate DAUD procedure at any time, so whenever ASP 
> > > changes
> 
> > > state from Inactive to Active Just generate DAUD.
> > > 
> > > Thanks & Regards
> > > Ashish Nagar
> > > 
> > > 
> > > -----Original Message-----
> > > From: Samuel Dur D. Jeyaseelan 
> > > [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > Sent: Thursday, February 06, 2003 11:27 AM
> > > To: sigtran@ietf.org
> > > Subject: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > Hi all,
> > > 
> > > Currently RFC 3332 says that SG sends SSNM messages to all
> > > 'concerned ASPs'(means only ACTIVE ASPs).
> > > 
> > > +----+           +---+           +-----+
> > > |SEP | ----------|SG |-----------| ASP1|
> > > +----+           +---+           +-----+
> > >                    |             
> > >                    |             +-----+
> > >                    +-------------| ASP2|
> > >                                  +-----+
> > > 
> > > (ASP1 is active and ASP2 is incative - override mode)
> > > 
> > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > But SEP status is not changed in ASP2.There are two ways to solve
> > > this problem
> > > 
> > > 1. ASP1 can equalize its database with ASP2
> > > or
> > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > 
> > > IMO, SG can send SSNM messages to ASP (either ASP is in active or
> > > inact
> > > state)
> > > 
> > > Any comments?
> > > 
> > > -----
> > > 
> > > section 5.4.2.1
> > > 
> > > ...
> > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> > >         concerned Network Appearance.
> > > 
> > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER 
> > > request
> 
> > > is sent to MTP3.
> > > 
> > > But there is no primitive availabe between M3UA and MTP3.
> > > 
> > > I think it should be corrected.
> > > 
> > > ----
> > > Presently ASP is unaware of SG's MTP3 restart,
> > > ASP receives only DUNA for all SS7 point codes which can be 
> > > reachable via that SG.But in case of peer to peer protocol SG and 
> > > AS point codes are different, So why dont we send DUNA with SG's 
> > > point code to ASP? So that just by getting single DUNA(with SG's 
> > > PC) ASP will not route any message via that
> SG.
> > > This can avoid sending multiple DUNAs to ASP.
> > > 
> > > 
> > > Any comments?
> > > 
> > > -----
> > > section 1.4.1
> > > 
> > > ...
> > >    Where a particular AS can be reached via more than one SGP, the
> > >    corresponding Routing Keys in the SGPs should be identical.
> (Note:
> > >    It is possible for the SGP Routing Key configuration data to be
> > >    temporarily out-of-sync during configuration updates).
> > >  
> > >                               +--------+
> > >                               |        |
> > >                  +------------+  SG 1  +--------------+
> > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > >      |  SEP  +---+            +--------+              +---/      \
> > >      |   or  |                    |*                      | ASPs
|
> > >      |  STP  +---+            +--------+              +---\      /
> > >      +-------+   |            |        |              |     ----
> > >                  +------------+  SG 2  +--------------+
> > >                               | "STP"  |
> > >                               +--------+
> > >  
> > >                     Figure 1  Example with mated SGs
> > > 
> > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are
> > > located geographically in different places.The ideas is when SG1 
> > > goes down in CITY1(due to earthquake...) , all the traffice still 
> > > can be handled by SG2 which is in CITY2.
> > > 
> > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will 
> > > be
> > > routed from SGP1 to SGP2.
> > > 
> > > Does all these things mean to say that SGP1's mate 'should be'
> > > present in SG2 and not in SG1 itself ?
> > >  
> > > 
> > > Comments are welcome.
> > > 
> > > --Samuel
> > > 
> > > 
> > > 
> > > 
> > >   Alcatel USA, Inc.                  Internet:
> > <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, 
> > > Inc.
> > *******
> > > 
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> > 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 03:25:01 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08178
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 03:25:01 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h168VMr10979
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 03:31:22 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168V8p10955;
	Thu, 6 Feb 2003 03:31:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168UPp10907
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 03:30:25 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08155
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 03:23:31 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h168R2eY017843;
	Thu, 6 Feb 2003 02:27:08 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h168QQM27256;
	Thu, 6 Feb 2003 02:26:26 -0600 (CST)
Date: Thu, 6 Feb 2003 02:26:26 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206004111.B8281@openss7.org>
Message-ID: <Pine.SOL.4.10.10302060219010.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 01:34:49)
> > 
> > Ashish,
> > 
> > But when inactive ASP changes state to active, first of all
> > it does not know whether to send DAUD or not.
> 
> It simply sends DAUD whenever it goes active.
> 
> > 
> > It will try to send traffic - but actually destination is unavailable.
> 
> No, becuase it will send DAUD.

It is not mandatory to send DAUD when ASP becomes ACTIVE.

see section 4.5.3

...
      - Isolation. The ASP is newly ASP-ACTIVE or has been
                   isolated from an SGP for an extended period.  The ASP
                   MAY request the availability/congestion status of one
                   or more SS7 destinations to which it expects to
                   communicate.

Note the word "MAY".

> 
> > 
> > The worst case is when SG triggers restart procedures this case all the
> > SS7 destinations are unavailable via restarted-SG.But this poor ASP
> > (newly come-up ASP) keeps sending traffic via restarted-SG to reach the
> > SS7 destination and for every message ASP receives DUNA.
> 
> It does not send traffic until DAUD is has completed because it does
> not send MTP-RESUME or MTP-RESTART-ENDS to the user until DAUD completes.
> 

Finally you missed my point, you are providing indirect solution.
My point is avoid sending DAUD to SG because SG has got capability
to send DUNA to inactive ASPs.Nothing will prevent SG to send DUNA.

> > 
> > Are we not doing double-work here?
> 
> 
> You have not considered my example where the ASP is down.

Again misunderstanding.I said when the ASP is in inactive state it can
receive DUNA.

--Samuel

> 
> --brian
> 
> > 
> > --Samuel
> > 
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > 
> > > Hi Samuel
> > > 
> > > My point is why to do any processing if ASP is in Inactive state.
> > > Think about ASP is in Inactive state for longer period and receiving lots of
> > > DUNA and DAVA, while what matter is the present only when ASP state changes
> > > to ASP-ACTIVE.
> > > 
> > > Thanks & Regards
> > > Ashish Nagar
> > > 
> > > 
> > > -----Original Message-----
> > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> > > Sent: Thursday, February 06, 2003 12:30 PM
> > > To: Nagar, Ashish
> > > Cc: sigtran@ietf.org
> > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > Ashish,
> > > 
> > > My point is if SG is able to send DUNA to all inact and active ASPs,
> > > we should not restrict it.
> > > 
> > > Sending DAUD is an alternative way ,If SG has direct way of sending
> > > DUNA then Why do you need alternative way?
> > > 
> > > --Samuel
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > 
> > > > You can generate DAUD procedure at any time, so whenever ASP changes state
> > > > from Inactive to Active Just generate DAUD.
> > > > 
> > > > Thanks & Regards
> > > > Ashish Nagar
> > > > 
> > > > 
> > > > -----Original Message-----
> > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > To: sigtran@ietf.org
> > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > 
> > > > 
> > > > Hi all,
> > > > 
> > > > Currently RFC 3332 says that SG sends SSNM messages to all 'concerned
> > > > ASPs'(means only ACTIVE ASPs).
> > > > 
> > > > +----+           +---+           +-----+
> > > > |SEP | ----------|SG |-----------| ASP1|
> > > > +----+           +---+           +-----+
> > > >                    |             
> > > >                    |             +-----+
> > > >                    +-------------| ASP2|
> > > >                                  +-----+
> > > > 
> > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > 
> > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > But SEP status is not changed in ASP2.There are two ways to solve this
> > > > problem
> > > > 
> > > > 1. ASP1 can equalize its database with ASP2
> > > > or
> > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > 
> > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or inact
> > > > state)
> > > > 
> > > > Any comments?
> > > > 
> > > > -----
> > > > 
> > > > section 5.4.2.1
> > > > 
> > > > ...
> > > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> > > >         concerned Network Appearance.
> > > > 
> > > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER request is
> > > > sent to MTP3.
> > > > 
> > > > But there is no primitive availabe between M3UA and MTP3.
> > > > 
> > > > I think it should be corrected.
> > > > 
> > > > ----
> > > > Presently ASP is unaware of SG's MTP3 restart, 
> > > > ASP receives only DUNA for all SS7 point codes which can be
> > > > reachable via that SG.But in case of peer to peer protocol
> > > > SG and AS point codes are different, So why dont we send
> > > > DUNA with SG's point code to ASP? So that just by getting
> > > > single DUNA(with SG's PC) ASP will not route any message via that SG.
> > > > This can avoid sending multiple DUNAs to ASP.
> > > > 
> > > > 
> > > > Any comments?
> > > > 
> > > > -----
> > > > section 1.4.1
> > > > 
> > > > ...
> > > >    Where a particular AS can be reached via more than one SGP, the
> > > >    corresponding Routing Keys in the SGPs should be identical.  (Note:
> > > >    It is possible for the SGP Routing Key configuration data to be
> > > >    temporarily out-of-sync during configuration updates).
> > > >  
> > > >                               +--------+
> > > >                               |        |
> > > >                  +------------+  SG 1  +--------------+
> > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > >      |  SEP  +---+            +--------+              +---/      \
> > > >      |   or  |                    |*                      | ASPs  |
> > > >      |  STP  +---+            +--------+              +---\      /
> > > >      +-------+   |            |        |              |     ----
> > > >                  +------------+  SG 2  +--------------+
> > > >                               | "STP"  |
> > > >                               +--------+
> > > >  
> > > >                     Figure 1  Example with mated SGs
> > > > 
> > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are located
> > > > geographically in different places.The ideas is when SG1 goes down in
> > > > CITY1(due to earthquake...) , all the traffice still can be handled by SG2
> > > > which is in CITY2.
> > > > 
> > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will be
> > > > routed from SGP1 to SGP2.
> > > > 
> > > > Does all these things mean to say that SGP1's mate 'should be' present in
> > > > SG2 and not in SG1 itself ?
> > > >  
> > > > 
> > > > Comments are welcome.
> > > > 
> > > > --Samuel
> > > > 
> > > > 
> > > > 
> > > > 
> > > >   Alcatel USA, Inc.                  Internet:
> > > <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > *******
> > > > 
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > 
> > > 
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 03:26:59 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08246
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 03:26:59 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h168XJ211082
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 03:33:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168X6p11065;
	Thu, 6 Feb 2003 03:33:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168W4p11024
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 03:32:04 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08191
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 03:25:12 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:fIE/xo/yO7o5sHlj3ogxea4K6ujKeMsG@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h168SoD13722;
	Thu, 6 Feb 2003 01:28:50 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h168Snn14741;
	Thu, 6 Feb 2003 01:28:49 -0700
Date: Thu, 6 Feb 2003 01:28:49 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206012849.A14210@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	sigtran@ietf.org
References: <20030206003840.A8281@openss7.org> <Pine.SOL.4.10.10302060215470.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302060215470.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 02:18:28AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

It will decrease M3UA efficiency.

Consider 1 active ASP and 200,000 inactive ASPs.

For each SSNM the SG must send 200,001 DUNA messages.

With the existing concensus DAUD procedure, the SG only
sends 1 DUNA message.

When an ASP activates for an AS it sends one DAUD message
to avoid the SG sending many useless SSNM messages.

The activating ASP need not even send DAUD.  It can just
start sending messages and use responsive DUNAs.

This is most efficient because SS7 destinations are almost
never unavailable.

Sending SSNM to inactive ASPs would be a BAD thing.  It is
good that the RFC does not permit this.

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 02:18:28)
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > IMO sending SSNM to inactive ASPs is a BAD idea.
> > 
> Which is bad idea sending DAUD and get the DUNA from SG
> or SG itself can send DUNA to inact ASP?
> 
> IMO my idea is not BAD.It will increase the M3UA stack
> efficency.
> 
> > --brian
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 00:30:30)
> > > 
> > > Brian,
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Wed, 5 Feb 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Samuel,
> > > > 
> > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Wed, 05 Feb 2003 23:57:10)
> > > > > 
> > > > > Hi all,
> > > > > 
> > > > > Currently RFC 3332 says that SG sends SSNM messages to all 'concerned
> > > > > ASPs'(means only ACTIVE ASPs).
> > > > > 
> > > > > +----+           +---+           +-----+
> > > > > |SEP | ----------|SG |-----------| ASP1|
> > > > > +----+           +---+           +-----+
> > > > >                    |             
> > > > >                    |             +-----+
> > > > >                    +-------------| ASP2|
> > > > >                                  +-----+
> > > > > 
> > > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > > 
> > > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > > But SEP status is not changed in ASP2.There are two ways to solve this
> > > > > problem
> > > > > 
> > > > > 1. ASP1 can equalize its database with ASP2
> > > > > or
> > > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > > 
> > > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or inact
> > > > > state)
> > > > > 
> > > > > Any comments?
> > > > 
> > > > The audit procedure is provided to allow any ASP to determine the
> > > > current accessibility of destinations from the SG.
> > > > 
> > > > The prosrcibed procedure in RFC 3332 is as follows:
> > > > 
> > > >   If ASP2 can know from ASP1 that the destination is inaccessible
> > > >   then nothing is required.
> > > This is out of the scope of RFC.
> > > 
> > > > 
> > > >   If ASP2 does not know from ASP1 that destinations are inaccessible,
> > > >   it issues DAUD when it goes active and the SG tells it which
> > > >   destinations are accessible.
> > > I think getting DUNA from SG is better than sending DAUD and receiving
> > > response.If SG is able to do it ,let SG send DUNA to inactive ASPs.
> > > IMO there is no harm.
> > > > 
> > > > SSNM are NOT sent to inactive ASPs.
> > > > 
> > > > The reason for this is as follows:
> > > > 
> > > > Consider that ASP1 is active and ASP2 is down.  SG sends DUNA
> > > > to ASP1.  APS2 does not receive this because APS2 is not even
> > > > running.
> > > > 
> > > > ASP2 goes ASPUP and moves to the inactive state for the AS for
> > > > which the DUNA was sent.
> > > > 
> > > > Can the SG be expected to then send this backlog of information?
> > > > 
> > > 
> > > I mean to say that when SG receives TFP from SEP and if the state
> > > of ASP is ACTIVE or INACT then sending DUNA is good idea.
> > > 
> > > If ASP is in DOWN state then SG will not send DUNA.
> > > IMO, practically speaking backup ASP will not be in DOWN state,
> > > It should be in ASP INACT state.
> > > 
> > > My point is if SG is able to send DUNA , we should not put any
> > > restriction.
> > > 
> > > --Samuel
> > > 
> > > > No, it is the repsonsibility of ASP2 to send DAUD to discover
> > > > the information.
> > > > 
> > > > This is also the case for the inactive state.
> > > > 
> > > > --brian
> > > > 
> > > > -- 
> > > > Brian F. G. Bidulock
> > > > bidulock@openss7.org
> > > > http://www.openss7.org/
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > 
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 03:27:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08281
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 03:27:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h168YH811125
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 03:34:17 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168Y2p11115;
	Thu, 6 Feb 2003 03:34:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168Xqp11100
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 03:33:52 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08245
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 03:26:59 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:42CXisEGJQWmRHf5u7SV8A+V7A4nzzqR@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h168UbD13814;
	Thu, 6 Feb 2003 01:30:37 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h168UaN14971;
	Thu, 6 Feb 2003 01:30:36 -0700
Date: Thu, 6 Feb 2003 01:30:36 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206013036.B14210@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206004111.B8281@openss7.org> <Pine.SOL.4.10.10302060219010.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302060219010.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 02:26:26AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 02:26:26)
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 01:34:49)
> > > 
> > > Ashish,
> > > 
> > > But when inactive ASP changes state to active, first of all
> > > it does not know whether to send DAUD or not.
> > 
> > It simply sends DAUD whenever it goes active.
> > 
> > > 
> > > It will try to send traffic - but actually destination is unavailable.
> > 
> > No, becuase it will send DAUD.
> 
> It is not mandatory to send DAUD when ASP becomes ACTIVE.
> 
> see section 4.5.3
> 
> ...
>       - Isolation. The ASP is newly ASP-ACTIVE or has been
>                    isolated from an SGP for an extended period.  The ASP
>                    MAY request the availability/congestion status of one
>                    or more SS7 destinations to which it expects to
>                    communicate.
> 
> Note the word "MAY".

So don't.

> 
> > 
> > > 
> > > The worst case is when SG triggers restart procedures this case all the
> > > SS7 destinations are unavailable via restarted-SG.But this poor ASP
> > > (newly come-up ASP) keeps sending traffic via restarted-SG to reach the
> > > SS7 destination and for every message ASP receives DUNA.
> > 
> > It does not send traffic until DAUD is has completed because it does
> > not send MTP-RESUME or MTP-RESTART-ENDS to the user until DAUD completes.
> > 
> 
> Finally you missed my point, you are providing indirect solution.
> My point is avoid sending DAUD to SG because SG has got capability
> to send DUNA to inactive ASPs.Nothing will prevent SG to send DUNA.

You missed my point.  There is no MTP user until the ASP is active.

> 
> > > 
> > > Are we not doing double-work here?
> > 
> > 
> > You have not considered my example where the ASP is down.
> 
> Again misunderstanding.I said when the ASP is in inactive state it can
> receive DUNA.

But it can do nothing with it, so it is useless to send SSNM to inactive
ASPs.

--brian

> 
> --Samuel
> 
> > 
> > --brian
> > 
> > > 
> > > --Samuel
> > > 
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > 
> > > > Hi Samuel
> > > > 
> > > > My point is why to do any processing if ASP is in Inactive state.
> > > > Think about ASP is in Inactive state for longer period and receiving lots of
> > > > DUNA and DAVA, while what matter is the present only when ASP state changes
> > > > to ASP-ACTIVE.
> > > > 
> > > > Thanks & Regards
> > > > Ashish Nagar
> > > > 
> > > > 
> > > > -----Original Message-----
> > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> > > > Sent: Thursday, February 06, 2003 12:30 PM
> > > > To: Nagar, Ashish
> > > > Cc: sigtran@ietf.org
> > > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > > 
> > > > 
> > > > Ashish,
> > > > 
> > > > My point is if SG is able to send DUNA to all inact and active ASPs,
> > > > we should not restrict it.
> > > > 
> > > > Sending DAUD is an alternative way ,If SG has direct way of sending
> > > > DUNA then Why do you need alternative way?
> > > > 
> > > > --Samuel
> > > > 
> > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > 
> > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > 
> > > > > You can generate DAUD procedure at any time, so whenever ASP changes state
> > > > > from Inactive to Active Just generate DAUD.
> > > > > 
> > > > > Thanks & Regards
> > > > > Ashish Nagar
> > > > > 
> > > > > 
> > > > > -----Original Message-----
> > > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> > > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > > To: sigtran@ietf.org
> > > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > > 
> > > > > 
> > > > > Hi all,
> > > > > 
> > > > > Currently RFC 3332 says that SG sends SSNM messages to all 'concerned
> > > > > ASPs'(means only ACTIVE ASPs).
> > > > > 
> > > > > +----+           +---+           +-----+
> > > > > |SEP | ----------|SG |-----------| ASP1|
> > > > > +----+           +---+           +-----+
> > > > >                    |             
> > > > >                    |             +-----+
> > > > >                    +-------------| ASP2|
> > > > >                                  +-----+
> > > > > 
> > > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > > 
> > > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > > But SEP status is not changed in ASP2.There are two ways to solve this
> > > > > problem
> > > > > 
> > > > > 1. ASP1 can equalize its database with ASP2
> > > > > or
> > > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > > 
> > > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or inact
> > > > > state)
> > > > > 
> > > > > Any comments?
> > > > > 
> > > > > -----
> > > > > 
> > > > > section 5.4.2.1
> > > > > 
> > > > > ...
> > > > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> > > > >         concerned Network Appearance.
> > > > > 
> > > > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER request is
> > > > > sent to MTP3.
> > > > > 
> > > > > But there is no primitive availabe between M3UA and MTP3.
> > > > > 
> > > > > I think it should be corrected.
> > > > > 
> > > > > ----
> > > > > Presently ASP is unaware of SG's MTP3 restart, 
> > > > > ASP receives only DUNA for all SS7 point codes which can be
> > > > > reachable via that SG.But in case of peer to peer protocol
> > > > > SG and AS point codes are different, So why dont we send
> > > > > DUNA with SG's point code to ASP? So that just by getting
> > > > > single DUNA(with SG's PC) ASP will not route any message via that SG.
> > > > > This can avoid sending multiple DUNAs to ASP.
> > > > > 
> > > > > 
> > > > > Any comments?
> > > > > 
> > > > > -----
> > > > > section 1.4.1
> > > > > 
> > > > > ...
> > > > >    Where a particular AS can be reached via more than one SGP, the
> > > > >    corresponding Routing Keys in the SGPs should be identical.  (Note:
> > > > >    It is possible for the SGP Routing Key configuration data to be
> > > > >    temporarily out-of-sync during configuration updates).
> > > > >  
> > > > >                               +--------+
> > > > >                               |        |
> > > > >                  +------------+  SG 1  +--------------+
> > > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > > >      |  SEP  +---+            +--------+              +---/      \
> > > > >      |   or  |                    |*                      | ASPs  |
> > > > >      |  STP  +---+            +--------+              +---\      /
> > > > >      +-------+   |            |        |              |     ----
> > > > >                  +------------+  SG 2  +--------------+
> > > > >                               | "STP"  |
> > > > >                               +--------+
> > > > >  
> > > > >                     Figure 1  Example with mated SGs
> > > > > 
> > > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are located
> > > > > geographically in different places.The ideas is when SG1 goes down in
> > > > > CITY1(due to earthquake...) , all the traffice still can be handled by SG2
> > > > > which is in CITY2.
> > > > > 
> > > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will be
> > > > > routed from SGP1 to SGP2.
> > > > > 
> > > > > Does all these things mean to say that SGP1's mate 'should be' present in
> > > > > SG2 and not in SG1 itself ?
> > > > >  
> > > > > 
> > > > > Comments are welcome.
> > > > > 
> > > > > --Samuel
> > > > > 
> > > > > 
> > > > > 
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet:
> > > > <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > > *******
> > > > > 
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > 
> > > > 
> > > 
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 03:31:10 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08360
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 03:31:10 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h168bVS11644
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 03:37:31 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168bHp11414;
	Thu, 6 Feb 2003 03:37:17 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168aYp11224
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 03:36:41 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08312
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 03:29:28 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h168X2eW018959;
	Thu, 6 Feb 2003 02:33:02 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h168WUM29243;
	Thu, 6 Feb 2003 02:32:30 -0600 (CST)
Date: Thu, 6 Feb 2003 02:32:30 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206011513.A12848@openss7.org>
Message-ID: <Pine.SOL.4.10.10302060227250.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Zannin,,
> 
> That is contrary to the conensus of the RFC.
> 

What is contrary and what is conensus?
RFC always welcomes all modifications.

> Consider 1 ASP active, 200,000 ASPs inactive.
> 
> The RFC cannot require an SG to send SSNM to an inactive ASP.
> 

Consider 200,000 nodes with SG1 and if SG1 restarts, Will you send
200,000 DUNA after 200,000 DAUD.

--Samuel
> --brian
> 
> Zannin, Francois wrote:                                                            (Thu, 06 Feb 2003 09:02:41)
> > Samuel,
> > 
> > IMO, it's a good idea to inform inactive (BUT UP) ASP of the state of
> > the ss7 network especially in the override mode where an ASP can be
> > ready to recover an AS as soon as it receive an AS-Pending notification
> > from the SGp. In this condition, the ASP is ready to operate quickly and
> > minimize the risk of buffered message lost during the pending timer.
> > 
> > If someone does not want to receive SSNM because it is not able to
> > recover an AS, it just stays in an ASP-Down state.
> > 
> > Because SSNM reflects ss7 Point Code state that is not "SI" dependant,
> > it affects all AS for a network and then, unless SG that plays multiple
> > networks, the ASP UP is sufficient to filter the undesired SSNM.
> > 
> > Francois.
> > 
> > -----Original Message-----
> > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> > Sent: Thursday, February 06, 2003 8:35 AM
> > To: Nagar, Ashish
> > Cc: sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > 
> > Ashish,
> > 
> > But when inactive ASP changes state to active, first of all
> > it does not know whether to send DAUD or not.
> > 
> > It will try to send traffic - but actually destination is unavailable.
> > 
> > The worst case is when SG triggers restart procedures this case all the
> > SS7 destinations are unavailable via restarted-SG.But this poor ASP
> > (newly come-up ASP) keeps sending traffic via restarted-SG to reach the
> > SS7 destination and for every message ASP receives DUNA.
> > 
> > Are we not doing double-work here?
> > 
> > --Samuel
> > 
> > 
> >   Alcatel USA, Inc.                  Internet:
> > <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > *******
> > 
> > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > 
> > > Hi Samuel
> > > 
> > > My point is why to do any processing if ASP is in Inactive state. 
> > > Think about ASP is in Inactive state for longer period and receiving 
> > > lots of DUNA and DAVA, while what matter is the present only when ASP 
> > > state changes to ASP-ACTIVE.
> > > 
> > > Thanks & Regards
> > > Ashish Nagar
> > > 
> > > 
> > > -----Original Message-----
> > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > Sent: Thursday, February 06, 2003 12:30 PM
> > > To: Nagar, Ashish
> > > Cc: sigtran@ietf.org
> > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > Ashish,
> > > 
> > > My point is if SG is able to send DUNA to all inact and active ASPs, 
> > > we should not restrict it.
> > > 
> > > Sending DAUD is an alternative way ,If SG has direct way of sending 
> > > DUNA then Why do you need alternative way?
> > > 
> > > --Samuel
> > > 
> > >   Alcatel USA, Inc.                  Internet:
> > <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. 
> > > *******
> > > 
> > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > 
> > > > You can generate DAUD procedure at any time, so whenever ASP changes
> > 
> > > > state from Inactive to Active Just generate DAUD.
> > > > 
> > > > Thanks & Regards
> > > > Ashish Nagar
> > > > 
> > > > 
> > > > -----Original Message-----
> > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > To: sigtran@ietf.org
> > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > 
> > > > 
> > > > Hi all,
> > > > 
> > > > Currently RFC 3332 says that SG sends SSNM messages to all 
> > > > 'concerned ASPs'(means only ACTIVE ASPs).
> > > > 
> > > > +----+           +---+           +-----+
> > > > |SEP | ----------|SG |-----------| ASP1|
> > > > +----+           +---+           +-----+
> > > >                    |             
> > > >                    |             +-----+
> > > >                    +-------------| ASP2|
> > > >                                  +-----+
> > > > 
> > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > 
> > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > But SEP status is not changed in ASP2.There are two ways to solve 
> > > > this problem
> > > > 
> > > > 1. ASP1 can equalize its database with ASP2
> > > > or
> > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > 
> > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or 
> > > > inact
> > > > state)
> > > > 
> > > > Any comments?
> > > > 
> > > > -----
> > > > 
> > > > section 5.4.2.1
> > > > 
> > > > ...
> > > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> > > >         concerned Network Appearance.
> > > > 
> > > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER request
> > 
> > > > is sent to MTP3.
> > > > 
> > > > But there is no primitive availabe between M3UA and MTP3.
> > > > 
> > > > I think it should be corrected.
> > > > 
> > > > ----
> > > > Presently ASP is unaware of SG's MTP3 restart,
> > > > ASP receives only DUNA for all SS7 point codes which can be
> > > > reachable via that SG.But in case of peer to peer protocol
> > > > SG and AS point codes are different, So why dont we send
> > > > DUNA with SG's point code to ASP? So that just by getting
> > > > single DUNA(with SG's PC) ASP will not route any message via that
> > SG.
> > > > This can avoid sending multiple DUNAs to ASP.
> > > > 
> > > > 
> > > > Any comments?
> > > > 
> > > > -----
> > > > section 1.4.1
> > > > 
> > > > ...
> > > >    Where a particular AS can be reached via more than one SGP, the
> > > >    corresponding Routing Keys in the SGPs should be identical.
> > (Note:
> > > >    It is possible for the SGP Routing Key configuration data to be
> > > >    temporarily out-of-sync during configuration updates).
> > > >  
> > > >                               +--------+
> > > >                               |        |
> > > >                  +------------+  SG 1  +--------------+
> > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > >      |  SEP  +---+            +--------+              +---/      \
> > > >      |   or  |                    |*                      | ASPs  |
> > > >      |  STP  +---+            +--------+              +---\      /
> > > >      +-------+   |            |        |              |     ----
> > > >                  +------------+  SG 2  +--------------+
> > > >                               | "STP"  |
> > > >                               +--------+
> > > >  
> > > >                     Figure 1  Example with mated SGs
> > > > 
> > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are 
> > > > located geographically in different places.The ideas is when SG1 
> > > > goes down in CITY1(due to earthquake...) , all the traffice still 
> > > > can be handled by SG2 which is in CITY2.
> > > > 
> > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will be 
> > > > routed from SGP1 to SGP2.
> > > > 
> > > > Does all these things mean to say that SGP1's mate 'should be' 
> > > > present in SG2 and not in SG1 itself ?
> > > >  
> > > > 
> > > > Comments are welcome.
> > > > 
> > > > --Samuel
> > > > 
> > > > 
> > > > 
> > > > 
> > > >   Alcatel USA, Inc.                  Internet:
> > > <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > *******
> > > > 
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > 
> > > 
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 03:34:02 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08427
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 03:34:02 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h168eNn12046
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 03:40:23 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168e6p12034;
	Thu, 6 Feb 2003 03:40:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168d1p11973
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 03:39:01 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08388
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 03:32:06 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:PYqaKAosM8Q59N4oMIjL1W+t+L44KmuR@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h168ZhD13874;
	Thu, 6 Feb 2003 01:35:43 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h168ZhJ15574;
	Thu, 6 Feb 2003 01:35:43 -0700
Date: Thu, 6 Feb 2003 01:35:42 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Zannin, Francois" <Francois.Zannin@hp.com>
Cc: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206013542.C14210@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Zannin, Francois" <Francois.Zannin@hp.com>,
	"Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <06CAA4D3753C53408D2CE6340B68444214C40C@vbeexc02.emea.cpqcorp.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <06CAA4D3753C53408D2CE6340B68444214C40C@vbeexc02.emea.cpqcorp.net>; from Francois.Zannin@hp.com on Thu, Feb 06, 2003 at 09:27:21AM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Zannin,,

Zannin, Francois wrote:                                                            (Thu, 06 Feb 2003 09:27:21)
> Bidulock,,,
> 
> I think your example of 200,000 ASP "UP" as backup for an AS is really
> realistic for a network!

200?  20?  It makes no difference.  The SG cannot be required to
send useless messages to bunch of ASPs that can do nothing with
the information.

> 
> The goal is to agree on procedure that can be used for the
> implementation of M3UA. There were a lot of threads discussing on this
> subject and not always concluded. The role of the IG is to clarify the
> RFC in order to make it usable.
> 
> This procedure similar to the ss7 network (TFP inform all the SP of the
> inaccessibility). If a SG is unable to reach all the ss7 SP (200000 PC,
> means 200000 ungrouped SSNM) it can use the DOWN_ACK or INACTIVE_ACK
> procedure for its own isolation.

DUNA corresponds to MTP-PAUSE not TFP.

Consider 100 APSs inactive, zero (0) ASPs active.

The SG cannot be required to send DUNA to inactive ASPs.  When the SG
NIF has no user to send MTP-PAUSE to, it is not possible for the SG
to send DUNA.

The SG cannot be required to send SSNM to inactive ASPs.

Besides, it is unnecessary.

--brian


> 
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> Sent: Thursday, February 06, 2003 9:15 AM
> To: Zannin, Francois
> Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> Subject: Re: [Sigtran] M3UA - ASP database equalization
> 
> 
> Zannin,,
> 
> That is contrary to the conensus of the RFC.
> 
> Consider 1 ASP active, 200,000 ASPs inactive.
> 
> The RFC cannot require an SG to send SSNM to an inactive ASP.
> 
> --brian
> 
> Zannin, Francois wrote:
> (Thu, 06 Feb 2003 09:02:41)
> > Samuel,
> > 
> > IMO, it's a good idea to inform inactive (BUT UP) ASP of the state of 
> > the ss7 network especially in the override mode where an ASP can be 
> > ready to recover an AS as soon as it receive an AS-Pending 
> > notification from the SGp. In this condition, the ASP is ready to 
> > operate quickly and minimize the risk of buffered message lost during 
> > the pending timer.
> > 
> > If someone does not want to receive SSNM because it is not able to 
> > recover an AS, it just stays in an ASP-Down state.
> > 
> > Because SSNM reflects ss7 Point Code state that is not "SI" dependant,
> 
> > it affects all AS for a network and then, unless SG that plays 
> > multiple networks, the ASP UP is sufficient to filter the undesired 
> > SSNM.
> > 
> > Francois.
> > 
> > -----Original Message-----
> > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > Sent: Thursday, February 06, 2003 8:35 AM
> > To: Nagar, Ashish
> > Cc: sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > 
> > Ashish,
> > 
> > But when inactive ASP changes state to active, first of all it does 
> > not know whether to send DAUD or not.
> > 
> > It will try to send traffic - but actually destination is unavailable.
> > 
> > The worst case is when SG triggers restart procedures this case all 
> > the SS7 destinations are unavailable via restarted-SG.But this poor 
> > ASP (newly come-up ASP) keeps sending traffic via restarted-SG to 
> > reach the SS7 destination and for every message ASP receives DUNA.
> > 
> > Are we not doing double-work here?
> > 
> > --Samuel
> > 
> > 
> >   Alcatel USA, Inc.                  Internet:
> > <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > *******
> > 
> > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > 
> > > Hi Samuel
> > > 
> > > My point is why to do any processing if ASP is in Inactive state.
> > > Think about ASP is in Inactive state for longer period and receiving
> 
> > > lots of DUNA and DAVA, while what matter is the present only when
> ASP 
> > > state changes to ASP-ACTIVE.
> > > 
> > > Thanks & Regards
> > > Ashish Nagar
> > > 
> > > 
> > > -----Original Message-----
> > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > Sent: Thursday, February 06, 2003 12:30 PM
> > > To: Nagar, Ashish
> > > Cc: sigtran@ietf.org
> > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > Ashish,
> > > 
> > > My point is if SG is able to send DUNA to all inact and active ASPs,
> > > we should not restrict it.
> > > 
> > > Sending DAUD is an alternative way ,If SG has direct way of sending
> > > DUNA then Why do you need alternative way?
> > > 
> > > --Samuel
> > > 
> > >   Alcatel USA, Inc.                  Internet:
> > <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > *******
> > > 
> > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > 
> > > > You can generate DAUD procedure at any time, so whenever ASP 
> > > > changes
> > 
> > > > state from Inactive to Active Just generate DAUD.
> > > > 
> > > > Thanks & Regards
> > > > Ashish Nagar
> > > > 
> > > > 
> > > > -----Original Message-----
> > > > From: Samuel Dur D. Jeyaseelan 
> > > > [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > To: sigtran@ietf.org
> > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > 
> > > > 
> > > > Hi all,
> > > > 
> > > > Currently RFC 3332 says that SG sends SSNM messages to all
> > > > 'concerned ASPs'(means only ACTIVE ASPs).
> > > > 
> > > > +----+           +---+           +-----+
> > > > |SEP | ----------|SG |-----------| ASP1|
> > > > +----+           +---+           +-----+
> > > >                    |             
> > > >                    |             +-----+
> > > >                    +-------------| ASP2|
> > > >                                  +-----+
> > > > 
> > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > 
> > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > But SEP status is not changed in ASP2.There are two ways to solve
> > > > this problem
> > > > 
> > > > 1. ASP1 can equalize its database with ASP2
> > > > or
> > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > 
> > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or
> > > > inact
> > > > state)
> > > > 
> > > > Any comments?
> > > > 
> > > > -----
> > > > 
> > > > section 5.4.2.1
> > > > 
> > > > ...
> > > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> > > >         concerned Network Appearance.
> > > > 
> > > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER 
> > > > request
> > 
> > > > is sent to MTP3.
> > > > 
> > > > But there is no primitive availabe between M3UA and MTP3.
> > > > 
> > > > I think it should be corrected.
> > > > 
> > > > ----
> > > > Presently ASP is unaware of SG's MTP3 restart,
> > > > ASP receives only DUNA for all SS7 point codes which can be 
> > > > reachable via that SG.But in case of peer to peer protocol SG and 
> > > > AS point codes are different, So why dont we send DUNA with SG's 
> > > > point code to ASP? So that just by getting single DUNA(with SG's 
> > > > PC) ASP will not route any message via that
> > SG.
> > > > This can avoid sending multiple DUNAs to ASP.
> > > > 
> > > > 
> > > > Any comments?
> > > > 
> > > > -----
> > > > section 1.4.1
> > > > 
> > > > ...
> > > >    Where a particular AS can be reached via more than one SGP, the
> > > >    corresponding Routing Keys in the SGPs should be identical.
> > (Note:
> > > >    It is possible for the SGP Routing Key configuration data to be
> > > >    temporarily out-of-sync during configuration updates).
> > > >  
> > > >                               +--------+
> > > >                               |        |
> > > >                  +------------+  SG 1  +--------------+
> > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > >      |  SEP  +---+            +--------+              +---/      \
> > > >      |   or  |                    |*                      | ASPs
> |
> > > >      |  STP  +---+            +--------+              +---\      /
> > > >      +-------+   |            |        |              |     ----
> > > >                  +------------+  SG 2  +--------------+
> > > >                               | "STP"  |
> > > >                               +--------+
> > > >  
> > > >                     Figure 1  Example with mated SGs
> > > > 
> > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are
> > > > located geographically in different places.The ideas is when SG1 
> > > > goes down in CITY1(due to earthquake...) , all the traffice still 
> > > > can be handled by SG2 which is in CITY2.
> > > > 
> > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will 
> > > > be
> > > > routed from SGP1 to SGP2.
> > > > 
> > > > Does all these things mean to say that SGP1's mate 'should be'
> > > > present in SG2 and not in SG1 itself ?
> > > >  
> > > > 
> > > > Comments are welcome.
> > > > 
> > > > --Samuel
> > > > 
> > > > 
> > > > 
> > > > 
> > > >   Alcatel USA, Inc.                  Internet:
> > > <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, 
> > > > Inc.
> > > *******
> > > > 
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > 
> > > 
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 03:35:00 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08453
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 03:35:00 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h168fJQ12112
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 03:41:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168f5p12103;
	Thu, 6 Feb 2003 03:41:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168egp12068
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 03:40:42 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08422
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 03:33:49 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:RaXi4z4OiLVAO2gbDXvyJfnm3OXWXVCT@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h168bQD13901;
	Thu, 6 Feb 2003 01:37:26 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h168bQ415801;
	Thu, 6 Feb 2003 01:37:26 -0700
Date: Thu, 6 Feb 2003 01:37:26 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206013726.D14210@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206011513.A12848@openss7.org> <Pine.SOL.4.10.10302060227250.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302060227250.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 02:32:30AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 02:32:30)
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Zannin,,
> > 
> > That is contrary to the conensus of the RFC.
> > 
> 
> What is contrary and what is conensus?
> RFC always welcomes all modifications.

All concensus modifications.

> 
> > Consider 1 ASP active, 200,000 ASPs inactive.
> > 
> > The RFC cannot require an SG to send SSNM to an inactive ASP.
> > 
> 
> Consider 200,000 nodes with SG1 and if SG1 restarts, Will you send
> 200,000 DUNA after 200,000 DAUD.

No, because 200,000 of them are inactive and don't need to send DAUD.

--brian

> 
> --Samuel
> > --brian
> > 
> > Zannin, Francois wrote:                                                            (Thu, 06 Feb 2003 09:02:41)
> > > Samuel,
> > > 
> > > IMO, it's a good idea to inform inactive (BUT UP) ASP of the state of
> > > the ss7 network especially in the override mode where an ASP can be
> > > ready to recover an AS as soon as it receive an AS-Pending notification
> > > from the SGp. In this condition, the ASP is ready to operate quickly and
> > > minimize the risk of buffered message lost during the pending timer.
> > > 
> > > If someone does not want to receive SSNM because it is not able to
> > > recover an AS, it just stays in an ASP-Down state.
> > > 
> > > Because SSNM reflects ss7 Point Code state that is not "SI" dependant,
> > > it affects all AS for a network and then, unless SG that plays multiple
> > > networks, the ASP UP is sufficient to filter the undesired SSNM.
> > > 
> > > Francois.
> > > 
> > > -----Original Message-----
> > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> > > Sent: Thursday, February 06, 2003 8:35 AM
> > > To: Nagar, Ashish
> > > Cc: sigtran@ietf.org
> > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > 
> > > Ashish,
> > > 
> > > But when inactive ASP changes state to active, first of all
> > > it does not know whether to send DAUD or not.
> > > 
> > > It will try to send traffic - but actually destination is unavailable.
> > > 
> > > The worst case is when SG triggers restart procedures this case all the
> > > SS7 destinations are unavailable via restarted-SG.But this poor ASP
> > > (newly come-up ASP) keeps sending traffic via restarted-SG to reach the
> > > SS7 destination and for every message ASP receives DUNA.
> > > 
> > > Are we not doing double-work here?
> > > 
> > > --Samuel
> > > 
> > > 
> > >   Alcatel USA, Inc.                  Internet:
> > > <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > *******
> > > 
> > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > 
> > > > Hi Samuel
> > > > 
> > > > My point is why to do any processing if ASP is in Inactive state. 
> > > > Think about ASP is in Inactive state for longer period and receiving 
> > > > lots of DUNA and DAVA, while what matter is the present only when ASP 
> > > > state changes to ASP-ACTIVE.
> > > > 
> > > > Thanks & Regards
> > > > Ashish Nagar
> > > > 
> > > > 
> > > > -----Original Message-----
> > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > Sent: Thursday, February 06, 2003 12:30 PM
> > > > To: Nagar, Ashish
> > > > Cc: sigtran@ietf.org
> > > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > > 
> > > > 
> > > > Ashish,
> > > > 
> > > > My point is if SG is able to send DUNA to all inact and active ASPs, 
> > > > we should not restrict it.
> > > > 
> > > > Sending DAUD is an alternative way ,If SG has direct way of sending 
> > > > DUNA then Why do you need alternative way?
> > > > 
> > > > --Samuel
> > > > 
> > > >   Alcatel USA, Inc.                  Internet:
> > > <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. 
> > > > *******
> > > > 
> > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > 
> > > > > You can generate DAUD procedure at any time, so whenever ASP changes
> > > 
> > > > > state from Inactive to Active Just generate DAUD.
> > > > > 
> > > > > Thanks & Regards
> > > > > Ashish Nagar
> > > > > 
> > > > > 
> > > > > -----Original Message-----
> > > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > > To: sigtran@ietf.org
> > > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > > 
> > > > > 
> > > > > Hi all,
> > > > > 
> > > > > Currently RFC 3332 says that SG sends SSNM messages to all 
> > > > > 'concerned ASPs'(means only ACTIVE ASPs).
> > > > > 
> > > > > +----+           +---+           +-----+
> > > > > |SEP | ----------|SG |-----------| ASP1|
> > > > > +----+           +---+           +-----+
> > > > >                    |             
> > > > >                    |             +-----+
> > > > >                    +-------------| ASP2|
> > > > >                                  +-----+
> > > > > 
> > > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > > 
> > > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > > But SEP status is not changed in ASP2.There are two ways to solve 
> > > > > this problem
> > > > > 
> > > > > 1. ASP1 can equalize its database with ASP2
> > > > > or
> > > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > > 
> > > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or 
> > > > > inact
> > > > > state)
> > > > > 
> > > > > Any comments?
> > > > > 
> > > > > -----
> > > > > 
> > > > > section 5.4.2.1
> > > > > 
> > > > > ...
> > > > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> > > > >         concerned Network Appearance.
> > > > > 
> > > > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER request
> > > 
> > > > > is sent to MTP3.
> > > > > 
> > > > > But there is no primitive availabe between M3UA and MTP3.
> > > > > 
> > > > > I think it should be corrected.
> > > > > 
> > > > > ----
> > > > > Presently ASP is unaware of SG's MTP3 restart,
> > > > > ASP receives only DUNA for all SS7 point codes which can be
> > > > > reachable via that SG.But in case of peer to peer protocol
> > > > > SG and AS point codes are different, So why dont we send
> > > > > DUNA with SG's point code to ASP? So that just by getting
> > > > > single DUNA(with SG's PC) ASP will not route any message via that
> > > SG.
> > > > > This can avoid sending multiple DUNAs to ASP.
> > > > > 
> > > > > 
> > > > > Any comments?
> > > > > 
> > > > > -----
> > > > > section 1.4.1
> > > > > 
> > > > > ...
> > > > >    Where a particular AS can be reached via more than one SGP, the
> > > > >    corresponding Routing Keys in the SGPs should be identical.
> > > (Note:
> > > > >    It is possible for the SGP Routing Key configuration data to be
> > > > >    temporarily out-of-sync during configuration updates).
> > > > >  
> > > > >                               +--------+
> > > > >                               |        |
> > > > >                  +------------+  SG 1  +--------------+
> > > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > > >      |  SEP  +---+            +--------+              +---/      \
> > > > >      |   or  |                    |*                      | ASPs  |
> > > > >      |  STP  +---+            +--------+              +---\      /
> > > > >      +-------+   |            |        |              |     ----
> > > > >                  +------------+  SG 2  +--------------+
> > > > >                               | "STP"  |
> > > > >                               +--------+
> > > > >  
> > > > >                     Figure 1  Example with mated SGs
> > > > > 
> > > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are 
> > > > > located geographically in different places.The ideas is when SG1 
> > > > > goes down in CITY1(due to earthquake...) , all the traffice still 
> > > > > can be handled by SG2 which is in CITY2.
> > > > > 
> > > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will be 
> > > > > routed from SGP1 to SGP2.
> > > > > 
> > > > > Does all these things mean to say that SGP1's mate 'should be' 
> > > > > present in SG2 and not in SG1 itself ?
> > > > >  
> > > > > 
> > > > > Comments are welcome.
> > > > > 
> > > > > --Samuel
> > > > > 
> > > > > 
> > > > > 
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet:
> > > > <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > > *******
> > > > > 
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > 
> > > > 
> > > 
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 03:39:03 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08561
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 03:39:03 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h168jOo12264
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 03:45:24 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168j8p12251;
	Thu, 6 Feb 2003 03:45:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168iIp12210
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 03:44:19 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08531
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 03:37:26 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h168f2eY019531;
	Thu, 6 Feb 2003 02:41:03 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h168eSM01958;
	Thu, 6 Feb 2003 02:40:28 -0600 (CST)
Date: Thu, 6 Feb 2003 02:40:28 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206013036.B14210@openss7.org>
Message-ID: <Pine.SOL.4.10.10302060234360.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>



Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 02:26:26)
> > 
> > Brian,
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 01:34:49)
> > > > 
> > > > Ashish,
> > > > 
> > > > But when inactive ASP changes state to active, first of all
> > > > it does not know whether to send DAUD or not.
> > > 
> > > It simply sends DAUD whenever it goes active.
> > > 
> > > > 
> > > > It will try to send traffic - but actually destination is unavailable.
> > > 
> > > No, becuase it will send DAUD.
> > 
> > It is not mandatory to send DAUD when ASP becomes ACTIVE.
> > 
> > see section 4.5.3
> > 
> > ...
> >       - Isolation. The ASP is newly ASP-ACTIVE or has been
> >                    isolated from an SGP for an extended period.  The ASP
> >                    MAY request the availability/congestion status of one
> >                    or more SS7 destinations to which it expects to
> >                    communicate.
> > 
> > Note the word "MAY".
> 
> So don't.
Again i have to start from beginning. newly active ASP sends traffic
to SS7 destination via re-started SG. But poor ASP does not know that all
SS7 destinations are unavailable, So he keeps sending traffice and DAUD.
I think this very bad idea.

--Samuel



> 
> > 
> > > 
> > > > 
> > > > The worst case is when SG triggers restart procedures this case all the
> > > > SS7 destinations are unavailable via restarted-SG.But this poor ASP
> > > > (newly come-up ASP) keeps sending traffic via restarted-SG to reach the
> > > > SS7 destination and for every message ASP receives DUNA.
> > > 
> > > It does not send traffic until DAUD is has completed because it does
> > > not send MTP-RESUME or MTP-RESTART-ENDS to the user until DAUD completes.
> > > 
> > 
> > Finally you missed my point, you are providing indirect solution.
> > My point is avoid sending DAUD to SG because SG has got capability
> > to send DUNA to inactive ASPs.Nothing will prevent SG to send DUNA.
> 
> You missed my point.  There is no MTP user until the ASP is active.
> 
> > 
> > > > 
> > > > Are we not doing double-work here?
> > > 
> > > 
> > > You have not considered my example where the ASP is down.
> > 
> > Again misunderstanding.I said when the ASP is in inactive state it can
> > receive DUNA.
> 
> But it can do nothing with it, so it is useless to send SSNM to inactive
> ASPs.
> 
No it is logical.

--Samuel
> --brian
> 
> > 
> > --Samuel
> > 
> > > 
> > > --brian
> > > 
> > > > 
> > > > --Samuel
> > > > 
> > > > 
> > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > 
> > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > 
> > > > > Hi Samuel
> > > > > 
> > > > > My point is why to do any processing if ASP is in Inactive state.
> > > > > Think about ASP is in Inactive state for longer period and receiving lots of
> > > > > DUNA and DAVA, while what matter is the present only when ASP state changes
> > > > > to ASP-ACTIVE.
> > > > > 
> > > > > Thanks & Regards
> > > > > Ashish Nagar
> > > > > 
> > > > > 
> > > > > -----Original Message-----
> > > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> > > > > Sent: Thursday, February 06, 2003 12:30 PM
> > > > > To: Nagar, Ashish
> > > > > Cc: sigtran@ietf.org
> > > > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > > > 
> > > > > 
> > > > > Ashish,
> > > > > 
> > > > > My point is if SG is able to send DUNA to all inact and active ASPs,
> > > > > we should not restrict it.
> > > > > 
> > > > > Sending DAUD is an alternative way ,If SG has direct way of sending
> > > > > DUNA then Why do you need alternative way?
> > > > > 
> > > > > --Samuel
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > > 
> > > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > > 
> > > > > > You can generate DAUD procedure at any time, so whenever ASP changes state
> > > > > > from Inactive to Active Just generate DAUD.
> > > > > > 
> > > > > > Thanks & Regards
> > > > > > Ashish Nagar
> > > > > > 
> > > > > > 
> > > > > > -----Original Message-----
> > > > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> > > > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > > > To: sigtran@ietf.org
> > > > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > > > 
> > > > > > 
> > > > > > Hi all,
> > > > > > 
> > > > > > Currently RFC 3332 says that SG sends SSNM messages to all 'concerned
> > > > > > ASPs'(means only ACTIVE ASPs).
> > > > > > 
> > > > > > +----+           +---+           +-----+
> > > > > > |SEP | ----------|SG |-----------| ASP1|
> > > > > > +----+           +---+           +-----+
> > > > > >                    |             
> > > > > >                    |             +-----+
> > > > > >                    +-------------| ASP2|
> > > > > >                                  +-----+
> > > > > > 
> > > > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > > > 
> > > > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > > > But SEP status is not changed in ASP2.There are two ways to solve this
> > > > > > problem
> > > > > > 
> > > > > > 1. ASP1 can equalize its database with ASP2
> > > > > > or
> > > > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > > > 
> > > > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or inact
> > > > > > state)
> > > > > > 
> > > > > > Any comments?
> > > > > > 
> > > > > > -----
> > > > > > 
> > > > > > section 5.4.2.1
> > > > > > 
> > > > > > ...
> > > > > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> > > > > >         concerned Network Appearance.
> > > > > > 
> > > > > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER request is
> > > > > > sent to MTP3.
> > > > > > 
> > > > > > But there is no primitive availabe between M3UA and MTP3.
> > > > > > 
> > > > > > I think it should be corrected.
> > > > > > 
> > > > > > ----
> > > > > > Presently ASP is unaware of SG's MTP3 restart, 
> > > > > > ASP receives only DUNA for all SS7 point codes which can be
> > > > > > reachable via that SG.But in case of peer to peer protocol
> > > > > > SG and AS point codes are different, So why dont we send
> > > > > > DUNA with SG's point code to ASP? So that just by getting
> > > > > > single DUNA(with SG's PC) ASP will not route any message via that SG.
> > > > > > This can avoid sending multiple DUNAs to ASP.
> > > > > > 
> > > > > > 
> > > > > > Any comments?
> > > > > > 
> > > > > > -----
> > > > > > section 1.4.1
> > > > > > 
> > > > > > ...
> > > > > >    Where a particular AS can be reached via more than one SGP, the
> > > > > >    corresponding Routing Keys in the SGPs should be identical.  (Note:
> > > > > >    It is possible for the SGP Routing Key configuration data to be
> > > > > >    temporarily out-of-sync during configuration updates).
> > > > > >  
> > > > > >                               +--------+
> > > > > >                               |        |
> > > > > >                  +------------+  SG 1  +--------------+
> > > > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > > > >      |  SEP  +---+            +--------+              +---/      \
> > > > > >      |   or  |                    |*                      | ASPs  |
> > > > > >      |  STP  +---+            +--------+              +---\      /
> > > > > >      +-------+   |            |        |              |     ----
> > > > > >                  +------------+  SG 2  +--------------+
> > > > > >                               | "STP"  |
> > > > > >                               +--------+
> > > > > >  
> > > > > >                     Figure 1  Example with mated SGs
> > > > > > 
> > > > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are located
> > > > > > geographically in different places.The ideas is when SG1 goes down in
> > > > > > CITY1(due to earthquake...) , all the traffice still can be handled by SG2
> > > > > > which is in CITY2.
> > > > > > 
> > > > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will be
> > > > > > routed from SGP1 to SGP2.
> > > > > > 
> > > > > > Does all these things mean to say that SGP1's mate 'should be' present in
> > > > > > SG2 and not in SG1 itself ?
> > > > > >  
> > > > > > 
> > > > > > Comments are welcome.
> > > > > > 
> > > > > > --Samuel
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > >   Alcatel USA, Inc.                  Internet:
> > > > > <userid>@ssd.usa.alcatel.com
> > > > > >   1000 Coit Road, Plano, Texas 75075
> > > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > > > *******
> > > > > > 
> > > > > > _______________________________________________
> > > > > > Sigtran mailing list
> > > > > > Sigtran@ietf.org
> > > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > > _______________________________________________
> > > > > > Sigtran mailing list
> > > > > > Sigtran@ietf.org
> > > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > > 
> > > > > 
> > > > 
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> > > -- 
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 03:43:58 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08663
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 03:43:58 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h168oJT12456
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 03:50:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168o4p12448;
	Thu, 6 Feb 2003 03:50:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168nRp12411
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 03:49:27 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08649
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 03:42:34 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h168k2ew019905;
	Thu, 6 Feb 2003 02:46:11 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h168j8M03550;
	Thu, 6 Feb 2003 02:45:08 -0600 (CST)
Date: Thu, 6 Feb 2003 02:45:08 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206013726.D14210@openss7.org>
Message-ID: <Pine.SOL.4.10.10302060242030.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,


  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 02:32:30)
> > 
> > Brian,
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > 
> > > Zannin,,
> > > 
> > > That is contrary to the conensus of the RFC.
> > > 
> > 
> > What is contrary and what is conensus?
> > RFC always welcomes all modifications.
> 
> All concensus modifications.
> 
> > 
> > > Consider 1 ASP active, 200,000 ASPs inactive.
> > > 
> > > The RFC cannot require an SG to send SSNM to an inactive ASP.
> > > 
> > 
> > Consider 200,000 nodes with SG1 and if SG1 restarts, Will you send
> > 200,000 DUNA after 200,000 DAUD.
> 
> No, because 200,000 of them are inactive and don't need to send DAUD.

Misinterpretation. I said 200,000 of them are inactive with respect to
ASP1 but ASP2 thinks 200,000 of them are active for ASP2.

All our mail discussions are withrespect to inactive ASP.


--Samuel
> 
> --brian
> 
> > 
> > --Samuel
> > > --brian
> > > 
> > > Zannin, Francois wrote:                                                            (Thu, 06 Feb 2003 09:02:41)
> > > > Samuel,
> > > > 
> > > > IMO, it's a good idea to inform inactive (BUT UP) ASP of the state of
> > > > the ss7 network especially in the override mode where an ASP can be
> > > > ready to recover an AS as soon as it receive an AS-Pending notification
> > > > from the SGp. In this condition, the ASP is ready to operate quickly and
> > > > minimize the risk of buffered message lost during the pending timer.
> > > > 
> > > > If someone does not want to receive SSNM because it is not able to
> > > > recover an AS, it just stays in an ASP-Down state.
> > > > 
> > > > Because SSNM reflects ss7 Point Code state that is not "SI" dependant,
> > > > it affects all AS for a network and then, unless SG that plays multiple
> > > > networks, the ASP UP is sufficient to filter the undesired SSNM.
> > > > 
> > > > Francois.
> > > > 
> > > > -----Original Message-----
> > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> > > > Sent: Thursday, February 06, 2003 8:35 AM
> > > > To: Nagar, Ashish
> > > > Cc: sigtran@ietf.org
> > > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > > 
> > > > 
> > > > 
> > > > Ashish,
> > > > 
> > > > But when inactive ASP changes state to active, first of all
> > > > it does not know whether to send DAUD or not.
> > > > 
> > > > It will try to send traffic - but actually destination is unavailable.
> > > > 
> > > > The worst case is when SG triggers restart procedures this case all the
> > > > SS7 destinations are unavailable via restarted-SG.But this poor ASP
> > > > (newly come-up ASP) keeps sending traffic via restarted-SG to reach the
> > > > SS7 destination and for every message ASP receives DUNA.
> > > > 
> > > > Are we not doing double-work here?
> > > > 
> > > > --Samuel
> > > > 
> > > > 
> > > >   Alcatel USA, Inc.                  Internet:
> > > > <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > > *******
> > > > 
> > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > 
> > > > > Hi Samuel
> > > > > 
> > > > > My point is why to do any processing if ASP is in Inactive state. 
> > > > > Think about ASP is in Inactive state for longer period and receiving 
> > > > > lots of DUNA and DAVA, while what matter is the present only when ASP 
> > > > > state changes to ASP-ACTIVE.
> > > > > 
> > > > > Thanks & Regards
> > > > > Ashish Nagar
> > > > > 
> > > > > 
> > > > > -----Original Message-----
> > > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > > Sent: Thursday, February 06, 2003 12:30 PM
> > > > > To: Nagar, Ashish
> > > > > Cc: sigtran@ietf.org
> > > > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > > > 
> > > > > 
> > > > > Ashish,
> > > > > 
> > > > > My point is if SG is able to send DUNA to all inact and active ASPs, 
> > > > > we should not restrict it.
> > > > > 
> > > > > Sending DAUD is an alternative way ,If SG has direct way of sending 
> > > > > DUNA then Why do you need alternative way?
> > > > > 
> > > > > --Samuel
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet:
> > > > <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. 
> > > > > *******
> > > > > 
> > > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > > 
> > > > > > You can generate DAUD procedure at any time, so whenever ASP changes
> > > > 
> > > > > > state from Inactive to Active Just generate DAUD.
> > > > > > 
> > > > > > Thanks & Regards
> > > > > > Ashish Nagar
> > > > > > 
> > > > > > 
> > > > > > -----Original Message-----
> > > > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > > > To: sigtran@ietf.org
> > > > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > > > 
> > > > > > 
> > > > > > Hi all,
> > > > > > 
> > > > > > Currently RFC 3332 says that SG sends SSNM messages to all 
> > > > > > 'concerned ASPs'(means only ACTIVE ASPs).
> > > > > > 
> > > > > > +----+           +---+           +-----+
> > > > > > |SEP | ----------|SG |-----------| ASP1|
> > > > > > +----+           +---+           +-----+
> > > > > >                    |             
> > > > > >                    |             +-----+
> > > > > >                    +-------------| ASP2|
> > > > > >                                  +-----+
> > > > > > 
> > > > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > > > 
> > > > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > > > But SEP status is not changed in ASP2.There are two ways to solve 
> > > > > > this problem
> > > > > > 
> > > > > > 1. ASP1 can equalize its database with ASP2
> > > > > > or
> > > > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > > > 
> > > > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or 
> > > > > > inact
> > > > > > state)
> > > > > > 
> > > > > > Any comments?
> > > > > > 
> > > > > > -----
> > > > > > 
> > > > > > section 5.4.2.1
> > > > > > 
> > > > > > ...
> > > > > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> > > > > >         concerned Network Appearance.
> > > > > > 
> > > > > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER request
> > > > 
> > > > > > is sent to MTP3.
> > > > > > 
> > > > > > But there is no primitive availabe between M3UA and MTP3.
> > > > > > 
> > > > > > I think it should be corrected.
> > > > > > 
> > > > > > ----
> > > > > > Presently ASP is unaware of SG's MTP3 restart,
> > > > > > ASP receives only DUNA for all SS7 point codes which can be
> > > > > > reachable via that SG.But in case of peer to peer protocol
> > > > > > SG and AS point codes are different, So why dont we send
> > > > > > DUNA with SG's point code to ASP? So that just by getting
> > > > > > single DUNA(with SG's PC) ASP will not route any message via that
> > > > SG.
> > > > > > This can avoid sending multiple DUNAs to ASP.
> > > > > > 
> > > > > > 
> > > > > > Any comments?
> > > > > > 
> > > > > > -----
> > > > > > section 1.4.1
> > > > > > 
> > > > > > ...
> > > > > >    Where a particular AS can be reached via more than one SGP, the
> > > > > >    corresponding Routing Keys in the SGPs should be identical.
> > > > (Note:
> > > > > >    It is possible for the SGP Routing Key configuration data to be
> > > > > >    temporarily out-of-sync during configuration updates).
> > > > > >  
> > > > > >                               +--------+
> > > > > >                               |        |
> > > > > >                  +------------+  SG 1  +--------------+
> > > > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > > > >      |  SEP  +---+            +--------+              +---/      \
> > > > > >      |   or  |                    |*                      | ASPs  |
> > > > > >      |  STP  +---+            +--------+              +---\      /
> > > > > >      +-------+   |            |        |              |     ----
> > > > > >                  +------------+  SG 2  +--------------+
> > > > > >                               | "STP"  |
> > > > > >                               +--------+
> > > > > >  
> > > > > >                     Figure 1  Example with mated SGs
> > > > > > 
> > > > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are 
> > > > > > located geographically in different places.The ideas is when SG1 
> > > > > > goes down in CITY1(due to earthquake...) , all the traffice still 
> > > > > > can be handled by SG2 which is in CITY2.
> > > > > > 
> > > > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will be 
> > > > > > routed from SGP1 to SGP2.
> > > > > > 
> > > > > > Does all these things mean to say that SGP1's mate 'should be' 
> > > > > > present in SG2 and not in SG1 itself ?
> > > > > >  
> > > > > > 
> > > > > > Comments are welcome.
> > > > > > 
> > > > > > --Samuel
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > >   Alcatel USA, Inc.                  Internet:
> > > > > <userid>@ssd.usa.alcatel.com
> > > > > >   1000 Coit Road, Plano, Texas 75075
> > > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > > > *******
> > > > > > 
> > > > > > _______________________________________________
> > > > > > Sigtran mailing list
> > > > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > > _______________________________________________
> > > > > > Sigtran mailing list
> > > > > > Sigtran@ietf.org
> > > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > > 
> > > > > 
> > > > 
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> > > -- 
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > 
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 03:45:01 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08699
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 03:45:01 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h168pMq12515
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 03:51:22 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168p8p12506;
	Thu, 6 Feb 2003 03:51:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168oNp12473
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 03:50:23 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08659
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 03:43:30 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:woNxiBtLaP9nSuWULlF4t0SbZnrj9cq3@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h168l7D14023;
	Thu, 6 Feb 2003 01:47:07 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h168l7i17022;
	Thu, 6 Feb 2003 01:47:07 -0700
Date: Thu, 6 Feb 2003 01:47:07 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206014707.A15998@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206013036.B14210@openss7.org> <Pine.SOL.4.10.10302060234360.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302060234360.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 02:40:28AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

If an ASP sends data to an inaccessible desintation it will receive DUNA
in response.

Besides, destinations are almost never inaccessible, so it would be a
good assumption on the ASPs part to assume that the destination is
acessible if it doesn't want to send one message (DAUD) before sending
many messages.  Which seems rather foolish to me.  There is a solution:
if the ASP requires to know the status of destinations when it goes
active, it can send DAUD.  If it doesn't need or want to know it does
not send DAUD.  So what.  It still doesn't need to receive SSNM while
inactive.  It only needs to receive them once it becomes active.

This is no different than the MTP/MTP-User interface.  When an MTP-User
attaches to an MTP-SAP, how it obtains information about inaccessible
destinations is implementation dependent.  Either the MTP User can ask
for them or the MTP can immediately provide them (DAUD) or it can simply
send data, relying on MTP responsive procedures to MTP users to send
MTP-PAUSE.  M3UA is identical to this.  MTP never sends MTP-PAUSE when
there is no user to send it to.  SG never sends SSNM to inactive ASPs.

--brian


Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 02:40:28)
> 
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 02:26:26)
> > > 
> > > Brian,
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Samuel,
> > > > 
> > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 01:34:49)
> > > > > 
> > > > > Ashish,
> > > > > 
> > > > > But when inactive ASP changes state to active, first of all
> > > > > it does not know whether to send DAUD or not.
> > > > 
> > > > It simply sends DAUD whenever it goes active.
> > > > 
> > > > > 
> > > > > It will try to send traffic - but actually destination is unavailable.
> > > > 
> > > > No, becuase it will send DAUD.
> > > 
> > > It is not mandatory to send DAUD when ASP becomes ACTIVE.
> > > 
> > > see section 4.5.3
> > > 
> > > ...
> > >       - Isolation. The ASP is newly ASP-ACTIVE or has been
> > >                    isolated from an SGP for an extended period.  The ASP
> > >                    MAY request the availability/congestion status of one
> > >                    or more SS7 destinations to which it expects to
> > >                    communicate.
> > > 
> > > Note the word "MAY".
> > 
> > So don't.
> Again i have to start from beginning. newly active ASP sends traffic
> to SS7 destination via re-started SG. But poor ASP does not know that all
> SS7 destinations are unavailable, So he keeps sending traffice and DAUD.
> I think this very bad idea.
> 
> --Samuel
> 
> 
> 
> > 
> > > 
> > > > 
> > > > > 
> > > > > The worst case is when SG triggers restart procedures this case all the
> > > > > SS7 destinations are unavailable via restarted-SG.But this poor ASP
> > > > > (newly come-up ASP) keeps sending traffic via restarted-SG to reach the
> > > > > SS7 destination and for every message ASP receives DUNA.
> > > > 
> > > > It does not send traffic until DAUD is has completed because it does
> > > > not send MTP-RESUME or MTP-RESTART-ENDS to the user until DAUD completes.
> > > > 
> > > 
> > > Finally you missed my point, you are providing indirect solution.
> > > My point is avoid sending DAUD to SG because SG has got capability
> > > to send DUNA to inactive ASPs.Nothing will prevent SG to send DUNA.
> > 
> > You missed my point.  There is no MTP user until the ASP is active.
> > 
> > > 
> > > > > 
> > > > > Are we not doing double-work here?
> > > > 
> > > > 
> > > > You have not considered my example where the ASP is down.
> > > 
> > > Again misunderstanding.I said when the ASP is in inactive state it can
> > > receive DUNA.
> > 
> > But it can do nothing with it, so it is useless to send SSNM to inactive
> > ASPs.
> > 
> No it is logical.
> 
> --Samuel
> > --brian
> > 
> > > 
> > > --Samuel
> > > 
> > > > 
> > > > --brian
> > > > 
> > > > > 
> > > > > --Samuel
> > > > > 
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > > 
> > > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > > 
> > > > > > Hi Samuel
> > > > > > 
> > > > > > My point is why to do any processing if ASP is in Inactive state.
> > > > > > Think about ASP is in Inactive state for longer period and receiving lots of
> > > > > > DUNA and DAVA, while what matter is the present only when ASP state changes
> > > > > > to ASP-ACTIVE.
> > > > > > 
> > > > > > Thanks & Regards
> > > > > > Ashish Nagar
> > > > > > 
> > > > > > 
> > > > > > -----Original Message-----
> > > > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> > > > > > Sent: Thursday, February 06, 2003 12:30 PM
> > > > > > To: Nagar, Ashish
> > > > > > Cc: sigtran@ietf.org
> > > > > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > > > > 
> > > > > > 
> > > > > > Ashish,
> > > > > > 
> > > > > > My point is if SG is able to send DUNA to all inact and active ASPs,
> > > > > > we should not restrict it.
> > > > > > 
> > > > > > Sending DAUD is an alternative way ,If SG has direct way of sending
> > > > > > DUNA then Why do you need alternative way?
> > > > > > 
> > > > > > --Samuel
> > > > > > 
> > > > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > > >   1000 Coit Road, Plano, Texas 75075
> > > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > > > 
> > > > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > > > 
> > > > > > > You can generate DAUD procedure at any time, so whenever ASP changes state
> > > > > > > from Inactive to Active Just generate DAUD.
> > > > > > > 
> > > > > > > Thanks & Regards
> > > > > > > Ashish Nagar
> > > > > > > 
> > > > > > > 
> > > > > > > -----Original Message-----
> > > > > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> > > > > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > > > > To: sigtran@ietf.org
> > > > > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > > > > 
> > > > > > > 
> > > > > > > Hi all,
> > > > > > > 
> > > > > > > Currently RFC 3332 says that SG sends SSNM messages to all 'concerned
> > > > > > > ASPs'(means only ACTIVE ASPs).
> > > > > > > 
> > > > > > > +----+           +---+           +-----+
> > > > > > > |SEP | ----------|SG |-----------| ASP1|
> > > > > > > +----+           +---+           +-----+
> > > > > > >                    |             
> > > > > > >                    |             +-----+
> > > > > > >                    +-------------| ASP2|
> > > > > > >                                  +-----+
> > > > > > > 
> > > > > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > > > > 
> > > > > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > > > > But SEP status is not changed in ASP2.There are two ways to solve this
> > > > > > > problem
> > > > > > > 
> > > > > > > 1. ASP1 can equalize its database with ASP2
> > > > > > > or
> > > > > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > > > > 
> > > > > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or inact
> > > > > > > state)
> > > > > > > 
> > > > > > > Any comments?
> > > > > > > 
> > > > > > > -----
> > > > > > > 
> > > > > > > section 5.4.2.1
> > > > > > > 
> > > > > > > ...
> > > > > > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> > > > > > >         concerned Network Appearance.
> > > > > > > 
> > > > > > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER request is
> > > > > > > sent to MTP3.
> > > > > > > 
> > > > > > > But there is no primitive availabe between M3UA and MTP3.
> > > > > > > 
> > > > > > > I think it should be corrected.
> > > > > > > 
> > > > > > > ----
> > > > > > > Presently ASP is unaware of SG's MTP3 restart, 
> > > > > > > ASP receives only DUNA for all SS7 point codes which can be
> > > > > > > reachable via that SG.But in case of peer to peer protocol
> > > > > > > SG and AS point codes are different, So why dont we send
> > > > > > > DUNA with SG's point code to ASP? So that just by getting
> > > > > > > single DUNA(with SG's PC) ASP will not route any message via that SG.
> > > > > > > This can avoid sending multiple DUNAs to ASP.
> > > > > > > 
> > > > > > > 
> > > > > > > Any comments?
> > > > > > > 
> > > > > > > -----
> > > > > > > section 1.4.1
> > > > > > > 
> > > > > > > ...
> > > > > > >    Where a particular AS can be reached via more than one SGP, the
> > > > > > >    corresponding Routing Keys in the SGPs should be identical.  (Note:
> > > > > > >    It is possible for the SGP Routing Key configuration data to be
> > > > > > >    temporarily out-of-sync during configuration updates).
> > > > > > >  
> > > > > > >                               +--------+
> > > > > > >                               |        |
> > > > > > >                  +------------+  SG 1  +--------------+
> > > > > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > > > > >      |  SEP  +---+            +--------+              +---/      \
> > > > > > >      |   or  |                    |*                      | ASPs  |
> > > > > > >      |  STP  +---+            +--------+              +---\      /
> > > > > > >      +-------+   |            |        |              |     ----
> > > > > > >                  +------------+  SG 2  +--------------+
> > > > > > >                               | "STP"  |
> > > > > > >                               +--------+
> > > > > > >  
> > > > > > >                     Figure 1  Example with mated SGs
> > > > > > > 
> > > > > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are located
> > > > > > > geographically in different places.The ideas is when SG1 goes down in
> > > > > > > CITY1(due to earthquake...) , all the traffice still can be handled by SG2
> > > > > > > which is in CITY2.
> > > > > > > 
> > > > > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will be
> > > > > > > routed from SGP1 to SGP2.
> > > > > > > 
> > > > > > > Does all these things mean to say that SGP1's mate 'should be' present in
> > > > > > > SG2 and not in SG1 itself ?
> > > > > > >  
> > > > > > > 
> > > > > > > Comments are welcome.
> > > > > > > 
> > > > > > > --Samuel
> > > > > > > 
> > > > > > > 
> > > > > > > 
> > > > > > > 
> > > > > > >   Alcatel USA, Inc.                  Internet:
> > > > > > <userid>@ssd.usa.alcatel.com
> > > > > > >   1000 Coit Road, Plano, Texas 75075
> > > > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > > > > *******
> > > > > > > 
> > > > > > > _______________________________________________
> > > > > > > Sigtran mailing list
> > > > > > > Sigtran@ietf.org
> > > > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > > > _______________________________________________
> > > > > > > Sigtran mailing list
> > > > > > > Sigtran@ietf.org
> > > > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > > > 
> > > > > > 
> > > > > 
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > 
> > > > -- 
> > > > Brian F. G. Bidulock
> > > > bidulock@openss7.org
> > > > http://www.openss7.org/
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > 
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 03:46:58 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08754
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 03:46:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h168rIS12642
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 03:53:18 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168r3p12610;
	Thu, 6 Feb 2003 03:53:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168qHp12559
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 03:52:17 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08721
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 03:45:23 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:4oXHMpEx+d6NkRnMsiAodkbzNa7R2a5B@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h168n1D14047;
	Thu, 6 Feb 2003 01:49:01 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h168n0H17251;
	Thu, 6 Feb 2003 01:49:00 -0700
Date: Thu, 6 Feb 2003 01:49:00 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206014900.B15998@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206013726.D14210@openss7.org> <Pine.SOL.4.10.10302060242030.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302060242030.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 02:45:08AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 02:45:08)
> 
> Brian,
> 
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 02:32:30)
> > > 
> > > Brian,
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Zannin,,
> > > > 
> > > > That is contrary to the conensus of the RFC.
> > > > 
> > > 
> > > What is contrary and what is conensus?
> > > RFC always welcomes all modifications.
> > 
> > All concensus modifications.
> > 
> > > 
> > > > Consider 1 ASP active, 200,000 ASPs inactive.
> > > > 
> > > > The RFC cannot require an SG to send SSNM to an inactive ASP.
> > > > 
> > > 
> > > Consider 200,000 nodes with SG1 and if SG1 restarts, Will you send
> > > 200,000 DUNA after 200,000 DAUD.
> > 
> > No, because 200,000 of them are inactive and don't need to send DAUD.
> 
> Misinterpretation. I said 200,000 of them are inactive with respect to
> ASP1 but ASP2 thinks 200,000 of them are active for ASP2.
> 
> All our mail discussions are withrespect to inactive ASP.

You misunderstand, to 200,001 DUNA messages are for one destination:
sent to 200,001 ASPs, one of which is active and can use the message,
200,000 of which are inactive and can do nothing with the message.

--brian

> 
> 
> --Samuel
> > 
> > --brian
> > 
> > > 
> > > --Samuel
> > > > --brian
> > > > 
> > > > Zannin, Francois wrote:                                                            (Thu, 06 Feb 2003 09:02:41)
> > > > > Samuel,
> > > > > 
> > > > > IMO, it's a good idea to inform inactive (BUT UP) ASP of the state of
> > > > > the ss7 network especially in the override mode where an ASP can be
> > > > > ready to recover an AS as soon as it receive an AS-Pending notification
> > > > > from the SGp. In this condition, the ASP is ready to operate quickly and
> > > > > minimize the risk of buffered message lost during the pending timer.
> > > > > 
> > > > > If someone does not want to receive SSNM because it is not able to
> > > > > recover an AS, it just stays in an ASP-Down state.
> > > > > 
> > > > > Because SSNM reflects ss7 Point Code state that is not "SI" dependant,
> > > > > it affects all AS for a network and then, unless SG that plays multiple
> > > > > networks, the ASP UP is sufficient to filter the undesired SSNM.
> > > > > 
> > > > > Francois.
> > > > > 
> > > > > -----Original Message-----
> > > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> > > > > Sent: Thursday, February 06, 2003 8:35 AM
> > > > > To: Nagar, Ashish
> > > > > Cc: sigtran@ietf.org
> > > > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > > > 
> > > > > 
> > > > > 
> > > > > Ashish,
> > > > > 
> > > > > But when inactive ASP changes state to active, first of all
> > > > > it does not know whether to send DAUD or not.
> > > > > 
> > > > > It will try to send traffic - but actually destination is unavailable.
> > > > > 
> > > > > The worst case is when SG triggers restart procedures this case all the
> > > > > SS7 destinations are unavailable via restarted-SG.But this poor ASP
> > > > > (newly come-up ASP) keeps sending traffic via restarted-SG to reach the
> > > > > SS7 destination and for every message ASP receives DUNA.
> > > > > 
> > > > > Are we not doing double-work here?
> > > > > 
> > > > > --Samuel
> > > > > 
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet:
> > > > > <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > > > *******
> > > > > 
> > > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > > 
> > > > > > Hi Samuel
> > > > > > 
> > > > > > My point is why to do any processing if ASP is in Inactive state. 
> > > > > > Think about ASP is in Inactive state for longer period and receiving 
> > > > > > lots of DUNA and DAVA, while what matter is the present only when ASP 
> > > > > > state changes to ASP-ACTIVE.
> > > > > > 
> > > > > > Thanks & Regards
> > > > > > Ashish Nagar
> > > > > > 
> > > > > > 
> > > > > > -----Original Message-----
> > > > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > > > Sent: Thursday, February 06, 2003 12:30 PM
> > > > > > To: Nagar, Ashish
> > > > > > Cc: sigtran@ietf.org
> > > > > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > > > > 
> > > > > > 
> > > > > > Ashish,
> > > > > > 
> > > > > > My point is if SG is able to send DUNA to all inact and active ASPs, 
> > > > > > we should not restrict it.
> > > > > > 
> > > > > > Sending DAUD is an alternative way ,If SG has direct way of sending 
> > > > > > DUNA then Why do you need alternative way?
> > > > > > 
> > > > > > --Samuel
> > > > > > 
> > > > > >   Alcatel USA, Inc.                  Internet:
> > > > > <userid>@ssd.usa.alcatel.com
> > > > > >   1000 Coit Road, Plano, Texas 75075
> > > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. 
> > > > > > *******
> > > > > > 
> > > > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > > > 
> > > > > > > You can generate DAUD procedure at any time, so whenever ASP changes
> > > > > 
> > > > > > > state from Inactive to Active Just generate DAUD.
> > > > > > > 
> > > > > > > Thanks & Regards
> > > > > > > Ashish Nagar
> > > > > > > 
> > > > > > > 
> > > > > > > -----Original Message-----
> > > > > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > > > > To: sigtran@ietf.org
> > > > > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > > > > 
> > > > > > > 
> > > > > > > Hi all,
> > > > > > > 
> > > > > > > Currently RFC 3332 says that SG sends SSNM messages to all 
> > > > > > > 'concerned ASPs'(means only ACTIVE ASPs).
> > > > > > > 
> > > > > > > +----+           +---+           +-----+
> > > > > > > |SEP | ----------|SG |-----------| ASP1|
> > > > > > > +----+           +---+           +-----+
> > > > > > >                    |             
> > > > > > >                    |             +-----+
> > > > > > >                    +-------------| ASP2|
> > > > > > >                                  +-----+
> > > > > > > 
> > > > > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > > > > 
> > > > > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > > > > But SEP status is not changed in ASP2.There are two ways to solve 
> > > > > > > this problem
> > > > > > > 
> > > > > > > 1. ASP1 can equalize its database with ASP2
> > > > > > > or
> > > > > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > > > > 
> > > > > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or 
> > > > > > > inact
> > > > > > > state)
> > > > > > > 
> > > > > > > Any comments?
> > > > > > > 
> > > > > > > -----
> > > > > > > 
> > > > > > > section 5.4.2.1
> > > > > > > 
> > > > > > > ...
> > > > > > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> > > > > > >         concerned Network Appearance.
> > > > > > > 
> > > > > > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER request
> > > > > 
> > > > > > > is sent to MTP3.
> > > > > > > 
> > > > > > > But there is no primitive availabe between M3UA and MTP3.
> > > > > > > 
> > > > > > > I think it should be corrected.
> > > > > > > 
> > > > > > > ----
> > > > > > > Presently ASP is unaware of SG's MTP3 restart,
> > > > > > > ASP receives only DUNA for all SS7 point codes which can be
> > > > > > > reachable via that SG.But in case of peer to peer protocol
> > > > > > > SG and AS point codes are different, So why dont we send
> > > > > > > DUNA with SG's point code to ASP? So that just by getting
> > > > > > > single DUNA(with SG's PC) ASP will not route any message via that
> > > > > SG.
> > > > > > > This can avoid sending multiple DUNAs to ASP.
> > > > > > > 
> > > > > > > 
> > > > > > > Any comments?
> > > > > > > 
> > > > > > > -----
> > > > > > > section 1.4.1
> > > > > > > 
> > > > > > > ...
> > > > > > >    Where a particular AS can be reached via more than one SGP, the
> > > > > > >    corresponding Routing Keys in the SGPs should be identical.
> > > > > (Note:
> > > > > > >    It is possible for the SGP Routing Key configuration data to be
> > > > > > >    temporarily out-of-sync during configuration updates).
> > > > > > >  
> > > > > > >                               +--------+
> > > > > > >                               |        |
> > > > > > >                  +------------+  SG 1  +--------------+
> > > > > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > > > > >      |  SEP  +---+            +--------+              +---/      \
> > > > > > >      |   or  |                    |*                      | ASPs  |
> > > > > > >      |  STP  +---+            +--------+              +---\      /
> > > > > > >      +-------+   |            |        |              |     ----
> > > > > > >                  +------------+  SG 2  +--------------+
> > > > > > >                               | "STP"  |
> > > > > > >                               +--------+
> > > > > > >  
> > > > > > >                     Figure 1  Example with mated SGs
> > > > > > > 
> > > > > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are 
> > > > > > > located geographically in different places.The ideas is when SG1 
> > > > > > > goes down in CITY1(due to earthquake...) , all the traffice still 
> > > > > > > can be handled by SG2 which is in CITY2.
> > > > > > > 
> > > > > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will be 
> > > > > > > routed from SGP1 to SGP2.
> > > > > > > 
> > > > > > > Does all these things mean to say that SGP1's mate 'should be' 
> > > > > > > present in SG2 and not in SG1 itself ?
> > > > > > >  
> > > > > > > 
> > > > > > > Comments are welcome.
> > > > > > > 
> > > > > > > --Samuel
> > > > > > > 
> > > > > > > 
> > > > > > > 
> > > > > > > 
> > > > > > >   Alcatel USA, Inc.                  Internet:
> > > > > > <userid>@ssd.usa.alcatel.com
> > > > > > >   1000 Coit Road, Plano, Texas 75075
> > > > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > > > > *******
> > > > > > > 
> > > > > > > _______________________________________________
> > > > > > > Sigtran mailing list
> > > > > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > > > _______________________________________________
> > > > > > > Sigtran mailing list
> > > > > > > Sigtran@ietf.org
> > > > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > > > 
> > > > > > 
> > > > > 
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > 
> > > > -- 
> > > > Brian F. G. Bidulock
> > > > bidulock@openss7.org
> > > > http://www.openss7.org/
> > > > 
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 03:49:59 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08835
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 03:49:59 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h168uKP12822
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 03:56:20 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168u5p12794;
	Thu, 6 Feb 2003 03:56:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168tMp12752
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 03:55:22 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08803
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 03:48:29 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h168q2eW020483;
	Thu, 6 Feb 2003 02:52:02 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h168pqM05900;
	Thu, 6 Feb 2003 02:51:52 -0600 (CST)
Date: Thu, 6 Feb 2003 02:51:52 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206013542.C14210@openss7.org>
Message-ID: <Pine.SOL.4.10.10302060245350.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Zannin,,
> 
> Zannin, Francois wrote:                                                            (Thu, 06 Feb 2003 09:27:21)
> > Bidulock,,,
> > 
> > I think your example of 200,000 ASP "UP" as backup for an AS is really
> > realistic for a network!
> 
> 200?  20?  It makes no difference.  The SG cannot be required to
> send useless messages to bunch of ASPs that can do nothing with
> the information.
> 
Everything is useless at one point of time, we should see always
the 'usefullness of many times".



> > 
> > The goal is to agree on procedure that can be used for the
> > implementation of M3UA. There were a lot of threads discussing on this
> > subject and not always concluded. The role of the IG is to clarify the
> > RFC in order to make it usable.
> > 
> > This procedure similar to the ss7 network (TFP inform all the SP of the
> > inaccessibility). If a SG is unable to reach all the ss7 SP (200000 PC,
> > means 200000 ungrouped SSNM) it can use the DOWN_ACK or INACTIVE_ACK
> > procedure for its own isolation.
> 
> DUNA corresponds to MTP-PAUSE not TFP.
> 
> Consider 100 APSs inactive, zero (0) ASPs active.

I can say 100 ASPs inactive and 100 ASPs active with
override mode.

> 
> The SG cannot be required to send DUNA to inactive ASPs.  When the SG
> NIF has no user to send MTP-PAUSE to, it is not possible for the SG
> to send DUNA.
You are saying user-part is unavailabe at inact ASP.
But logically it is available.

> 
> The SG cannot be required to send SSNM to inactive ASPs.

It is required.Sending DUNA to inactive ASP is meaningfull.
 
> Besides, it is unnecessary.

I cant accept.

--Samuel

> 
> --brian
> 
> 
> > 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> > Sent: Thursday, February 06, 2003 9:15 AM
> > To: Zannin, Francois
> > Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> > Subject: Re: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > Zannin,,
> > 
> > That is contrary to the conensus of the RFC.
> > 
> > Consider 1 ASP active, 200,000 ASPs inactive.
> > 
> > The RFC cannot require an SG to send SSNM to an inactive ASP.
> > 
> > --brian
> > 
> > Zannin, Francois wrote:
> > (Thu, 06 Feb 2003 09:02:41)
> > > Samuel,
> > > 
> > > IMO, it's a good idea to inform inactive (BUT UP) ASP of the state of 
> > > the ss7 network especially in the override mode where an ASP can be 
> > > ready to recover an AS as soon as it receive an AS-Pending 
> > > notification from the SGp. In this condition, the ASP is ready to 
> > > operate quickly and minimize the risk of buffered message lost during 
> > > the pending timer.
> > > 
> > > If someone does not want to receive SSNM because it is not able to 
> > > recover an AS, it just stays in an ASP-Down state.
> > > 
> > > Because SSNM reflects ss7 Point Code state that is not "SI" dependant,
> > 
> > > it affects all AS for a network and then, unless SG that plays 
> > > multiple networks, the ASP UP is sufficient to filter the undesired 
> > > SSNM.
> > > 
> > > Francois.
> > > 
> > > -----Original Message-----
> > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > Sent: Thursday, February 06, 2003 8:35 AM
> > > To: Nagar, Ashish
> > > Cc: sigtran@ietf.org
> > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > 
> > > Ashish,
> > > 
> > > But when inactive ASP changes state to active, first of all it does 
> > > not know whether to send DAUD or not.
> > > 
> > > It will try to send traffic - but actually destination is unavailable.
> > > 
> > > The worst case is when SG triggers restart procedures this case all 
> > > the SS7 destinations are unavailable via restarted-SG.But this poor 
> > > ASP (newly come-up ASP) keeps sending traffic via restarted-SG to 
> > > reach the SS7 destination and for every message ASP receives DUNA.
> > > 
> > > Are we not doing double-work here?
> > > 
> > > --Samuel
> > > 
> > > 
> > >   Alcatel USA, Inc.                  Internet:
> > > <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > *******
> > > 
> > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > 
> > > > Hi Samuel
> > > > 
> > > > My point is why to do any processing if ASP is in Inactive state.
> > > > Think about ASP is in Inactive state for longer period and receiving
> > 
> > > > lots of DUNA and DAVA, while what matter is the present only when
> > ASP 
> > > > state changes to ASP-ACTIVE.
> > > > 
> > > > Thanks & Regards
> > > > Ashish Nagar
> > > > 
> > > > 
> > > > -----Original Message-----
> > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > Sent: Thursday, February 06, 2003 12:30 PM
> > > > To: Nagar, Ashish
> > > > Cc: sigtran@ietf.org
> > > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > > 
> > > > 
> > > > Ashish,
> > > > 
> > > > My point is if SG is able to send DUNA to all inact and active ASPs,
> > > > we should not restrict it.
> > > > 
> > > > Sending DAUD is an alternative way ,If SG has direct way of sending
> > > > DUNA then Why do you need alternative way?
> > > > 
> > > > --Samuel
> > > > 
> > > >   Alcatel USA, Inc.                  Internet:
> > > <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > > *******
> > > > 
> > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > 
> > > > > You can generate DAUD procedure at any time, so whenever ASP 
> > > > > changes
> > > 
> > > > > state from Inactive to Active Just generate DAUD.
> > > > > 
> > > > > Thanks & Regards
> > > > > Ashish Nagar
> > > > > 
> > > > > 
> > > > > -----Original Message-----
> > > > > From: Samuel Dur D. Jeyaseelan 
> > > > > [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > > To: sigtran@ietf.org
> > > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > > 
> > > > > 
> > > > > Hi all,
> > > > > 
> > > > > Currently RFC 3332 says that SG sends SSNM messages to all
> > > > > 'concerned ASPs'(means only ACTIVE ASPs).
> > > > > 
> > > > > +----+           +---+           +-----+
> > > > > |SEP | ----------|SG |-----------| ASP1|
> > > > > +----+           +---+           +-----+
> > > > >                    |             
> > > > >                    |             +-----+
> > > > >                    +-------------| ASP2|
> > > > >                                  +-----+
> > > > > 
> > > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > > 
> > > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > > But SEP status is not changed in ASP2.There are two ways to solve
> > > > > this problem
> > > > > 
> > > > > 1. ASP1 can equalize its database with ASP2
> > > > > or
> > > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > > 
> > > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or
> > > > > inact
> > > > > state)
> > > > > 
> > > > > Any comments?
> > > > > 
> > > > > -----
> > > > > 
> > > > > section 5.4.2.1
> > > > > 
> > > > > ...
> > > > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> > > > >         concerned Network Appearance.
> > > > > 
> > > > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER 
> > > > > request
> > > 
> > > > > is sent to MTP3.
> > > > > 
> > > > > But there is no primitive availabe between M3UA and MTP3.
> > > > > 
> > > > > I think it should be corrected.
> > > > > 
> > > > > ----
> > > > > Presently ASP is unaware of SG's MTP3 restart,
> > > > > ASP receives only DUNA for all SS7 point codes which can be 
> > > > > reachable via that SG.But in case of peer to peer protocol SG and 
> > > > > AS point codes are different, So why dont we send DUNA with SG's 
> > > > > point code to ASP? So that just by getting single DUNA(with SG's 
> > > > > PC) ASP will not route any message via that
> > > SG.
> > > > > This can avoid sending multiple DUNAs to ASP.
> > > > > 
> > > > > 
> > > > > Any comments?
> > > > > 
> > > > > -----
> > > > > section 1.4.1
> > > > > 
> > > > > ...
> > > > >    Where a particular AS can be reached via more than one SGP, the
> > > > >    corresponding Routing Keys in the SGPs should be identical.
> > > (Note:
> > > > >    It is possible for the SGP Routing Key configuration data to be
> > > > >    temporarily out-of-sync during configuration updates).
> > > > >  
> > > > >                               +--------+
> > > > >                               |        |
> > > > >                  +------------+  SG 1  +--------------+
> > > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > > >      |  SEP  +---+            +--------+              +---/      \
> > > > >      |   or  |                    |*                      | ASPs
> > |
> > > > >      |  STP  +---+            +--------+              +---\      /
> > > > >      +-------+   |            |        |              |     ----
> > > > >                  +------------+  SG 2  +--------------+
> > > > >                               | "STP"  |
> > > > >                               +--------+
> > > > >  
> > > > >                     Figure 1  Example with mated SGs
> > > > > 
> > > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are
> > > > > located geographically in different places.The ideas is when SG1 
> > > > > goes down in CITY1(due to earthquake...) , all the traffice still 
> > > > > can be handled by SG2 which is in CITY2.
> > > > > 
> > > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will 
> > > > > be
> > > > > routed from SGP1 to SGP2.
> > > > > 
> > > > > Does all these things mean to say that SGP1's mate 'should be'
> > > > > present in SG2 and not in SG1 itself ?
> > > > >  
> > > > > 
> > > > > Comments are welcome.
> > > > > 
> > > > > --Samuel
> > > > > 
> > > > > 
> > > > > 
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet:
> > > > <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA, 
> > > > > Inc.
> > > > *******
> > > > > 
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > 
> > > > 
> > > 
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 03:50:58 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08870
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 03:50:58 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h168vJ812882
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 03:57:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168v4p12864;
	Thu, 6 Feb 2003 03:57:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168udp12840
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 03:56:39 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08830
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 03:49:45 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:Mz8DzbFvc6uHj9eVw2i6xa/jjhykCP64@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h168rND14102;
	Thu, 6 Feb 2003 01:53:23 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h168rMP17813;
	Thu, 6 Feb 2003 01:53:22 -0700
Date: Thu, 6 Feb 2003 01:53:22 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Richard Adams <Richard.Adams@etsi.org>
Cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206015322.C15998@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Richard Adams <Richard.Adams@etsi.org>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <4091553999CBE4409CC2B562152B5A33EDDF58@email10.etsihq.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <4091553999CBE4409CC2B562152B5A33EDDF58@email10.etsihq.org>; from Richard.Adams@etsi.org on Thu, Feb 06, 2003 at 09:41:41AM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Richard,

Let's say it is an override AS...  In that case only one can be
active at a time.

I'm glad you agree with the concensus in the RFC.  We have discussed
this many times over the last several years and have always arrived
at the same conclusion: SG cannot be required to send SSNM to inactive
ASPs.

It is even worse on an SG when dynamic registration is not performed.
In this case, every ASP is considered to be ASP-INACTIVE for every
AS when it comes up.  Sending SSNM to inactive ASPs would send every
SSNM to every ASP, regardless of the more restricted set of AS that
the ASP is actually configured to ever be active for.

--brian

Richard Adams wrote:                                                               (Thu, 06 Feb 2003 09:41:41)
> Brian,
> I agree with you. I think when the ASP becomes active, it should do an
> audit to update its knowledge of the SS7 network. The ASP needs to take
> the initiative to find out about the SS7 nodes etc. in which it is
> interested. Any other method of working would lead to impossible loading
> on the SG. 
> Of course, I hope 199,999 of your inactive ASPs do not decide to become
> active at the same time.....! 
> Kind regards,
> Dick Adams,
> richard.adams@etsi.org
> 
>  
> 
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: 06 February 2003 09:15
> To: Zannin, Francois
> Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> Subject: Re: [Sigtran] M3UA - ASP database equalization
> 
> 
> Zannin,,
> 
> That is contrary to the conensus of the RFC.
> 
> Consider 1 ASP active, 200,000 ASPs inactive.
> 
> The RFC cannot require an SG to send SSNM to an inactive ASP.
> 
> --brian
> 
> Zannin, Francois wrote:
> (Thu, 06 Feb 2003 09:02:41)
> > Samuel,
> > 
> > IMO, it's a good idea to inform inactive (BUT UP) ASP of the state of
> > the ss7 network especially in the override mode where an ASP can be
> > ready to recover an AS as soon as it receive an AS-Pending
> notification
> > from the SGp. In this condition, the ASP is ready to operate quickly
> and
> > minimize the risk of buffered message lost during the pending timer.
> > 
> > If someone does not want to receive SSNM because it is not able to
> > recover an AS, it just stays in an ASP-Down state.
> > 
> > Because SSNM reflects ss7 Point Code state that is not "SI" dependant,
> > it affects all AS for a network and then, unless SG that plays
> multiple
> > networks, the ASP UP is sufficient to filter the undesired SSNM.
> > 
> > Francois.
> > 
> > -----Original Message-----
> > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> > Sent: Thursday, February 06, 2003 8:35 AM
> > To: Nagar, Ashish
> > Cc: sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > 
> > Ashish,
> > 
> > But when inactive ASP changes state to active, first of all
> > it does not know whether to send DAUD or not.
> > 
> > It will try to send traffic - but actually destination is unavailable.
> > 
> > The worst case is when SG triggers restart procedures this case all
> the
> > SS7 destinations are unavailable via restarted-SG.But this poor ASP
> > (newly come-up ASP) keeps sending traffic via restarted-SG to reach
> the
> > SS7 destination and for every message ASP receives DUNA.
> > 
> > Are we not doing double-work here?
> > 
> > --Samuel
> > 
> > 
> >   Alcatel USA, Inc.                  Internet:
> > <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > *******
> > 
> > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > 
> > > Hi Samuel
> > > 
> > > My point is why to do any processing if ASP is in Inactive state. 
> > > Think about ASP is in Inactive state for longer period and receiving
> 
> > > lots of DUNA and DAVA, while what matter is the present only when
> ASP 
> > > state changes to ASP-ACTIVE.
> > > 
> > > Thanks & Regards
> > > Ashish Nagar
> > > 
> > > 
> > > -----Original Message-----
> > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > Sent: Thursday, February 06, 2003 12:30 PM
> > > To: Nagar, Ashish
> > > Cc: sigtran@ietf.org
> > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > Ashish,
> > > 
> > > My point is if SG is able to send DUNA to all inact and active ASPs,
> 
> > > we should not restrict it.
> > > 
> > > Sending DAUD is an alternative way ,If SG has direct way of sending 
> > > DUNA then Why do you need alternative way?
> > > 
> > > --Samuel
> > > 
> > >   Alcatel USA, Inc.                  Internet:
> > <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. 
> > > *******
> > > 
> > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > 
> > > > You can generate DAUD procedure at any time, so whenever ASP
> changes
> > 
> > > > state from Inactive to Active Just generate DAUD.
> > > > 
> > > > Thanks & Regards
> > > > Ashish Nagar
> > > > 
> > > > 
> > > > -----Original Message-----
> > > > From: Samuel Dur D. Jeyaseelan
> [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > To: sigtran@ietf.org
> > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > 
> > > > 
> > > > Hi all,
> > > > 
> > > > Currently RFC 3332 says that SG sends SSNM messages to all 
> > > > 'concerned ASPs'(means only ACTIVE ASPs).
> > > > 
> > > > +----+           +---+           +-----+
> > > > |SEP | ----------|SG |-----------| ASP1|
> > > > +----+           +---+           +-----+
> > > >                    |             
> > > >                    |             +-----+
> > > >                    +-------------| ASP2|
> > > >                                  +-----+
> > > > 
> > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > 
> > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > But SEP status is not changed in ASP2.There are two ways to solve 
> > > > this problem
> > > > 
> > > > 1. ASP1 can equalize its database with ASP2
> > > > or
> > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > 
> > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or 
> > > > inact
> > > > state)
> > > > 
> > > > Any comments?
> > > > 
> > > > -----
> > > > 
> > > > section 5.4.2.1
> > > > 
> > > > ...
> > > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> > > >         concerned Network Appearance.
> > > > 
> > > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER
> request
> > 
> > > > is sent to MTP3.
> > > > 
> > > > But there is no primitive availabe between M3UA and MTP3.
> > > > 
> > > > I think it should be corrected.
> > > > 
> > > > ----
> > > > Presently ASP is unaware of SG's MTP3 restart,
> > > > ASP receives only DUNA for all SS7 point codes which can be
> > > > reachable via that SG.But in case of peer to peer protocol
> > > > SG and AS point codes are different, So why dont we send
> > > > DUNA with SG's point code to ASP? So that just by getting
> > > > single DUNA(with SG's PC) ASP will not route any message via that
> > SG.
> > > > This can avoid sending multiple DUNAs to ASP.
> > > > 
> > > > 
> > > > Any comments?
> > > > 
> > > > -----
> > > > section 1.4.1
> > > > 
> > > > ...
> > > >    Where a particular AS can be reached via more than one SGP, the
> > > >    corresponding Routing Keys in the SGPs should be identical.
> > (Note:
> > > >    It is possible for the SGP Routing Key configuration data to be
> > > >    temporarily out-of-sync during configuration updates).
> > > >  
> > > >                               +--------+
> > > >                               |        |
> > > >                  +------------+  SG 1  +--------------+
> > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > >      |  SEP  +---+            +--------+              +---/      \
> > > >      |   or  |                    |*                      | ASPs
> |
> > > >      |  STP  +---+            +--------+              +---\      /
> > > >      +-------+   |            |        |              |     ----
> > > >                  +------------+  SG 2  +--------------+
> > > >                               | "STP"  |
> > > >                               +--------+
> > > >  
> > > >                     Figure 1  Example with mated SGs
> > > > 
> > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are 
> > > > located geographically in different places.The ideas is when SG1 
> > > > goes down in CITY1(due to earthquake...) , all the traffice still 
> > > > can be handled by SG2 which is in CITY2.
> > > > 
> > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will
> be 
> > > > routed from SGP1 to SGP2.
> > > > 
> > > > Does all these things mean to say that SGP1's mate 'should be' 
> > > > present in SG2 and not in SG1 itself ?
> > > >  
> > > > 
> > > > Comments are welcome.
> > > > 
> > > > --Samuel
> > > > 
> > > > 
> > > > 
> > > > 
> > > >   Alcatel USA, Inc.                  Internet:
> > > <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA,
> Inc.
> > > *******
> > > > 
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > 
> > > 
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 03:54:58 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08956
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 03:54:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1691Jg13112
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 04:01:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16913p13088;
	Thu, 6 Feb 2003 04:01:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16901p13011
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 04:00:01 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08924
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 03:53:07 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:dXK3ym304/WmX8NM/k6zcA7t6R1g72E0@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h168ujD14148;
	Thu, 6 Feb 2003 01:56:45 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h168uix18244;
	Thu, 6 Feb 2003 01:56:44 -0700
Date: Thu, 6 Feb 2003 01:56:44 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206015644.D15998@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206013542.C14210@openss7.org> <Pine.SOL.4.10.10302060245350.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302060245350.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 02:51:52AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 02:51:52)
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Zannin,,
> > 
> > Zannin, Francois wrote:                                                            (Thu, 06 Feb 2003 09:27:21)
> > > Bidulock,,,
> > > 
> > > I think your example of 200,000 ASP "UP" as backup for an AS is really
> > > realistic for a network!
> > 
> > 200?  20?  It makes no difference.  The SG cannot be required to
> > send useless messages to bunch of ASPs that can do nothing with
> > the information.
> > 
> Everything is useless at one point of time, we should see always
> the 'usefullness of many times".

In this case it is useless all the time to send SSNM to inactive ASPs.

> > > 
> > > The goal is to agree on procedure that can be used for the
> > > implementation of M3UA. There were a lot of threads discussing on this
> > > subject and not always concluded. The role of the IG is to clarify the
> > > RFC in order to make it usable.
> > > 
> > > This procedure similar to the ss7 network (TFP inform all the SP of the
> > > inaccessibility). If a SG is unable to reach all the ss7 SP (200000 PC,
> > > means 200000 ungrouped SSNM) it can use the DOWN_ACK or INACTIVE_ACK
> > > procedure for its own isolation.
> > 
> > DUNA corresponds to MTP-PAUSE not TFP.
> > 
> > Consider 100 APSs inactive, zero (0) ASPs active.
> 
> I can say 100 ASPs inactive and 100 ASPs active with
> override mode.

100 ASP cannot be active for the same AS in override mode: only 1 can.

> 
> > 
> > The SG cannot be required to send DUNA to inactive ASPs.  When the SG
> > NIF has no user to send MTP-PAUSE to, it is not possible for the SG
> > to send DUNA.
> You are saying user-part is unavailabe at inact ASP.
> But logically it is available.
> 
> > 
> > The SG cannot be required to send SSNM to inactive ASPs.
> 
> It is required.Sending DUNA to inactive ASP is meaningfull.

It is not required and cannot be required.

>  
> > Besides, it is unnecessary.
> 
> I cant accept.

Well, as we can't agree, we will have to stay with the RFC, which
says that SSNM is not sent to inactive ASPs.

It is not the WG's responsibiltiy to convince you that the RFC is
proper, it is your responsibility to convince the WG otherwise.

--brian

> 
> --Samuel
> 
> > 
> > --brian
> > 
> > 
> > > 
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> > > Sent: Thursday, February 06, 2003 9:15 AM
> > > To: Zannin, Francois
> > > Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> > > Subject: Re: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > Zannin,,
> > > 
> > > That is contrary to the conensus of the RFC.
> > > 
> > > Consider 1 ASP active, 200,000 ASPs inactive.
> > > 
> > > The RFC cannot require an SG to send SSNM to an inactive ASP.
> > > 
> > > --brian
> > > 
> > > Zannin, Francois wrote:
> > > (Thu, 06 Feb 2003 09:02:41)
> > > > Samuel,
> > > > 
> > > > IMO, it's a good idea to inform inactive (BUT UP) ASP of the state of 
> > > > the ss7 network especially in the override mode where an ASP can be 
> > > > ready to recover an AS as soon as it receive an AS-Pending 
> > > > notification from the SGp. In this condition, the ASP is ready to 
> > > > operate quickly and minimize the risk of buffered message lost during 
> > > > the pending timer.
> > > > 
> > > > If someone does not want to receive SSNM because it is not able to 
> > > > recover an AS, it just stays in an ASP-Down state.
> > > > 
> > > > Because SSNM reflects ss7 Point Code state that is not "SI" dependant,
> > > 
> > > > it affects all AS for a network and then, unless SG that plays 
> > > > multiple networks, the ASP UP is sufficient to filter the undesired 
> > > > SSNM.
> > > > 
> > > > Francois.
> > > > 
> > > > -----Original Message-----
> > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > Sent: Thursday, February 06, 2003 8:35 AM
> > > > To: Nagar, Ashish
> > > > Cc: sigtran@ietf.org
> > > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > > 
> > > > 
> > > > 
> > > > Ashish,
> > > > 
> > > > But when inactive ASP changes state to active, first of all it does 
> > > > not know whether to send DAUD or not.
> > > > 
> > > > It will try to send traffic - but actually destination is unavailable.
> > > > 
> > > > The worst case is when SG triggers restart procedures this case all 
> > > > the SS7 destinations are unavailable via restarted-SG.But this poor 
> > > > ASP (newly come-up ASP) keeps sending traffic via restarted-SG to 
> > > > reach the SS7 destination and for every message ASP receives DUNA.
> > > > 
> > > > Are we not doing double-work here?
> > > > 
> > > > --Samuel
> > > > 
> > > > 
> > > >   Alcatel USA, Inc.                  Internet:
> > > > <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > > *******
> > > > 
> > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > 
> > > > > Hi Samuel
> > > > > 
> > > > > My point is why to do any processing if ASP is in Inactive state.
> > > > > Think about ASP is in Inactive state for longer period and receiving
> > > 
> > > > > lots of DUNA and DAVA, while what matter is the present only when
> > > ASP 
> > > > > state changes to ASP-ACTIVE.
> > > > > 
> > > > > Thanks & Regards
> > > > > Ashish Nagar
> > > > > 
> > > > > 
> > > > > -----Original Message-----
> > > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > > Sent: Thursday, February 06, 2003 12:30 PM
> > > > > To: Nagar, Ashish
> > > > > Cc: sigtran@ietf.org
> > > > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > > > 
> > > > > 
> > > > > Ashish,
> > > > > 
> > > > > My point is if SG is able to send DUNA to all inact and active ASPs,
> > > > > we should not restrict it.
> > > > > 
> > > > > Sending DAUD is an alternative way ,If SG has direct way of sending
> > > > > DUNA then Why do you need alternative way?
> > > > > 
> > > > > --Samuel
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet:
> > > > <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > > > *******
> > > > > 
> > > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > > 
> > > > > > You can generate DAUD procedure at any time, so whenever ASP 
> > > > > > changes
> > > > 
> > > > > > state from Inactive to Active Just generate DAUD.
> > > > > > 
> > > > > > Thanks & Regards
> > > > > > Ashish Nagar
> > > > > > 
> > > > > > 
> > > > > > -----Original Message-----
> > > > > > From: Samuel Dur D. Jeyaseelan 
> > > > > > [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > > > To: sigtran@ietf.org
> > > > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > > > 
> > > > > > 
> > > > > > Hi all,
> > > > > > 
> > > > > > Currently RFC 3332 says that SG sends SSNM messages to all
> > > > > > 'concerned ASPs'(means only ACTIVE ASPs).
> > > > > > 
> > > > > > +----+           +---+           +-----+
> > > > > > |SEP | ----------|SG |-----------| ASP1|
> > > > > > +----+           +---+           +-----+
> > > > > >                    |             
> > > > > >                    |             +-----+
> > > > > >                    +-------------| ASP2|
> > > > > >                                  +-----+
> > > > > > 
> > > > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > > > 
> > > > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > > > But SEP status is not changed in ASP2.There are two ways to solve
> > > > > > this problem
> > > > > > 
> > > > > > 1. ASP1 can equalize its database with ASP2
> > > > > > or
> > > > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > > > 
> > > > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or
> > > > > > inact
> > > > > > state)
> > > > > > 
> > > > > > Any comments?
> > > > > > 
> > > > > > -----
> > > > > > 
> > > > > > section 5.4.2.1
> > > > > > 
> > > > > > ...
> > > > > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> > > > > >         concerned Network Appearance.
> > > > > > 
> > > > > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER 
> > > > > > request
> > > > 
> > > > > > is sent to MTP3.
> > > > > > 
> > > > > > But there is no primitive availabe between M3UA and MTP3.
> > > > > > 
> > > > > > I think it should be corrected.
> > > > > > 
> > > > > > ----
> > > > > > Presently ASP is unaware of SG's MTP3 restart,
> > > > > > ASP receives only DUNA for all SS7 point codes which can be 
> > > > > > reachable via that SG.But in case of peer to peer protocol SG and 
> > > > > > AS point codes are different, So why dont we send DUNA with SG's 
> > > > > > point code to ASP? So that just by getting single DUNA(with SG's 
> > > > > > PC) ASP will not route any message via that
> > > > SG.
> > > > > > This can avoid sending multiple DUNAs to ASP.
> > > > > > 
> > > > > > 
> > > > > > Any comments?
> > > > > > 
> > > > > > -----
> > > > > > section 1.4.1
> > > > > > 
> > > > > > ...
> > > > > >    Where a particular AS can be reached via more than one SGP, the
> > > > > >    corresponding Routing Keys in the SGPs should be identical.
> > > > (Note:
> > > > > >    It is possible for the SGP Routing Key configuration data to be
> > > > > >    temporarily out-of-sync during configuration updates).
> > > > > >  
> > > > > >                               +--------+
> > > > > >                               |        |
> > > > > >                  +------------+  SG 1  +--------------+
> > > > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > > > >      |  SEP  +---+            +--------+              +---/      \
> > > > > >      |   or  |                    |*                      | ASPs
> > > |
> > > > > >      |  STP  +---+            +--------+              +---\      /
> > > > > >      +-------+   |            |        |              |     ----
> > > > > >                  +------------+  SG 2  +--------------+
> > > > > >                               | "STP"  |
> > > > > >                               +--------+
> > > > > >  
> > > > > >                     Figure 1  Example with mated SGs
> > > > > > 
> > > > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are
> > > > > > located geographically in different places.The ideas is when SG1 
> > > > > > goes down in CITY1(due to earthquake...) , all the traffice still 
> > > > > > can be handled by SG2 which is in CITY2.
> > > > > > 
> > > > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will 
> > > > > > be
> > > > > > routed from SGP1 to SGP2.
> > > > > > 
> > > > > > Does all these things mean to say that SGP1's mate 'should be'
> > > > > > present in SG2 and not in SG1 itself ?
> > > > > >  
> > > > > > 
> > > > > > Comments are welcome.
> > > > > > 
> > > > > > --Samuel
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > >   Alcatel USA, Inc.                  Internet:
> > > > > <userid>@ssd.usa.alcatel.com
> > > > > >   1000 Coit Road, Plano, Texas 75075
> > > > > >   ******* The opinions expressed are not those of Alcatel USA, 
> > > > > > Inc.
> > > > > *******
> > > > > > 
> > > > > > _______________________________________________
> > > > > > Sigtran mailing list
> > > > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > > _______________________________________________
> > > > > > Sigtran mailing list
> > > > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > > 
> > > > > 
> > > > 
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> > > -- 
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 03:56:58 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA09011
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 03:56:58 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1693J713486
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 04:03:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16934p13470;
	Thu, 6 Feb 2003 04:03:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16926p13153
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 04:02:06 -0500
Received: from zdemail03.zdem.compaq.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08962
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 03:55:11 -0500 (EST)
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail03.zdem.compaq.com (Postfix) with ESMTP
	id D0736690; Thu,  6 Feb 2003 09:58:48 +0100 (CET)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.2966);
	 Thu, 6 Feb 2003 09:58:48 +0100
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Subject: RE: [Sigtran] M3UA - ASP database equalization
Date: Thu, 6 Feb 2003 09:58:46 +0100
Message-ID: <06CAA4D3753C53408D2CE6340B68444214C40D@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA - ASP database equalization
Thread-Index: AcLNuujCznq4F9XvS0mTUP+Dp5vEKgAAKuKA
From: "Zannin, Francois" <Francois.Zannin@hp.com>
To: <bidulock@openss7.org>
Cc: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 06 Feb 2003 08:58:48.0701 (UTC) FILETIME=[F6728AD0:01C2CDBD]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h16926p13154
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Bidulock,,

Does it make sense to have "a lot" of ASP UP connected to a SGP and not
ready to do anything?

If an ASP is UP for a SGP, it certainly must be ready to activate an AS
and then concerned by the SSNM, and again, for an emergency AS recovery
(AS-Pending notification) the ASP must be ready to operate after the
Active-ACK.

Only the "ready to operate for an AS" ASP have a reason to be "UP" on a
SG.
I don't think that it is useful to connect "UP" a lot of ASP with no
reason!

You are right, DUNA is not the same thing as MTP-PAUSE, but the sense is
the same, a user (AS) cannot use a ROUTE (or SG) if the remote SP is
inaccessible because of TFP, TCP and then MTP-PAUSE which is the
consolidation of all the routes for a destination.

The reason to wait for a DAUD for the ss7 state on an ASP is to have a
selective use of the ss7 network SP by an AS (then ASP). In that case,
the number of DAUD is limited to the minimum after the ACTIVE procedure.
But as soon as the ASP is active, there is no filter in the SG for ASP
concerned by the SSNM for specific destinations.

-----Original Message-----
From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
Sent: Thursday, February 06, 2003 9:36 AM
To: Zannin, Francois
Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization


Zannin,,

Zannin, Francois wrote:
(Thu, 06 Feb 2003 09:27:21)
> Bidulock,,,
> 
> I think your example of 200,000 ASP "UP" as backup for an AS is really

> realistic for a network!

200?  20?  It makes no difference.  The SG cannot be required to send
useless messages to bunch of ASPs that can do nothing with the
information.

> 
> The goal is to agree on procedure that can be used for the 
> implementation of M3UA. There were a lot of threads discussing on this

> subject and not always concluded. The role of the IG is to clarify the

> RFC in order to make it usable.
> 
> This procedure similar to the ss7 network (TFP inform all the SP of 
> the inaccessibility). If a SG is unable to reach all the ss7 SP 
> (200000 PC, means 200000 ungrouped SSNM) it can use the DOWN_ACK or 
> INACTIVE_ACK procedure for its own isolation.

DUNA corresponds to MTP-PAUSE not TFP.

Consider 100 APSs inactive, zero (0) ASPs active.

The SG cannot be required to send DUNA to inactive ASPs.  When the SG
NIF has no user to send MTP-PAUSE to, it is not possible for the SG to
send DUNA.

The SG cannot be required to send SSNM to inactive ASPs.

Besides, it is unnecessary.

--brian


> 
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Thursday, February 06, 2003 9:15 AM
> To: Zannin, Francois
> Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> Subject: Re: [Sigtran] M3UA - ASP database equalization
> 
> 
> Zannin,,
> 
> That is contrary to the conensus of the RFC.
> 
> Consider 1 ASP active, 200,000 ASPs inactive.
> 
> The RFC cannot require an SG to send SSNM to an inactive ASP.
> 
> --brian
> 
> Zannin, Francois wrote:
> (Thu, 06 Feb 2003 09:02:41)
> > Samuel,
> > 
> > IMO, it's a good idea to inform inactive (BUT UP) ASP of the state 
> > of
> > the ss7 network especially in the override mode where an ASP can be 
> > ready to recover an AS as soon as it receive an AS-Pending 
> > notification from the SGp. In this condition, the ASP is ready to 
> > operate quickly and minimize the risk of buffered message lost
during 
> > the pending timer.
> > 
> > If someone does not want to receive SSNM because it is not able to
> > recover an AS, it just stays in an ASP-Down state.
> > 
> > Because SSNM reflects ss7 Point Code state that is not "SI" 
> > dependant,
> 
> > it affects all AS for a network and then, unless SG that plays
> > multiple networks, the ASP UP is sufficient to filter the undesired 
> > SSNM.
> > 
> > Francois.
> > 
> > -----Original Message-----
> > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > Sent: Thursday, February 06, 2003 8:35 AM
> > To: Nagar, Ashish
> > Cc: sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > 
> > Ashish,
> > 
> > But when inactive ASP changes state to active, first of all it does
> > not know whether to send DAUD or not.
> > 
> > It will try to send traffic - but actually destination is 
> > unavailable.
> > 
> > The worst case is when SG triggers restart procedures this case all
> > the SS7 destinations are unavailable via restarted-SG.But this poor 
> > ASP (newly come-up ASP) keeps sending traffic via restarted-SG to 
> > reach the SS7 destination and for every message ASP receives DUNA.
> > 
> > Are we not doing double-work here?
> > 
> > --Samuel
> > 
> > 
> >   Alcatel USA, Inc.                  Internet:
> > <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > *******
> > 
> > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > 
> > > Hi Samuel
> > > 
> > > My point is why to do any processing if ASP is in Inactive state. 
> > > Think about ASP is in Inactive state for longer period and 
> > > receiving
> 
> > > lots of DUNA and DAVA, while what matter is the present only when
> ASP
> > > state changes to ASP-ACTIVE.
> > > 
> > > Thanks & Regards
> > > Ashish Nagar
> > > 
> > > 
> > > -----Original Message-----
> > > From: Samuel Dur D. Jeyaseelan 
> > > [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > Sent: Thursday, February 06, 2003 12:30 PM
> > > To: Nagar, Ashish
> > > Cc: sigtran@ietf.org
> > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > Ashish,
> > > 
> > > My point is if SG is able to send DUNA to all inact and active 
> > > ASPs, we should not restrict it.
> > > 
> > > Sending DAUD is an alternative way ,If SG has direct way of 
> > > sending DUNA then Why do you need alternative way?
> > > 
> > > --Samuel
> > > 
> > >   Alcatel USA, Inc.                  Internet:
> > <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, 
> > > Inc.
> > > *******
> > > 
> > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > 
> > > > You can generate DAUD procedure at any time, so whenever ASP
> > > > changes
> > 
> > > > state from Inactive to Active Just generate DAUD.
> > > > 
> > > > Thanks & Regards
> > > > Ashish Nagar
> > > > 
> > > > 
> > > > -----Original Message-----
> > > > From: Samuel Dur D. Jeyaseelan
> > > > [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > To: sigtran@ietf.org
> > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > 
> > > > 
> > > > Hi all,
> > > > 
> > > > Currently RFC 3332 says that SG sends SSNM messages to all 
> > > > 'concerned ASPs'(means only ACTIVE ASPs).
> > > > 
> > > > +----+           +---+           +-----+
> > > > |SEP | ----------|SG |-----------| ASP1|
> > > > +----+           +---+           +-----+
> > > >                    |             
> > > >                    |             +-----+
> > > >                    +-------------| ASP2|
> > > >                                  +-----+
> > > > 
> > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > 
> > > > When SEP is getting restarted ,SG sends DUNA to ASP1. But SEP 
> > > > status is not changed in ASP2.There are two ways to solve this 
> > > > problem
> > > > 
> > > > 1. ASP1 can equalize its database with ASP2
> > > > or
> > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is 
> > > > inactive)
> > > > 
> > > > IMO, SG can send SSNM messages to ASP (either ASP is in active 
> > > > or inact
> > > > state)
> > > > 
> > > > Any comments?
> > > > 
> > > > -----
> > > > 
> > > > section 5.4.2.1
> > > > 
> > > > ...
> > > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of
the
> > > >         concerned Network Appearance.
> > > > 
> > > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER
> > > > request
> > 
> > > > is sent to MTP3.
> > > > 
> > > > But there is no primitive availabe between M3UA and MTP3.
> > > > 
> > > > I think it should be corrected.
> > > > 
> > > > ----
> > > > Presently ASP is unaware of SG's MTP3 restart,
> > > > ASP receives only DUNA for all SS7 point codes which can be
> > > > reachable via that SG.But in case of peer to peer protocol SG
and 
> > > > AS point codes are different, So why dont we send DUNA with SG's

> > > > point code to ASP? So that just by getting single DUNA(with SG's

> > > > PC) ASP will not route any message via that
> > SG.
> > > > This can avoid sending multiple DUNAs to ASP.
> > > > 
> > > > 
> > > > Any comments?
> > > > 
> > > > -----
> > > > section 1.4.1
> > > > 
> > > > ...
> > > >    Where a particular AS can be reached via more than one SGP,
the
> > > >    corresponding Routing Keys in the SGPs should be identical.
> > (Note:
> > > >    It is possible for the SGP Routing Key configuration data to
be
> > > >    temporarily out-of-sync during configuration updates).
> > > >  
> > > >                               +--------+
> > > >                               |        |
> > > >                  +------------+  SG 1  +--------------+
> > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > >      |  SEP  +---+            +--------+              +---/
\
> > > >      |   or  |                    |*                      | ASPs
> |
> > > >      |  STP  +---+            +--------+              +---\
/
> > > >      +-------+   |            |        |              |     ----
> > > >                  +------------+  SG 2  +--------------+
> > > >                               | "STP"  |
> > > >                               +--------+
> > > >  
> > > >                     Figure 1  Example with mated SGs
> > > > 
> > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are 
> > > > located geographically in different places.The ideas is when SG1

> > > > goes down in CITY1(due to earthquake...) , all the traffice 
> > > > still can be handled by SG2 which is in CITY2.
> > > > 
> > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will
> > > > be
> > > > routed from SGP1 to SGP2.
> > > > 
> > > > Does all these things mean to say that SGP1's mate 'should be' 
> > > > present in SG2 and not in SG1 itself ?
> > > >  
> > > > 
> > > > Comments are welcome.
> > > > 
> > > > --Samuel
> > > > 
> > > > 
> > > > 
> > > > 
> > > >   Alcatel USA, Inc.                  Internet:
> > > <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA,
> > > > Inc.
> > > *******
> > > > 
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > 
> > > 
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> 
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 04:00:04 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA09127
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 04:00:04 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1696PB13737
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 04:06:25 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16968p13723;
	Thu, 6 Feb 2003 04:06:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1695Pp13673
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 04:05:25 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA09065
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 03:58:32 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h16922qg015160;
	Thu, 6 Feb 2003 03:02:03 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h1691dM09314;
	Thu, 6 Feb 2003 03:01:39 -0600 (CST)
Date: Thu, 6 Feb 2003 03:01:39 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: Richard Adams <Richard.Adams@etsi.org>
cc: bidulock@openss7.org, "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: RE: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <4091553999CBE4409CC2B562152B5A33EDDF58@email10.etsihq.org>
Message-ID: <Pine.SOL.4.10.10302060256200.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Richard,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Richard Adams wrote:

> Brian,
> I agree with you. I think when the ASP becomes active, it should do an
> audit to update its knowledge of the SS7 network. The ASP needs to take
> the initiative to find out about the SS7 nodes etc. in which it is
> interested. Any other method of working would lead to impossible loading
> on the SG.

When SG itself is re-starting , it means no ASP can reach any SS7
destinations via re-started SG.Always it is good to tell my ASP
who is in inactive state about SS7 destinations unavailability.

--Samuel

> Of course, I hope 199,999 of your inactive ASPs do not decide to become
> active at the same time.....! 
> Kind regards,
> Dick Adams,
> richard.adams@etsi.org
> 
>  
> 
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: 06 February 2003 09:15
> To: Zannin, Francois
> Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> Subject: Re: [Sigtran] M3UA - ASP database equalization
> 
> 
> Zannin,,
> 
> That is contrary to the conensus of the RFC.
> 
> Consider 1 ASP active, 200,000 ASPs inactive.
> 
> The RFC cannot require an SG to send SSNM to an inactive ASP.
> 
> --brian
> 
> Zannin, Francois wrote:
> (Thu, 06 Feb 2003 09:02:41)
> > Samuel,
> > 
> > IMO, it's a good idea to inform inactive (BUT UP) ASP of the state of
> > the ss7 network especially in the override mode where an ASP can be
> > ready to recover an AS as soon as it receive an AS-Pending
> notification
> > from the SGp. In this condition, the ASP is ready to operate quickly
> and
> > minimize the risk of buffered message lost during the pending timer.
> > 
> > If someone does not want to receive SSNM because it is not able to
> > recover an AS, it just stays in an ASP-Down state.
> > 
> > Because SSNM reflects ss7 Point Code state that is not "SI" dependant,
> > it affects all AS for a network and then, unless SG that plays
> multiple
> > networks, the ASP UP is sufficient to filter the undesired SSNM.
> > 
> > Francois.
> > 
> > -----Original Message-----
> > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> > Sent: Thursday, February 06, 2003 8:35 AM
> > To: Nagar, Ashish
> > Cc: sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > 
> > Ashish,
> > 
> > But when inactive ASP changes state to active, first of all
> > it does not know whether to send DAUD or not.
> > 
> > It will try to send traffic - but actually destination is unavailable.
> > 
> > The worst case is when SG triggers restart procedures this case all
> the
> > SS7 destinations are unavailable via restarted-SG.But this poor ASP
> > (newly come-up ASP) keeps sending traffic via restarted-SG to reach
> the
> > SS7 destination and for every message ASP receives DUNA.
> > 
> > Are we not doing double-work here?
> > 
> > --Samuel
> > 
> > 
> >   Alcatel USA, Inc.                  Internet:
> > <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > *******
> > 
> > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > 
> > > Hi Samuel
> > > 
> > > My point is why to do any processing if ASP is in Inactive state. 
> > > Think about ASP is in Inactive state for longer period and receiving
> 
> > > lots of DUNA and DAVA, while what matter is the present only when
> ASP 
> > > state changes to ASP-ACTIVE.
> > > 
> > > Thanks & Regards
> > > Ashish Nagar
> > > 
> > > 
> > > -----Original Message-----
> > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > Sent: Thursday, February 06, 2003 12:30 PM
> > > To: Nagar, Ashish
> > > Cc: sigtran@ietf.org
> > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > Ashish,
> > > 
> > > My point is if SG is able to send DUNA to all inact and active ASPs,
> 
> > > we should not restrict it.
> > > 
> > > Sending DAUD is an alternative way ,If SG has direct way of sending 
> > > DUNA then Why do you need alternative way?
> > > 
> > > --Samuel
> > > 
> > >   Alcatel USA, Inc.                  Internet:
> > <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. 
> > > *******
> > > 
> > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > 
> > > > You can generate DAUD procedure at any time, so whenever ASP
> changes
> > 
> > > > state from Inactive to Active Just generate DAUD.
> > > > 
> > > > Thanks & Regards
> > > > Ashish Nagar
> > > > 
> > > > 
> > > > -----Original Message-----
> > > > From: Samuel Dur D. Jeyaseelan
> [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > To: sigtran@ietf.org
> > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > 
> > > > 
> > > > Hi all,
> > > > 
> > > > Currently RFC 3332 says that SG sends SSNM messages to all 
> > > > 'concerned ASPs'(means only ACTIVE ASPs).
> > > > 
> > > > +----+           +---+           +-----+
> > > > |SEP | ----------|SG |-----------| ASP1|
> > > > +----+           +---+           +-----+
> > > >                    |             
> > > >                    |             +-----+
> > > >                    +-------------| ASP2|
> > > >                                  +-----+
> > > > 
> > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > 
> > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > But SEP status is not changed in ASP2.There are two ways to solve 
> > > > this problem
> > > > 
> > > > 1. ASP1 can equalize its database with ASP2
> > > > or
> > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > 
> > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or 
> > > > inact
> > > > state)
> > > > 
> > > > Any comments?
> > > > 
> > > > -----
> > > > 
> > > > section 5.4.2.1
> > > > 
> > > > ...
> > > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> > > >         concerned Network Appearance.
> > > > 
> > > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER
> request
> > 
> > > > is sent to MTP3.
> > > > 
> > > > But there is no primitive availabe between M3UA and MTP3.
> > > > 
> > > > I think it should be corrected.
> > > > 
> > > > ----
> > > > Presently ASP is unaware of SG's MTP3 restart,
> > > > ASP receives only DUNA for all SS7 point codes which can be
> > > > reachable via that SG.But in case of peer to peer protocol
> > > > SG and AS point codes are different, So why dont we send
> > > > DUNA with SG's point code to ASP? So that just by getting
> > > > single DUNA(with SG's PC) ASP will not route any message via that
> > SG.
> > > > This can avoid sending multiple DUNAs to ASP.
> > > > 
> > > > 
> > > > Any comments?
> > > > 
> > > > -----
> > > > section 1.4.1
> > > > 
> > > > ...
> > > >    Where a particular AS can be reached via more than one SGP, the
> > > >    corresponding Routing Keys in the SGPs should be identical.
> > (Note:
> > > >    It is possible for the SGP Routing Key configuration data to be
> > > >    temporarily out-of-sync during configuration updates).
> > > >  
> > > >                               +--------+
> > > >                               |        |
> > > >                  +------------+  SG 1  +--------------+
> > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > >      |  SEP  +---+            +--------+              +---/      \
> > > >      |   or  |                    |*                      | ASPs
> |
> > > >      |  STP  +---+            +--------+              +---\      /
> > > >      +-------+   |            |        |              |     ----
> > > >                  +------------+  SG 2  +--------------+
> > > >                               | "STP"  |
> > > >                               +--------+
> > > >  
> > > >                     Figure 1  Example with mated SGs
> > > > 
> > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are 
> > > > located geographically in different places.The ideas is when SG1 
> > > > goes down in CITY1(due to earthquake...) , all the traffice still 
> > > > can be handled by SG2 which is in CITY2.
> > > > 
> > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will
> be 
> > > > routed from SGP1 to SGP2.
> > > > 
> > > > Does all these things mean to say that SGP1's mate 'should be' 
> > > > present in SG2 and not in SG1 itself ?
> > > >  
> > > > 
> > > > Comments are welcome.
> > > > 
> > > > --Samuel
> > > > 
> > > > 
> > > > 
> > > > 
> > > >   Alcatel USA, Inc.                  Internet:
> > > <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA,
> Inc.
> > > *******
> > > > 
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > 
> > > 
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 04:12:14 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA09403
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 04:12:13 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h169IZB14801
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 04:18:35 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169IJp14787;
	Thu, 6 Feb 2003 04:18:20 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169HLp14759
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 04:17:21 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA09341
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 04:10:27 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h169E2qe016042;
	Thu, 6 Feb 2003 03:14:02 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h169DuM13717;
	Thu, 6 Feb 2003 03:13:56 -0600 (CST)
Date: Thu, 6 Feb 2003 03:13:56 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206015644.D15998@openss7.org>
Message-ID: <Pine.SOL.4.10.10302060308190.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 02:51:52)
> > 
> > Brian,
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > 
> > > Zannin,,
> > > 
> > > Zannin, Francois wrote:                                                            (Thu, 06 Feb 2003 09:27:21)
> > > > Bidulock,,,
> > > > 
> > > > I think your example of 200,000 ASP "UP" as backup for an AS is really
> > > > realistic for a network!
> > > 
> > > 200?  20?  It makes no difference.  The SG cannot be required to
> > > send useless messages to bunch of ASPs that can do nothing with
> > > the information.
> > > 
> > Everything is useless at one point of time, we should see always
> > the 'usefullness of many times".
> 
> In this case it is useless all the time to send SSNM to inactive ASPs.
> 
> > > > 
> > > > The goal is to agree on procedure that can be used for the
> > > > implementation of M3UA. There were a lot of threads discussing on this
> > > > subject and not always concluded. The role of the IG is to clarify the
> > > > RFC in order to make it usable.
> > > > 
> > > > This procedure similar to the ss7 network (TFP inform all the SP of the
> > > > inaccessibility). If a SG is unable to reach all the ss7 SP (200000 PC,
> > > > means 200000 ungrouped SSNM) it can use the DOWN_ACK or INACTIVE_ACK
> > > > procedure for its own isolation.
> > > 
> > > DUNA corresponds to MTP-PAUSE not TFP.
> > > 
> > > Consider 100 APSs inactive, zero (0) ASPs active.
> > 
> > I can say 100 ASPs inactive and 100 ASPs active with
> > override mode.
> 
> 100 ASP cannot be active for the same AS in override mode: only 1 can.
> 

Total misinterpretation.ASP1 is active ASP2 is inactive....
ASP100 is active and ASP100 is inactive - VIA RE-STARTED SG.

Now 100 ASPs are active and 100 ASPs are inactive from the point
of re-started SG.

> > 
> > > 
> > > The SG cannot be required to send DUNA to inactive ASPs.  When the SG
> > > NIF has no user to send MTP-PAUSE to, it is not possible for the SG
> > > to send DUNA.
> > You are saying user-part is unavailabe at inact ASP.
> > But logically it is available.
> > 
> > > 
> > > The SG cannot be required to send SSNM to inactive ASPs.
> > 
> > It is required.Sending DUNA to inactive ASP is meaningfull.
> 
> It is not required and cannot be required.
> 
> >  
> > > Besides, it is unnecessary.
> > 
> > I cant accept.
> 
> Well, as we can't agree, we will have to stay with the RFC, which
> says that SSNM is not sent to inactive ASPs.

You know so far 'disagreement with RFC' only changes RFC.
(with good reason).
> 
> It is not the WG's responsibiltiy to convince you that the RFC is
> proper, it is your responsibility to convince the WG otherwise.

no comments.

--Samuel
 
> --brian
> 
> > 
> > --Samuel
> > 
> > > 
> > > --brian
> > > 
> > > 
> > > > 
> > > > -----Original Message-----
> > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> > > > Sent: Thursday, February 06, 2003 9:15 AM
> > > > To: Zannin, Francois
> > > > Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> > > > Subject: Re: [Sigtran] M3UA - ASP database equalization
> > > > 
> > > > 
> > > > Zannin,,
> > > > 
> > > > That is contrary to the conensus of the RFC.
> > > > 
> > > > Consider 1 ASP active, 200,000 ASPs inactive.
> > > > 
> > > > The RFC cannot require an SG to send SSNM to an inactive ASP.
> > > > 
> > > > --brian
> > > > 
> > > > Zannin, Francois wrote:
> > > > (Thu, 06 Feb 2003 09:02:41)
> > > > > Samuel,
> > > > > 
> > > > > IMO, it's a good idea to inform inactive (BUT UP) ASP of the state of 
> > > > > the ss7 network especially in the override mode where an ASP can be 
> > > > > ready to recover an AS as soon as it receive an AS-Pending 
> > > > > notification from the SGp. In this condition, the ASP is ready to 
> > > > > operate quickly and minimize the risk of buffered message lost during 
> > > > > the pending timer.
> > > > > 
> > > > > If someone does not want to receive SSNM because it is not able to 
> > > > > recover an AS, it just stays in an ASP-Down state.
> > > > > 
> > > > > Because SSNM reflects ss7 Point Code state that is not "SI" dependant,
> > > > 
> > > > > it affects all AS for a network and then, unless SG that plays 
> > > > > multiple networks, the ASP UP is sufficient to filter the undesired 
> > > > > SSNM.
> > > > > 
> > > > > Francois.
> > > > > 
> > > > > -----Original Message-----
> > > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > > Sent: Thursday, February 06, 2003 8:35 AM
> > > > > To: Nagar, Ashish
> > > > > Cc: sigtran@ietf.org
> > > > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > > > 
> > > > > 
> > > > > 
> > > > > Ashish,
> > > > > 
> > > > > But when inactive ASP changes state to active, first of all it does 
> > > > > not know whether to send DAUD or not.
> > > > > 
> > > > > It will try to send traffic - but actually destination is unavailable.
> > > > > 
> > > > > The worst case is when SG triggers restart procedures this case all 
> > > > > the SS7 destinations are unavailable via restarted-SG.But this poor 
> > > > > ASP (newly come-up ASP) keeps sending traffic via restarted-SG to 
> > > > > reach the SS7 destination and for every message ASP receives DUNA.
> > > > > 
> > > > > Are we not doing double-work here?
> > > > > 
> > > > > --Samuel
> > > > > 
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet:
> > > > > <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > > > *******
> > > > > 
> > > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > > 
> > > > > > Hi Samuel
> > > > > > 
> > > > > > My point is why to do any processing if ASP is in Inactive state.
> > > > > > Think about ASP is in Inactive state for longer period and receiving
> > > > 
> > > > > > lots of DUNA and DAVA, while what matter is the present only when
> > > > ASP 
> > > > > > state changes to ASP-ACTIVE.
> > > > > > 
> > > > > > Thanks & Regards
> > > > > > Ashish Nagar
> > > > > > 
> > > > > > 
> > > > > > -----Original Message-----
> > > > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > > > Sent: Thursday, February 06, 2003 12:30 PM
> > > > > > To: Nagar, Ashish
> > > > > > Cc: sigtran@ietf.org
> > > > > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > > > > 
> > > > > > 
> > > > > > Ashish,
> > > > > > 
> > > > > > My point is if SG is able to send DUNA to all inact and active ASPs,
> > > > > > we should not restrict it.
> > > > > > 
> > > > > > Sending DAUD is an alternative way ,If SG has direct way of sending
> > > > > > DUNA then Why do you need alternative way?
> > > > > > 
> > > > > > --Samuel
> > > > > > 
> > > > > >   Alcatel USA, Inc.                  Internet:
> > > > > <userid>@ssd.usa.alcatel.com
> > > > > >   1000 Coit Road, Plano, Texas 75075
> > > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > > > > *******
> > > > > > 
> > > > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > > > 
> > > > > > > You can generate DAUD procedure at any time, so whenever ASP 
> > > > > > > changes
> > > > > 
> > > > > > > state from Inactive to Active Just generate DAUD.
> > > > > > > 
> > > > > > > Thanks & Regards
> > > > > > > Ashish Nagar
> > > > > > > 
> > > > > > > 
> > > > > > > -----Original Message-----
> > > > > > > From: Samuel Dur D. Jeyaseelan 
> > > > > > > [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > > > > To: sigtran@ietf.org
> > > > > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > > > > 
> > > > > > > 
> > > > > > > Hi all,
> > > > > > > 
> > > > > > > Currently RFC 3332 says that SG sends SSNM messages to all
> > > > > > > 'concerned ASPs'(means only ACTIVE ASPs).
> > > > > > > 
> > > > > > > +----+           +---+           +-----+
> > > > > > > |SEP | ----------|SG |-----------| ASP1|
> > > > > > > +----+           +---+           +-----+
> > > > > > >                    |             
> > > > > > >                    |             +-----+
> > > > > > >                    +-------------| ASP2|
> > > > > > >                                  +-----+
> > > > > > > 
> > > > > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > > > > 
> > > > > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > > > > But SEP status is not changed in ASP2.There are two ways to solve
> > > > > > > this problem
> > > > > > > 
> > > > > > > 1. ASP1 can equalize its database with ASP2
> > > > > > > or
> > > > > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > > > > 
> > > > > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or
> > > > > > > inact
> > > > > > > state)
> > > > > > > 
> > > > > > > Any comments?
> > > > > > > 
> > > > > > > -----
> > > > > > > 
> > > > > > > section 5.4.2.1
> > > > > > > 
> > > > > > > ...
> > > > > > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> > > > > > >         concerned Network Appearance.
> > > > > > > 
> > > > > > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER 
> > > > > > > request
> > > > > 
> > > > > > > is sent to MTP3.
> > > > > > > 
> > > > > > > But there is no primitive availabe between M3UA and MTP3.
> > > > > > > 
> > > > > > > I think it should be corrected.
> > > > > > > 
> > > > > > > ----
> > > > > > > Presently ASP is unaware of SG's MTP3 restart,
> > > > > > > ASP receives only DUNA for all SS7 point codes which can be 
> > > > > > > reachable via that SG.But in case of peer to peer protocol SG and 
> > > > > > > AS point codes are different, So why dont we send DUNA with SG's 
> > > > > > > point code to ASP? So that just by getting single DUNA(with SG's 
> > > > > > > PC) ASP will not route any message via that
> > > > > SG.
> > > > > > > This can avoid sending multiple DUNAs to ASP.
> > > > > > > 
> > > > > > > 
> > > > > > > Any comments?
> > > > > > > 
> > > > > > > -----
> > > > > > > section 1.4.1
> > > > > > > 
> > > > > > > ...
> > > > > > >    Where a particular AS can be reached via more than one SGP, the
> > > > > > >    corresponding Routing Keys in the SGPs should be identical.
> > > > > (Note:
> > > > > > >    It is possible for the SGP Routing Key configuration data to be
> > > > > > >    temporarily out-of-sync during configuration updates).
> > > > > > >  
> > > > > > >                               +--------+
> > > > > > >                               |        |
> > > > > > >                  +------------+  SG 1  +--------------+
> > > > > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > > > > >      |  SEP  +---+            +--------+              +---/      \
> > > > > > >      |   or  |                    |*                      | ASPs
> > > > |
> > > > > > >      |  STP  +---+            +--------+              +---\      /
> > > > > > >      +-------+   |            |        |              |     ----
> > > > > > >                  +------------+  SG 2  +--------------+
> > > > > > >                               | "STP"  |
> > > > > > >                               +--------+
> > > > > > >  
> > > > > > >                     Figure 1  Example with mated SGs
> > > > > > > 
> > > > > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are
> > > > > > > located geographically in different places.The ideas is when SG1 
> > > > > > > goes down in CITY1(due to earthquake...) , all the traffice still 
> > > > > > > can be handled by SG2 which is in CITY2.
> > > > > > > 
> > > > > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will 
> > > > > > > be
> > > > > > > routed from SGP1 to SGP2.
> > > > > > > 
> > > > > > > Does all these things mean to say that SGP1's mate 'should be'
> > > > > > > present in SG2 and not in SG1 itself ?
> > > > > > >  
> > > > > > > 
> > > > > > > Comments are welcome.
> > > > > > > 
> > > > > > > --Samuel
> > > > > > > 
> > > > > > > 
> > > > > > > 
> > > > > > > 
> > > > > > >   Alcatel USA, Inc.                  Internet:
> > > > > > <userid>@ssd.usa.alcatel.com
> > > > > > >   1000 Coit Road, Plano, Texas 75075
> > > > > > >   ******* The opinions expressed are not those of Alcatel USA, 
> > > > > > > Inc.
> > > > > > *******
> > > > > > > 
> > > > > > > _______________________________________________
> > > > > > > Sigtran mailing list
> > > > > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > > > _______________________________________________
> > > > > > > Sigtran mailing list
> > > > > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > > > 
> > > > > > 
> > > > > 
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > 
> > > > -- 
> > > > Brian F. G. Bidulock
> > > > bidulock@openss7.org
> > > > http://www.openss7.org/
> > > 
> > > -- 
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 04:19:00 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA09559
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 04:18:59 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h169PLl15140
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 04:25:21 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169P7p15131;
	Thu, 6 Feb 2003 04:25:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169Olp15086
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 04:24:47 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA09537
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 04:17:53 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:OCv7WUH+HatsXHF1P8+mOZOaGVRoSzEW@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h169LVD14559;
	Thu, 6 Feb 2003 02:21:31 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h169LUx21333;
	Thu, 6 Feb 2003 02:21:30 -0700
Date: Thu, 6 Feb 2003 02:21:30 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Zannin, Francois" <Francois.Zannin@hp.com>
Cc: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206022130.A18441@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Zannin, Francois" <Francois.Zannin@hp.com>,
	"Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <06CAA4D3753C53408D2CE6340B68444214C40D@vbeexc02.emea.cpqcorp.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <06CAA4D3753C53408D2CE6340B68444214C40D@vbeexc02.emea.cpqcorp.net>; from Francois.Zannin@hp.com on Thu, Feb 06, 2003 at 09:58:46AM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Zannin,,

Zannin, Francois wrote:                                                            (Thu, 06 Feb 2003 09:58:46)
> Bidulock,,
> 
> Does it make sense to have "a lot" of ASP UP connected to a SGP and not
> ready to do anything?

They could be active for different AS.

> If an ASP is UP for a SGP, it certainly must be ready to activate an AS
> and then concerned by the SSNM, and again, for an emergency AS recovery
> (AS-Pending notification) the ASP must be ready to operate after the
> Active-ACK.
> 
> Only the "ready to operate for an AS" ASP have a reason to be "UP" on a
> SG.

Wrong.  Whenever an ASP goes ASP UP it becomes ASP-INACTIVE for all
AS for which it is configured to operate.  Depending on SG configuration,
this could mean that the ASP is ASP-INACTIVE for all AS served by the
SG, regarless of whether the ASP will ever go ASP-ACTIVE for all of
the AS.  There is no way for an ASP to tell the SG that it is ASP-DOWN
for some AS and ASP-INACTIVE for others.

Also, the purpose of the ASP-INACTIVE state is to receive NTFY messages
for the AS.  In override mode, one ASP is active for the AS and the
others are inactive.  The inactive ASPs have no intention of going
active unless the active ASP fails, or unless manual management action is
taken.

> I don't think that it is useful to connect "UP" a lot of ASP with no
> reason!

See above.

> 
> You are right, DUNA is not the same thing as MTP-PAUSE, but the sense is
> the same, a user (AS) cannot use a ROUTE (or SG) if the remote SP is
> inaccessible because of TFP, TCP and then MTP-PAUSE which is the
> consolidation of all the routes for a destination.
> 
> The reason to wait for a DAUD for the ss7 state on an ASP is to have a
> selective use of the ss7 network SP by an AS (then ASP). In that case,
> the number of DAUD is limited to the minimum after the ACTIVE procedure.
> But as soon as the ASP is active, there is no filter in the SG for ASP
> concerned by the SSNM for specific destinations.

MTP procedures require MTP-PAUSE to be sent to MTP users that are
sending to inaccessible destinations.  An MTP conforming SG will send
DUNA in response to messages sent to an inacessible destination.

--brian

> 
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> Sent: Thursday, February 06, 2003 9:36 AM
> To: Zannin, Francois
> Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> Subject: Re: [Sigtran] M3UA - ASP database equalization
> 
> 
> Zannin,,
> 
> Zannin, Francois wrote:
> (Thu, 06 Feb 2003 09:27:21)
> > Bidulock,,,
> > 
> > I think your example of 200,000 ASP "UP" as backup for an AS is really
> 
> > realistic for a network!
> 
> 200?  20?  It makes no difference.  The SG cannot be required to send
> useless messages to bunch of ASPs that can do nothing with the
> information.
> 
> > 
> > The goal is to agree on procedure that can be used for the 
> > implementation of M3UA. There were a lot of threads discussing on this
> 
> > subject and not always concluded. The role of the IG is to clarify the
> 
> > RFC in order to make it usable.
> > 
> > This procedure similar to the ss7 network (TFP inform all the SP of 
> > the inaccessibility). If a SG is unable to reach all the ss7 SP 
> > (200000 PC, means 200000 ungrouped SSNM) it can use the DOWN_ACK or 
> > INACTIVE_ACK procedure for its own isolation.
> 
> DUNA corresponds to MTP-PAUSE not TFP.
> 
> Consider 100 APSs inactive, zero (0) ASPs active.
> 
> The SG cannot be required to send DUNA to inactive ASPs.  When the SG
> NIF has no user to send MTP-PAUSE to, it is not possible for the SG to
> send DUNA.
> 
> The SG cannot be required to send SSNM to inactive ASPs.
> 
> Besides, it is unnecessary.
> 
> --brian
> 
> 
> > 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: Thursday, February 06, 2003 9:15 AM
> > To: Zannin, Francois
> > Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> > Subject: Re: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > Zannin,,
> > 
> > That is contrary to the conensus of the RFC.
> > 
> > Consider 1 ASP active, 200,000 ASPs inactive.
> > 
> > The RFC cannot require an SG to send SSNM to an inactive ASP.
> > 
> > --brian
> > 
> > Zannin, Francois wrote:
> > (Thu, 06 Feb 2003 09:02:41)
> > > Samuel,
> > > 
> > > IMO, it's a good idea to inform inactive (BUT UP) ASP of the state 
> > > of
> > > the ss7 network especially in the override mode where an ASP can be 
> > > ready to recover an AS as soon as it receive an AS-Pending 
> > > notification from the SGp. In this condition, the ASP is ready to 
> > > operate quickly and minimize the risk of buffered message lost
> during 
> > > the pending timer.
> > > 
> > > If someone does not want to receive SSNM because it is not able to
> > > recover an AS, it just stays in an ASP-Down state.
> > > 
> > > Because SSNM reflects ss7 Point Code state that is not "SI" 
> > > dependant,
> > 
> > > it affects all AS for a network and then, unless SG that plays
> > > multiple networks, the ASP UP is sufficient to filter the undesired 
> > > SSNM.
> > > 
> > > Francois.
> > > 
> > > -----Original Message-----
> > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > Sent: Thursday, February 06, 2003 8:35 AM
> > > To: Nagar, Ashish
> > > Cc: sigtran@ietf.org
> > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > 
> > > Ashish,
> > > 
> > > But when inactive ASP changes state to active, first of all it does
> > > not know whether to send DAUD or not.
> > > 
> > > It will try to send traffic - but actually destination is 
> > > unavailable.
> > > 
> > > The worst case is when SG triggers restart procedures this case all
> > > the SS7 destinations are unavailable via restarted-SG.But this poor 
> > > ASP (newly come-up ASP) keeps sending traffic via restarted-SG to 
> > > reach the SS7 destination and for every message ASP receives DUNA.
> > > 
> > > Are we not doing double-work here?
> > > 
> > > --Samuel
> > > 
> > > 
> > >   Alcatel USA, Inc.                  Internet:
> > > <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > *******
> > > 
> > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > 
> > > > Hi Samuel
> > > > 
> > > > My point is why to do any processing if ASP is in Inactive state. 
> > > > Think about ASP is in Inactive state for longer period and 
> > > > receiving
> > 
> > > > lots of DUNA and DAVA, while what matter is the present only when
> > ASP
> > > > state changes to ASP-ACTIVE.
> > > > 
> > > > Thanks & Regards
> > > > Ashish Nagar
> > > > 
> > > > 
> > > > -----Original Message-----
> > > > From: Samuel Dur D. Jeyaseelan 
> > > > [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > Sent: Thursday, February 06, 2003 12:30 PM
> > > > To: Nagar, Ashish
> > > > Cc: sigtran@ietf.org
> > > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > > 
> > > > 
> > > > Ashish,
> > > > 
> > > > My point is if SG is able to send DUNA to all inact and active 
> > > > ASPs, we should not restrict it.
> > > > 
> > > > Sending DAUD is an alternative way ,If SG has direct way of 
> > > > sending DUNA then Why do you need alternative way?
> > > > 
> > > > --Samuel
> > > > 
> > > >   Alcatel USA, Inc.                  Internet:
> > > <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, 
> > > > Inc.
> > > > *******
> > > > 
> > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > 
> > > > > You can generate DAUD procedure at any time, so whenever ASP
> > > > > changes
> > > 
> > > > > state from Inactive to Active Just generate DAUD.
> > > > > 
> > > > > Thanks & Regards
> > > > > Ashish Nagar
> > > > > 
> > > > > 
> > > > > -----Original Message-----
> > > > > From: Samuel Dur D. Jeyaseelan
> > > > > [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > > To: sigtran@ietf.org
> > > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > > 
> > > > > 
> > > > > Hi all,
> > > > > 
> > > > > Currently RFC 3332 says that SG sends SSNM messages to all 
> > > > > 'concerned ASPs'(means only ACTIVE ASPs).
> > > > > 
> > > > > +----+           +---+           +-----+
> > > > > |SEP | ----------|SG |-----------| ASP1|
> > > > > +----+           +---+           +-----+
> > > > >                    |             
> > > > >                    |             +-----+
> > > > >                    +-------------| ASP2|
> > > > >                                  +-----+
> > > > > 
> > > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > > 
> > > > > When SEP is getting restarted ,SG sends DUNA to ASP1. But SEP 
> > > > > status is not changed in ASP2.There are two ways to solve this 
> > > > > problem
> > > > > 
> > > > > 1. ASP1 can equalize its database with ASP2
> > > > > or
> > > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is 
> > > > > inactive)
> > > > > 
> > > > > IMO, SG can send SSNM messages to ASP (either ASP is in active 
> > > > > or inact
> > > > > state)
> > > > > 
> > > > > Any comments?
> > > > > 
> > > > > -----
> > > > > 
> > > > > section 5.4.2.1
> > > > > 
> > > > > ...
> > > > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of
> the
> > > > >         concerned Network Appearance.
> > > > > 
> > > > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER
> > > > > request
> > > 
> > > > > is sent to MTP3.
> > > > > 
> > > > > But there is no primitive availabe between M3UA and MTP3.
> > > > > 
> > > > > I think it should be corrected.
> > > > > 
> > > > > ----
> > > > > Presently ASP is unaware of SG's MTP3 restart,
> > > > > ASP receives only DUNA for all SS7 point codes which can be
> > > > > reachable via that SG.But in case of peer to peer protocol SG
> and 
> > > > > AS point codes are different, So why dont we send DUNA with SG's
> 
> > > > > point code to ASP? So that just by getting single DUNA(with SG's
> 
> > > > > PC) ASP will not route any message via that
> > > SG.
> > > > > This can avoid sending multiple DUNAs to ASP.
> > > > > 
> > > > > 
> > > > > Any comments?
> > > > > 
> > > > > -----
> > > > > section 1.4.1
> > > > > 
> > > > > ...
> > > > >    Where a particular AS can be reached via more than one SGP,
> the
> > > > >    corresponding Routing Keys in the SGPs should be identical.
> > > (Note:
> > > > >    It is possible for the SGP Routing Key configuration data to
> be
> > > > >    temporarily out-of-sync during configuration updates).
> > > > >  
> > > > >                               +--------+
> > > > >                               |        |
> > > > >                  +------------+  SG 1  +--------------+
> > > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > > >      |  SEP  +---+            +--------+              +---/
> \
> > > > >      |   or  |                    |*                      | ASPs
> > |
> > > > >      |  STP  +---+            +--------+              +---\
> /
> > > > >      +-------+   |            |        |              |     ----
> > > > >                  +------------+  SG 2  +--------------+
> > > > >                               | "STP"  |
> > > > >                               +--------+
> > > > >  
> > > > >                     Figure 1  Example with mated SGs
> > > > > 
> > > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are 
> > > > > located geographically in different places.The ideas is when SG1
> 
> > > > > goes down in CITY1(due to earthquake...) , all the traffice 
> > > > > still can be handled by SG2 which is in CITY2.
> > > > > 
> > > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will
> > > > > be
> > > > > routed from SGP1 to SGP2.
> > > > > 
> > > > > Does all these things mean to say that SGP1's mate 'should be' 
> > > > > present in SG2 and not in SG1 itself ?
> > > > >  
> > > > > 
> > > > > Comments are welcome.
> > > > > 
> > > > > --Samuel
> > > > > 
> > > > > 
> > > > > 
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet:
> > > > <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA,
> > > > > Inc.
> > > > *******
> > > > > 
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > 
> > > > 
> > > 
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > 
> > --
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 04:21:56 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA09643
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 04:21:56 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h169SIj15295
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 04:28:18 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169S4p15273;
	Thu, 6 Feb 2003 04:28:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169R1p15232
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 04:27:01 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA09597
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 04:20:06 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:EWGowdnUdcbQL+d59xyNuIOjmDoD3tey@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h169NiD14586;
	Thu, 6 Feb 2003 02:23:44 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h169Ni921610;
	Thu, 6 Feb 2003 02:23:44 -0700
Date: Thu, 6 Feb 2003 02:23:44 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: Richard Adams <Richard.Adams@etsi.org>,
        "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206022344.B18441@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	Richard Adams <Richard.Adams@etsi.org>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <4091553999CBE4409CC2B562152B5A33EDDF58@email10.etsihq.org> <Pine.SOL.4.10.10302060256200.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302060256200.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 03:01:39AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 03:01:39)
> 
> Richard,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Richard Adams wrote:
> 
> > Brian,
> > I agree with you. I think when the ASP becomes active, it should do an
> > audit to update its knowledge of the SS7 network. The ASP needs to take
> > the initiative to find out about the SS7 nodes etc. in which it is
> > interested. Any other method of working would lead to impossible loading
> > on the SG.
> 
> When SG itself is re-starting , it means no ASP can reach any SS7
> destinations via re-started SG.Always it is good to tell my ASP
> who is in inactive state about SS7 destinations unavailability.

It is unneccessary.  Iff the ASP goes active, the SG can send DUNA(*)
to it in one message to indicate that it is restarting.  The SG isn't
required to wait for DAUD to do this, it could send DUNA(*) just because
the ASP hase become active, or it could send DUNA(*) in response to the
first DATA message from the ASP.

There is no need to send SSNM to inactive ASPs.

--brian


> 
> --Samuel
> 
> > Of course, I hope 199,999 of your inactive ASPs do not decide to become
> > active at the same time.....! 
> > Kind regards,
> > Dick Adams,
> > richard.adams@etsi.org
> > 
> >  
> > 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: 06 February 2003 09:15
> > To: Zannin, Francois
> > Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> > Subject: Re: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > Zannin,,
> > 
> > That is contrary to the conensus of the RFC.
> > 
> > Consider 1 ASP active, 200,000 ASPs inactive.
> > 
> > The RFC cannot require an SG to send SSNM to an inactive ASP.
> > 
> > --brian
> > 
> > Zannin, Francois wrote:
> > (Thu, 06 Feb 2003 09:02:41)
> > > Samuel,
> > > 
> > > IMO, it's a good idea to inform inactive (BUT UP) ASP of the state of
> > > the ss7 network especially in the override mode where an ASP can be
> > > ready to recover an AS as soon as it receive an AS-Pending
> > notification
> > > from the SGp. In this condition, the ASP is ready to operate quickly
> > and
> > > minimize the risk of buffered message lost during the pending timer.
> > > 
> > > If someone does not want to receive SSNM because it is not able to
> > > recover an AS, it just stays in an ASP-Down state.
> > > 
> > > Because SSNM reflects ss7 Point Code state that is not "SI" dependant,
> > > it affects all AS for a network and then, unless SG that plays
> > multiple
> > > networks, the ASP UP is sufficient to filter the undesired SSNM.
> > > 
> > > Francois.
> > > 
> > > -----Original Message-----
> > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> > > Sent: Thursday, February 06, 2003 8:35 AM
> > > To: Nagar, Ashish
> > > Cc: sigtran@ietf.org
> > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > 
> > > Ashish,
> > > 
> > > But when inactive ASP changes state to active, first of all
> > > it does not know whether to send DAUD or not.
> > > 
> > > It will try to send traffic - but actually destination is unavailable.
> > > 
> > > The worst case is when SG triggers restart procedures this case all
> > the
> > > SS7 destinations are unavailable via restarted-SG.But this poor ASP
> > > (newly come-up ASP) keeps sending traffic via restarted-SG to reach
> > the
> > > SS7 destination and for every message ASP receives DUNA.
> > > 
> > > Are we not doing double-work here?
> > > 
> > > --Samuel
> > > 
> > > 
> > >   Alcatel USA, Inc.                  Internet:
> > > <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > *******
> > > 
> > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > 
> > > > Hi Samuel
> > > > 
> > > > My point is why to do any processing if ASP is in Inactive state. 
> > > > Think about ASP is in Inactive state for longer period and receiving
> > 
> > > > lots of DUNA and DAVA, while what matter is the present only when
> > ASP 
> > > > state changes to ASP-ACTIVE.
> > > > 
> > > > Thanks & Regards
> > > > Ashish Nagar
> > > > 
> > > > 
> > > > -----Original Message-----
> > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > Sent: Thursday, February 06, 2003 12:30 PM
> > > > To: Nagar, Ashish
> > > > Cc: sigtran@ietf.org
> > > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > > 
> > > > 
> > > > Ashish,
> > > > 
> > > > My point is if SG is able to send DUNA to all inact and active ASPs,
> > 
> > > > we should not restrict it.
> > > > 
> > > > Sending DAUD is an alternative way ,If SG has direct way of sending 
> > > > DUNA then Why do you need alternative way?
> > > > 
> > > > --Samuel
> > > > 
> > > >   Alcatel USA, Inc.                  Internet:
> > > <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. 
> > > > *******
> > > > 
> > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > 
> > > > > You can generate DAUD procedure at any time, so whenever ASP
> > changes
> > > 
> > > > > state from Inactive to Active Just generate DAUD.
> > > > > 
> > > > > Thanks & Regards
> > > > > Ashish Nagar
> > > > > 
> > > > > 
> > > > > -----Original Message-----
> > > > > From: Samuel Dur D. Jeyaseelan
> > [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > > To: sigtran@ietf.org
> > > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > > 
> > > > > 
> > > > > Hi all,
> > > > > 
> > > > > Currently RFC 3332 says that SG sends SSNM messages to all 
> > > > > 'concerned ASPs'(means only ACTIVE ASPs).
> > > > > 
> > > > > +----+           +---+           +-----+
> > > > > |SEP | ----------|SG |-----------| ASP1|
> > > > > +----+           +---+           +-----+
> > > > >                    |             
> > > > >                    |             +-----+
> > > > >                    +-------------| ASP2|
> > > > >                                  +-----+
> > > > > 
> > > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > > 
> > > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > > But SEP status is not changed in ASP2.There are two ways to solve 
> > > > > this problem
> > > > > 
> > > > > 1. ASP1 can equalize its database with ASP2
> > > > > or
> > > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > > 
> > > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or 
> > > > > inact
> > > > > state)
> > > > > 
> > > > > Any comments?
> > > > > 
> > > > > -----
> > > > > 
> > > > > section 5.4.2.1
> > > > > 
> > > > > ...
> > > > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> > > > >         concerned Network Appearance.
> > > > > 
> > > > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER
> > request
> > > 
> > > > > is sent to MTP3.
> > > > > 
> > > > > But there is no primitive availabe between M3UA and MTP3.
> > > > > 
> > > > > I think it should be corrected.
> > > > > 
> > > > > ----
> > > > > Presently ASP is unaware of SG's MTP3 restart,
> > > > > ASP receives only DUNA for all SS7 point codes which can be
> > > > > reachable via that SG.But in case of peer to peer protocol
> > > > > SG and AS point codes are different, So why dont we send
> > > > > DUNA with SG's point code to ASP? So that just by getting
> > > > > single DUNA(with SG's PC) ASP will not route any message via that
> > > SG.
> > > > > This can avoid sending multiple DUNAs to ASP.
> > > > > 
> > > > > 
> > > > > Any comments?
> > > > > 
> > > > > -----
> > > > > section 1.4.1
> > > > > 
> > > > > ...
> > > > >    Where a particular AS can be reached via more than one SGP, the
> > > > >    corresponding Routing Keys in the SGPs should be identical.
> > > (Note:
> > > > >    It is possible for the SGP Routing Key configuration data to be
> > > > >    temporarily out-of-sync during configuration updates).
> > > > >  
> > > > >                               +--------+
> > > > >                               |        |
> > > > >                  +------------+  SG 1  +--------------+
> > > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > > >      |  SEP  +---+            +--------+              +---/      \
> > > > >      |   or  |                    |*                      | ASPs
> > |
> > > > >      |  STP  +---+            +--------+              +---\      /
> > > > >      +-------+   |            |        |              |     ----
> > > > >                  +------------+  SG 2  +--------------+
> > > > >                               | "STP"  |
> > > > >                               +--------+
> > > > >  
> > > > >                     Figure 1  Example with mated SGs
> > > > > 
> > > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are 
> > > > > located geographically in different places.The ideas is when SG1 
> > > > > goes down in CITY1(due to earthquake...) , all the traffice still 
> > > > > can be handled by SG2 which is in CITY2.
> > > > > 
> > > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will
> > be 
> > > > > routed from SGP1 to SGP2.
> > > > > 
> > > > > Does all these things mean to say that SGP1's mate 'should be' 
> > > > > present in SG2 and not in SG1 itself ?
> > > > >  
> > > > > 
> > > > > Comments are welcome.
> > > > > 
> > > > > --Samuel
> > > > > 
> > > > > 
> > > > > 
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet:
> > > > <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA,
> > Inc.
> > > > *******
> > > > > 
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > 
> > > > 
> > > 
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 04:22:11 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA09665
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 04:22:11 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h169SXB15323
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 04:28:33 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169S6p15288;
	Thu, 6 Feb 2003 04:28:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169Rrp15255
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 04:27:53 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA09612
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 04:20:59 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:EgrwZcCHwkbiHIMBlQPMbGYoLUXQwZYT@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h169ObD14599;
	Thu, 6 Feb 2003 02:24:37 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h169Obx21714;
	Thu, 6 Feb 2003 02:24:37 -0700
Date: Thu, 6 Feb 2003 02:24:37 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206022437.C18441@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206015644.D15998@openss7.org> <Pine.SOL.4.10.10302060308190.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302060308190.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 03:13:56AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 03:13:56)
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 02:51:52)
> > > 
> > > Brian,
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Zannin,,
> > > > 
> > > > Zannin, Francois wrote:                                                            (Thu, 06 Feb 2003 09:27:21)
> > > > > Bidulock,,,
> > > > > 
> > > > > I think your example of 200,000 ASP "UP" as backup for an AS is really
> > > > > realistic for a network!
> > > > 
> > > > 200?  20?  It makes no difference.  The SG cannot be required to
> > > > send useless messages to bunch of ASPs that can do nothing with
> > > > the information.
> > > > 
> > > Everything is useless at one point of time, we should see always
> > > the 'usefullness of many times".
> > 
> > In this case it is useless all the time to send SSNM to inactive ASPs.
> > 
> > > > > 
> > > > > The goal is to agree on procedure that can be used for the
> > > > > implementation of M3UA. There were a lot of threads discussing on this
> > > > > subject and not always concluded. The role of the IG is to clarify the
> > > > > RFC in order to make it usable.
> > > > > 
> > > > > This procedure similar to the ss7 network (TFP inform all the SP of the
> > > > > inaccessibility). If a SG is unable to reach all the ss7 SP (200000 PC,
> > > > > means 200000 ungrouped SSNM) it can use the DOWN_ACK or INACTIVE_ACK
> > > > > procedure for its own isolation.
> > > > 
> > > > DUNA corresponds to MTP-PAUSE not TFP.
> > > > 
> > > > Consider 100 APSs inactive, zero (0) ASPs active.
> > > 
> > > I can say 100 ASPs inactive and 100 ASPs active with
> > > override mode.
> > 
> > 100 ASP cannot be active for the same AS in override mode: only 1 can.
> > 
> 
> Total misinterpretation.ASP1 is active ASP2 is inactive....
> ASP100 is active and ASP100 is inactive - VIA RE-STARTED SG.
> 
> Now 100 ASPs are active and 100 ASPs are inactive from the point
> of re-started SG.

100 ASP cannot be active for the same AS in override mode: only 1 can.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 04:28:11 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA09767
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 04:28:11 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h169YX215599
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 04:34:33 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169YIp15588;
	Thu, 6 Feb 2003 04:34:18 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169XJp15543
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 04:33:19 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA09740
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 04:26:25 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h169U2qe017241;
	Thu, 6 Feb 2003 03:30:02 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h169TmM19082;
	Thu, 6 Feb 2003 03:29:48 -0600 (CST)
Date: Thu, 6 Feb 2003 03:29:48 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206012849.A14210@openss7.org>
Message-ID: <Pine.SOL.4.10.10302060319350.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> It will decrease M3UA efficiency.
> 
> Consider 1 active ASP and 200,000 inactive ASPs.
> 
You are deviating from original thread.

ASP1 is active and ASP2 is inactive -overrid mode (via SG1)
.
.
.
ASP200,000 is active and ASP200,000 is inactive -overrid mode(via SG1)
 
Now we could see that 200,000 ASPs are active and 200,000 ASPs are
inactive via RE-STARTED SG1.But all active 200,000 ASPs database
is not same as inactive 200,000 ASPs.
 
According to you , SG1 will receive 200,000 DAUD msgs when inact ASPs
come up and again 200,000 DUNA are being sent from SG1.

According to me only 200,000 DUNAs are sent from SG1 if we allow SSNM
messages to be processing inact state.So that we can avoid sending 200,000
DAUD messages.




> For each SSNM the SG must send 200,001 DUNA messages.
> 
> With the existing concensus DAUD procedure, the SG only
> sends 1 DUNA message.
> 
> When an ASP activates for an AS it sends one DAUD message
> to avoid the SG sending many useless SSNM messages.
> 
> The activating ASP need not even send DAUD.  It can just
> start sending messages and use responsive DUNAs.
> 
> This is most efficient because SS7 destinations are almost
> never unavailable.
> 
> Sending SSNM to inactive ASPs would be a BAD thing.  It is
> good that the RFC does not permit this.
No comment

--Samuel
> 
> --brian
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 02:18:28)
> > 
> > Brian,
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > IMO sending SSNM to inactive ASPs is a BAD idea.
> > > 
> > Which is bad idea sending DAUD and get the DUNA from SG
> > or SG itself can send DUNA to inact ASP?
> > 
> > IMO my idea is not BAD.It will increase the M3UA stack
> > efficency.
> > 
> > > --brian
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 00:30:30)
> > > > 
> > > > Brian,
> > > > 
> > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > 
> > > > On Wed, 5 Feb 2003, Brian F. G. Bidulock wrote:
> > > > 
> > > > > Samuel,
> > > > > 
> > > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Wed, 05 Feb 2003 23:57:10)
> > > > > > 
> > > > > > Hi all,
> > > > > > 
> > > > > > Currently RFC 3332 says that SG sends SSNM messages to all 'concerned
> > > > > > ASPs'(means only ACTIVE ASPs).
> > > > > > 
> > > > > > +----+           +---+           +-----+
> > > > > > |SEP | ----------|SG |-----------| ASP1|
> > > > > > +----+           +---+           +-----+
> > > > > >                    |             
> > > > > >                    |             +-----+
> > > > > >                    +-------------| ASP2|
> > > > > >                                  +-----+
> > > > > > 
> > > > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > > > 
> > > > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > > > But SEP status is not changed in ASP2.There are two ways to solve this
> > > > > > problem
> > > > > > 
> > > > > > 1. ASP1 can equalize its database with ASP2
> > > > > > or
> > > > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > > > 
> > > > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or inact
> > > > > > state)
> > > > > > 
> > > > > > Any comments?
> > > > > 
> > > > > The audit procedure is provided to allow any ASP to determine the
> > > > > current accessibility of destinations from the SG.
> > > > > 
> > > > > The prosrcibed procedure in RFC 3332 is as follows:
> > > > > 
> > > > >   If ASP2 can know from ASP1 that the destination is inaccessible
> > > > >   then nothing is required.
> > > > This is out of the scope of RFC.
> > > > 
> > > > > 
> > > > >   If ASP2 does not know from ASP1 that destinations are inaccessible,
> > > > >   it issues DAUD when it goes active and the SG tells it which
> > > > >   destinations are accessible.
> > > > I think getting DUNA from SG is better than sending DAUD and receiving
> > > > response.If SG is able to do it ,let SG send DUNA to inactive ASPs.
> > > > IMO there is no harm.
> > > > > 
> > > > > SSNM are NOT sent to inactive ASPs.
> > > > > 
> > > > > The reason for this is as follows:
> > > > > 
> > > > > Consider that ASP1 is active and ASP2 is down.  SG sends DUNA
> > > > > to ASP1.  APS2 does not receive this because APS2 is not even
> > > > > running.
> > > > > 
> > > > > ASP2 goes ASPUP and moves to the inactive state for the AS for
> > > > > which the DUNA was sent.
> > > > > 
> > > > > Can the SG be expected to then send this backlog of information?
> > > > > 
> > > > 
> > > > I mean to say that when SG receives TFP from SEP and if the state
> > > > of ASP is ACTIVE or INACT then sending DUNA is good idea.
> > > > 
> > > > If ASP is in DOWN state then SG will not send DUNA.
> > > > IMO, practically speaking backup ASP will not be in DOWN state,
> > > > It should be in ASP INACT state.
> > > > 
> > > > My point is if SG is able to send DUNA , we should not put any
> > > > restriction.
> > > > 
> > > > --Samuel
> > > > 
> > > > > No, it is the repsonsibility of ASP2 to send DAUD to discover
> > > > > the information.
> > > > > 
> > > > > This is also the case for the inactive state.
> > > > > 
> > > > > --brian
> > > > > 
> > > > > -- 
> > > > > Brian F. G. Bidulock
> > > > > bidulock@openss7.org
> > > > > http://www.openss7.org/
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > 
> > > 
> > > -- 
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 04:37:00 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA09959
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 04:37:00 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h169hM616651
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 04:43:22 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169h9p16618;
	Thu, 6 Feb 2003 04:43:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169g4p16577
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 04:42:04 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA09922
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 04:35:10 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:gCBc4f/kl8xbr6AmhJUJAwuq05qDtpXJ@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h169cmD14904;
	Thu, 6 Feb 2003 02:38:48 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h169cmM23515;
	Thu, 6 Feb 2003 02:38:48 -0700
Date: Thu, 6 Feb 2003 02:38:48 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206023848.D18441@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	sigtran@ietf.org
References: <20030206012849.A14210@openss7.org> <Pine.SOL.4.10.10302060319350.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302060319350.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 03:29:48AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

I was my example.

1 ASP is active for AS1, 200,000 ASPs are inactive for AS1.

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 03:29:48)
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > It will decrease M3UA efficiency.
> > 
> > Consider 1 active ASP and 200,000 inactive ASPs.
> > 
> You are deviating from original thread.
> 
> ASP1 is active and ASP2 is inactive -overrid mode (via SG1)
> .
> .
> .
> ASP200,000 is active and ASP200,000 is inactive -overrid mode(via SG1)
>  
> Now we could see that 200,000 ASPs are active and 200,000 ASPs are
> inactive via RE-STARTED SG1.But all active 200,000 ASPs database
> is not same as inactive 200,000 ASPs.
>  
> According to you , SG1 will receive 200,000 DAUD msgs when inact ASPs
> come up and again 200,000 DUNA are being sent from SG1.
> 
> According to me only 200,000 DUNAs are sent from SG1 if we allow SSNM
> messages to be processing inact state.So that we can avoid sending 200,000
> DAUD messages.
> 
> 
> 
> 
> > For each SSNM the SG must send 200,001 DUNA messages.
> > 
> > With the existing concensus DAUD procedure, the SG only
> > sends 1 DUNA message.
> > 
> > When an ASP activates for an AS it sends one DAUD message
> > to avoid the SG sending many useless SSNM messages.
> > 
> > The activating ASP need not even send DAUD.  It can just
> > start sending messages and use responsive DUNAs.
> > 
> > This is most efficient because SS7 destinations are almost
> > never unavailable.
> > 
> > Sending SSNM to inactive ASPs would be a BAD thing.  It is
> > good that the RFC does not permit this.
> No comment
> 
> --Samuel
> > 
> > --brian
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 02:18:28)
> > > 
> > > Brian,
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Samuel,
> > > > 
> > > > IMO sending SSNM to inactive ASPs is a BAD idea.
> > > > 
> > > Which is bad idea sending DAUD and get the DUNA from SG
> > > or SG itself can send DUNA to inact ASP?
> > > 
> > > IMO my idea is not BAD.It will increase the M3UA stack
> > > efficency.
> > > 
> > > > --brian
> > > > 
> > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 00:30:30)
> > > > > 
> > > > > Brian,
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > > 
> > > > > On Wed, 5 Feb 2003, Brian F. G. Bidulock wrote:
> > > > > 
> > > > > > Samuel,
> > > > > > 
> > > > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Wed, 05 Feb 2003 23:57:10)
> > > > > > > 
> > > > > > > Hi all,
> > > > > > > 
> > > > > > > Currently RFC 3332 says that SG sends SSNM messages to all 'concerned
> > > > > > > ASPs'(means only ACTIVE ASPs).
> > > > > > > 
> > > > > > > +----+           +---+           +-----+
> > > > > > > |SEP | ----------|SG |-----------| ASP1|
> > > > > > > +----+           +---+           +-----+
> > > > > > >                    |             
> > > > > > >                    |             +-----+
> > > > > > >                    +-------------| ASP2|
> > > > > > >                                  +-----+
> > > > > > > 
> > > > > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > > > > 
> > > > > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > > > > But SEP status is not changed in ASP2.There are two ways to solve this
> > > > > > > problem
> > > > > > > 
> > > > > > > 1. ASP1 can equalize its database with ASP2
> > > > > > > or
> > > > > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > > > > 
> > > > > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or inact
> > > > > > > state)
> > > > > > > 
> > > > > > > Any comments?
> > > > > > 
> > > > > > The audit procedure is provided to allow any ASP to determine the
> > > > > > current accessibility of destinations from the SG.
> > > > > > 
> > > > > > The prosrcibed procedure in RFC 3332 is as follows:
> > > > > > 
> > > > > >   If ASP2 can know from ASP1 that the destination is inaccessible
> > > > > >   then nothing is required.
> > > > > This is out of the scope of RFC.
> > > > > 
> > > > > > 
> > > > > >   If ASP2 does not know from ASP1 that destinations are inaccessible,
> > > > > >   it issues DAUD when it goes active and the SG tells it which
> > > > > >   destinations are accessible.
> > > > > I think getting DUNA from SG is better than sending DAUD and receiving
> > > > > response.If SG is able to do it ,let SG send DUNA to inactive ASPs.
> > > > > IMO there is no harm.
> > > > > > 
> > > > > > SSNM are NOT sent to inactive ASPs.
> > > > > > 
> > > > > > The reason for this is as follows:
> > > > > > 
> > > > > > Consider that ASP1 is active and ASP2 is down.  SG sends DUNA
> > > > > > to ASP1.  APS2 does not receive this because APS2 is not even
> > > > > > running.
> > > > > > 
> > > > > > ASP2 goes ASPUP and moves to the inactive state for the AS for
> > > > > > which the DUNA was sent.
> > > > > > 
> > > > > > Can the SG be expected to then send this backlog of information?
> > > > > > 
> > > > > 
> > > > > I mean to say that when SG receives TFP from SEP and if the state
> > > > > of ASP is ACTIVE or INACT then sending DUNA is good idea.
> > > > > 
> > > > > If ASP is in DOWN state then SG will not send DUNA.
> > > > > IMO, practically speaking backup ASP will not be in DOWN state,
> > > > > It should be in ASP INACT state.
> > > > > 
> > > > > My point is if SG is able to send DUNA , we should not put any
> > > > > restriction.
> > > > > 
> > > > > --Samuel
> > > > > 
> > > > > > No, it is the repsonsibility of ASP2 to send DAUD to discover
> > > > > > the information.
> > > > > > 
> > > > > > This is also the case for the inactive state.
> > > > > > 
> > > > > > --brian
> > > > > > 
> > > > > > -- 
> > > > > > Brian F. G. Bidulock
> > > > > > bidulock@openss7.org
> > > > > > http://www.openss7.org/
> > > > > > _______________________________________________
> > > > > > Sigtran mailing list
> > > > > > Sigtran@ietf.org
> > > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > > 
> > > > 
> > > > -- 
> > > > Brian F. G. Bidulock
> > > > bidulock@openss7.org
> > > > http://www.openss7.org/
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > 
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 04:37:07 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA09963
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 04:37:07 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h169hTI16668
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 04:43:29 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169hBp16641;
	Thu, 6 Feb 2003 04:43:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169gLp16589
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 04:42:21 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA09928
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 04:35:26 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h169d2qg017810;
	Thu, 6 Feb 2003 03:39:04 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h169cVM22066;
	Thu, 6 Feb 2003 03:38:31 -0600 (CST)
Date: Thu, 6 Feb 2003 03:38:31 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <Pine.SOL.4.10.10302060319350.22291-100000@sun3271.ssd.usa.alcatel.com>
Message-ID: <Pine.SOL.4.10.10302060336310.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian (last mail had some problem, i am re-sending),


  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Samuel Dur D. Jeyaseelan wrote:

> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > It will decrease M3UA efficiency.
> > 
> > Consider 1 active ASP and 200,000 inactive ASPs.
> > 
 You are deviating from original thread.
 
 ASP1 is active and ASP2 is inactive -overrid mode (via SG1)
 .
 .
 .
 ASP200,000 is active and ASP200,000 is inactive -overrid mode(via SG1)
  
 Now we could see that 200,000 ASPs are active and 200,000 ASPs are
 inactive via RE-STARTED SG1.But all active 200,000 ASPs database
 is not same as inactive 200,000 ASPs.
  
 According to you , SG1 will receive 200,000 DAUD msgs when inact ASPs
 come up and again 200,000 DUNA are being sent from SG1.
 
 According to me only 200,000 DUNAs are sent from SG1 if we allow SSNM
 messages to be processing inact state.So that we can avoid sending 200,000
 DAUD messages.
> 
> > For each SSNM the SG must send 200,001 DUNA messages.
> > 
> > With the existing concensus DAUD procedure, the SG only
> > sends 1 DUNA message.
> > 
> > When an ASP activates for an AS it sends one DAUD message
> > to avoid the SG sending many useless SSNM messages.
> > 
> > The activating ASP need not even send DAUD.  It can just
> > start sending messages and use responsive DUNAs.
> > 
> > This is most efficient because SS7 destinations are almost
> > never unavailable.
> > 
> > Sending SSNM to inactive ASPs would be a BAD thing.  It is
> > good that the RFC does not permit this.

 No comment
> 
> --Samuel
> > 
> > --brian
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 02:18:28)
> > > 
> > > Brian,
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Samuel,
> > > > 
> > > > IMO sending SSNM to inactive ASPs is a BAD idea.
> > > > 
> > > Which is bad idea sending DAUD and get the DUNA from SG
> > > or SG itself can send DUNA to inact ASP?
> > > 
> > > IMO my idea is not BAD.It will increase the M3UA stack
> > > efficency.
> > > 
> > > > --brian
> > > > 
> > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 00:30:30)
> > > > > 
> > > > > Brian,
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > > 
> > > > > On Wed, 5 Feb 2003, Brian F. G. Bidulock wrote:
> > > > > 
> > > > > > Samuel,
> > > > > > 
> > > > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Wed, 05 Feb 2003 23:57:10)
> > > > > > > 
> > > > > > > Hi all,
> > > > > > > 
> > > > > > > Currently RFC 3332 says that SG sends SSNM messages to all 'concerned
> > > > > > > ASPs'(means only ACTIVE ASPs).
> > > > > > > 
> > > > > > > +----+           +---+           +-----+
> > > > > > > |SEP | ----------|SG |-----------| ASP1|
> > > > > > > +----+           +---+           +-----+
> > > > > > >                    |             
> > > > > > >                    |             +-----+
> > > > > > >                    +-------------| ASP2|
> > > > > > >                                  +-----+
> > > > > > > 
> > > > > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > > > > 
> > > > > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > > > > But SEP status is not changed in ASP2.There are two ways to solve this
> > > > > > > problem
> > > > > > > 
> > > > > > > 1. ASP1 can equalize its database with ASP2
> > > > > > > or
> > > > > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > > > > 
> > > > > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or inact
> > > > > > > state)
> > > > > > > 
> > > > > > > Any comments?
> > > > > > 
> > > > > > The audit procedure is provided to allow any ASP to determine the
> > > > > > current accessibility of destinations from the SG.
> > > > > > 
> > > > > > The prosrcibed procedure in RFC 3332 is as follows:
> > > > > > 
> > > > > >   If ASP2 can know from ASP1 that the destination is inaccessible
> > > > > >   then nothing is required.
> > > > > This is out of the scope of RFC.
> > > > > 
> > > > > > 
> > > > > >   If ASP2 does not know from ASP1 that destinations are inaccessible,
> > > > > >   it issues DAUD when it goes active and the SG tells it which
> > > > > >   destinations are accessible.
> > > > > I think getting DUNA from SG is better than sending DAUD and receiving
> > > > > response.If SG is able to do it ,let SG send DUNA to inactive ASPs.
> > > > > IMO there is no harm.
> > > > > > 
> > > > > > SSNM are NOT sent to inactive ASPs.
> > > > > > 
> > > > > > The reason for this is as follows:
> > > > > > 
> > > > > > Consider that ASP1 is active and ASP2 is down.  SG sends DUNA
> > > > > > to ASP1.  APS2 does not receive this because APS2 is not even
> > > > > > running.
> > > > > > 
> > > > > > ASP2 goes ASPUP and moves to the inactive state for the AS for
> > > > > > which the DUNA was sent.
> > > > > > 
> > > > > > Can the SG be expected to then send this backlog of information?
> > > > > > 
> > > > > 
> > > > > I mean to say that when SG receives TFP from SEP and if the state
> > > > > of ASP is ACTIVE or INACT then sending DUNA is good idea.
> > > > > 
> > > > > If ASP is in DOWN state then SG will not send DUNA.
> > > > > IMO, practically speaking backup ASP will not be in DOWN state,
> > > > > It should be in ASP INACT state.
> > > > > 
> > > > > My point is if SG is able to send DUNA , we should not put any
> > > > > restriction.
> > > > > 
> > > > > --Samuel
> > > > > 
> > > > > > No, it is the repsonsibility of ASP2 to send DAUD to discover
> > > > > > the information.
> > > > > > 
> > > > > > This is also the case for the inactive state.
> > > > > > 
> > > > > > --brian
> > > > > > 
> > > > > > -- 
> > > > > > Brian F. G. Bidulock
> > > > > > bidulock@openss7.org
> > > > > > http://www.openss7.org/
> > > > > > _______________________________________________
> > > > > > Sigtran mailing list
> > > > > > Sigtran@ietf.org
> > > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > > 
> > > > 
> > > > -- 
> > > > Brian F. G. Bidulock
> > > > bidulock@openss7.org
> > > > http://www.openss7.org/
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > 
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > 
> 
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 04:41:58 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA10097
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 04:41:58 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h169mKI16870
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 04:48:20 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169m3p16859;
	Thu, 6 Feb 2003 04:48:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169lSp16830
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 04:47:29 -0500
Received: from zdemail03.zdem.compaq.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA10053
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 04:40:33 -0500 (EST)
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail03.zdem.compaq.com (Postfix) with ESMTP
	id 9E6195A0; Thu,  6 Feb 2003 10:44:10 +0100 (CET)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.2966);
	 Thu, 6 Feb 2003 10:44:09 +0100
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Subject: RE: [Sigtran] M3UA - ASP database equalization
Date: Thu, 6 Feb 2003 10:44:08 +0100
Message-ID: <06CAA4D3753C53408D2CE6340B68444214C40E@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA - ASP database equalization
Thread-Index: AcLNwTLedn79QXW0SzuB8W7YseqKKwAAMgow
From: "Zannin, Francois" <Francois.Zannin@hp.com>
To: <bidulock@openss7.org>
Cc: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 06 Feb 2003 09:44:09.0279 (UTC) FILETIME=[4C09CCF0:01C2CDC4]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h169lTp16831
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Bidulock,,

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> Sent: Thursday, February 06, 2003 10:22 AM
> To: Zannin, Francois
> Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> Subject: Re: [Sigtran] M3UA - ASP database equalization
> 
> 
> Zannin,,
> 
> Zannin, Francois wrote:                                       
>                      (Thu, 06 Feb 2003 09:58:46)
> > Bidulock,,
> > 
> > Does it make sense to have "a lot" of ASP UP connected to a SGP and 
> > not ready to do anything?
> 
> They could be active for different AS.

In that case, because SSNM are for ss7 SP and concern all SI (AS), it is
not useless to receive (and you will receive) these SSNM. Now, the
question is about the RC in these SSNM.

> 
> > If an ASP is UP for a SGP, it certainly must be ready to 
> activate an 
> > AS and then concerned by the SSNM, and again, for an emergency AS 
> > recovery (AS-Pending notification) the ASP must be ready to operate 
> > after the Active-ACK.
> > 
> > Only the "ready to operate for an AS" ASP have a reason to 
> be "UP" on 
> > a SG.
> 
> Wrong.  Whenever an ASP goes ASP UP it becomes ASP-INACTIVE 
> for all AS for which it is configured to operate.  Depending 
> on SG configuration, this could mean that the ASP is 
> ASP-INACTIVE for all AS served by the SG, regarless of 
> whether the ASP will ever go ASP-ACTIVE for all of the AS.  
> There is no way for an ASP to tell the SG that it is ASP-DOWN 
> for some AS and ASP-INACTIVE for others.
> 
> Also, the purpose of the ASP-INACTIVE state is to receive 
> NTFY messages for the AS.  In override mode, one ASP is 
> active for the AS and the others are inactive.  The inactive 
> ASPs have no intention of going active unless the active ASP 
> fails, or unless manual management action is taken.
> 

A ss7 state concern all SI (AS) for a NA, the ASP receive the SSNM for
the ACTIVE (or activable) AS.
If no AS is "activable" no reason to send an ASP-UP...

This can be less efficent for SG that plays multiple network!

> > I don't think that it is useful to connect "UP" a lot of 
> ASP with no 
> > reason!
> 
> See above.
> 
> > 
> > You are right, DUNA is not the same thing as MTP-PAUSE, but 
> the sense 
> > is the same, a user (AS) cannot use a ROUTE (or SG) if the 
> remote SP 
> > is inaccessible because of TFP, TCP and then MTP-PAUSE which is the 
> > consolidation of all the routes for a destination.
> > 
> > The reason to wait for a DAUD for the ss7 state on an ASP 
> is to have a 
> > selective use of the ss7 network SP by an AS (then ASP). In 
> that case, 
> > the number of DAUD is limited to the minimum after the ACTIVE 
> > procedure. But as soon as the ASP is active, there is no 
> filter in the 
> > SG for ASP concerned by the SSNM for specific destinations.
> 
> MTP procedures require MTP-PAUSE to be sent to MTP users that 
> are sending to inaccessible destinations.  An MTP conforming 
> SG will send DUNA in response to messages sent to an 
> inacessible destination.
> 

Do you mean that when a SG receive a TFP on the last route for a
destination, it does not send the DUNA to all the ASP???
If yes, that a NEW behavior for a SG?????

> --brian
> 
> > 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: Thursday, February 06, 2003 9:36 AM
> > To: Zannin, Francois
> > Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> > Subject: Re: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > Zannin,,
> > 
> > Zannin, Francois wrote:
> > (Thu, 06 Feb 2003 09:27:21)
> > > Bidulock,,,
> > > 
> > > I think your example of 200,000 ASP "UP" as backup for an AS is 
> > > really
> > 
> > > realistic for a network!
> > 
> > 200?  20?  It makes no difference.  The SG cannot be 
> required to send 
> > useless messages to bunch of ASPs that can do nothing with the 
> > information.
> > 
> > > 
> > > The goal is to agree on procedure that can be used for the
> > > implementation of M3UA. There were a lot of threads 
> discussing on this
> > 
> > > subject and not always concluded. The role of the IG is 
> to clarify 
> > > the
> > 
> > > RFC in order to make it usable.
> > > 
> > > This procedure similar to the ss7 network (TFP inform all 
> the SP of
> > > the inaccessibility). If a SG is unable to reach all the ss7 SP 
> > > (200000 PC, means 200000 ungrouped SSNM) it can use the 
> DOWN_ACK or 
> > > INACTIVE_ACK procedure for its own isolation.
> > 
> > DUNA corresponds to MTP-PAUSE not TFP.
> > 
> > Consider 100 APSs inactive, zero (0) ASPs active.
> > 
> > The SG cannot be required to send DUNA to inactive ASPs.  
> When the SG 
> > NIF has no user to send MTP-PAUSE to, it is not possible 
> for the SG to 
> > send DUNA.
> > 
> > The SG cannot be required to send SSNM to inactive ASPs.
> > 
> > Besides, it is unnecessary.
> > 
> > --brian
> > 
> > 
> > > 
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: Thursday, February 06, 2003 9:15 AM
> > > To: Zannin, Francois
> > > Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> > > Subject: Re: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > Zannin,,
> > > 
> > > That is contrary to the conensus of the RFC.
> > > 
> > > Consider 1 ASP active, 200,000 ASPs inactive.
> > > 
> > > The RFC cannot require an SG to send SSNM to an inactive ASP.
> > > 
> > > --brian
> > > 
> > > Zannin, Francois wrote:
> > > (Thu, 06 Feb 2003 09:02:41)
> > > > Samuel,
> > > > 
> > > > IMO, it's a good idea to inform inactive (BUT UP) ASP 
> of the state
> > > > of
> > > > the ss7 network especially in the override mode where 
> an ASP can be 
> > > > ready to recover an AS as soon as it receive an AS-Pending 
> > > > notification from the SGp. In this condition, the ASP 
> is ready to 
> > > > operate quickly and minimize the risk of buffered message lost
> > during
> > > > the pending timer.
> > > > 
> > > > If someone does not want to receive SSNM because it is 
> not able to 
> > > > recover an AS, it just stays in an ASP-Down state.
> > > > 
> > > > Because SSNM reflects ss7 Point Code state that is not "SI"
> > > > dependant,
> > > 
> > > > it affects all AS for a network and then, unless SG that plays 
> > > > multiple networks, the ASP UP is sufficient to filter the 
> > > > undesired SSNM.
> > > > 
> > > > Francois.
> > > > 
> > > > -----Original Message-----
> > > > From: Samuel Dur D. Jeyaseelan 
> > > > [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > Sent: Thursday, February 06, 2003 8:35 AM
> > > > To: Nagar, Ashish
> > > > Cc: sigtran@ietf.org
> > > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > > 
> > > > 
> > > > 
> > > > Ashish,
> > > > 
> > > > But when inactive ASP changes state to active, first of all it 
> > > > does not know whether to send DAUD or not.
> > > > 
> > > > It will try to send traffic - but actually destination is
> > > > unavailable.
> > > > 
> > > > The worst case is when SG triggers restart procedures this case 
> > > > all the SS7 destinations are unavailable via 
> restarted-SG.But this 
> > > > poor ASP (newly come-up ASP) keeps sending traffic via 
> > > > restarted-SG to reach the SS7 destination and for every message 
> > > > ASP receives DUNA.
> > > > 
> > > > Are we not doing double-work here?
> > > > 
> > > > --Samuel
> > > > 
> > > > 
> > > >   Alcatel USA, Inc.                  Internet:
> > > > <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, 
> > > > Inc.
> > > > *******
> > > > 
> > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > 
> > > > > Hi Samuel
> > > > > 
> > > > > My point is why to do any processing if ASP is in Inactive 
> > > > > state.
> > > > > Think about ASP is in Inactive state for longer period and 
> > > > > receiving
> > > 
> > > > > lots of DUNA and DAVA, while what matter is the present only 
> > > > > when
> > > ASP
> > > > > state changes to ASP-ACTIVE.
> > > > > 
> > > > > Thanks & Regards
> > > > > Ashish Nagar
> > > > > 
> > > > > 
> > > > > -----Original Message-----
> > > > > From: Samuel Dur D. Jeyaseelan
> > > > > [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > > Sent: Thursday, February 06, 2003 12:30 PM
> > > > > To: Nagar, Ashish
> > > > > Cc: sigtran@ietf.org
> > > > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > > > 
> > > > > 
> > > > > Ashish,
> > > > > 
> > > > > My point is if SG is able to send DUNA to all inact and active
> > > > > ASPs, we should not restrict it.
> > > > > 
> > > > > Sending DAUD is an alternative way ,If SG has direct way of
> > > > > sending DUNA then Why do you need alternative way?
> > > > > 
> > > > > --Samuel
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet:
> > > > <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA,
> > > > > Inc.
> > > > > *******
> > > > > 
> > > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > > 
> > > > > > You can generate DAUD procedure at any time, so 
> whenever ASP 
> > > > > > changes
> > > > 
> > > > > > state from Inactive to Active Just generate DAUD.
> > > > > > 
> > > > > > Thanks & Regards
> > > > > > Ashish Nagar
> > > > > > 
> > > > > > 
> > > > > > -----Original Message-----
> > > > > > From: Samuel Dur D. Jeyaseelan 
> > > > > > [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > > > To: sigtran@ietf.org
> > > > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > > > 
> > > > > > 
> > > > > > Hi all,
> > > > > > 
> > > > > > Currently RFC 3332 says that SG sends SSNM messages to all
> > > > > > 'concerned ASPs'(means only ACTIVE ASPs).
> > > > > > 
> > > > > > +----+           +---+           +-----+
> > > > > > |SEP | ----------|SG |-----------| ASP1|
> > > > > > +----+           +---+           +-----+
> > > > > >                    |             
> > > > > >                    |             +-----+
> > > > > >                    +-------------| ASP2|
> > > > > >                                  +-----+
> > > > > > 
> > > > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > > > 
> > > > > > When SEP is getting restarted ,SG sends DUNA to 
> ASP1. But SEP
> > > > > > status is not changed in ASP2.There are two ways to 
> solve this 
> > > > > > problem
> > > > > > 
> > > > > > 1. ASP1 can equalize its database with ASP2
> > > > > > or
> > > > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is
> > > > > > inactive)
> > > > > > 
> > > > > > IMO, SG can send SSNM messages to ASP (either ASP 
> is in active
> > > > > > or inact
> > > > > > state)
> > > > > > 
> > > > > > Any comments?
> > > > > > 
> > > > > > -----
> > > > > > 
> > > > > > section 5.4.2.1
> > > > > > 
> > > > > > ...
> > > > > >       - Pass the MTP-TRANSFER request primitive to 
> the MTP3 of
> > the
> > > > > >         concerned Network Appearance.
> > > > > > 
> > > > > > It conveys the meaning that M3UA layer (at SGP) 
> MTP-TRANSFER 
> > > > > > request
> > > > 
> > > > > > is sent to MTP3.
> > > > > > 
> > > > > > But there is no primitive availabe between M3UA and MTP3.
> > > > > > 
> > > > > > I think it should be corrected.
> > > > > > 
> > > > > > ----
> > > > > > Presently ASP is unaware of SG's MTP3 restart,
> > > > > > ASP receives only DUNA for all SS7 point codes which can be 
> > > > > > reachable via that SG.But in case of peer to peer 
> protocol SG
> > and
> > > > > > AS point codes are different, So why dont we send DUNA with 
> > > > > > SG's
> > 
> > > > > > point code to ASP? So that just by getting single DUNA(with 
> > > > > > SG's
> > 
> > > > > > PC) ASP will not route any message via that
> > > > SG.
> > > > > > This can avoid sending multiple DUNAs to ASP.
> > > > > > 
> > > > > > 
> > > > > > Any comments?
> > > > > > 
> > > > > > -----
> > > > > > section 1.4.1
> > > > > > 
> > > > > > ...
> > > > > >    Where a particular AS can be reached via more 
> than one SGP,
> > the
> > > > > >    corresponding Routing Keys in the SGPs should be 
> identical.
> > > > (Note:
> > > > > >    It is possible for the SGP Routing Key 
> configuration data 
> > > > > > to
> > be
> > > > > >    temporarily out-of-sync during configuration updates).
> > > > > >  
> > > > > >                               +--------+
> > > > > >                               |        |
> > > > > >                  +------------+  SG 1  +--------------+
> > > > > >      +-------+   |  SS7 links | "STP"  |  IP 
> network  |     ----
> > > > > >      |  SEP  +---+            +--------+              +---/
> > \
> > > > > >      |   or  |                    |*                
>       | ASPs
> > > |
> > > > > >      |  STP  +---+            +--------+              +---\
> > /
> > > > > >      +-------+   |            |        |            
>   |     ----
> > > > > >                  +------------+  SG 2  +--------------+
> > > > > >                               | "STP"  |
> > > > > >                               +--------+
> > > > > >  
> > > > > >                     Figure 1  Example with mated SGs
> > > > > > 
> > > > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs 
> > > > > > are
> > > > > > located geographically in different places.The 
> ideas is when SG1
> > 
> > > > > > goes down in CITY1(due to earthquake...) , all the traffice
> > > > > > still can be handled by SG2 which is in CITY2.
> > > > > > 
> > > > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages 
> > > > > > will be routed from SGP1 to SGP2.
> > > > > > 
> > > > > > Does all these things mean to say that SGP1's mate 
> 'should be'
> > > > > > present in SG2 and not in SG1 itself ?
> > > > > >  
> > > > > > 
> > > > > > Comments are welcome.
> > > > > > 
> > > > > > --Samuel
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > >   Alcatel USA, Inc.                  Internet:
> > > > > <userid>@ssd.usa.alcatel.com
> > > > > >   1000 Coit Road, Plano, Texas 75075
> > > > > >   ******* The opinions expressed are not those of 
> Alcatel USA, 
> > > > > > Inc.
> > > > > *******
> > > > > > 
> > > > > > _______________________________________________
> > > > > > Sigtran mailing list
> > > > > > Sigtran@ietf.org 
> > > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > > _______________________________________________
> > > > > > Sigtran mailing list
> > > > > > Sigtran@ietf.org 
> https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > > 
> > > 
> > > 
> > > > 
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> > > --
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > 
> > --
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 04:47:58 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA10213
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 04:47:58 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h169sL217097
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 04:54:21 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169s7p17087;
	Thu, 6 Feb 2003 04:54:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169rJp17042
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 04:53:19 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA10159
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 04:46:25 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h169o2qe018699;
	Thu, 6 Feb 2003 03:50:02 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h169nfM25914;
	Thu, 6 Feb 2003 03:49:41 -0600 (CST)
Date: Thu, 6 Feb 2003 03:49:41 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206023848.D18441@openss7.org>
Message-ID: <Pine.SOL.4.10.10302060343370.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

Your example is correct only theoritically but logically it sounds
NONE.

 ASP1 is active and ASP2 is inactive -overrid mode (via SG1)
 . .
 ASP200,000 is active and ASP200,000 is inactive -overrid mode(via SG1)
  
 Now we could see that 200,000 ASPs are active and 200,000 ASPs are
 inactive via RE-STARTED SG1.But all active 200,000 ASPs database
 is not same as inactive 200,000 ASPs.  

 According to you , SG1 will receive 200,000 DAUD msgs when inact ASPs
 come up and again 200,000 DUNA are being sent from SG1.

 According to me only 200,000 DUNAs are sent from SG1 if we allow SSNM
 messages to be processed in  inact state.So that we can avoid sending
200,000  DAUD messages.

--Samuel

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> I was my example.
> 
> 1 ASP is active for AS1, 200,000 ASPs are inactive for AS1.
> 
> --brian
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 03:29:48)
> > 
> > Brian,
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > It will decrease M3UA efficiency.
> > > 
> > > Consider 1 active ASP and 200,000 inactive ASPs.
> > > 
> > You are deviating from original thread.
> > 
> > ASP1 is active and ASP2 is inactive -overrid mode (via SG1)
> > .
> > .
> > .
> > ASP200,000 is active and ASP200,000 is inactive -overrid mode(via SG1)
> >  
> > Now we could see that 200,000 ASPs are active and 200,000 ASPs are
> > inactive via RE-STARTED SG1.But all active 200,000 ASPs database
> > is not same as inactive 200,000 ASPs.
> >  
> > According to you , SG1 will receive 200,000 DAUD msgs when inact ASPs
> > come up and again 200,000 DUNA are being sent from SG1.
> > 
> > According to me only 200,000 DUNAs are sent from SG1 if we allow SSNM
> > messages to be processing inact state.So that we can avoid sending 200,000
> > DAUD messages.
> > 
> > 
> > 
> > 
> > > For each SSNM the SG must send 200,001 DUNA messages.
> > > 
> > > With the existing concensus DAUD procedure, the SG only
> > > sends 1 DUNA message.
> > > 
> > > When an ASP activates for an AS it sends one DAUD message
> > > to avoid the SG sending many useless SSNM messages.
> > > 
> > > The activating ASP need not even send DAUD.  It can just
> > > start sending messages and use responsive DUNAs.
> > > 
> > > This is most efficient because SS7 destinations are almost
> > > never unavailable.
> > > 
> > > Sending SSNM to inactive ASPs would be a BAD thing.  It is
> > > good that the RFC does not permit this.
> > No comment
> > 
> > --Samuel
> > > 
> > > --brian
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 02:18:28)
> > > > 
> > > > Brian,
> > > > 
> > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > 
> > > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > > 
> > > > > Samuel,
> > > > > 
> > > > > IMO sending SSNM to inactive ASPs is a BAD idea.
> > > > > 
> > > > Which is bad idea sending DAUD and get the DUNA from SG
> > > > or SG itself can send DUNA to inact ASP?
> > > > 
> > > > IMO my idea is not BAD.It will increase the M3UA stack
> > > > efficency.
> > > > 
> > > > > --brian
> > > > > 
> > > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 00:30:30)
> > > > > > 
> > > > > > Brian,
> > > > > > 
> > > > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > > >   1000 Coit Road, Plano, Texas 75075
> > > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > > > 
> > > > > > On Wed, 5 Feb 2003, Brian F. G. Bidulock wrote:
> > > > > > 
> > > > > > > Samuel,
> > > > > > > 
> > > > > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Wed, 05 Feb 2003 23:57:10)
> > > > > > > > 
> > > > > > > > Hi all,
> > > > > > > > 
> > > > > > > > Currently RFC 3332 says that SG sends SSNM messages to all 'concerned
> > > > > > > > ASPs'(means only ACTIVE ASPs).
> > > > > > > > 
> > > > > > > > +----+           +---+           +-----+
> > > > > > > > |SEP | ----------|SG |-----------| ASP1|
> > > > > > > > +----+           +---+           +-----+
> > > > > > > >                    |             
> > > > > > > >                    |             +-----+
> > > > > > > >                    +-------------| ASP2|
> > > > > > > >                                  +-----+
> > > > > > > > 
> > > > > > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > > > > > 
> > > > > > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > > > > > But SEP status is not changed in ASP2.There are two ways to solve this
> > > > > > > > problem
> > > > > > > > 
> > > > > > > > 1. ASP1 can equalize its database with ASP2
> > > > > > > > or
> > > > > > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > > > > > 
> > > > > > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or inact
> > > > > > > > state)
> > > > > > > > 
> > > > > > > > Any comments?
> > > > > > > 
> > > > > > > The audit procedure is provided to allow any ASP to determine the
> > > > > > > current accessibility of destinations from the SG.
> > > > > > > 
> > > > > > > The prosrcibed procedure in RFC 3332 is as follows:
> > > > > > > 
> > > > > > >   If ASP2 can know from ASP1 that the destination is inaccessible
> > > > > > >   then nothing is required.
> > > > > > This is out of the scope of RFC.
> > > > > > 
> > > > > > > 
> > > > > > >   If ASP2 does not know from ASP1 that destinations are inaccessible,
> > > > > > >   it issues DAUD when it goes active and the SG tells it which
> > > > > > >   destinations are accessible.
> > > > > > I think getting DUNA from SG is better than sending DAUD and receiving
> > > > > > response.If SG is able to do it ,let SG send DUNA to inactive ASPs.
> > > > > > IMO there is no harm.
> > > > > > > 
> > > > > > > SSNM are NOT sent to inactive ASPs.
> > > > > > > 
> > > > > > > The reason for this is as follows:
> > > > > > > 
> > > > > > > Consider that ASP1 is active and ASP2 is down.  SG sends DUNA
> > > > > > > to ASP1.  APS2 does not receive this because APS2 is not even
> > > > > > > running.
> > > > > > > 
> > > > > > > ASP2 goes ASPUP and moves to the inactive state for the AS for
> > > > > > > which the DUNA was sent.
> > > > > > > 
> > > > > > > Can the SG be expected to then send this backlog of information?
> > > > > > > 
> > > > > > 
> > > > > > I mean to say that when SG receives TFP from SEP and if the state
> > > > > > of ASP is ACTIVE or INACT then sending DUNA is good idea.
> > > > > > 
> > > > > > If ASP is in DOWN state then SG will not send DUNA.
> > > > > > IMO, practically speaking backup ASP will not be in DOWN state,
> > > > > > It should be in ASP INACT state.
> > > > > > 
> > > > > > My point is if SG is able to send DUNA , we should not put any
> > > > > > restriction.
> > > > > > 
> > > > > > --Samuel
> > > > > > 
> > > > > > > No, it is the repsonsibility of ASP2 to send DAUD to discover
> > > > > > > the information.
> > > > > > > 
> > > > > > > This is also the case for the inactive state.
> > > > > > > 
> > > > > > > --brian
> > > > > > > 
> > > > > > > -- 
> > > > > > > Brian F. G. Bidulock
> > > > > > > bidulock@openss7.org
> > > > > > > http://www.openss7.org/
> > > > > > > _______________________________________________
> > > > > > > Sigtran mailing list
> > > > > > > Sigtran@ietf.org
> > > > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > > > 
> > > > > 
> > > > > -- 
> > > > > Brian F. G. Bidulock
> > > > > bidulock@openss7.org
> > > > > http://www.openss7.org/
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > 
> > > 
> > > -- 
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > 
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 04:51:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA10329
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 04:51:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h169wKP17233
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 04:58:20 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169w3p17218;
	Thu, 6 Feb 2003 04:58:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169v3p17193
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 04:57:03 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA10281
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 04:50:08 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:Z8buz7tNkyJ3+z+7LlqpQY3RgUB0LGvp@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h169rkD15085;
	Thu, 6 Feb 2003 02:53:46 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h169rkm25420;
	Thu, 6 Feb 2003 02:53:46 -0700
Date: Thu, 6 Feb 2003 02:53:46 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Zannin, Francois" <Francois.Zannin@hp.com>
Cc: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206025346.E18441@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Zannin, Francois" <Francois.Zannin@hp.com>,
	"Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <06CAA4D3753C53408D2CE6340B68444214C40E@vbeexc02.emea.cpqcorp.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <06CAA4D3753C53408D2CE6340B68444214C40E@vbeexc02.emea.cpqcorp.net>; from Francois.Zannin@hp.com on Thu, Feb 06, 2003 at 10:44:08AM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Zannin,,

Zannin, Francois wrote:                                                            (Thu, 06 Feb 2003 10:44:08)
> Bidulock,,
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> > Sent: Thursday, February 06, 2003 10:22 AM
> > To: Zannin, Francois
> > Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> > Subject: Re: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > Zannin,,
> > 
> > Zannin, Francois wrote:                                       
> >                      (Thu, 06 Feb 2003 09:58:46)
> > > Bidulock,,
> > > 
> > > Does it make sense to have "a lot" of ASP UP connected to a SGP and 
> > > not ready to do anything?
> > 
> > They could be active for different AS.
> 
> In that case, because SSNM are for ss7 SP and concern all SI (AS), it is
> not useless to receive (and you will receive) these SSNM. Now, the
> question is about the RC in these SSNM.

AS is not synonymous with SI.  AS can be for different point codes.

> 
> > 
> > > If an ASP is UP for a SGP, it certainly must be ready to 
> > activate an 
> > > AS and then concerned by the SSNM, and again, for an emergency AS 
> > > recovery (AS-Pending notification) the ASP must be ready to operate 
> > > after the Active-ACK.
> > > 
> > > Only the "ready to operate for an AS" ASP have a reason to 
> > be "UP" on 
> > > a SG.
> > 
> > Wrong.  Whenever an ASP goes ASP UP it becomes ASP-INACTIVE 
> > for all AS for which it is configured to operate.  Depending 
> > on SG configuration, this could mean that the ASP is 
> > ASP-INACTIVE for all AS served by the SG, regarless of 
> > whether the ASP will ever go ASP-ACTIVE for all of the AS.  
> > There is no way for an ASP to tell the SG that it is ASP-DOWN 
> > for some AS and ASP-INACTIVE for others.
> > 
> > Also, the purpose of the ASP-INACTIVE state is to receive 
> > NTFY messages for the AS.  In override mode, one ASP is 
> > active for the AS and the others are inactive.  The inactive 
> > ASPs have no intention of going active unless the active ASP 
> > fails, or unless manual management action is taken.
> > 
> 
> A ss7 state concern all SI (AS) for a NA, the ASP receive the SSNM for
> the ACTIVE (or activable) AS.
> If no AS is "activable" no reason to send an ASP-UP...

Yes, the reason is to act as standby for one active AS.

> 
> This can be less efficent for SG that plays multiple network!

Actually it is very efficient for an SG to consider all ASPs
as ASP-INACTIVE for all AS that it serves: particularly if, as
in the RFC, it is not required to send SSNM to inactive ASPs.

> 
> > > I don't think that it is useful to connect "UP" a lot of 
> > ASP with no 
> > > reason!
> > 
> > See above.
> > 
> > > 
> > > You are right, DUNA is not the same thing as MTP-PAUSE, but 
> > the sense 
> > > is the same, a user (AS) cannot use a ROUTE (or SG) if the 
> > remote SP 
> > > is inaccessible because of TFP, TCP and then MTP-PAUSE which is the 
> > > consolidation of all the routes for a destination.
> > > 
> > > The reason to wait for a DAUD for the ss7 state on an ASP 
> > is to have a 
> > > selective use of the ss7 network SP by an AS (then ASP). In 
> > that case, 
> > > the number of DAUD is limited to the minimum after the ACTIVE 
> > > procedure. But as soon as the ASP is active, there is no 
> > filter in the 
> > > SG for ASP concerned by the SSNM for specific destinations.
> > 
> > MTP procedures require MTP-PAUSE to be sent to MTP users that 
> > are sending to inaccessible destinations.  An MTP conforming 
> > SG will send DUNA in response to messages sent to an 
> > inacessible destination.
> > 
> 
> Do you mean that when a SG receive a TFP on the last route for a
> destination, it does not send the DUNA to all the ASP???
> If yes, that a NEW behavior for a SG?????

I did not say that.  I said: "An MTP conforming SG will send DUNA
in response to messages sent to an inacessible destination."

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 04:52:56 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA10382
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 04:52:56 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h169xJN17271
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 04:59:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169x5p17262;
	Thu, 6 Feb 2003 04:59:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h169wSp17246
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 04:58:28 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA10321
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 04:51:34 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:w1bQGY/obYs4LzWp3VG3xcFlTofaWOoY@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h169tCD15115;
	Thu, 6 Feb 2003 02:55:12 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h169tCr25605;
	Thu, 6 Feb 2003 02:55:12 -0700
Date: Thu, 6 Feb 2003 02:55:12 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206025512.F18441@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	sigtran@ietf.org
References: <20030206023848.D18441@openss7.org> <Pine.SOL.4.10.10302060343370.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302060343370.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 03:49:41AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

ASP1 is active for AS1.

ASP2 is inactive for AS1.
ASP3 is inactive for AS1.
.
.
.
ASP200,001 is inactive for AS1.

Understand?

--brian.


Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 03:49:41)
> 
> Brian,
> 
> Your example is correct only theoritically but logically it sounds
> NONE.
> 
>  ASP1 is active and ASP2 is inactive -overrid mode (via SG1)
>  . .
>  ASP200,000 is active and ASP200,000 is inactive -overrid mode(via SG1)
>   
>  Now we could see that 200,000 ASPs are active and 200,000 ASPs are
>  inactive via RE-STARTED SG1.But all active 200,000 ASPs database
>  is not same as inactive 200,000 ASPs.  
> 
>  According to you , SG1 will receive 200,000 DAUD msgs when inact ASPs
>  come up and again 200,000 DUNA are being sent from SG1.
> 
>  According to me only 200,000 DUNAs are sent from SG1 if we allow SSNM
>  messages to be processed in  inact state.So that we can avoid sending
> 200,000  DAUD messages.
> 
> --Samuel
> 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 04:59:01 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA10637
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 04:59:01 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16A5Ot17584
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 05:05:24 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16A58p17563;
	Thu, 6 Feb 2003 05:05:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16A4ap17523
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 05:04:36 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA10579
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 04:57:41 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h16A12JX015656;
	Thu, 6 Feb 2003 04:01:04 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h16A0kM29762;
	Thu, 6 Feb 2003 04:00:46 -0600 (CST)
Date: Thu, 6 Feb 2003 04:00:46 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: Richard Adams <Richard.Adams@etsi.org>,
        "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206022344.B18441@openss7.org>
Message-ID: <Pine.SOL.4.10.10302060352390.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 03:01:39)
> > 
> > Richard,
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Thu, 6 Feb 2003, Richard Adams wrote:
> > 
> > > Brian,
> > > I agree with you. I think when the ASP becomes active, it should do an
> > > audit to update its knowledge of the SS7 network. The ASP needs to take
> > > the initiative to find out about the SS7 nodes etc. in which it is
> > > interested. Any other method of working would lead to impossible loading
> > > on the SG.
> > 
> > When SG itself is re-starting , it means no ASP can reach any SS7
> > destinations via re-started SG.Always it is good to tell my ASP
> > who is in inactive state about SS7 destinations unavailability.
> 
> It is unneccessary.  Iff the ASP goes active, the SG can send DUNA(*)
> to it in one message to indicate that it is restarting.  The SG isn't
> required to wait for DAUD to do this, it could send DUNA(*) just because
> the ASP hase become active, or it could send DUNA(*) in response to the
> first DATA message from the ASP.
> 

Alternative ASP has to find out destination status by sending DAUD or
DATA.This is unnecessary traffic if the ASP had already received DUNA(*).

If there are 100 ASPs coming up ,  100 DAUD or DATA should be sent from
ASP, obviously here the problem with respect to number of un-necessary
messages - POTENTIAL.

--Samuel  

> There is no need to send SSNM to inactive ASPs.
> 
> --brian
> 
> 
> > 
> > --Samuel
> > 
> > > Of course, I hope 199,999 of your inactive ASPs do not decide to become
> > > active at the same time.....! 
> > > Kind regards,
> > > Dick Adams,
> > > richard.adams@etsi.org
> > > 
> > >  
> > > 
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: 06 February 2003 09:15
> > > To: Zannin, Francois
> > > Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> > > Subject: Re: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > Zannin,,
> > > 
> > > That is contrary to the conensus of the RFC.
> > > 
> > > Consider 1 ASP active, 200,000 ASPs inactive.
> > > 
> > > The RFC cannot require an SG to send SSNM to an inactive ASP.
> > > 
> > > --brian
> > > 
> > > Zannin, Francois wrote:
> > > (Thu, 06 Feb 2003 09:02:41)
> > > > Samuel,
> > > > 
> > > > IMO, it's a good idea to inform inactive (BUT UP) ASP of the state of
> > > > the ss7 network especially in the override mode where an ASP can be
> > > > ready to recover an AS as soon as it receive an AS-Pending
> > > notification
> > > > from the SGp. In this condition, the ASP is ready to operate quickly
> > > and
> > > > minimize the risk of buffered message lost during the pending timer.
> > > > 
> > > > If someone does not want to receive SSNM because it is not able to
> > > > recover an AS, it just stays in an ASP-Down state.
> > > > 
> > > > Because SSNM reflects ss7 Point Code state that is not "SI" dependant,
> > > > it affects all AS for a network and then, unless SG that plays
> > > multiple
> > > > networks, the ASP UP is sufficient to filter the undesired SSNM.
> > > > 
> > > > Francois.
> > > > 
> > > > -----Original Message-----
> > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> > > > Sent: Thursday, February 06, 2003 8:35 AM
> > > > To: Nagar, Ashish
> > > > Cc: sigtran@ietf.org
> > > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > > 
> > > > 
> > > > 
> > > > Ashish,
> > > > 
> > > > But when inactive ASP changes state to active, first of all
> > > > it does not know whether to send DAUD or not.
> > > > 
> > > > It will try to send traffic - but actually destination is unavailable.
> > > > 
> > > > The worst case is when SG triggers restart procedures this case all
> > > the
> > > > SS7 destinations are unavailable via restarted-SG.But this poor ASP
> > > > (newly come-up ASP) keeps sending traffic via restarted-SG to reach
> > > the
> > > > SS7 destination and for every message ASP receives DUNA.
> > > > 
> > > > Are we not doing double-work here?
> > > > 
> > > > --Samuel
> > > > 
> > > > 
> > > >   Alcatel USA, Inc.                  Internet:
> > > > <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > > *******
> > > > 
> > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > 
> > > > > Hi Samuel
> > > > > 
> > > > > My point is why to do any processing if ASP is in Inactive state. 
> > > > > Think about ASP is in Inactive state for longer period and receiving
> > > 
> > > > > lots of DUNA and DAVA, while what matter is the present only when
> > > ASP 
> > > > > state changes to ASP-ACTIVE.
> > > > > 
> > > > > Thanks & Regards
> > > > > Ashish Nagar
> > > > > 
> > > > > 
> > > > > -----Original Message-----
> > > > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > > Sent: Thursday, February 06, 2003 12:30 PM
> > > > > To: Nagar, Ashish
> > > > > Cc: sigtran@ietf.org
> > > > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > > > 
> > > > > 
> > > > > Ashish,
> > > > > 
> > > > > My point is if SG is able to send DUNA to all inact and active ASPs,
> > > 
> > > > > we should not restrict it.
> > > > > 
> > > > > Sending DAUD is an alternative way ,If SG has direct way of sending 
> > > > > DUNA then Why do you need alternative way?
> > > > > 
> > > > > --Samuel
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet:
> > > > <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. 
> > > > > *******
> > > > > 
> > > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > > 
> > > > > > You can generate DAUD procedure at any time, so whenever ASP
> > > changes
> > > > 
> > > > > > state from Inactive to Active Just generate DAUD.
> > > > > > 
> > > > > > Thanks & Regards
> > > > > > Ashish Nagar
> > > > > > 
> > > > > > 
> > > > > > -----Original Message-----
> > > > > > From: Samuel Dur D. Jeyaseelan
> > > [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > > > To: sigtran@ietf.org
> > > > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > > > 
> > > > > > 
> > > > > > Hi all,
> > > > > > 
> > > > > > Currently RFC 3332 says that SG sends SSNM messages to all 
> > > > > > 'concerned ASPs'(means only ACTIVE ASPs).
> > > > > > 
> > > > > > +----+           +---+           +-----+
> > > > > > |SEP | ----------|SG |-----------| ASP1|
> > > > > > +----+           +---+           +-----+
> > > > > >                    |             
> > > > > >                    |             +-----+
> > > > > >                    +-------------| ASP2|
> > > > > >                                  +-----+
> > > > > > 
> > > > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > > > 
> > > > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > > > But SEP status is not changed in ASP2.There are two ways to solve 
> > > > > > this problem
> > > > > > 
> > > > > > 1. ASP1 can equalize its database with ASP2
> > > > > > or
> > > > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > > > 
> > > > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or 
> > > > > > inact
> > > > > > state)
> > > > > > 
> > > > > > Any comments?
> > > > > > 
> > > > > > -----
> > > > > > 
> > > > > > section 5.4.2.1
> > > > > > 
> > > > > > ...
> > > > > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> > > > > >         concerned Network Appearance.
> > > > > > 
> > > > > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER
> > > request
> > > > 
> > > > > > is sent to MTP3.
> > > > > > 
> > > > > > But there is no primitive availabe between M3UA and MTP3.
> > > > > > 
> > > > > > I think it should be corrected.
> > > > > > 
> > > > > > ----
> > > > > > Presently ASP is unaware of SG's MTP3 restart,
> > > > > > ASP receives only DUNA for all SS7 point codes which can be
> > > > > > reachable via that SG.But in case of peer to peer protocol
> > > > > > SG and AS point codes are different, So why dont we send
> > > > > > DUNA with SG's point code to ASP? So that just by getting
> > > > > > single DUNA(with SG's PC) ASP will not route any message via that
> > > > SG.
> > > > > > This can avoid sending multiple DUNAs to ASP.
> > > > > > 
> > > > > > 
> > > > > > Any comments?
> > > > > > 
> > > > > > -----
> > > > > > section 1.4.1
> > > > > > 
> > > > > > ...
> > > > > >    Where a particular AS can be reached via more than one SGP, the
> > > > > >    corresponding Routing Keys in the SGPs should be identical.
> > > > (Note:
> > > > > >    It is possible for the SGP Routing Key configuration data to be
> > > > > >    temporarily out-of-sync during configuration updates).
> > > > > >  
> > > > > >                               +--------+
> > > > > >                               |        |
> > > > > >                  +------------+  SG 1  +--------------+
> > > > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > > > >      |  SEP  +---+            +--------+              +---/      \
> > > > > >      |   or  |                    |*                      | ASPs
> > > |
> > > > > >      |  STP  +---+            +--------+              +---\      /
> > > > > >      +-------+   |            |        |              |     ----
> > > > > >                  +------------+  SG 2  +--------------+
> > > > > >                               | "STP"  |
> > > > > >                               +--------+
> > > > > >  
> > > > > >                     Figure 1  Example with mated SGs
> > > > > > 
> > > > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are 
> > > > > > located geographically in different places.The ideas is when SG1 
> > > > > > goes down in CITY1(due to earthquake...) , all the traffice still 
> > > > > > can be handled by SG2 which is in CITY2.
> > > > > > 
> > > > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will
> > > be 
> > > > > > routed from SGP1 to SGP2.
> > > > > > 
> > > > > > Does all these things mean to say that SGP1's mate 'should be' 
> > > > > > present in SG2 and not in SG1 itself ?
> > > > > >  
> > > > > > 
> > > > > > Comments are welcome.
> > > > > > 
> > > > > > --Samuel
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > >   Alcatel USA, Inc.                  Internet:
> > > > > <userid>@ssd.usa.alcatel.com
> > > > > >   1000 Coit Road, Plano, Texas 75075
> > > > > >   ******* The opinions expressed are not those of Alcatel USA,
> > > Inc.
> > > > > *******
> > > > > > 
> > > > > > _______________________________________________
> > > > > > Sigtran mailing list
> > > > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > > _______________________________________________
> > > > > > Sigtran mailing list
> > > > > > Sigtran@ietf.org
> > > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > > 
> > > > > 
> > > > 
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> > > -- 
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 05:00:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA10729
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 05:00:56 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16A7JE18249
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 05:07:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16A79p17994;
	Thu, 6 Feb 2003 05:07:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16A6Tp17653
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 05:06:29 -0500
Received: from zdemail03.zdem.compaq.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA10661
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 04:59:34 -0500 (EST)
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail03.zdem.compaq.com (Postfix) with ESMTP
	id 356516C6; Thu,  6 Feb 2003 11:03:12 +0100 (CET)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.2966);
	 Thu, 6 Feb 2003 11:03:02 +0100
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Subject: RE: [Sigtran] M3UA - ASP database equalization
Date: Thu, 6 Feb 2003 11:02:54 +0100
Message-ID: <06CAA4D3753C53408D2CE6340B684442014888D0@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA - ASP database equalization
Thread-Index: AcLNxbapczXOg25NTy+siaq2KqBnLQAALUCA
From: "Zannin, Francois" <Francois.Zannin@hp.com>
To: <bidulock@openss7.org>
Cc: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 06 Feb 2003 10:03:02.0170 (UTC) FILETIME=[EF4B3FA0:01C2CDC6]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h16A6Tp17654
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Bidulock,,

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> Sent: Thursday, February 06, 2003 10:54 AM
> To: Zannin, Francois
> Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> Subject: Re: [Sigtran] M3UA - ASP database equalization
> 
> 
> Zannin,,
> 
> Zannin, Francois wrote:                                       
>                      (Thu, 06 Feb 2003 10:44:08)
> > Bidulock,,
> > 
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: Thursday, February 06, 2003 10:22 AM
> > > To: Zannin, Francois
> > > Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> > > Subject: Re: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > Zannin,,
> > > 
> > > Zannin, Francois wrote:                                       
> > >                      (Thu, 06 Feb 2003 09:58:46)
> > > > Bidulock,,
> > > > 
> > > > Does it make sense to have "a lot" of ASP UP connected to a SGP 
> > > > and
> > > > not ready to do anything?
> > > 
> > > They could be active for different AS.
> > 
> > In that case, because SSNM are for ss7 SP and concern all 
> SI (AS), it 
> > is not useless to receive (and you will receive) these 
> SSNM. Now, the 
> > question is about the RC in these SSNM.
> 
> AS is not synonymous with SI.  AS can be for different point codes.

You are right, but it doesn't change anything...

> 
> > 
> > > 
> > > > If an ASP is UP for a SGP, it certainly must be ready to
> > > activate an
> > > > AS and then concerned by the SSNM, and again, for an 
> emergency AS
> > > > recovery (AS-Pending notification) the ASP must be 
> ready to operate 
> > > > after the Active-ACK.
> > > > 
> > > > Only the "ready to operate for an AS" ASP have a reason to
> > > be "UP" on
> > > > a SG.
> > > 
> > > Wrong.  Whenever an ASP goes ASP UP it becomes ASP-INACTIVE
> > > for all AS for which it is configured to operate.  Depending 
> > > on SG configuration, this could mean that the ASP is 
> > > ASP-INACTIVE for all AS served by the SG, regarless of 
> > > whether the ASP will ever go ASP-ACTIVE for all of the AS.  
> > > There is no way for an ASP to tell the SG that it is ASP-DOWN 
> > > for some AS and ASP-INACTIVE for others.
> > > 
> > > Also, the purpose of the ASP-INACTIVE state is to receive
> > > NTFY messages for the AS.  In override mode, one ASP is 
> > > active for the AS and the others are inactive.  The inactive 
> > > ASPs have no intention of going active unless the active ASP 
> > > fails, or unless manual management action is taken.
> > > 
> > 
> > A ss7 state concern all SI (AS) for a NA, the ASP receive 
> the SSNM for 
> > the ACTIVE (or activable) AS. If no AS is "activable" no reason to 
> > send an ASP-UP...
> 
> Yes, the reason is to act as standby for one active AS.

So, on AS IS "activable"...

> 
> > 
> > This can be less efficent for SG that plays multiple network!
> 
> Actually it is very efficient for an SG to consider all ASPs
> as ASP-INACTIVE for all AS that it serves: particularly if, 
> as in the RFC, it is not required to send SSNM to inactive ASPs.
> 
> > 
> > > > I don't think that it is useful to connect "UP" a lot of
> > > ASP with no
> > > > reason!
> > > 
> > > See above.
> > > 
> > > > 
> > > > You are right, DUNA is not the same thing as MTP-PAUSE, but
> > > the sense
> > > > is the same, a user (AS) cannot use a ROUTE (or SG) if the
> > > remote SP
> > > > is inaccessible because of TFP, TCP and then MTP-PAUSE which is 
> > > > the
> > > > consolidation of all the routes for a destination.
> > > > 
> > > > The reason to wait for a DAUD for the ss7 state on an ASP
> > > is to have a
> > > > selective use of the ss7 network SP by an AS (then ASP). In
> > > that case,
> > > > the number of DAUD is limited to the minimum after the ACTIVE
> > > > procedure. But as soon as the ASP is active, there is no 
> > > filter in the
> > > > SG for ASP concerned by the SSNM for specific destinations.
> > > 
> > > MTP procedures require MTP-PAUSE to be sent to MTP users that
> > > are sending to inaccessible destinations.  An MTP conforming 
> > > SG will send DUNA in response to messages sent to an 
> > > inacessible destination.
> > > 
> > 
> > Do you mean that when a SG receive a TFP on the last route for a 
> > destination, it does not send the DUNA to all the ASP??? If 
> yes, that 
> > a NEW behavior for a SG?????
> 
> I did not say that.  I said: "An MTP conforming SG will send 
> DUNA in response to messages sent to an inacessible destination."
> 
> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 05:03:54 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA10828
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 05:03:54 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16AAHh18461
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 05:10:17 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16AA4p18437;
	Thu, 6 Feb 2003 05:10:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16A9Lp18404
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 05:09:21 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA10793
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 05:02:27 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h16A64JV016206;
	Thu, 6 Feb 2003 04:06:05 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h16A5qM01638;
	Thu, 6 Feb 2003 04:05:52 -0600 (CST)
Date: Thu, 6 Feb 2003 04:05:52 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206025512.F18441@openss7.org>
Message-ID: <Pine.SOL.4.10.10302060403180.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

I said you are giving all theoritical examples which is IMPOSSIBLE
practically.

Nobody will configure 1 ASP as active (20,000-1) ASP as inactive.
Your example is total deviation from this thread.

--Samuel

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> ASP1 is active for AS1.
> 
> ASP2 is inactive for AS1.
> ASP3 is inactive for AS1.
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 05:15:15 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA11091
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 05:15:14 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16ALcn18984
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 05:21:38 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16ALJp18934;
	Thu, 6 Feb 2003 05:21:20 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16AK1p18867
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 05:20:01 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA11053
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 05:13:06 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:nbNo2wtP4gEMabN47tnxo/n2fwJC2S0w@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h16AGiD15456;
	Thu, 6 Feb 2003 03:16:44 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h16AGh828349;
	Thu, 6 Feb 2003 03:16:43 -0700
Date: Thu, 6 Feb 2003 03:16:43 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: Richard Adams <Richard.Adams@etsi.org>,
        "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206031643.G18441@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	Richard Adams <Richard.Adams@etsi.org>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206022344.B18441@openss7.org> <Pine.SOL.4.10.10302060352390.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302060352390.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 04:00:46AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 04:00:46)
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 03:01:39)
> > > 
> > > Richard,
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Thu, 6 Feb 2003, Richard Adams wrote:
> > > 
> > > > Brian,
> > > > I agree with you. I think when the ASP becomes active, it should do an
> > > > audit to update its knowledge of the SS7 network. The ASP needs to take
> > > > the initiative to find out about the SS7 nodes etc. in which it is
> > > > interested. Any other method of working would lead to impossible loading
> > > > on the SG.
> > > 
> > > When SG itself is re-starting , it means no ASP can reach any SS7
> > > destinations via re-started SG.Always it is good to tell my ASP
> > > who is in inactive state about SS7 destinations unavailability.
> > 
> > It is unneccessary.  Iff the ASP goes active, the SG can send DUNA(*)
> > to it in one message to indicate that it is restarting.  The SG isn't
> > required to wait for DAUD to do this, it could send DUNA(*) just because
> > the ASP hase become active, or it could send DUNA(*) in response to the
> > first DATA message from the ASP.
> > 
> 
> Alternative ASP has to find out destination status by sending DAUD or
> DATA.This is unnecessary traffic if the ASP had already received DUNA(*).

The alternate ASP does not have to send DAUD.

> 
> If there are 100 ASPs coming up ,  100 DAUD or DATA should be sent from
> ASP, obviously here the problem with respect to number of un-necessary
> messages - POTENTIAL.

100 ASPAC Req mesages and 100 ASPUP Ack messages have to be sent for 100 ASPs
to come up.  There is little difficulty bundling a 100 DAUDs on the 100
ASPACs.  There is no loss of efficiency.

If SSNM were to be sent to inactive ASPs, 100 messages would be sent for
every SSNM.  There is no ASPAC or ASPAC Ack to bundle with these and they
could be sent repeatedly to inactive ASPs before they ever activate.

So, for the current situation:

  SG sends 50 SSNM to the active ASP.
  100 ASPs go active sending 100 ASPAC and 100 DAUD and receiving
  100 ASPAC Ack and 100 SSNM (including all 50 indications).

  SG sends a total of 250 messages, bundled in 150 packets.
  Each ASP sends 2 messages bundled in 1 packet.
  Total 150 packets.

For your unnecessary situation:

  SG sends 50 SSNM to each of 101 ASPs (1 active and 100 inactive).
  100 ASPs go active sending 100 ASPAC and receiving 100 ASPAC Ack.

  SG sends a total of 5150 messages, in 5150 packets.
  Each ASP sends 1 message in 1 packet.
  Total 5250 packets.

You suggestion would have us send 5150 packets where 150 will do.

The SG cannot be required to send SSNM to inactive ASPs.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 05:15:11 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA11089
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 05:15:11 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16ALYN18967
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 05:21:34 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16ALNp18951;
	Thu, 6 Feb 2003 05:21:23 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16AKep18912
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 05:20:40 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA11064
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 05:13:46 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:X+l7ttw7y1vZ5nV8gsu97wKRRHNJ17SV@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h16AHOD15469;
	Thu, 6 Feb 2003 03:17:24 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h16AHOK28433;
	Thu, 6 Feb 2003 03:17:24 -0700
Date: Thu, 6 Feb 2003 03:17:23 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Zannin, Francois" <Francois.Zannin@hp.com>
Cc: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206031723.H18441@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Zannin, Francois" <Francois.Zannin@hp.com>,
	"Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <06CAA4D3753C53408D2CE6340B684442014888D0@vbeexc02.emea.cpqcorp.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <06CAA4D3753C53408D2CE6340B684442014888D0@vbeexc02.emea.cpqcorp.net>; from Francois.Zannin@hp.com on Thu, Feb 06, 2003 at 11:02:54AM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Zannin,,

Zannin, Francois wrote:                                                            (Thu, 06 Feb 2003 11:02:54)
> Bidulock,,
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> > Sent: Thursday, February 06, 2003 10:54 AM
> > To: Zannin, Francois
> > Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> > Subject: Re: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > Zannin,,
> > 
> > Zannin, Francois wrote:                                       
> >                      (Thu, 06 Feb 2003 10:44:08)
> > > Bidulock,,
> > > 
> > > > -----Original Message-----
> > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > Sent: Thursday, February 06, 2003 10:22 AM
> > > > To: Zannin, Francois
> > > > Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> > > > Subject: Re: [Sigtran] M3UA - ASP database equalization
> > > > 
> > > > 
> > > > Zannin,,
> > > > 
> > > > Zannin, Francois wrote:                                       
> > > >                      (Thu, 06 Feb 2003 09:58:46)
> > > > > Bidulock,,
> > > > > 
> > > > > Does it make sense to have "a lot" of ASP UP connected to a SGP 
> > > > > and
> > > > > not ready to do anything?
> > > > 
> > > > They could be active for different AS.
> > > 
> > > In that case, because SSNM are for ss7 SP and concern all 
> > SI (AS), it 
> > > is not useless to receive (and you will receive) these 
> > SSNM. Now, the 
> > > question is about the RC in these SSNM.
> > 
> > AS is not synonymous with SI.  AS can be for different point codes.
> 
> You are right, but it doesn't change anything...
> 
> > 
> > > 
> > > > 
> > > > > If an ASP is UP for a SGP, it certainly must be ready to
> > > > activate an
> > > > > AS and then concerned by the SSNM, and again, for an 
> > emergency AS
> > > > > recovery (AS-Pending notification) the ASP must be 
> > ready to operate 
> > > > > after the Active-ACK.
> > > > > 
> > > > > Only the "ready to operate for an AS" ASP have a reason to
> > > > be "UP" on
> > > > > a SG.
> > > > 
> > > > Wrong.  Whenever an ASP goes ASP UP it becomes ASP-INACTIVE
> > > > for all AS for which it is configured to operate.  Depending 
> > > > on SG configuration, this could mean that the ASP is 
> > > > ASP-INACTIVE for all AS served by the SG, regarless of 
> > > > whether the ASP will ever go ASP-ACTIVE for all of the AS.  
> > > > There is no way for an ASP to tell the SG that it is ASP-DOWN 
> > > > for some AS and ASP-INACTIVE for others.
> > > > 
> > > > Also, the purpose of the ASP-INACTIVE state is to receive
> > > > NTFY messages for the AS.  In override mode, one ASP is 
> > > > active for the AS and the others are inactive.  The inactive 
> > > > ASPs have no intention of going active unless the active ASP 
> > > > fails, or unless manual management action is taken.
> > > > 
> > > 
> > > A ss7 state concern all SI (AS) for a NA, the ASP receive 
> > the SSNM for 
> > > the ACTIVE (or activable) AS. If no AS is "activable" no reason to 
> > > send an ASP-UP...
> > 
> > Yes, the reason is to act as standby for one active AS.
> 
> So, on AS IS "activable"...

What's "activalble"?

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 05:15:53 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA11107
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 05:15:53 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16AMGY19057
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 05:22:16 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16AM2p19040;
	Thu, 6 Feb 2003 05:22:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16ALsp19026
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 05:21:54 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA11087
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 05:14:59 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:kJVnLiLUFzC5Y2divGXr9WmO9dN4cugg@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h16AIbD15488;
	Thu, 6 Feb 2003 03:18:37 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h16AIbB28602;
	Thu, 6 Feb 2003 03:18:37 -0700
Date: Thu, 6 Feb 2003 03:18:37 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206031837.I18441@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	sigtran@ietf.org
References: <20030206025512.F18441@openss7.org> <Pine.SOL.4.10.10302060403180.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302060403180.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 04:05:52AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

200? 20?

Its still the same.  For an example of 100, your approach would have
the SG and ASP exchange 5150 packets were 150 will do.

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 04:05:52)
> 
> Brian,
> 
> I said you are giving all theoritical examples which is IMPOSSIBLE
> practically.
> 
> Nobody will configure 1 ASP as active (20,000-1) ASP as inactive.
> Your example is total deviation from this thread.
> 
> --Samuel
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > ASP1 is active for AS1.
> > 
> > ASP2 is inactive for AS1.
> > ASP3 is inactive for AS1.
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 05:17:55 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA11266
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 05:17:55 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16AOIw19381
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 05:24:18 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16AO7p19330;
	Thu, 6 Feb 2003 05:24:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16ANhp19185
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 05:23:43 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA11198
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 05:16:48 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:G18TdNPkUYX8VpOKWD7PEa7l0HqZ2hla@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h16AKQD15524;
	Thu, 6 Feb 2003 03:20:26 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h16AKQK28886;
	Thu, 6 Feb 2003 03:20:26 -0700
Date: Thu, 6 Feb 2003 03:20:26 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        Richard Adams <Richard.Adams@etsi.org>,
        "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206032025.J18441@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	Richard Adams <Richard.Adams@etsi.org>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206022344.B18441@openss7.org> <Pine.SOL.4.10.10302060352390.22291-100000@sun3271.ssd.usa.alcatel.com> <20030206031643.G18441@openss7.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20030206031643.G18441@openss7.org>; from bidulock@openss7.org on Thu, Feb 06, 2003 at 03:16:43AM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Sorry 250 packets vs. 5250.  The 5000 unnecessary packets are the useless
messages sent to inactive ASP.

There is no way the SG can be required to send SSNM to inactive ASP.

--brian

Brian F. G. Bidulock wrote:                                                         (Thu, 06 Feb 2003 03:16:43)
> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 04:00:46)
> > 
> > Brian,
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 03:01:39)
> > > > 
> > > > Richard,
> > > > 
> > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > 
> > > > On Thu, 6 Feb 2003, Richard Adams wrote:
> > > > 
> > > > > Brian,
> > > > > I agree with you. I think when the ASP becomes active, it should do an
> > > > > audit to update its knowledge of the SS7 network. The ASP needs to take
> > > > > the initiative to find out about the SS7 nodes etc. in which it is
> > > > > interested. Any other method of working would lead to impossible loading
> > > > > on the SG.
> > > > 
> > > > When SG itself is re-starting , it means no ASP can reach any SS7
> > > > destinations via re-started SG.Always it is good to tell my ASP
> > > > who is in inactive state about SS7 destinations unavailability.
> > > 
> > > It is unneccessary.  Iff the ASP goes active, the SG can send DUNA(*)
> > > to it in one message to indicate that it is restarting.  The SG isn't
> > > required to wait for DAUD to do this, it could send DUNA(*) just because
> > > the ASP hase become active, or it could send DUNA(*) in response to the
> > > first DATA message from the ASP.
> > > 
> > 
> > Alternative ASP has to find out destination status by sending DAUD or
> > DATA.This is unnecessary traffic if the ASP had already received DUNA(*).
> 
> The alternate ASP does not have to send DAUD.
> 
> > 
> > If there are 100 ASPs coming up ,  100 DAUD or DATA should be sent from
> > ASP, obviously here the problem with respect to number of un-necessary
> > messages - POTENTIAL.
> 
> 100 ASPAC Req mesages and 100 ASPUP Ack messages have to be sent for 100 ASPs
> to come up.  There is little difficulty bundling a 100 DAUDs on the 100
> ASPACs.  There is no loss of efficiency.
> 
> If SSNM were to be sent to inactive ASPs, 100 messages would be sent for
> every SSNM.  There is no ASPAC or ASPAC Ack to bundle with these and they
> could be sent repeatedly to inactive ASPs before they ever activate.
> 
> So, for the current situation:
> 
>   SG sends 50 SSNM to the active ASP.
>   100 ASPs go active sending 100 ASPAC and 100 DAUD and receiving
>   100 ASPAC Ack and 100 SSNM (including all 50 indications).
> 
>   SG sends a total of 250 messages, bundled in 150 packets.
>   Each ASP sends 2 messages bundled in 1 packet.
>   Total 150 packets.
> 
> For your unnecessary situation:
> 
>   SG sends 50 SSNM to each of 101 ASPs (1 active and 100 inactive).
>   100 ASPs go active sending 100 ASPAC and receiving 100 ASPAC Ack.
> 
>   SG sends a total of 5150 messages, in 5150 packets.
>   Each ASP sends 1 message in 1 packet.
>   Total 5250 packets.
> 
> You suggestion would have us send 5150 packets where 150 will do.
> 
> The SG cannot be required to send SSNM to inactive ASPs.
> 
> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 05:44:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA12051
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 05:44:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16ApLu21314
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 05:51:21 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16Ap8p21299;
	Thu, 6 Feb 2003 05:51:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16Anpp21148
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 05:49:51 -0500
Received: from zdemail04.zdem.compaq.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA11991
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 05:42:55 -0500 (EST)
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail04.zdem.compaq.com (Postfix) with ESMTP
	id 52956E30; Thu,  6 Feb 2003 11:46:33 +0100 (CET)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.2966);
	 Thu, 6 Feb 2003 11:46:29 +0100
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Subject: RE: [Sigtran] M3UA - ASP database equalization
Date: Thu, 6 Feb 2003 11:46:28 +0100
Message-ID: <06CAA4D3753C53408D2CE6340B684442014888D1@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA - ASP database equalization
Thread-Index: AcLNyQwEzzr4cg+KSZGQubURm0g9wAAA2+kw
From: "Zannin, Francois" <Francois.Zannin@hp.com>
To: <bidulock@openss7.org>
Cc: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 06 Feb 2003 10:46:29.0592 (UTC) FILETIME=[01704580:01C2CDCD]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h16Anpp21149
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

See below,

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> Sent: Thursday, February 06, 2003 11:17 AM
> To: Zannin, Francois
> Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> Subject: Re: [Sigtran] M3UA - ASP database equalization
> 
> 
> Zannin,,
> 
> Zannin, Francois wrote:                                       
>                      (Thu, 06 Feb 2003 11:02:54)
> > Bidulock,,
> > 
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: Thursday, February 06, 2003 10:54 AM
> > > To: Zannin, Francois
> > > Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> > > Subject: Re: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > Zannin,,
> > > 
> > > Zannin, Francois wrote:                                       
> > >                      (Thu, 06 Feb 2003 10:44:08)
> > > > Bidulock,,
> > > > 
> > > > > -----Original Message-----
> > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > Sent: Thursday, February 06, 2003 10:22 AM
> > > > > To: Zannin, Francois
> > > > > Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> > > > > Subject: Re: [Sigtran] M3UA - ASP database equalization
> > > > > 
> > > > > 
> > > > > Zannin,,
> > > > > 
> > > > > Zannin, Francois wrote:                                       
> > > > >                      (Thu, 06 Feb 2003 09:58:46)
> > > > > > Bidulock,,
> > > > > > 
> > > > > > Does it make sense to have "a lot" of ASP UP connected to a 
> > > > > > SGP
> > > > > > and
> > > > > > not ready to do anything?
> > > > > 
> > > > > They could be active for different AS.
> > > > 
> > > > In that case, because SSNM are for ss7 SP and concern all
> > > SI (AS), it
> > > > is not useless to receive (and you will receive) these
> > > SSNM. Now, the
> > > > question is about the RC in these SSNM.
> > > 
> > > AS is not synonymous with SI.  AS can be for different 
> point codes.
> > 
> > You are right, but it doesn't change anything...
> > 
> > > 
> > > > 
> > > > > 
> > > > > > If an ASP is UP for a SGP, it certainly must be ready to
> > > > > activate an
> > > > > > AS and then concerned by the SSNM, and again, for an
> > > emergency AS
> > > > > > recovery (AS-Pending notification) the ASP must be
> > > ready to operate
> > > > > > after the Active-ACK.
> > > > > > 
> > > > > > Only the "ready to operate for an AS" ASP have a reason to
> > > > > be "UP" on
> > > > > > a SG.
> > > > > 
> > > > > Wrong.  Whenever an ASP goes ASP UP it becomes 
> ASP-INACTIVE for 
> > > > > all AS for which it is configured to operate.  
> Depending on SG 
> > > > > configuration, this could mean that the ASP is 
> ASP-INACTIVE for 
> > > > > all AS served by the SG, regarless of whether the ASP 
> will ever 
> > > > > go ASP-ACTIVE for all of the AS.
> > > > > There is no way for an ASP to tell the SG that it is ASP-DOWN 
> > > > > for some AS and ASP-INACTIVE for others.
> > > > > 
> > > > > Also, the purpose of the ASP-INACTIVE state is to 
> receive NTFY 
> > > > > messages for the AS.  In override mode, one ASP is active for 
> > > > > the AS and the others are inactive.  The inactive 
> ASPs have no 
> > > > > intention of going active unless the active ASP 
> fails, or unless 
> > > > > manual management action is taken.
> > > > > 
> > > > 
> > > > A ss7 state concern all SI (AS) for a NA, the ASP receive
> > > the SSNM for
> > > > the ACTIVE (or activable) AS. If no AS is "activable" 
> no reason to
> > > > send an ASP-UP...
> > > 
> > > Yes, the reason is to act as standby for one active AS.
> > 
> > So, on AS IS "activable"...
> 
> What's "activalble"?

"Activable" is an AS that is ready to be activated when it receive a
pending notification or a activate management command.
If NO AS is ready to work on an ASP, there is no need to send an ASP-UP
to the SG, and then no SSNM is received.

If at least one AS is ready to work (ASP_ACTIVE is coming soon) it make
sense to receive SSNM from the SG

> 
> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 05:46:52 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA12098
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 05:46:52 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16ArGw21414
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 05:53:16 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16Ar4p21403;
	Thu, 6 Feb 2003 05:53:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16AqNp21359
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 05:52:23 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA12073
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 05:45:27 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h16An2eW029312;
	Thu, 6 Feb 2003 04:49:03 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h16AmbM22071;
	Thu, 6 Feb 2003 04:48:37 -0600 (CST)
Date: Thu, 6 Feb 2003 04:48:37 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: Richard Adams <Richard.Adams@etsi.org>,
        "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206031643.G18441@openss7.org>
Message-ID: <Pine.SOL.4.10.10302060423540.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,


  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 04:00:46)
> > 
> > Brian,
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 03:01:39)
> > > > 
> > > > Richard,
> > > > 
> > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > 
> > > > On Thu, 6 Feb 2003, Richard Adams wrote:
> > > > 
> > > > > Brian,
> > > > > I agree with you. I think when the ASP becomes active, it should do an
> > > > > audit to update its knowledge of the SS7 network. The ASP needs to take
> > > > > the initiative to find out about the SS7 nodes etc. in which it is
> > > > > interested. Any other method of working would lead to impossible loading
> > > > > on the SG.
> > > > 
> > > > When SG itself is re-starting , it means no ASP can reach any SS7
> > > > destinations via re-started SG.Always it is good to tell my ASP
> > > > who is in inactive state about SS7 destinations unavailability.
> > > 
> > > It is unneccessary.  Iff the ASP goes active, the SG can send DUNA(*)
> > > to it in one message to indicate that it is restarting.  The SG isn't
> > > required to wait for DAUD to do this, it could send DUNA(*) just because
> > > the ASP hase become active, or it could send DUNA(*) in response to the
> > > first DATA message from the ASP.
> > > 
> > 
> > Alternative ASP has to find out destination status by sending DAUD or
> > DATA.This is unnecessary traffic if the ASP had already received DUNA(*).
> 
> The alternate ASP does not have to send DAUD.

You have not read properly my mail. I said "DAUD or DATA"..
see above.

> 
> > 
> > If there are 100 ASPs coming up ,  100 DAUD or DATA should be sent from
> > ASP, obviously here the problem with respect to number of un-necessary
> > messages - POTENTIAL.
> 
> 100 ASPAC Req mesages and 100 ASPUP Ack messages have to be sent for 100 ASPs
> to come up.  There is little difficulty bundling a 100 DAUDs on the 100
> ASPACs.  There is no loss of efficiency.

Bad concept. Only 5% of your idea is good remaining 95% is BAD.
This concept cannot be used to develop a good M3Ua stack.

> 
> If SSNM were to be sent to inactive ASPs, 100 messages would be sent for
> every SSNM.  There is no ASPAC or ASPAC Ack to bundle with these and they
> could be sent repeatedly to inactive ASPs before they ever activate.
> 


> So, for the current situation:
> 
>   SG sends 50 SSNM to the active ASP.
>   100 ASPs go active sending 100 ASPAC and 100 DAUD and receiving
>   100 ASPAC Ack and 100 SSNM (including all 50 indications).
> 
>   SG sends a total of 250 messages, bundled in 150 packets.
>   Each ASP sends 2 messages bundled in 1 packet.
>   Total 150 packets.
> 
> For your unnecessary situation:
> 
>   SG sends 50 SSNM to each of 101 ASPs (1 active and 100 inactive).
Again your example is wrong.PRACTICALLY NOBODY WILL NEVER EVER CONFIGURE
1 ASP and 100 ASPs INACTIVE.AND THERE IS NO NEED FOR THIS BAD
CONFUGURATION WHICH IS WASTING RESOURCES.

>   100 ASPs go active sending 100 ASPAC and receiving 100 ASPAC Ack.
> 
>   SG sends a total of 5150 messages, in 5150 packets.
>   Each ASP sends 1 message in 1 packet.
>   Total 5250 packets.
> 
> You suggestion would have us send 5150 packets where 150 will do.
> 
According to your example 5150 packets are used instead of 150 packets.
Poor example and poor solution.

--Samuel

> The SG cannot be required to send SSNM to inactive ASPs.
> 
> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 05:49:52 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA12315
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 05:49:52 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16AuGg21603
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 05:56:16 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16Au4p21579;
	Thu, 6 Feb 2003 05:56:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16AtQp21543
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 05:55:26 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA12263
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 05:48:30 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h16Aq3eW029544;
	Thu, 6 Feb 2003 04:52:03 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h16ApoM23211;
	Thu, 6 Feb 2003 04:51:50 -0600 (CST)
Date: Thu, 6 Feb 2003 04:51:50 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: Richard Adams <Richard.Adams@etsi.org>,
        "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206032025.J18441@openss7.org>
Message-ID: <Pine.SOL.4.10.10302060448570.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

I have never and ever used these numbers '250, 5250, and 5000'
they all are your example .Since your example is poor so your suggestion
is also wrong.

--Samuel

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Sorry 250 packets vs. 5250.  The 5000 unnecessary packets are the useless
> messages sent to inactive ASP.
> 
> There is no way the SG can be required to send SSNM to inactive ASP.
> 
> --brian
> 
> Brian F. G. Bidulock wrote:                                                         (Thu, 06 Feb 2003 03:16:43)
> > Samuel,
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 04:00:46)
> > > 
> > > Brian,
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Samuel,
> > > > 
> > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 03:01:39)
> > > > > 
> > > > > Richard,
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > > 
> > > > > On Thu, 6 Feb 2003, Richard Adams wrote:
> > > > > 
> > > > > > Brian,
> > > > > > I agree with you. I think when the ASP becomes active, it should do an
> > > > > > audit to update its knowledge of the SS7 network. The ASP needs to take
> > > > > > the initiative to find out about the SS7 nodes etc. in which it is
> > > > > > interested. Any other method of working would lead to impossible loading
> > > > > > on the SG.
> > > > > 
> > > > > When SG itself is re-starting , it means no ASP can reach any SS7
> > > > > destinations via re-started SG.Always it is good to tell my ASP
> > > > > who is in inactive state about SS7 destinations unavailability.
> > > > 
> > > > It is unneccessary.  Iff the ASP goes active, the SG can send DUNA(*)
> > > > to it in one message to indicate that it is restarting.  The SG isn't
> > > > required to wait for DAUD to do this, it could send DUNA(*) just because
> > > > the ASP hase become active, or it could send DUNA(*) in response to the
> > > > first DATA message from the ASP.
> > > > 
> > > 
> > > Alternative ASP has to find out destination status by sending DAUD or
> > > DATA.This is unnecessary traffic if the ASP had already received DUNA(*).
> > 
> > The alternate ASP does not have to send DAUD.
> > 
> > > 
> > > If there are 100 ASPs coming up ,  100 DAUD or DATA should be sent from
> > > ASP, obviously here the problem with respect to number of un-necessary
> > > messages - POTENTIAL.
> > 
> > 100 ASPAC Req mesages and 100 ASPUP Ack messages have to be sent for 100 ASPs
> > to come up.  There is little difficulty bundling a 100 DAUDs on the 100
> > ASPACs.  There is no loss of efficiency.
> > 
> > If SSNM were to be sent to inactive ASPs, 100 messages would be sent for
> > every SSNM.  There is no ASPAC or ASPAC Ack to bundle with these and they
> > could be sent repeatedly to inactive ASPs before they ever activate.
> > 
> > So, for the current situation:
> > 
> >   SG sends 50 SSNM to the active ASP.
> >   100 ASPs go active sending 100 ASPAC and 100 DAUD and receiving
> >   100 ASPAC Ack and 100 SSNM (including all 50 indications).
> > 
> >   SG sends a total of 250 messages, bundled in 150 packets.
> >   Each ASP sends 2 messages bundled in 1 packet.
> >   Total 150 packets.
> > 
> > For your unnecessary situation:
> > 
> >   SG sends 50 SSNM to each of 101 ASPs (1 active and 100 inactive).
> >   100 ASPs go active sending 100 ASPAC and receiving 100 ASPAC Ack.
> > 
> >   SG sends a total of 5150 messages, in 5150 packets.
> >   Each ASP sends 1 message in 1 packet.
> >   Total 5250 packets.
> > 
> > You suggestion would have us send 5150 packets where 150 will do.
> > 
> > The SG cannot be required to send SSNM to inactive ASPs.
> > 
> > --brian
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 06:15:01 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA13175
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 06:15:01 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16BLQ323640
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 06:21:26 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16BLDp23605;
	Thu, 6 Feb 2003 06:21:13 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16BIDp23402
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 06:18:13 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA13074
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 06:11:17 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:1i0oI/gYxyqF6e9EnnAf0XLTt/01UHVR@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h16BEtD16648;
	Thu, 6 Feb 2003 04:14:55 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h16BEsV03347;
	Thu, 6 Feb 2003 04:14:54 -0700
Date: Thu, 6 Feb 2003 04:14:53 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Zannin, Francois" <Francois.Zannin@hp.com>
Cc: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206041453.K18441@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Zannin, Francois" <Francois.Zannin@hp.com>,
	"Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <06CAA4D3753C53408D2CE6340B684442014888D1@vbeexc02.emea.cpqcorp.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <06CAA4D3753C53408D2CE6340B684442014888D1@vbeexc02.emea.cpqcorp.net>; from Francois.Zannin@hp.com on Thu, Feb 06, 2003 at 11:46:28AM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Zannin,,

Zannin, Francois wrote:                                                            (Thu, 06 Feb 2003 11:46:28)
> > 
> > What's "activalble"?
> 
> "Activable" is an AS that is ready to be activated when it receive a
> pending notification or a activate management command.
> If NO AS is ready to work on an ASP, there is no need to send an ASP-UP
> to the SG, and then no SSNM is received.
> 
> If at least one AS is ready to work (ASP_ACTIVE is coming soon) it make
> sense to receive SSNM from the SG

Going ASP UP with the intentions of going ASP-ACTIVE for one AS
can make an ASP ASP-INACTIVE for all AS.  There is no need to send
SSNM to inactive ASPs.

--brian


-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 06:16:02 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA13248
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 06:16:02 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16BMRs23717
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 06:22:27 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16BM7p23679;
	Thu, 6 Feb 2003 06:22:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16BLGp23621
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 06:21:16 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA13148
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 06:14:19 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:UaDgFMjX1FNR4eJ24KRl6KOGZsFMFcfA@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h16BHwD16686;
	Thu, 6 Feb 2003 04:17:58 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h16BHvR03682;
	Thu, 6 Feb 2003 04:17:57 -0700
Date: Thu, 6 Feb 2003 04:17:57 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: Richard Adams <Richard.Adams@etsi.org>,
        "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206041757.L18441@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	Richard Adams <Richard.Adams@etsi.org>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206031643.G18441@openss7.org> <Pine.SOL.4.10.10302060423540.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302060423540.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 04:48:37AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 04:48:37)
> 
> Brian,
> 
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 04:00:46)
> > > 
> > > Brian,
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Samuel,
> > > > 
> > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 03:01:39)
> > > > > 
> > > > > Richard,
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > > 
> > > > > On Thu, 6 Feb 2003, Richard Adams wrote:
> > > > > 
> > > > > > Brian,
> > > > > > I agree with you. I think when the ASP becomes active, it should do an
> > > > > > audit to update its knowledge of the SS7 network. The ASP needs to take
> > > > > > the initiative to find out about the SS7 nodes etc. in which it is
> > > > > > interested. Any other method of working would lead to impossible loading
> > > > > > on the SG.
> > > > > 
> > > > > When SG itself is re-starting , it means no ASP can reach any SS7
> > > > > destinations via re-started SG.Always it is good to tell my ASP
> > > > > who is in inactive state about SS7 destinations unavailability.
> > > > 
> > > > It is unneccessary.  Iff the ASP goes active, the SG can send DUNA(*)
> > > > to it in one message to indicate that it is restarting.  The SG isn't
> > > > required to wait for DAUD to do this, it could send DUNA(*) just because
> > > > the ASP hase become active, or it could send DUNA(*) in response to the
> > > > first DATA message from the ASP.
> > > > 
> > > 
> > > Alternative ASP has to find out destination status by sending DAUD or
> > > DATA.This is unnecessary traffic if the ASP had already received DUNA(*).
> > 
> > The alternate ASP does not have to send DAUD.
> 
> You have not read properly my mail. I said "DAUD or DATA"..
> see above.

You did not read above.  I said that the SG can send DUNA(*) to the ASP
when it goes active.  The ASP does not have to send DAUD or DATA.

> 
> > 
> > > 
> > > If there are 100 ASPs coming up ,  100 DAUD or DATA should be sent from
> > > ASP, obviously here the problem with respect to number of un-necessary
> > > messages - POTENTIAL.
> > 
> > 100 ASPAC Req mesages and 100 ASPUP Ack messages have to be sent for 100 ASPs
> > to come up.  There is little difficulty bundling a 100 DAUDs on the 100
> > ASPACs.  There is no loss of efficiency.
> 
> Bad concept. Only 5% of your idea is good remaining 95% is BAD.
> This concept cannot be used to develop a good M3Ua stack.

Explain.

> 
> > 
> > If SSNM were to be sent to inactive ASPs, 100 messages would be sent for
> > every SSNM.  There is no ASPAC or ASPAC Ack to bundle with these and they
> > could be sent repeatedly to inactive ASPs before they ever activate.
> > 
> 
> 
> > So, for the current situation:
> > 
> >   SG sends 50 SSNM to the active ASP.
> >   100 ASPs go active sending 100 ASPAC and 100 DAUD and receiving
> >   100 ASPAC Ack and 100 SSNM (including all 50 indications).
> > 
> >   SG sends a total of 250 messages, bundled in 150 packets.
> >   Each ASP sends 2 messages bundled in 1 packet.
> >   Total 150 packets.
> > 
> > For your unnecessary situation:
> > 
> >   SG sends 50 SSNM to each of 101 ASPs (1 active and 100 inactive).
> Again your example is wrong.PRACTICALLY NOBODY WILL NEVER EVER CONFIGURE
> 1 ASP and 100 ASPs INACTIVE.AND THERE IS NO NEED FOR THIS BAD
> CONFUGURATION WHICH IS WASTING RESOURCES.
> 
> >   100 ASPs go active sending 100 ASPAC and receiving 100 ASPAC Ack.
> > 
> >   SG sends a total of 5150 messages, in 5150 packets.
> >   Each ASP sends 1 message in 1 packet.
> >   Total 5250 packets.
> > 
> > You suggestion would have us send 5150 packets where 150 will do.
> > 
> According to your example 5150 packets are used instead of 150 packets.
> Poor example and poor solution.

It was your example.

And your solution is demostrably poor for your example.

I'll stick with the RFC approach, thank you.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 06:16:51 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA13283
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 06:16:51 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16BNGt23824
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 06:23:16 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16BN3p23803;
	Thu, 6 Feb 2003 06:23:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16BMkp23738
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 06:22:46 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA13243
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 06:15:49 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:orYORrcg/Ammq5tzUydgcKQ+I0U5It1G@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h16BJSD16716;
	Thu, 6 Feb 2003 04:19:28 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h16BJRP03888;
	Thu, 6 Feb 2003 04:19:27 -0700
Date: Thu, 6 Feb 2003 04:19:27 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: Richard Adams <Richard.Adams@etsi.org>,
        "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206041927.M18441@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	Richard Adams <Richard.Adams@etsi.org>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206032025.J18441@openss7.org> <Pine.SOL.4.10.10302060448570.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302060448570.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 04:51:50AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Your example, my count.

You have yet to show any efficiencies gained by deviating from the RFC
in any circumstance.

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 04:51:50)
> 
> Brian,
> 
> I have never and ever used these numbers '250, 5250, and 5000'
> they all are your example .Since your example is poor so your suggestion
> is also wrong.
> 
> --Samuel
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Sorry 250 packets vs. 5250.  The 5000 unnecessary packets are the useless
> > messages sent to inactive ASP.
> > 
> > There is no way the SG can be required to send SSNM to inactive ASP.
> > 
> > --brian
> > 
> > Brian F. G. Bidulock wrote:                                                         (Thu, 06 Feb 2003 03:16:43)
> > > Samuel,
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 04:00:46)
> > > > 
> > > > Brian,
> > > > 
> > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > 
> > > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > > 
> > > > > Samuel,
> > > > > 
> > > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 03:01:39)
> > > > > > 
> > > > > > Richard,
> > > > > > 
> > > > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > > >   1000 Coit Road, Plano, Texas 75075
> > > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > > > 
> > > > > > On Thu, 6 Feb 2003, Richard Adams wrote:
> > > > > > 
> > > > > > > Brian,
> > > > > > > I agree with you. I think when the ASP becomes active, it should do an
> > > > > > > audit to update its knowledge of the SS7 network. The ASP needs to take
> > > > > > > the initiative to find out about the SS7 nodes etc. in which it is
> > > > > > > interested. Any other method of working would lead to impossible loading
> > > > > > > on the SG.
> > > > > > 
> > > > > > When SG itself is re-starting , it means no ASP can reach any SS7
> > > > > > destinations via re-started SG.Always it is good to tell my ASP
> > > > > > who is in inactive state about SS7 destinations unavailability.
> > > > > 
> > > > > It is unneccessary.  Iff the ASP goes active, the SG can send DUNA(*)
> > > > > to it in one message to indicate that it is restarting.  The SG isn't
> > > > > required to wait for DAUD to do this, it could send DUNA(*) just because
> > > > > the ASP hase become active, or it could send DUNA(*) in response to the
> > > > > first DATA message from the ASP.
> > > > > 
> > > > 
> > > > Alternative ASP has to find out destination status by sending DAUD or
> > > > DATA.This is unnecessary traffic if the ASP had already received DUNA(*).
> > > 
> > > The alternate ASP does not have to send DAUD.
> > > 
> > > > 
> > > > If there are 100 ASPs coming up ,  100 DAUD or DATA should be sent from
> > > > ASP, obviously here the problem with respect to number of un-necessary
> > > > messages - POTENTIAL.
> > > 
> > > 100 ASPAC Req mesages and 100 ASPUP Ack messages have to be sent for 100 ASPs
> > > to come up.  There is little difficulty bundling a 100 DAUDs on the 100
> > > ASPACs.  There is no loss of efficiency.
> > > 
> > > If SSNM were to be sent to inactive ASPs, 100 messages would be sent for
> > > every SSNM.  There is no ASPAC or ASPAC Ack to bundle with these and they
> > > could be sent repeatedly to inactive ASPs before they ever activate.
> > > 
> > > So, for the current situation:
> > > 
> > >   SG sends 50 SSNM to the active ASP.
> > >   100 ASPs go active sending 100 ASPAC and 100 DAUD and receiving
> > >   100 ASPAC Ack and 100 SSNM (including all 50 indications).
> > > 
> > >   SG sends a total of 250 messages, bundled in 150 packets.
> > >   Each ASP sends 2 messages bundled in 1 packet.
> > >   Total 150 packets.
> > > 
> > > For your unnecessary situation:
> > > 
> > >   SG sends 50 SSNM to each of 101 ASPs (1 active and 100 inactive).
> > >   100 ASPs go active sending 100 ASPAC and receiving 100 ASPAC Ack.
> > > 
> > >   SG sends a total of 5150 messages, in 5150 packets.
> > >   Each ASP sends 1 message in 1 packet.
> > >   Total 5250 packets.
> > > 
> > > You suggestion would have us send 5150 packets where 150 will do.
> > > 
> > > The SG cannot be required to send SSNM to inactive ASPs.
> > > 
> > > --brian
> > > 
> > > -- 
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 06:22:56 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA13419
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 06:22:56 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16BTLV24076
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 06:29:21 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16BT6p24060;
	Thu, 6 Feb 2003 06:29:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16BSTp24030
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 06:28:29 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA13397
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 06:21:32 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:WnzuMnJTIGUAKoFF0kxlHok+qCS6Y3aA@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h16BPBD16787;
	Thu, 6 Feb 2003 04:25:11 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h16BPAc04594;
	Thu, 6 Feb 2003 04:25:10 -0700
Date: Thu, 6 Feb 2003 04:25:09 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: Richard Adams <Richard.Adams@etsi.org>,
        "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206042509.N18441@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	Richard Adams <Richard.Adams@etsi.org>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206032025.J18441@openss7.org> <Pine.SOL.4.10.10302060448570.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302060448570.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 04:51:50AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Mine is not a suggestion.  It is the RFC.

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 04:51:50)
> 
> Brian,
> 
> I have never and ever used these numbers '250, 5250, and 5000'
> they all are your example .Since your example is poor so your suggestion
> is also wrong.
> 
> --Samuel
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Sorry 250 packets vs. 5250.  The 5000 unnecessary packets are the useless
> > messages sent to inactive ASP.
> > 
> > There is no way the SG can be required to send SSNM to inactive ASP.
> > 
> > --brian
> > 
> > Brian F. G. Bidulock wrote:                                                         (Thu, 06 Feb 2003 03:16:43)
> > > Samuel,
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 04:00:46)
> > > > 
> > > > Brian,
> > > > 
> > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > 
> > > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > > 
> > > > > Samuel,
> > > > > 
> > > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 03:01:39)
> > > > > > 
> > > > > > Richard,
> > > > > > 
> > > > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > > >   1000 Coit Road, Plano, Texas 75075
> > > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > > > 
> > > > > > On Thu, 6 Feb 2003, Richard Adams wrote:
> > > > > > 
> > > > > > > Brian,
> > > > > > > I agree with you. I think when the ASP becomes active, it should do an
> > > > > > > audit to update its knowledge of the SS7 network. The ASP needs to take
> > > > > > > the initiative to find out about the SS7 nodes etc. in which it is
> > > > > > > interested. Any other method of working would lead to impossible loading
> > > > > > > on the SG.
> > > > > > 
> > > > > > When SG itself is re-starting , it means no ASP can reach any SS7
> > > > > > destinations via re-started SG.Always it is good to tell my ASP
> > > > > > who is in inactive state about SS7 destinations unavailability.
> > > > > 
> > > > > It is unneccessary.  Iff the ASP goes active, the SG can send DUNA(*)
> > > > > to it in one message to indicate that it is restarting.  The SG isn't
> > > > > required to wait for DAUD to do this, it could send DUNA(*) just because
> > > > > the ASP hase become active, or it could send DUNA(*) in response to the
> > > > > first DATA message from the ASP.
> > > > > 
> > > > 
> > > > Alternative ASP has to find out destination status by sending DAUD or
> > > > DATA.This is unnecessary traffic if the ASP had already received DUNA(*).
> > > 
> > > The alternate ASP does not have to send DAUD.
> > > 
> > > > 
> > > > If there are 100 ASPs coming up ,  100 DAUD or DATA should be sent from
> > > > ASP, obviously here the problem with respect to number of un-necessary
> > > > messages - POTENTIAL.
> > > 
> > > 100 ASPAC Req mesages and 100 ASPUP Ack messages have to be sent for 100 ASPs
> > > to come up.  There is little difficulty bundling a 100 DAUDs on the 100
> > > ASPACs.  There is no loss of efficiency.
> > > 
> > > If SSNM were to be sent to inactive ASPs, 100 messages would be sent for
> > > every SSNM.  There is no ASPAC or ASPAC Ack to bundle with these and they
> > > could be sent repeatedly to inactive ASPs before they ever activate.
> > > 
> > > So, for the current situation:
> > > 
> > >   SG sends 50 SSNM to the active ASP.
> > >   100 ASPs go active sending 100 ASPAC and 100 DAUD and receiving
> > >   100 ASPAC Ack and 100 SSNM (including all 50 indications).
> > > 
> > >   SG sends a total of 250 messages, bundled in 150 packets.
> > >   Each ASP sends 2 messages bundled in 1 packet.
> > >   Total 150 packets.
> > > 
> > > For your unnecessary situation:
> > > 
> > >   SG sends 50 SSNM to each of 101 ASPs (1 active and 100 inactive).
> > >   100 ASPs go active sending 100 ASPAC and receiving 100 ASPAC Ack.
> > > 
> > >   SG sends a total of 5150 messages, in 5150 packets.
> > >   Each ASP sends 1 message in 1 packet.
> > >   Total 5250 packets.
> > > 
> > > You suggestion would have us send 5150 packets where 150 will do.
> > > 
> > > The SG cannot be required to send SSNM to inactive ASPs.
> > > 
> > > --brian
> > > 
> > > -- 
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 06:24:50 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA13494
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 06:24:50 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16BVF124202
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 06:31:15 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16BV4p24185;
	Thu, 6 Feb 2003 06:31:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16BURp24120
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 06:30:27 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA13457
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 06:23:30 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h16BR7eY001783;
	Thu, 6 Feb 2003 05:27:07 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h16BQXM06259;
	Thu, 6 Feb 2003 05:26:33 -0600 (CST)
Date: Thu, 6 Feb 2003 05:26:33 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206041453.K18441@openss7.org>
Message-ID: <Pine.SOL.4.10.10302060517430.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Zannin,,
> 
> Zannin, Francois wrote:                                                            (Thu, 06 Feb 2003 11:46:28)
> > > 
> > > What's "activalble"?
> > 
> > "Activable" is an AS that is ready to be activated when it receive a
> > pending notification or a activate management command.
> > If NO AS is ready to work on an ASP, there is no need to send an ASP-UP
> > to the SG, and then no SSNM is received.
> > 
> > If at least one AS is ready to work (ASP_ACTIVE is coming soon) it make
> > sense to receive SSNM from the SG
> 
> Going ASP UP with the intentions of going ASP-ACTIVE for one AS
> can make an ASP ASP-INACTIVE for all AS.  There is no need to send
> SSNM to inactive ASPs.
> 

In override mode primary ASP will be in ASCTIVE state and backup ASP
will be in ASP INACT state.If it is DOWN state then we can declare that
M3UA stack has poor efficency.

Why ASP2 is in INACT state? the reason is anytime ASP1 may become
INACT and ASP2 will take the load by sending ASPAC message.

If the databases in ASP1 and ASP2 are same , this case ASP2 can easily
continue the traffic without any time delay.
How databases in ASP1 and ASP2 are same?
the answer is ASP2 should have received all the destination nodes statuses
when it was in inact state.

--Samuel
> --brian
> 
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 06:27:51 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA13592
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 06:27:51 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16BYG824382
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 06:34:16 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16BY5p24374;
	Thu, 6 Feb 2003 06:34:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16BXsp24351
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 06:33:54 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA13568
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 06:26:57 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:Aj/14tzsKp3OkdNM1wg1dDElP8Wk8KVn@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h16BUaD16915;
	Thu, 6 Feb 2003 04:30:36 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h16BUZ005261;
	Thu, 6 Feb 2003 04:30:35 -0700
Date: Thu, 6 Feb 2003 04:30:35 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206043035.O18441@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206041453.K18441@openss7.org> <Pine.SOL.4.10.10302060517430.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302060517430.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 05:26:33AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 05:26:33)
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Zannin,,
> > 
> > Zannin, Francois wrote:                                                            (Thu, 06 Feb 2003 11:46:28)
> > > > 
> > > > What's "activalble"?
> > > 
> > > "Activable" is an AS that is ready to be activated when it receive a
> > > pending notification or a activate management command.
> > > If NO AS is ready to work on an ASP, there is no need to send an ASP-UP
> > > to the SG, and then no SSNM is received.
> > > 
> > > If at least one AS is ready to work (ASP_ACTIVE is coming soon) it make
> > > sense to receive SSNM from the SG
> > 
> > Going ASP UP with the intentions of going ASP-ACTIVE for one AS
> > can make an ASP ASP-INACTIVE for all AS.  There is no need to send
> > SSNM to inactive ASPs.
> > 
> 
> In override mode primary ASP will be in ASCTIVE state and backup ASP
> will be in ASP INACT state.If it is DOWN state then we can declare that
> M3UA stack has poor efficency.

How so?  What poor efficiency?

> 
> Why ASP2 is in INACT state? the reason is anytime ASP1 may become
> INACT and ASP2 will take the load by sending ASPAC message.
> 
> If the databases in ASP1 and ASP2 are same , this case ASP2 can easily
> continue the traffic without any time delay.

The can continue traffic without any dely even if their databases
are different.  What delay do you think is required by the RFC to occur?

> How databases in ASP1 and ASP2 are same?

They don't have to be.

> the answer is ASP2 should have received all the destination nodes statuses
> when it was in inact state.

No, ASP2 should receive all the destination node statuses after it goes
active.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 06:33:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA13844
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 06:33:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16BeM425448
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 06:40:22 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16Be6p25420;
	Thu, 6 Feb 2003 06:40:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16BcOp25283
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 06:38:24 -0500
Received: from mx1.etsi.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA13776
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 06:31:26 -0500 (EST)
Received: from email10.etsihq.org ([172.27.1.62]) by mx1.etsi.org with Microsoft SMTPSVC(5.0.2195.5329);
	 Thu, 6 Feb 2003 12:35:03 +0100
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="US-ASCII"
Subject: RE: [Sigtran] M3UA - ASP database equalization
Date: Thu, 6 Feb 2003 12:35:02 +0100
Message-ID: <4091553999CBE4409CC2B562152B5A33EDDF59@email10.etsihq.org>
Thread-Topic: [Sigtran] M3UA - ASP database equalization
Thread-Index: AcLNvT63i7v8hYdnS5uO4fQJDQUcZQABi8ZQ
From: "Richard Adams" <Richard.Adams@etsi.org>
To: <bidulock@openss7.org>
Cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 06 Feb 2003 11:35:03.0277 (UTC) FILETIME=[CA2151D0:01C2CDD3]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h16BcOp25284
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Brian,
The discussion is similar to the arguments we had when defining MTP
restart. 
There, a restarting SP (SEP or STP) is sent, by all its adjacent running
STPs, TFPs referring to all 
SPs that each such adjacent STP can't currently reach. 
All other SPs in the restarting SP's routing tables are assumed by the
restarting SP to be accessible. 
This keeps the number of messages as low as possible. Then a TRA message
interchange signifies the
end of restart. 
TFPs and TFAs are sent only to those adjacent SPs that are available.

The M3UA messages defined for ASPs becoming active (ASPTM messages)
imply that the ASP part of the network does not act in the same way as a
collection of SS7 SEPs.
We don't have the M3UA equivalent of a TRA message (and the primitive
for the MTP notifying
its local users when restart has completed, was purposely left in Q.704
as implementation dependent).

Now, we have a question of ASP doing a restart, but also of a restarting
SG(P). 
For the SS7 network, a restarting SG presumably will do MTP 
restart as an SEP or an STP. Should it also then send out SSNMs to all
ASPs (that it presumably
thinks are inactive ?) The ASP part of the network could have been
treated as SEPs, but it wasn't.
In our case, the becoming active ASP itself could prompt for information
in DAUD(s), 
and receive DUNA(s) or DAVA(s) as appropriate. This is ~similar to the
Q.704 Red Book 
(section 12.4.2(b) I think) national option of an SP doing a route set
test to 
an adjacent STP when a linkset becomes available. 
And that worked, although it could lead to large numbers of messages in
the network.
Being able to aggregate PCs in DAUD and response messages helps avoid
this problem.

With the ASP active procedures, an alternative would have been for the
ASP Active message 
from ASP to SG to prompt the SG to send DUNA(s) about inaccessible SS7
nodes, followed
by the ASPAC Ack, but I don't read that in the M3UA RFC.     

So, IMO, whether or not a newly active ASP sends DAUD or not is up to it
(although in a managed IP
network maybe the network owner could state if it wanted this to be
done). The ASP could, entirely
dependent on its implementation, decide not to send DAUD and rely on
DUNAs, from the
SG  in response to DATA, if the destination is inaccessible. Or the ASP
could assume
some destinations to be available and others need testing (much as SCCP
does for
subsystems with which it is concerned on its own restart). 

So, I do NOT think that an SG should be required to send SSNM messages
to inactive ASPs. 

Kind regards,
Dick Adams,
richard.adams@etsi.org



-----Original Message-----
From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
Sent: 06 February 2003 09:53
To: Richard Adams
Cc: Zannin, Francois; Samuel Dur D. Jeyaseelan; Nagar, Ashish;
sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization


Richard,

Let's say it is an override AS...  In that case only one can be
active at a time.

I'm glad you agree with the concensus in the RFC.  We have discussed
this many times over the last several years and have always arrived
at the same conclusion: SG cannot be required to send SSNM to inactive
ASPs.

It is even worse on an SG when dynamic registration is not performed.
In this case, every ASP is considered to be ASP-INACTIVE for every
AS when it comes up.  Sending SSNM to inactive ASPs would send every
SSNM to every ASP, regardless of the more restricted set of AS that
the ASP is actually configured to ever be active for.

--brian

Richard Adams wrote:
(Thu, 06 Feb 2003 09:41:41)
> Brian,
> I agree with you. I think when the ASP becomes active, it should do an
> audit to update its knowledge of the SS7 network. The ASP needs to
take
> the initiative to find out about the SS7 nodes etc. in which it is
> interested. Any other method of working would lead to impossible
loading
> on the SG. 
> Of course, I hope 199,999 of your inactive ASPs do not decide to
become
> active at the same time.....! 
> Kind regards,
> Dick Adams,
> richard.adams@etsi.org
> 
>  
> 
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: 06 February 2003 09:15
> To: Zannin, Francois
> Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> Subject: Re: [Sigtran] M3UA - ASP database equalization
> 
> 
> Zannin,,
> 
> That is contrary to the conensus of the RFC.
> 
> Consider 1 ASP active, 200,000 ASPs inactive.
> 
> The RFC cannot require an SG to send SSNM to an inactive ASP.
> 
> --brian
> 
> Zannin, Francois wrote:
> (Thu, 06 Feb 2003 09:02:41)
> > Samuel,
> > 
> > IMO, it's a good idea to inform inactive (BUT UP) ASP of the state
of
> > the ss7 network especially in the override mode where an ASP can be
> > ready to recover an AS as soon as it receive an AS-Pending
> notification
> > from the SGp. In this condition, the ASP is ready to operate quickly
> and
> > minimize the risk of buffered message lost during the pending timer.
> > 
> > If someone does not want to receive SSNM because it is not able to
> > recover an AS, it just stays in an ASP-Down state.
> > 
> > Because SSNM reflects ss7 Point Code state that is not "SI"
dependant,
> > it affects all AS for a network and then, unless SG that plays
> multiple
> > networks, the ASP UP is sufficient to filter the undesired SSNM.
> > 
> > Francois.
> > 
> > -----Original Message-----
> > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]

> > Sent: Thursday, February 06, 2003 8:35 AM
> > To: Nagar, Ashish
> > Cc: sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > 
> > Ashish,
> > 
> > But when inactive ASP changes state to active, first of all
> > it does not know whether to send DAUD or not.
> > 
> > It will try to send traffic - but actually destination is
unavailable.
> > 
> > The worst case is when SG triggers restart procedures this case all
> the
> > SS7 destinations are unavailable via restarted-SG.But this poor ASP
> > (newly come-up ASP) keeps sending traffic via restarted-SG to reach
> the
> > SS7 destination and for every message ASP receives DUNA.
> > 
> > Are we not doing double-work here?
> > 
> > --Samuel
> > 
> > 
> >   Alcatel USA, Inc.                  Internet:
> > <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > *******
> > 
> > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > 
> > > Hi Samuel
> > > 
> > > My point is why to do any processing if ASP is in Inactive state. 
> > > Think about ASP is in Inactive state for longer period and
receiving
> 
> > > lots of DUNA and DAVA, while what matter is the present only when
> ASP 
> > > state changes to ASP-ACTIVE.
> > > 
> > > Thanks & Regards
> > > Ashish Nagar
> > > 
> > > 
> > > -----Original Message-----
> > > From: Samuel Dur D. Jeyaseelan
[mailto:sjeyasee@ssd.usa.alcatel.com]
> > > Sent: Thursday, February 06, 2003 12:30 PM
> > > To: Nagar, Ashish
> > > Cc: sigtran@ietf.org
> > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > Ashish,
> > > 
> > > My point is if SG is able to send DUNA to all inact and active
ASPs,
> 
> > > we should not restrict it.
> > > 
> > > Sending DAUD is an alternative way ,If SG has direct way of
sending 
> > > DUNA then Why do you need alternative way?
> > > 
> > > --Samuel
> > > 
> > >   Alcatel USA, Inc.                  Internet:
> > <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA,
Inc. 
> > > *******
> > > 
> > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > 
> > > > You can generate DAUD procedure at any time, so whenever ASP
> changes
> > 
> > > > state from Inactive to Active Just generate DAUD.
> > > > 
> > > > Thanks & Regards
> > > > Ashish Nagar
> > > > 
> > > > 
> > > > -----Original Message-----
> > > > From: Samuel Dur D. Jeyaseelan
> [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > To: sigtran@ietf.org
> > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > 
> > > > 
> > > > Hi all,
> > > > 
> > > > Currently RFC 3332 says that SG sends SSNM messages to all 
> > > > 'concerned ASPs'(means only ACTIVE ASPs).
> > > > 
> > > > +----+           +---+           +-----+
> > > > |SEP | ----------|SG |-----------| ASP1|
> > > > +----+           +---+           +-----+
> > > >                    |             
> > > >                    |             +-----+
> > > >                    +-------------| ASP2|
> > > >                                  +-----+
> > > > 
> > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > 
> > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > But SEP status is not changed in ASP2.There are two ways to
solve 
> > > > this problem
> > > > 
> > > > 1. ASP1 can equalize its database with ASP2
> > > > or
> > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is
inactive)
> > > > 
> > > > IMO, SG can send SSNM messages to ASP (either ASP is in active
or 
> > > > inact
> > > > state)
> > > > 
> > > > Any comments?
> > > > 
> > > > -----
> > > > 
> > > > section 5.4.2.1
> > > > 
> > > > ...
> > > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of
the
> > > >         concerned Network Appearance.
> > > > 
> > > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER
> request
> > 
> > > > is sent to MTP3.
> > > > 
> > > > But there is no primitive availabe between M3UA and MTP3.
> > > > 
> > > > I think it should be corrected.
> > > > 
> > > > ----
> > > > Presently ASP is unaware of SG's MTP3 restart,
> > > > ASP receives only DUNA for all SS7 point codes which can be
> > > > reachable via that SG.But in case of peer to peer protocol
> > > > SG and AS point codes are different, So why dont we send
> > > > DUNA with SG's point code to ASP? So that just by getting
> > > > single DUNA(with SG's PC) ASP will not route any message via
that
> > SG.
> > > > This can avoid sending multiple DUNAs to ASP.
> > > > 
> > > > 
> > > > Any comments?
> > > > 
> > > > -----
> > > > section 1.4.1
> > > > 
> > > > ...
> > > >    Where a particular AS can be reached via more than one SGP,
the
> > > >    corresponding Routing Keys in the SGPs should be identical.
> > (Note:
> > > >    It is possible for the SGP Routing Key configuration data to
be
> > > >    temporarily out-of-sync during configuration updates).
> > > >  
> > > >                               +--------+
> > > >                               |        |
> > > >                  +------------+  SG 1  +--------------+
> > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > >      |  SEP  +---+            +--------+              +---/
\
> > > >      |   or  |                    |*                      | ASPs
> |
> > > >      |  STP  +---+            +--------+              +---\
/
> > > >      +-------+   |            |        |              |     ----
> > > >                  +------------+  SG 2  +--------------+
> > > >                               | "STP"  |
> > > >                               +--------+
> > > >  
> > > >                     Figure 1  Example with mated SGs
> > > > 
> > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are 
> > > > located geographically in different places.The ideas is when SG1

> > > > goes down in CITY1(due to earthquake...) , all the traffice
still 
> > > > can be handled by SG2 which is in CITY2.
> > > > 
> > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will
> be 
> > > > routed from SGP1 to SGP2.
> > > > 
> > > > Does all these things mean to say that SGP1's mate 'should be' 
> > > > present in SG2 and not in SG1 itself ?
> > > >  
> > > > 
> > > > Comments are welcome.
> > > > 
> > > > --Samuel
> > > > 
> > > > 
> > > > 
> > > > 
> > > >   Alcatel USA, Inc.                  Internet:
> > > <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA,
> Inc.
> > > *******
> > > > 
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > 
> > > 
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 06:45:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA14254
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 06:45:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16BqND25996
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 06:52:23 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16Bq9p25964;
	Thu, 6 Feb 2003 06:52:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16BpPp25920
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 06:51:25 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA14200
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 06:44:27 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h16Bm2qe026666;
	Thu, 6 Feb 2003 05:48:02 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h16BloM14068;
	Thu, 6 Feb 2003 05:47:50 -0600 (CST)
Date: Thu, 6 Feb 2003 05:47:50 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: Richard Adams <Richard.Adams@etsi.org>,
        "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206041757.L18441@openss7.org>
Message-ID: <Pine.SOL.4.10.10302060534460.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 04:48:37)
> > 
> > Brian,
> > 
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 04:00:46)
> > > > 
> > > > Brian,
> > > > 
> > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > 
> > > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > > 
> > > > > Samuel,
> > > > > 
> > > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 03:01:39)
> > > > > > 
> > > > > > Richard,
> > > > > > 
> > > > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > > >   1000 Coit Road, Plano, Texas 75075
> > > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > > > 
> > > > > > On Thu, 6 Feb 2003, Richard Adams wrote:
> > > > > > 
> > > > > > > Brian,
> > > > > > > I agree with you. I think when the ASP becomes active, it should do an
> > > > > > > audit to update its knowledge of the SS7 network. The ASP needs to take
> > > > > > > the initiative to find out about the SS7 nodes etc. in which it is
> > > > > > > interested. Any other method of working would lead to impossible loading
> > > > > > > on the SG.
> > > > > > 
> > > > > > When SG itself is re-starting , it means no ASP can reach any SS7
> > > > > > destinations via re-started SG.Always it is good to tell my ASP
> > > > > > who is in inactive state about SS7 destinations unavailability.
> > > > > 
> > > > > It is unneccessary.  Iff the ASP goes active, the SG can send DUNA(*)
> > > > > to it in one message to indicate that it is restarting.  The SG isn't
> > > > > required to wait for DAUD to do this, it could send DUNA(*) just because
> > > > > the ASP hase become active, or it could send DUNA(*) in response to the
> > > > > first DATA message from the ASP.
> > > > > 
> > > > 
> > > > Alternative ASP has to find out destination status by sending DAUD or
> > > > DATA.This is unnecessary traffic if the ASP had already received DUNA(*).
> > > 
> > > The alternate ASP does not have to send DAUD.
> > 
> > You have not read properly my mail. I said "DAUD or DATA"..
> > see above.
> 
> You did not read above.  I said that the SG can send DUNA(*) to the ASP
> when it goes active.  The ASP does not have to send DAUD or DATA.
> 

WRONG. 

 What about DAVA, SCON,DRST? as soon as ASP comes up Will you send
these messages based on the destination status?
Again this is indirect solution.Finally stack will be implemented with
these concepts but we cant find efficentcy?


 
> > 
> > > 
> > > > 
> > > > If there are 100 ASPs coming up ,  100 DAUD or DATA should be sent from
> > > > ASP, obviously here the problem with respect to number of un-necessary
> > > > messages - POTENTIAL.
> > > 
> > > 100 ASPAC Req mesages and 100 ASPUP Ack messages have to be sent for 100 ASPs
> > > to come up.  There is little difficulty bundling a 100 DAUDs on the 100
> > > ASPACs.  There is no loss of efficiency.
> > 
> > Bad concept. Only 5% of your idea is good remaining 95% is BAD.
> > This concept cannot be used to develop a good M3Ua stack.
> 
> Explain.
> 
> > 
> > > 
> > > If SSNM were to be sent to inactive ASPs, 100 messages would be sent for
> > > every SSNM.  There is no ASPAC or ASPAC Ack to bundle with these and they
> > > could be sent repeatedly to inactive ASPs before they ever activate.
> > > 
> > 
> > 
> > > So, for the current situation:
> > > 
> > >   SG sends 50 SSNM to the active ASP.
> > >   100 ASPs go active sending 100 ASPAC and 100 DAUD and receiving
> > >   100 ASPAC Ack and 100 SSNM (including all 50 indications).
> > > 
> > >   SG sends a total of 250 messages, bundled in 150 packets.
> > >   Each ASP sends 2 messages bundled in 1 packet.
> > >   Total 150 packets.
> > > 
> > > For your unnecessary situation:
> > > 
> > >   SG sends 50 SSNM to each of 101 ASPs (1 active and 100 inactive).
> > Again your example is wrong.PRACTICALLY NOBODY WILL NEVER EVER CONFIGURE
> > 1 ASP and 100 ASPs INACTIVE.AND THERE IS NO NEED FOR THIS BAD
> > CONFUGURATION WHICH IS WASTING RESOURCES.
> > 
> > >   100 ASPs go active sending 100 ASPAC and receiving 100 ASPAC Ack.
> > > 
> > >   SG sends a total of 5150 messages, in 5150 packets.
> > >   Each ASP sends 1 message in 1 packet.
> > >   Total 5250 packets.
> > > 
> > > You suggestion would have us send 5150 packets where 150 will do.
> > > 
> > According to your example 5150 packets are used instead of 150 packets.
> > Poor example and poor solution.
> 
> It was your example.
> 
> And your solution is demostrably poor for your example.
> 
> I'll stick with the RFC approach, thank you.
Yes everybody wants to stick with RFC approach.But if RFC approach is
inefficent then anybody can suggest efficent approach.

--Samuel

> 
> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 06:46:51 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA14282
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 06:46:51 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16BrHg26078
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 06:53:17 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16Br2p26054;
	Thu, 6 Feb 2003 06:53:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16BqRp26009
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 06:52:27 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA14241
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 06:45:27 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:oSRPnmhTyjgz3GGVIdBPG6gs8NouuoSd@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h16Bn6D17130;
	Thu, 6 Feb 2003 04:49:06 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h16Bn6R07541;
	Thu, 6 Feb 2003 04:49:06 -0700
Date: Thu, 6 Feb 2003 04:49:05 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Richard Adams <Richard.Adams@etsi.org>
Cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206044905.P18441@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Richard Adams <Richard.Adams@etsi.org>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <4091553999CBE4409CC2B562152B5A33EDDF59@email10.etsihq.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <4091553999CBE4409CC2B562152B5A33EDDF59@email10.etsihq.org>; from Richard.Adams@etsi.org on Thu, Feb 06, 2003 at 12:35:02PM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Richard,

Yes, this is what we arrived at in the "MTP Restart" thread of
September 2001.

I agree with your analysis as I did then.

The simplest approach for the newly active ASP is to just send
traffic and rely on responsive DUNAs (there being very few
inaccessible destinations at any point in time with the exception
of MTP Restart).

The intelligent SG designer could, as you say, send preventative
DUNAs (or DUNA(*) during MTP Restart) immediately following
ASPAC Ack without violating any requirement of the RFC.  I would
consider that "sending SSNM to concerned ASPs."

I also agree with your conclusion: SG should not be required
to send SSNM to inactive ASPs.

--brian


Richard Adams wrote:                                                               (Thu, 06 Feb 2003 12:35:02)
> Brian,
> The discussion is similar to the arguments we had when defining MTP
> restart. 
> There, a restarting SP (SEP or STP) is sent, by all its adjacent running
> STPs, TFPs referring to all 
> SPs that each such adjacent STP can't currently reach. 
> All other SPs in the restarting SP's routing tables are assumed by the
> restarting SP to be accessible. 
> This keeps the number of messages as low as possible. Then a TRA message
> interchange signifies the
> end of restart. 
> TFPs and TFAs are sent only to those adjacent SPs that are available.
> 
> The M3UA messages defined for ASPs becoming active (ASPTM messages)
> imply that the ASP part of the network does not act in the same way as a
> collection of SS7 SEPs.
> We don't have the M3UA equivalent of a TRA message (and the primitive
> for the MTP notifying
> its local users when restart has completed, was purposely left in Q.704
> as implementation dependent).
> 
> Now, we have a question of ASP doing a restart, but also of a restarting
> SG(P). 
> For the SS7 network, a restarting SG presumably will do MTP 
> restart as an SEP or an STP. Should it also then send out SSNMs to all
> ASPs (that it presumably
> thinks are inactive ?) The ASP part of the network could have been
> treated as SEPs, but it wasn't.
> In our case, the becoming active ASP itself could prompt for information
> in DAUD(s), 
> and receive DUNA(s) or DAVA(s) as appropriate. This is ~similar to the
> Q.704 Red Book 
> (section 12.4.2(b) I think) national option of an SP doing a route set
> test to 
> an adjacent STP when a linkset becomes available. 
> And that worked, although it could lead to large numbers of messages in
> the network.
> Being able to aggregate PCs in DAUD and response messages helps avoid
> this problem.
> 
> With the ASP active procedures, an alternative would have been for the
> ASP Active message 
> from ASP to SG to prompt the SG to send DUNA(s) about inaccessible SS7
> nodes, followed
> by the ASPAC Ack, but I don't read that in the M3UA RFC.     
> 
> So, IMO, whether or not a newly active ASP sends DAUD or not is up to it
> (although in a managed IP
> network maybe the network owner could state if it wanted this to be
> done). The ASP could, entirely
> dependent on its implementation, decide not to send DAUD and rely on
> DUNAs, from the
> SG  in response to DATA, if the destination is inaccessible. Or the ASP
> could assume
> some destinations to be available and others need testing (much as SCCP
> does for
> subsystems with which it is concerned on its own restart). 
> 
> So, I do NOT think that an SG should be required to send SSNM messages
> to inactive ASPs. 
> 
> Kind regards,
> Dick Adams,
> richard.adams@etsi.org
> 
> 
> 
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: 06 February 2003 09:53
> To: Richard Adams
> Cc: Zannin, Francois; Samuel Dur D. Jeyaseelan; Nagar, Ashish;
> sigtran@ietf.org
> Subject: Re: [Sigtran] M3UA - ASP database equalization
> 
> 
> Richard,
> 
> Let's say it is an override AS...  In that case only one can be
> active at a time.
> 
> I'm glad you agree with the concensus in the RFC.  We have discussed
> this many times over the last several years and have always arrived
> at the same conclusion: SG cannot be required to send SSNM to inactive
> ASPs.
> 
> It is even worse on an SG when dynamic registration is not performed.
> In this case, every ASP is considered to be ASP-INACTIVE for every
> AS when it comes up.  Sending SSNM to inactive ASPs would send every
> SSNM to every ASP, regardless of the more restricted set of AS that
> the ASP is actually configured to ever be active for.
> 
> --brian
> 
> Richard Adams wrote:
> (Thu, 06 Feb 2003 09:41:41)
> > Brian,
> > I agree with you. I think when the ASP becomes active, it should do an
> > audit to update its knowledge of the SS7 network. The ASP needs to
> take
> > the initiative to find out about the SS7 nodes etc. in which it is
> > interested. Any other method of working would lead to impossible
> loading
> > on the SG. 
> > Of course, I hope 199,999 of your inactive ASPs do not decide to
> become
> > active at the same time.....! 
> > Kind regards,
> > Dick Adams,
> > richard.adams@etsi.org
> > 
> >  
> > 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: 06 February 2003 09:15
> > To: Zannin, Francois
> > Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> > Subject: Re: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > Zannin,,
> > 
> > That is contrary to the conensus of the RFC.
> > 
> > Consider 1 ASP active, 200,000 ASPs inactive.
> > 
> > The RFC cannot require an SG to send SSNM to an inactive ASP.
> > 
> > --brian
> > 
> > Zannin, Francois wrote:
> > (Thu, 06 Feb 2003 09:02:41)
> > > Samuel,
> > > 
> > > IMO, it's a good idea to inform inactive (BUT UP) ASP of the state
> of
> > > the ss7 network especially in the override mode where an ASP can be
> > > ready to recover an AS as soon as it receive an AS-Pending
> > notification
> > > from the SGp. In this condition, the ASP is ready to operate quickly
> > and
> > > minimize the risk of buffered message lost during the pending timer.
> > > 
> > > If someone does not want to receive SSNM because it is not able to
> > > recover an AS, it just stays in an ASP-Down state.
> > > 
> > > Because SSNM reflects ss7 Point Code state that is not "SI"
> dependant,
> > > it affects all AS for a network and then, unless SG that plays
> > multiple
> > > networks, the ASP UP is sufficient to filter the undesired SSNM.
> > > 
> > > Francois.
> > > 
> > > -----Original Message-----
> > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> 
> > > Sent: Thursday, February 06, 2003 8:35 AM
> > > To: Nagar, Ashish
> > > Cc: sigtran@ietf.org
> > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > 
> > > Ashish,
> > > 
> > > But when inactive ASP changes state to active, first of all
> > > it does not know whether to send DAUD or not.
> > > 
> > > It will try to send traffic - but actually destination is
> unavailable.
> > > 
> > > The worst case is when SG triggers restart procedures this case all
> > the
> > > SS7 destinations are unavailable via restarted-SG.But this poor ASP
> > > (newly come-up ASP) keeps sending traffic via restarted-SG to reach
> > the
> > > SS7 destination and for every message ASP receives DUNA.
> > > 
> > > Are we not doing double-work here?
> > > 
> > > --Samuel
> > > 
> > > 
> > >   Alcatel USA, Inc.                  Internet:
> > > <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > *******
> > > 
> > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > 
> > > > Hi Samuel
> > > > 
> > > > My point is why to do any processing if ASP is in Inactive state. 
> > > > Think about ASP is in Inactive state for longer period and
> receiving
> > 
> > > > lots of DUNA and DAVA, while what matter is the present only when
> > ASP 
> > > > state changes to ASP-ACTIVE.
> > > > 
> > > > Thanks & Regards
> > > > Ashish Nagar
> > > > 
> > > > 
> > > > -----Original Message-----
> > > > From: Samuel Dur D. Jeyaseelan
> [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > Sent: Thursday, February 06, 2003 12:30 PM
> > > > To: Nagar, Ashish
> > > > Cc: sigtran@ietf.org
> > > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > > 
> > > > 
> > > > Ashish,
> > > > 
> > > > My point is if SG is able to send DUNA to all inact and active
> ASPs,
> > 
> > > > we should not restrict it.
> > > > 
> > > > Sending DAUD is an alternative way ,If SG has direct way of
> sending 
> > > > DUNA then Why do you need alternative way?
> > > > 
> > > > --Samuel
> > > > 
> > > >   Alcatel USA, Inc.                  Internet:
> > > <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA,
> Inc. 
> > > > *******
> > > > 
> > > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > > 
> > > > > You can generate DAUD procedure at any time, so whenever ASP
> > changes
> > > 
> > > > > state from Inactive to Active Just generate DAUD.
> > > > > 
> > > > > Thanks & Regards
> > > > > Ashish Nagar
> > > > > 
> > > > > 
> > > > > -----Original Message-----
> > > > > From: Samuel Dur D. Jeyaseelan
> > [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > > To: sigtran@ietf.org
> > > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > > 
> > > > > 
> > > > > Hi all,
> > > > > 
> > > > > Currently RFC 3332 says that SG sends SSNM messages to all 
> > > > > 'concerned ASPs'(means only ACTIVE ASPs).
> > > > > 
> > > > > +----+           +---+           +-----+
> > > > > |SEP | ----------|SG |-----------| ASP1|
> > > > > +----+           +---+           +-----+
> > > > >                    |             
> > > > >                    |             +-----+
> > > > >                    +-------------| ASP2|
> > > > >                                  +-----+
> > > > > 
> > > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > > 
> > > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > > But SEP status is not changed in ASP2.There are two ways to
> solve 
> > > > > this problem
> > > > > 
> > > > > 1. ASP1 can equalize its database with ASP2
> > > > > or
> > > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is
> inactive)
> > > > > 
> > > > > IMO, SG can send SSNM messages to ASP (either ASP is in active
> or 
> > > > > inact
> > > > > state)
> > > > > 
> > > > > Any comments?
> > > > > 
> > > > > -----
> > > > > 
> > > > > section 5.4.2.1
> > > > > 
> > > > > ...
> > > > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of
> the
> > > > >         concerned Network Appearance.
> > > > > 
> > > > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER
> > request
> > > 
> > > > > is sent to MTP3.
> > > > > 
> > > > > But there is no primitive availabe between M3UA and MTP3.
> > > > > 
> > > > > I think it should be corrected.
> > > > > 
> > > > > ----
> > > > > Presently ASP is unaware of SG's MTP3 restart,
> > > > > ASP receives only DUNA for all SS7 point codes which can be
> > > > > reachable via that SG.But in case of peer to peer protocol
> > > > > SG and AS point codes are different, So why dont we send
> > > > > DUNA with SG's point code to ASP? So that just by getting
> > > > > single DUNA(with SG's PC) ASP will not route any message via
> that
> > > SG.
> > > > > This can avoid sending multiple DUNAs to ASP.
> > > > > 
> > > > > 
> > > > > Any comments?
> > > > > 
> > > > > -----
> > > > > section 1.4.1
> > > > > 
> > > > > ...
> > > > >    Where a particular AS can be reached via more than one SGP,
> the
> > > > >    corresponding Routing Keys in the SGPs should be identical.
> > > (Note:
> > > > >    It is possible for the SGP Routing Key configuration data to
> be
> > > > >    temporarily out-of-sync during configuration updates).
> > > > >  
> > > > >                               +--------+
> > > > >                               |        |
> > > > >                  +------------+  SG 1  +--------------+
> > > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > > >      |  SEP  +---+            +--------+              +---/
> \
> > > > >      |   or  |                    |*                      | ASPs
> > |
> > > > >      |  STP  +---+            +--------+              +---\
> /
> > > > >      +-------+   |            |        |              |     ----
> > > > >                  +------------+  SG 2  +--------------+
> > > > >                               | "STP"  |
> > > > >                               +--------+
> > > > >  
> > > > >                     Figure 1  Example with mated SGs
> > > > > 
> > > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are 
> > > > > located geographically in different places.The ideas is when SG1
> 
> > > > > goes down in CITY1(due to earthquake...) , all the traffice
> still 
> > > > > can be handled by SG2 which is in CITY2.
> > > > > 
> > > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will
> > be 
> > > > > routed from SGP1 to SGP2.
> > > > > 
> > > > > Does all these things mean to say that SGP1's mate 'should be' 
> > > > > present in SG2 and not in SG1 itself ?
> > > > >  
> > > > > 
> > > > > Comments are welcome.
> > > > > 
> > > > > --Samuel
> > > > > 
> > > > > 
> > > > > 
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet:
> > > > <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA,
> > Inc.
> > > > *******
> > > > > 
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > 
> > > > 
> > > 
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 06:48:50 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA14359
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 06:48:50 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16BtFt26195
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 06:55:15 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16Bt2p26183;
	Thu, 6 Feb 2003 06:55:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16BsGp26137
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 06:54:16 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA14312
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 06:47:18 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:UXARDhmmlB+w1hwhL0IYkfzNBbqougwu@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h16BovD17155;
	Thu, 6 Feb 2003 04:50:57 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h16Bovf07777;
	Thu, 6 Feb 2003 04:50:57 -0700
Date: Thu, 6 Feb 2003 04:50:57 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: Richard Adams <Richard.Adams@etsi.org>,
        "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206045057.Q18441@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	Richard Adams <Richard.Adams@etsi.org>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206041757.L18441@openss7.org> <Pine.SOL.4.10.10302060534460.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302060534460.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 05:47:50AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

What inefficiencies do you think exist in the current approach?

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 05:47:50)
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 04:48:37)
> > > 
> > > Brian,
> > > 
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Samuel,
> > > > 
> > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 04:00:46)
> > > > > 
> > > > > Brian,
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > > 
> > > > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > > > 
> > > > > > Samuel,
> > > > > > 
> > > > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 03:01:39)
> > > > > > > 
> > > > > > > Richard,
> > > > > > > 
> > > > > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > > > >   1000 Coit Road, Plano, Texas 75075
> > > > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > > > > 
> > > > > > > On Thu, 6 Feb 2003, Richard Adams wrote:
> > > > > > > 
> > > > > > > > Brian,
> > > > > > > > I agree with you. I think when the ASP becomes active, it should do an
> > > > > > > > audit to update its knowledge of the SS7 network. The ASP needs to take
> > > > > > > > the initiative to find out about the SS7 nodes etc. in which it is
> > > > > > > > interested. Any other method of working would lead to impossible loading
> > > > > > > > on the SG.
> > > > > > > 
> > > > > > > When SG itself is re-starting , it means no ASP can reach any SS7
> > > > > > > destinations via re-started SG.Always it is good to tell my ASP
> > > > > > > who is in inactive state about SS7 destinations unavailability.
> > > > > > 
> > > > > > It is unneccessary.  Iff the ASP goes active, the SG can send DUNA(*)
> > > > > > to it in one message to indicate that it is restarting.  The SG isn't
> > > > > > required to wait for DAUD to do this, it could send DUNA(*) just because
> > > > > > the ASP hase become active, or it could send DUNA(*) in response to the
> > > > > > first DATA message from the ASP.
> > > > > > 
> > > > > 
> > > > > Alternative ASP has to find out destination status by sending DAUD or
> > > > > DATA.This is unnecessary traffic if the ASP had already received DUNA(*).
> > > > 
> > > > The alternate ASP does not have to send DAUD.
> > > 
> > > You have not read properly my mail. I said "DAUD or DATA"..
> > > see above.
> > 
> > You did not read above.  I said that the SG can send DUNA(*) to the ASP
> > when it goes active.  The ASP does not have to send DAUD or DATA.
> > 
> 
> WRONG. 
> 
>  What about DAVA, SCON,DRST? as soon as ASP comes up Will you send
> these messages based on the destination status?
> Again this is indirect solution.Finally stack will be implemented with
> these concepts but we cant find efficentcy?
> 
> 
>  
> > > 
> > > > 
> > > > > 
> > > > > If there are 100 ASPs coming up ,  100 DAUD or DATA should be sent from
> > > > > ASP, obviously here the problem with respect to number of un-necessary
> > > > > messages - POTENTIAL.
> > > > 
> > > > 100 ASPAC Req mesages and 100 ASPUP Ack messages have to be sent for 100 ASPs
> > > > to come up.  There is little difficulty bundling a 100 DAUDs on the 100
> > > > ASPACs.  There is no loss of efficiency.
> > > 
> > > Bad concept. Only 5% of your idea is good remaining 95% is BAD.
> > > This concept cannot be used to develop a good M3Ua stack.
> > 
> > Explain.
> > 
> > > 
> > > > 
> > > > If SSNM were to be sent to inactive ASPs, 100 messages would be sent for
> > > > every SSNM.  There is no ASPAC or ASPAC Ack to bundle with these and they
> > > > could be sent repeatedly to inactive ASPs before they ever activate.
> > > > 
> > > 
> > > 
> > > > So, for the current situation:
> > > > 
> > > >   SG sends 50 SSNM to the active ASP.
> > > >   100 ASPs go active sending 100 ASPAC and 100 DAUD and receiving
> > > >   100 ASPAC Ack and 100 SSNM (including all 50 indications).
> > > > 
> > > >   SG sends a total of 250 messages, bundled in 150 packets.
> > > >   Each ASP sends 2 messages bundled in 1 packet.
> > > >   Total 150 packets.
> > > > 
> > > > For your unnecessary situation:
> > > > 
> > > >   SG sends 50 SSNM to each of 101 ASPs (1 active and 100 inactive).
> > > Again your example is wrong.PRACTICALLY NOBODY WILL NEVER EVER CONFIGURE
> > > 1 ASP and 100 ASPs INACTIVE.AND THERE IS NO NEED FOR THIS BAD
> > > CONFUGURATION WHICH IS WASTING RESOURCES.
> > > 
> > > >   100 ASPs go active sending 100 ASPAC and receiving 100 ASPAC Ack.
> > > > 
> > > >   SG sends a total of 5150 messages, in 5150 packets.
> > > >   Each ASP sends 1 message in 1 packet.
> > > >   Total 5250 packets.
> > > > 
> > > > You suggestion would have us send 5150 packets where 150 will do.
> > > > 
> > > According to your example 5150 packets are used instead of 150 packets.
> > > Poor example and poor solution.
> > 
> > It was your example.
> > 
> > And your solution is demostrably poor for your example.
> > 
> > I'll stick with the RFC approach, thank you.
> Yes everybody wants to stick with RFC approach.But if RFC approach is
> inefficent then anybody can suggest efficent approach.
> 
> --Samuel
> 
> > 
> > --brian
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 07:08:01 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA14979
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 07:08:01 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16CERE27545
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 07:14:27 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16CEAp27529;
	Thu, 6 Feb 2003 07:14:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16CDSp27505
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 07:13:28 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA14922
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 07:06:31 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h16CA2qg028500;
	Thu, 6 Feb 2003 06:10:03 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h16C9qM22615;
	Thu, 6 Feb 2003 06:09:52 -0600 (CST)
Date: Thu, 6 Feb 2003 06:09:52 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206043035.O18441@openss7.org>
Message-ID: <Pine.SOL.4.10.10302060558300.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 05:26:33)
> > 
> > Brian,
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > 
> > > Zannin,,
> > > 
> > > Zannin, Francois wrote:                                                            (Thu, 06 Feb 2003 11:46:28)
> > > > > 
> > > > > What's "activalble"?
> > > > 
> > > > "Activable" is an AS that is ready to be activated when it receive a
> > > > pending notification or a activate management command.
> > > > If NO AS is ready to work on an ASP, there is no need to send an ASP-UP
> > > > to the SG, and then no SSNM is received.
> > > > 
> > > > If at least one AS is ready to work (ASP_ACTIVE is coming soon) it make
> > > > sense to receive SSNM from the SG
> > > 
> > > Going ASP UP with the intentions of going ASP-ACTIVE for one AS
> > > can make an ASP ASP-INACTIVE for all AS.  There is no need to send
> > > SSNM to inactive ASPs.
> > > 
> > 
> > In override mode primary ASP will be in ASCTIVE state and backup ASP
> > will be in ASP INACT state.If it is DOWN state then we can declare that
> > M3UA stack has poor efficency.
> 
> How so?  What poor efficiency?
> 
I should have said 'null efficiency'

section 4.3.1 says
In DOWN state ASP can receive only Hearbeat,ASP DOWN ACK and Error
messages.
So when ASP2 is in down state it will discard NTFY message from SGP.
Override mode will collapse here.
A good stack will expect ASP2(backup) in INACT state.
 
> > 
> > Why ASP2 is in INACT state? the reason is anytime ASP1 may become
> > INACT and ASP2 will take the load by sending ASPAC message.
> > 
> > If the databases in ASP1 and ASP2 are same , this case ASP2 can easily
> > continue the traffic without any time delay.
> 
> The can continue traffic without any dely even if their databases
> are different.  What delay do you think is required by the RFC to occur?

They can't continue without delay.Because newly comeup ASP does not know
the current status of SS7 destinations with respect to the current route.
He has to spend some time by sending all un-necessary messages to SG.

> 
> > How databases in ASP1 and ASP2 are same?
> 
> They don't have to be.
No comment.
> 
> > the answer is ASP2 should have received all the destination nodes statuses
> > when it was in inact state.
> 
> No, ASP2 should receive all the destination node statuses after it goes
> active.
> 
No comment.

--Samuel
> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 07:18:15 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA15190
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 07:18:15 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16COfd27934
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 07:24:41 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16COQp27905;
	Thu, 6 Feb 2003 07:24:28 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16CNPp27871
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 07:23:25 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA15139
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 07:16:27 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:fCmp/ls+vdqGH0J2oy5SH/C7QUDLzahX@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h16CK6D17534;
	Thu, 6 Feb 2003 05:20:06 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h16CK5311303;
	Thu, 6 Feb 2003 05:20:05 -0700
Date: Thu, 6 Feb 2003 05:20:04 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206052004.R18441@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206043035.O18441@openss7.org> <Pine.SOL.4.10.10302060558300.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302060558300.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 06:09:52AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 06:09:52)
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 05:26:33)
> > > 
> > > Brian,
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Zannin,,
> > > > 
> > > > Zannin, Francois wrote:                                                            (Thu, 06 Feb 2003 11:46:28)
> > > > > > 
> > > > > > What's "activalble"?
> > > > > 
> > > > > "Activable" is an AS that is ready to be activated when it receive a
> > > > > pending notification or a activate management command.
> > > > > If NO AS is ready to work on an ASP, there is no need to send an ASP-UP
> > > > > to the SG, and then no SSNM is received.
> > > > > 
> > > > > If at least one AS is ready to work (ASP_ACTIVE is coming soon) it make
> > > > > sense to receive SSNM from the SG
> > > > 
> > > > Going ASP UP with the intentions of going ASP-ACTIVE for one AS
> > > > can make an ASP ASP-INACTIVE for all AS.  There is no need to send
> > > > SSNM to inactive ASPs.
> > > > 
> > > 
> > > In override mode primary ASP will be in ASCTIVE state and backup ASP
> > > will be in ASP INACT state.If it is DOWN state then we can declare that
> > > M3UA stack has poor efficency.
> > 
> > How so?  What poor efficiency?
> > 
> I should have said 'null efficiency'
> 
> section 4.3.1 says
> In DOWN state ASP can receive only Hearbeat,ASP DOWN ACK and Error
> messages.

An ASP can receive any message in any state.  Section 4.3.1 says that an SG
should not send other than these messages.

> So when ASP2 is in down state it will discard NTFY message from SGP.

It should not receive any NTFY message from the SGP.

> Override mode will collapse here.

Explain.

> A good stack will expect ASP2(backup) in INACT state.

It is not required by the RFC.  A backup ASP can be in any state.
If it is in the ASP-DOWN state, it will not receive any NTFY messages,
but that does not stop an ASP from determining the state of its
brethren by other means.

Where is the 'null efficiency'?

>  
> > > 
> > > Why ASP2 is in INACT state? the reason is anytime ASP1 may become
> > > INACT and ASP2 will take the load by sending ASPAC message.
> > > 
> > > If the databases in ASP1 and ASP2 are same , this case ASP2 can easily
> > > continue the traffic without any time delay.
> > 
> > The can continue traffic without any dely even if their databases
> > are different.  What delay do you think is required by the RFC to occur?
> 
> They can't continue without delay.Because newly comeup ASP does not know
> the current status of SS7 destinations with respect to the current route.
> He has to spend some time by sending all un-necessary messages to SG.

The newly active ASP must wait for ASPAC Ack from the SG before it sends
any DATA messages.  If the ASP sends DAUD with ASPAC and receives
the DAUD response with ASPAC Ack there is no additional delay whatsoever.

Where did you think there would be a delay?

> 
> > 
> > > How databases in ASP1 and ASP2 are same?
> > 
> > They don't have to be.
> No comment.

I take it you agree.

> > 
> > > the answer is ASP2 should have received all the destination nodes statuses
> > > when it was in inact state.
> > 
> > No, ASP2 should receive all the destination node statuses after it goes
> > active.
> > 
> No comment.

I take it you agree.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 07:23:49 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA15363
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 07:23:49 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16CUF928241
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 07:30:15 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16CU4p28205;
	Thu, 6 Feb 2003 07:30:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16CR1p28049
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 07:27:01 -0500
Received: from zdemail03.zdem.compaq.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA15248
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 07:20:03 -0500 (EST)
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail03.zdem.compaq.com (Postfix) with ESMTP
	id 8F4351C9; Thu,  6 Feb 2003 13:23:38 +0100 (CET)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.2966);
	 Thu, 6 Feb 2003 13:23:39 +0100
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Subject: RE: [Sigtran] M3UA - ASP database equalization
Date: Thu, 6 Feb 2003 13:23:37 +0100
Message-ID: <06CAA4D3753C53408D2CE6340B68444214C40F@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA - ASP database equalization
Thread-Index: AcLN0TGLEJMJt4FGQIaGJQiCrHpKbgACBKsA
From: "Zannin, Francois" <Francois.Zannin@hp.com>
To: <bidulock@openss7.org>
Cc: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 06 Feb 2003 12:23:39.0032 (UTC) FILETIME=[940E3980:01C2CDDA]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h16CR1p28050
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Bidulock,,

See below,

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> Sent: Thursday, February 06, 2003 12:15 PM
> To: Zannin, Francois
> Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> Subject: Re: [Sigtran] M3UA - ASP database equalization
> 
> 
> Zannin,,
> 
> Zannin, Francois wrote:                                       
>                      (Thu, 06 Feb 2003 11:46:28)
> > > 
> > > What's "activalble"?
> > 
> > "Activable" is an AS that is ready to be activated when it 
> receive a 
> > pending notification or a activate management command. If NO AS is 
> > ready to work on an ASP, there is no need to send an ASP-UP 
> to the SG, 
> > and then no SSNM is received.
> > 
> > If at least one AS is ready to work (ASP_ACTIVE is coming soon) it 
> > make sense to receive SSNM from the SG
> 
> Going ASP UP with the intentions of going ASP-ACTIVE for one 
> AS can make an ASP ASP-INACTIVE for all AS.  There is no need 
> to send SSNM to inactive ASPs.
> 

You are right, ASP UP WILL make ASP-INACTIVE for all AS, but because at
least one AS on this ASP is concerned by the SSNM, the ASP anyway will
receive the SSNM, maybe the RC list can change but these SSNM are valid
for all AS (for a NA).
Because one ASP is active for one AS, the number SSNM received by the
ASP do not change. 
If NO AS is "ready to work", you just do not send ASP UP to the SG. BUT
if you need to be ready for an AS recovery (pending notification), you
send the ASP UP and then you are concerned by the SSNM for an emergency
recovery.

> --brian
> 
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 07:31:56 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA15707
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 07:31:55 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16CcMh29227
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 07:38:22 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16Cc9p29215;
	Thu, 6 Feb 2003 07:38:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16Cbtp29184
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 07:37:55 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA15619
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 07:30:57 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:+4O2F/f07dSDnFwp9aDtErf4CczpbC5W@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h16CYaD17779;
	Thu, 6 Feb 2003 05:34:36 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h16CYZK13099;
	Thu, 6 Feb 2003 05:34:35 -0700
Date: Thu, 6 Feb 2003 05:34:35 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Zannin, Francois" <Francois.Zannin@hp.com>
Cc: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206053435.S18441@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Zannin, Francois" <Francois.Zannin@hp.com>,
	"Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <06CAA4D3753C53408D2CE6340B68444214C40F@vbeexc02.emea.cpqcorp.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <06CAA4D3753C53408D2CE6340B68444214C40F@vbeexc02.emea.cpqcorp.net>; from Francois.Zannin@hp.com on Thu, Feb 06, 2003 at 01:23:37PM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Zannin,,

So you see that there are ASP that do not want to receive SSNM
for inactive AS.  This is one reason why the SG cannot be required
to send such useless SSNM to inactive ASPs.

--brian

Zannin, Francois wrote:                                                            (Thu, 06 Feb 2003 13:23:37)
> Bidulock,,
> 
> See below,
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> > Sent: Thursday, February 06, 2003 12:15 PM
> > To: Zannin, Francois
> > Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> > Subject: Re: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > Zannin,,
> > 
> > Zannin, Francois wrote:                                       
> >                      (Thu, 06 Feb 2003 11:46:28)
> > > > 
> > > > What's "activalble"?
> > > 
> > > "Activable" is an AS that is ready to be activated when it 
> > receive a 
> > > pending notification or a activate management command. If NO AS is 
> > > ready to work on an ASP, there is no need to send an ASP-UP 
> > to the SG, 
> > > and then no SSNM is received.
> > > 
> > > If at least one AS is ready to work (ASP_ACTIVE is coming soon) it 
> > > make sense to receive SSNM from the SG
> > 
> > Going ASP UP with the intentions of going ASP-ACTIVE for one 
> > AS can make an ASP ASP-INACTIVE for all AS.  There is no need 
> > to send SSNM to inactive ASPs.
> > 
> 
> You are right, ASP UP WILL make ASP-INACTIVE for all AS, but because at
> least one AS on this ASP is concerned by the SSNM, the ASP anyway will
> receive the SSNM, maybe the RC list can change but these SSNM are valid
> for all AS (for a NA).
> Because one ASP is active for one AS, the number SSNM received by the
> ASP do not change. 
> If NO AS is "ready to work", you just do not send ASP UP to the SG. BUT
> if you need to be ready for an AS recovery (pending notification), you
> send the ASP UP and then you are concerned by the SSNM for an emergency
> recovery.
> 
> > --brian
> > 
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 07:43:12 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA16006
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 07:43:12 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16Cndh29663
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 07:49:39 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16CnCp29648;
	Thu, 6 Feb 2003 07:49:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16Cm3p29612
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 07:48:03 -0500
Received: from zdemail03.zdem.compaq.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA15948
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 07:41:04 -0500 (EST)
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail03.zdem.compaq.com (Postfix) with ESMTP
	id 594E35DF; Thu,  6 Feb 2003 13:44:43 +0100 (CET)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.2966);
	 Thu, 6 Feb 2003 13:44:43 +0100
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Subject: RE: [Sigtran] M3UA - ASP database equalization
Date: Thu, 6 Feb 2003 13:44:40 +0100
Message-ID: <06CAA4D3753C53408D2CE6340B68444214C410@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA - ASP database equalization
Thread-Index: AcLN3DQzDL2mnsHwT9GaSJIkHy6ibgAAGvpA
From: "Zannin, Francois" <Francois.Zannin@hp.com>
To: <bidulock@openss7.org>
Cc: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 06 Feb 2003 12:44:43.0760 (UTC) FILETIME=[85E46700:01C2CDDD]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h16Cm3p29613
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Bidolock,,

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> Sent: Thursday, February 06, 2003 1:35 PM
> To: Zannin, Francois
> Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> Subject: Re: [Sigtran] M3UA - ASP database equalization
> 
> 
> Zannin,,
> 
> So you see that there are ASP that do not want to receive 
> SSNM for inactive AS.  This is one reason why the SG cannot 
> be required to send such useless SSNM to inactive ASPs.

If an ASP do not need to receive SSNM, it just do not send the ASP UP.
As soon as the ASP is ready to activate an AS, it sends the ASP UP, and
then it is notified by the SSNM.
SSNM reflect the state of a ss7 SP and is valid for all AS of the same
NA, so receiving a SSNM for one or several AS is the same thing.
If you receive a SSNM for one AS, it is valid for all AS of the same NA.
If you don't need to receive SSNM,,,,  don't send the ASP UP.

> 
> --brian
> 
> Zannin, Francois wrote:                                       
>                      (Thu, 06 Feb 2003 13:23:37)
> > Bidulock,,
> > 
> > See below,
> > 
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: Thursday, February 06, 2003 12:15 PM
> > > To: Zannin, Francois
> > > Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> > > Subject: Re: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > Zannin,,
> > > 
> > > Zannin, Francois wrote:                                       
> > >                      (Thu, 06 Feb 2003 11:46:28)
> > > > > 
> > > > > What's "activalble"?
> > > > 
> > > > "Activable" is an AS that is ready to be activated when it
> > > receive a
> > > > pending notification or a activate management command. 
> If NO AS is
> > > > ready to work on an ASP, there is no need to send an ASP-UP 
> > > to the SG,
> > > > and then no SSNM is received.
> > > > 
> > > > If at least one AS is ready to work (ASP_ACTIVE is 
> coming soon) it
> > > > make sense to receive SSNM from the SG
> > > 
> > > Going ASP UP with the intentions of going ASP-ACTIVE for one
> > > AS can make an ASP ASP-INACTIVE for all AS.  There is no need 
> > > to send SSNM to inactive ASPs.
> > > 
> > 
> > You are right, ASP UP WILL make ASP-INACTIVE for all AS, 
> but because 
> > at least one AS on this ASP is concerned by the SSNM, the 
> ASP anyway 
> > will receive the SSNM, maybe the RC list can change but 
> these SSNM are 
> > valid for all AS (for a NA). Because one ASP is active for 
> one AS, the 
> > number SSNM received by the ASP do not change.
> > If NO AS is "ready to work", you just do not send ASP UP to 
> the SG. BUT
> > if you need to be ready for an AS recovery (pending 
> notification), you
> > send the ASP UP and then you are concerned by the SSNM for 
> an emergency
> > recovery.
> > 
> > > --brian
> > > 
> > > 
> > > --
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > 
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 07:58:58 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA16352
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 07:58:58 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16D5P730294
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 08:05:25 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16D5Cp30285;
	Thu, 6 Feb 2003 08:05:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16D4Kp30241
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 08:04:20 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA16317
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 07:57:21 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:412VkHB8beqKgcIV1OPLLnPT5i4MU3iu@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h16D10D18156;
	Thu, 6 Feb 2003 06:01:00 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h16D0xL16381;
	Thu, 6 Feb 2003 06:00:59 -0700
Date: Thu, 6 Feb 2003 06:00:59 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Zannin, Francois" <Francois.Zannin@hp.com>
Cc: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206060059.A14908@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Zannin, Francois" <Francois.Zannin@hp.com>,
	"Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <06CAA4D3753C53408D2CE6340B68444214C410@vbeexc02.emea.cpqcorp.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <06CAA4D3753C53408D2CE6340B68444214C410@vbeexc02.emea.cpqcorp.net>; from Francois.Zannin@hp.com on Thu, Feb 06, 2003 at 01:44:40PM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Zannin,,

Zannin, Francois wrote:                                                            (Thu, 06 Feb 2003 13:44:40)
> Bidolock,,
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> > Sent: Thursday, February 06, 2003 1:35 PM
> > To: Zannin, Francois
> > Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> > Subject: Re: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > Zannin,,
> > 
> > So you see that there are ASP that do not want to receive 
> > SSNM for inactive AS.  This is one reason why the SG cannot 
> > be required to send such useless SSNM to inactive ASPs.
> 
> If an ASP do not need to receive SSNM, it just do not send the ASP UP.
> As soon as the ASP is ready to activate an AS, it sends the ASP UP, and
> then it is notified by the SSNM.

There are ASP that need to receive NTFY for an AS but do not need
to receive SSNM for the AS.  In ASP-INACTIVE state, the ASP receives
NTFY, but does not receive SSNM.  If the ASP wishes to receive SSNM,
it moves to the ASP-ACTIVE state.

> SSNM reflect the state of a ss7 SP and is valid for all AS of the same
> NA, so receiving a SSNM for one or several AS is the same thing.

Not true.  Two AS in the same NA (but different point codes, not
just different SI) can have different routing.  One AS can be given
DUNA because it is not permitted to access the destination (has no
route to the destination), the other given DAVA because it is
permitted to access the destination (has a route to the destination).
So, where multiple AS share the same association, RC is required in
the SSNM message to indicate for which AS the SSNM message pertains.
All RCs to which the message pertains must be listed in the SSNM.
None of these RCs may be inactive.

> If you receive a SSNM for one AS, it is valid for all AS of the same NA.

That is not true.  See above.

> If you don't need to receive SSNM,,,,  don't send the ASP UP.

On the contrary, of you need to receive SSNM, send ASP Active.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 08:19:07 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA16990
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 08:19:07 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16DPZQ31710
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 08:25:35 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16DPJp31648;
	Thu, 6 Feb 2003 08:25:20 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16DOFp31587
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 08:24:15 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA16933
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 08:17:09 -0500 (EST)
Subject: RE: [Sigtran] M3UA - ASP database equalization
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OFCECAF790.3AA9C2B1-ONC2256CC5.0046F41F@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Thu, 6 Feb 2003 15:17:17 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/06/2003
 03:17:30 PM
MIME-Version: 1.0
Content-type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h16DOFp31588
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit


I see RC as the main construct of communication between ASP and SGP (which
mimics MTP3/MTP3-User relationship) So, I would prefer to keep RC for any
message between ASP and SGP. It can be thought (although maybe somehow
implementation dependent) that each AS will correspond to a MTP-SAP on ASP.
RC in the incoming messages helps to route those messages easily to the
corresponding MTP-SAP.

I think also the assumption that all AS will have the same destinatin
availability may not hold if AS have RKs with different DPC values.

Furthermore, if we consider dynamic registration, for an ASP, which did not
register yet, which RC value will be put to the SSNM, if we want to send
SSNM while ASP is INACTIVE.

I personally don´t see a significant benefit in either approach from
message exchange point of view. And for the argument that standby ASP
should start working properly as soon as possible, I would say:
a) Better use loadsharing(or even maybe broadcast) mode of operation
b) Synchronize also destination availability (in addition to other things
which need to be synchronized) between active/standby ASPs.
c) Start DAUD procedure when standby ASP goes ACTIVE or consider all
destinations available and update the database based on DUNA received from
SG(already presented during this thread a few times)

But I think RFC uses the word SHOULD NOT, about SG sending SSNM to INACTIVE
ASP. So, it is not a MUST.

I would think keeping RFC as it is now should be fine but I also agree that
this is quite gray area.



Bidolock,,

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Thursday, February 06, 2003 1:35 PM
> To: Zannin, Francois
> Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> Subject: Re: [Sigtran] M3UA - ASP database equalization
>
>
> Zannin,,
>
> So you see that there are ASP that do not want to receive
> SSNM for inactive AS.  This is one reason why the SG cannot
> be required to send such useless SSNM to inactive ASPs.

If an ASP do not need to receive SSNM, it just do not send the ASP UP.
As soon as the ASP is ready to activate an AS, it sends the ASP UP, and
then it is notified by the SSNM.
SSNM reflect the state of a ss7 SP and is valid for all AS of the same
NA, so receiving a SSNM for one or several AS is the same thing.
If you receive a SSNM for one AS, it is valid for all AS of the same NA.
If you don't need to receive SSNM,,,,  don't send the ASP UP.









_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 08:27:54 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA17222
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 08:27:54 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16DYMV31987
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 08:34:22 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16DY8p31976;
	Thu, 6 Feb 2003 08:34:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16DXgp31947
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 08:33:42 -0500
Received: from zdemail03.zdem.compaq.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA17203
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 08:26:42 -0500 (EST)
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail03.zdem.compaq.com (Postfix) with ESMTP
	id 7451A597; Thu,  6 Feb 2003 14:30:20 +0100 (CET)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.2966);
	 Thu, 6 Feb 2003 14:30:20 +0100
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Subject: RE: [Sigtran] M3UA - ASP database equalization
Date: Thu, 6 Feb 2003 14:30:15 +0100
Message-ID: <06CAA4D3753C53408D2CE6340B68444214C411@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA - ASP database equalization
Thread-Index: AcLN39zFj7ZcL0StTqi5FwZHIscd7QAAl9mQ
From: "Zannin, Francois" <Francois.Zannin@hp.com>
To: <bidulock@openss7.org>
Cc: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 06 Feb 2003 13:30:20.0375 (UTC) FILETIME=[E50AB670:01C2CDE3]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h16DXhp31948
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit



> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> Sent: Thursday, February 06, 2003 2:01 PM
> To: Zannin, Francois
> Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> Subject: Re: [Sigtran] M3UA - ASP database equalization
> 
> 
> Zannin,,
> 
> Zannin, Francois wrote:                                       
>                      (Thu, 06 Feb 2003 13:44:40)
> > Bidolock,,
> > 
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: Thursday, February 06, 2003 1:35 PM
> > > To: Zannin, Francois
> > > Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> > > Subject: Re: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > Zannin,,
> > > 
> > > So you see that there are ASP that do not want to receive
> > > SSNM for inactive AS.  This is one reason why the SG cannot 
> > > be required to send such useless SSNM to inactive ASPs.
> > 
> > If an ASP do not need to receive SSNM, it just do not send 
> the ASP UP. 
> > As soon as the ASP is ready to activate an AS, it sends the ASP UP, 
> > and then it is notified by the SSNM.
> 
> There are ASP that need to receive NTFY for an AS but do not 
> need to receive SSNM for the AS.  In ASP-INACTIVE state, the 
> ASP receives NTFY, but does not receive SSNM.  If the ASP 
> wishes to receive SSNM, it moves to the ASP-ACTIVE state.
> 
> > SSNM reflect the state of a ss7 SP and is valid for all AS 
> of the same 
> > NA, so receiving a SSNM for one or several AS is the same thing.
> 
> Not true.  Two AS in the same NA (but different point codes, 
> not just different SI) can have different routing.  One AS 
> can be given DUNA because it is not permitted to access the 
> destination (has no route to the destination), the other 
> given DAVA because it is permitted to access the destination 
> (has a route to the destination). So, where multiple AS share 
> the same association, RC is required in the SSNM message to 
> indicate for which AS the SSNM message pertains. All RCs to 
> which the message pertains must be listed in the SSNM. None 
> of these RCs may be inactive.
> 

The state of a ss7 network destination is the SAME on a SG for all users
(remote AS).
DUNA/DAVA/DRST reflects the state of the PC.
DUPU reflects the state of a SI on a SP.

UNLESS special SG that locally allows an AS to use or not a destination,
this behavior means that a SG can see a ss7 destination but prohibits an
AS to use it! In that case, the SSNM do not reflects the state of a
destination but a management state on a SG for a couple (M3UA-AS <-> ss7
Destination)! Very tricky SG to manage.

> > If you receive a SSNM for one AS, it is valid for all AS of 
> the same 
> > NA.
> 
> That is not true.  See above.
> 
> > If you don't need to receive SSNM,,,,  don't send the ASP UP.
> 
> On the contrary, of you need to receive SSNM, send ASP Active.
> 
> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 08:41:02 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA17566
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 08:41:02 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16DlUV00662
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 08:47:30 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16Dl9p00647;
	Thu, 6 Feb 2003 08:47:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16DkZp00621
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 08:46:35 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA17515
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 08:39:35 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:0WmTP+DMy5eMI7Ja5P/5lxnWLei+7hiy@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h16DhED18904;
	Thu, 6 Feb 2003 06:43:14 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h16DhDG21466;
	Thu, 6 Feb 2003 06:43:13 -0700
Date: Thu, 6 Feb 2003 06:43:13 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Zannin, Francois" <Francois.Zannin@hp.com>
Cc: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206064313.B14908@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Zannin, Francois" <Francois.Zannin@hp.com>,
	"Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <06CAA4D3753C53408D2CE6340B68444214C411@vbeexc02.emea.cpqcorp.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <06CAA4D3753C53408D2CE6340B68444214C411@vbeexc02.emea.cpqcorp.net>; from Francois.Zannin@hp.com on Thu, Feb 06, 2003 at 02:30:15PM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Zannin,,

Zannin, Francois wrote:                                                            (Thu, 06 Feb 2003 14:30:15)
> 
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> > Sent: Thursday, February 06, 2003 2:01 PM
> > To: Zannin, Francois
> > Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> > Subject: Re: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > Zannin,,
> > 
> > Zannin, Francois wrote:                                       
> >                      (Thu, 06 Feb 2003 13:44:40)
> > > Bidolock,,
> > > 
> > > > -----Original Message-----
> > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > Sent: Thursday, February 06, 2003 1:35 PM
> > > > To: Zannin, Francois
> > > > Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> > > > Subject: Re: [Sigtran] M3UA - ASP database equalization
> > > > 
> > > > 
> > > > Zannin,,
> > > > 
> > > > So you see that there are ASP that do not want to receive
> > > > SSNM for inactive AS.  This is one reason why the SG cannot 
> > > > be required to send such useless SSNM to inactive ASPs.
> > > 
> > > If an ASP do not need to receive SSNM, it just do not send 
> > the ASP UP. 
> > > As soon as the ASP is ready to activate an AS, it sends the ASP UP, 
> > > and then it is notified by the SSNM.
> > 
> > There are ASP that need to receive NTFY for an AS but do not 
> > need to receive SSNM for the AS.  In ASP-INACTIVE state, the 
> > ASP receives NTFY, but does not receive SSNM.  If the ASP 
> > wishes to receive SSNM, it moves to the ASP-ACTIVE state.
> > 
> > > SSNM reflect the state of a ss7 SP and is valid for all AS 
> > of the same 
> > > NA, so receiving a SSNM for one or several AS is the same thing.
> > 
> > Not true.  Two AS in the same NA (but different point codes, 
> > not just different SI) can have different routing.  One AS 
> > can be given DUNA because it is not permitted to access the 
> > destination (has no route to the destination), the other 
> > given DAVA because it is permitted to access the destination 
> > (has a route to the destination). So, where multiple AS share 
> > the same association, RC is required in the SSNM message to 
> > indicate for which AS the SSNM message pertains. All RCs to 
> > which the message pertains must be listed in the SSNM. None 
> > of these RCs may be inactive.
> > 
> 
> The state of a ss7 network destination is the SAME on a SG for all users
> (remote AS).
> DUNA/DAVA/DRST reflects the state of the PC.
> DUPU reflects the state of a SI on a SP.

No it does not.  See the thread "RC in DATA and SNMM".  Several example
network configurations were presented where this does not hold.

> 
> UNLESS special SG that locally allows an AS to use or not a destination,

One AS can have one set of routing, another a different set of routing.

> this behavior means that a SG can see a ss7 destination but prohibits an
> AS to use it!

The SG uses one set of routes for one AS, and a different set of routes
for another AS.  This is not a radical concept.

> In that case, the SSNM do not reflects the state of a
> destination but a management state on a SG for a couple (M3UA-AS <-> ss7
> Destination)! Very tricky SG to manage.

Not tricky at all.  STPs do this all day long.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 08:45:58 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA17698
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 08:45:58 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16DqQ000835
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 08:52:26 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16Dq3p00822;
	Thu, 6 Feb 2003 08:52:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16Dpap00806
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 08:51:36 -0500
Received: from mx1.etsi.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA17671
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 08:44:34 -0500 (EST)
Received: from email10.etsihq.org ([172.27.1.62]) by mx1.etsi.org with Microsoft SMTPSVC(5.0.2195.5329);
	 Thu, 6 Feb 2003 14:49:31 +0100
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="US-ASCII"
Subject: RE: [Sigtran] M3UA - ASP database equalization
Date: Thu, 6 Feb 2003 14:49:31 +0100
Message-ID: <4091553999CBE4409CC2B562152B5A33EDDF5A@email10.etsihq.org>
Thread-Topic: [Sigtran] M3UA - ASP database equalization
Thread-Index: AcLN39zFj7ZcL0StTqi5FwZHIscd7QAAl9mQAACJbPA=
From: "Richard Adams" <Richard.Adams@etsi.org>
To: "Zannin, Francois" <Francois.Zannin@hp.com>, <bidulock@openss7.org>
Cc: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 06 Feb 2003 13:49:31.0973 (UTC) FILETIME=[93729F50:01C2CDE6]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h16Dpbp00807
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Francois,
According to Q.704, section 2.3.1 last clause, routing to a destination
can depend upon the SI (unfortunately!).
Most existing SS7 MTP implementations I know of consider just the DPC,
not the DPC+SI, when selecting
the linkset or combined linkset to use towards the destination, but they
COULD also consider
the SI. So, the routeing towards the destination could be different for
different SIs. And there
is no imperative to use as alternative routes those used by other SIs.
The routeset for
DPC "D" + SI "S"  might be different (i.e. have other routes) to that
for "D" + all other SIs than "S".
It's awkward, probably pathological, but exists.

-----Original Message-----
From: Zannin, Francois [mailto:Francois.Zannin@hp.com]
Sent: 06 February 2003 14:30
To: bidulock@openss7.org
Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
Subject: RE: [Sigtran] M3UA - ASP database equalization

<<<snip>>>
The state of a ss7 network destination is the SAME on a SG for all users
(remote AS).
[RAA] Not quite.
DUNA/DAVA/DRST reflects the state of the PC.
[RAA] almost, but not quite. It reflects the state of the routeset used
by the SG to get to the DPC. [RAA]
DUPU reflects the state of a SI on a SP.
[RAA] Agreed.

UNLESS special SG that locally allows an AS to use or not a destination,
this behavior means that a SG can see a ss7 destination but prohibits an
AS to use it! In that case, the SSNM do not reflects the state of a
destination but a management state on a SG for a couple (M3UA-AS <-> ss7
Destination)! Very tricky SG to manage.

[RAA] Agreed. And I would hope network operators wouldn't ask for this,
but 
I have seen some requests in the past! [RAA]
Kind regards,
Dick Adams,
richard.adams@etsi.org



_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 09:00:15 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA18028
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 09:00:15 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16E6iR01492
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 09:06:44 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16E6Hp01439;
	Thu, 6 Feb 2003 09:06:17 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16E5Pp01379
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 09:05:25 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA17932
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 08:58:24 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h16E23eW023952;
	Thu, 6 Feb 2003 08:02:03 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h16E1cM02575;
	Thu, 6 Feb 2003 08:01:38 -0600 (CST)
Date: Thu, 6 Feb 2003 08:01:38 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206064313.B14908@openss7.org>
Message-ID: <Pine.SOL.4.10.10302060759390.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hi,
All i want to say - here i have precised.

    SG1
 +------+               +------+
 | SGP1 |---------------| ASP1 |
 +------+               +------+
     |
     |                  +------+
     +------------------| ASP2 |
                        +------+
Mode : over-ride
ASP1 - ACTIVE
ASP2 - INACTIVE

Say initially ss7 destination status - TABLE1 at ASP1 and ASP2.

PC   STATUS    DIRECT-ROUTE  ALT-ROUTE ...
--------------------------------------
111 available   SG1           SG2      ...
222 available   SG1           SG2
333 available   SG1           SG2
444 available   SG1           SG2
555 available   SG1           SG2

Now ASP1 is handling traffice.
ASP1 gets DRST(222) and DUNA(333,444,555) from SGP1.

So the TABLE1 at ASP1 looks

PC   STATUS    ROUTE-ROUTE  ALT-ROUTE ...
-------------------------------------
111 available   SG1          SG2      ...
222 restrct     SG1          SG2
333 unavail     SG1          SG2
444 unavail     SG1          SG2
555 restrct     SG1          SG2

But ASP2 keeps the old TABLE1.

Say ASP1 goes down so SGP1 sends NTFY to ASP2.
ASP2 becomes ACTIVE.

1.ASP2 looks at TABLE1 which says all PCs are available So tries sending 
DATA(222) but actually PC 222 is restricted so it receives DRST(222).

2.ASP2 tries sending DATA(333) but it receives DUNA(333)

3.ASP2 tries sending DATA(444) but it receives DUNA(444)

4.ASP2 tries sending DATA(555) but it receives DRST(555).

Here my point is If ASP2 already had received
DRST(222),DUNA(333),DUNA(444)
and DRST(555) from SGP1 when ASP2 was in INACT state, It would have
directly selected ALR-ROUTE SG2 to send the traffice. It does not waste its
precisous time by sending DATA or DAUD to find the dest node
status(according to the old TABLE1).


--Samuel

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 09:22:23 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA18608
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 09:22:22 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16ESou03055
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 09:28:50 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16ESPp03031;
	Thu, 6 Feb 2003 09:28:26 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16EPfp02963
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 09:25:41 -0500
Received: from domino2.nc.tekelec.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA18506
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 09:18:42 -0500 (EST)
From: jeffrey.craig@tekelec.com
Received: from domino4.nc.tekelec.com ([198.89.40.9])
          by domino2.nc.tekelec.com (Lotus Domino Release 5.0.10)
          with ESMTP id 2003020609160498:21047 ;
          Thu, 6 Feb 2003 09:16:04 -0500 
Subject: Re: [Sigtran] M3UA - ASP database equalization -- allow inactive ASP to
 audit?
To: sigtran@ietf.org
Cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, bidulock@openss7.org
X-Mailer: Lotus Notes Release 5.0.7  March 21, 2001
Message-ID: <OF481930D9.C91B03A2-ON85256CC5.004B005C@nc.tekelec.com>
Date: Thu, 6 Feb 2003 09:18:05 -0500
MIME-Version: 1.0
X-MIMETrack: Serialize by Router on Domino4/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/06/2003 09:18:06 AM,
	Itemize by SMTP Server on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/06/2003 09:16:05 AM,
	Serialize by Router on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/06/2003 09:16:08 AM,
	Serialize complete at 02/06/2003 09:16:08 AM
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hello All,

I agree that the SG must not be *required* to send *unsolicited* SSNMs to
inactive ASPs.

An idea not considered in the thread so far, however, is *allowing*
(optional) inactive ASPs
to audit and *requiring* an SG to respond to audits from inactive ASPs.

In the case of the override traffic mode example where one has one active
ASP and
possibly many inactive ASPs, perhaps it would make fail-over more
efficient, if the
inactive ASPs were able to audit. With this mechanism they could keep their
(NA) network
status fairly current and not have to depend on a blast of audits upon
going active. This
idea, however, does not improve 'efficiency', if efficiency is defined to
be 'less messages
communicated', nor does it address the issue of RC, which could not be used
in this
case, since the ASPs are inactive.

Regards,

Jeff


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 09:45:09 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA19740
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 09:45:09 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16Epbb04727
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 09:51:37 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16EpEp04701;
	Thu, 6 Feb 2003 09:51:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16Eojp04681
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 09:50:45 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA19704
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 09:43:46 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:H3ex0gtgSM5QISV9mPPiqaNNBMTyUyS2@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h16ElND19794;
	Thu, 6 Feb 2003 07:47:23 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h16ElMs29671;
	Thu, 6 Feb 2003 07:47:22 -0700
Date: Thu, 6 Feb 2003 07:47:22 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206074722.A21518@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206064313.B14908@openss7.org> <Pine.SOL.4.10.10302060759390.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302060759390.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 08:01:38AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 08:01:38)
> 
> Hi,
> All i want to say - here i have precised.
> 
>     SG1
>  +------+               +------+
>  | SGP1 |---------------| ASP1 |
>  +------+               +------+
>      |
>      |                  +------+
>      +------------------| ASP2 |
>                         +------+
> Mode : over-ride
> ASP1 - ACTIVE
> ASP2 - INACTIVE
> 
> Say initially ss7 destination status - TABLE1 at ASP1 and ASP2.
> 
> PC   STATUS    DIRECT-ROUTE  ALT-ROUTE ...
> --------------------------------------
> 111 available   SG1           SG2      ...
> 222 available   SG1           SG2
> 333 available   SG1           SG2
> 444 available   SG1           SG2
> 555 available   SG1           SG2
> 
> Now ASP1 is handling traffice.
> ASP1 gets DRST(222) and DUNA(333,444,555) from SGP1.
> 
> So the TABLE1 at ASP1 looks
> 
> PC   STATUS    ROUTE-ROUTE  ALT-ROUTE ...
> -------------------------------------
> 111 available   SG1          SG2      ...
> 222 restrct     SG1          SG2
> 333 unavail     SG1          SG2
> 444 unavail     SG1          SG2
> 555 restrct     SG1          SG2
> 
> But ASP2 keeps the old TABLE1.
> 
> Say ASP1 goes down so SGP1 sends NTFY to ASP2.
> ASP2 becomes ACTIVE.
> 
> 1.ASP2 looks at TABLE1 which says all PCs are available So tries sending 
> DATA(222) but actually PC 222 is restricted so it receives DRST(222).
> 
> 2.ASP2 tries sending DATA(333) but it receives DUNA(333)
> 
> 3.ASP2 tries sending DATA(444) but it receives DUNA(444)
> 
> 4.ASP2 tries sending DATA(555) but it receives DRST(555).
> 
> Here my point is If ASP2 already had received
> DRST(222),DUNA(333),DUNA(444)
> and DRST(555) from SGP1 when ASP2 was in INACT state, It would have
> directly selected ALR-ROUTE SG2 to send the traffice. It does not waste its
> precisous time by sending DATA or DAUD to find the dest node
> status(according to the old TABLE1).

Answer me this simple question:

   What time is wasted sending DAUD?

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 09:58:02 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA20188
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 09:58:02 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16F4Uq05320
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 10:04:30 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16F47p05298;
	Thu, 6 Feb 2003 10:04:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16F3ep05266
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 10:03:40 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA20107
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 09:56:40 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:dGU9jKncLHR17Y5uFMKmALPkxdSSqy/h@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h16F06D19988;
	Thu, 6 Feb 2003 08:00:06 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h16F05x31247;
	Thu, 6 Feb 2003 08:00:05 -0700
Date: Thu, 6 Feb 2003 08:00:04 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: jeffrey.craig@tekelec.com
Cc: sigtran@ietf.org, "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>
Subject: Re: [Sigtran] M3UA - ASP database equalization -- allow inactive ASP to audit?
Message-ID: <20030206080004.B21518@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: jeffrey.craig@tekelec.com, sigtran@ietf.org,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>
References: <OF481930D9.C91B03A2-ON85256CC5.004B005C@nc.tekelec.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF481930D9.C91B03A2-ON85256CC5.004B005C@nc.tekelec.com>; from jeffrey.craig@tekelec.com on Thu, Feb 06, 2003 at 09:18:05AM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

jeffrey.craig,

jeffrey.craig@tekelec.com wrote:                                             (Thu, 06 Feb 2003 09:18:05)
> 
> Hello All,
> 
> I agree that the SG must not be *required* to send *unsolicited* SSNMs to
> inactive ASPs.
> 
> An idea not considered in the thread so far, however, is *allowing*
> (optional) inactive ASPs to audit and *requiring* an SG to respond to audits
> from inactive ASPs.

An SG is never "required" to respond to audits.  An SG is permitted
to respond with "Error - Destination Status Unavailable", or, for
that matter: "Error - Unexpected Message".

Besides, what is the difference of an audit performed one microsecond
before going active and one performed one microsecond after going active?

Sending SSNM to inactive ASPs serves no purpose.


> In the case of the override traffic mode example where one has one active
> ASP and possibly many inactive ASPs, perhaps it would make fail-over more
> efficient, if the inactive ASPs were able to audit.

How do you think it would make fail-over "more efficient"?

> With this mechanism they could keep their (NA) network status fairly current
> and not have to depend on a blast of audits upon going active.

What blast of audits?

> This idea, however, does not improve 'efficiency', if efficiency is defined
> to be 'less messages communicated', nor does it address the issue of RC,
> which could not be used in this case, since the ASPs are inactive.
> 

BTW, what has the newly active ASP in an override AS done about circuit
(or transaction) state?  The failover can hardly be smooth if the
new ASP has no circuit or transaction state from the previously
active ASP.  If the ASPs share circuit or transaction state, they can
easily share destination status.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 10:00:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA20373
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 10:00:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16F7Po05960
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 10:07:25 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16F7Cp05699;
	Thu, 6 Feb 2003 10:07:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16F6qp05428
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 10:06:52 -0500
Received: from pt.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA20296
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 09:59:51 -0500 (EST)
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id JAA28480;
	Thu, 6 Feb 2003 09:59:36 -0500
Message-ID: <3E427929.5000703@pt.com>
Date: Thu, 06 Feb 2003 10:03:05 -0500
From: Andrew Booth <abooth@pt.com>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.1) Gecko/20020826
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: bidulock@openss7.org
CC: Tom George <Tom.George@alcatel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M2PA 7 questions
References: <3E4179D6.8080502@pt.com> <3E419CE5.99B40D2@alcatel.com> <20030205183611.A28979@openss7.org> <3E41E209.9010104@pt.com> <20030205230118.A26810@openss7.org>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Thanks Brian,

That makes sense to me.

Andrew

Brian F. G. Bidulock wrote:

>Andrew,
>
>Andrew Booth wrote:                                                                 (Wed, 05 Feb 2003 23:18:17)
>  
>
>>Tom, Brian,
>>
>>Thanks for the comments.  More questions below.
>>
>>Brian F. G. Bidulock wrote:
>>
>>    
>>
>>>Tom,
>>>
>>>Tom George wrote:                                                                      (Wed, 05 Feb 2003 17:23:17)
>>> 
>>>
>>>      
>>>
>>>>Andrew,
>>>>
>>>>Thanks for the comments. Replies below...
>>>>
>>>>Andrew Booth wrote:
>>>>   
>>>>
>>>>        
>>>>
>>>>>Hi All,
>>>>>
>>>>>I was looking over M2PA draft 7 and I have a few questions.
>>>>>
>>>>>3. M2PA Link State Control
>>>>>SCTP Restart is not mentioned in the state transistion diagram for SCTP
>>>>>associations in Figure 8.  Is this an oversight, or intentional?  There
>>>>>don't seem to be procedures for SCTP Restart mentioned elsewhere either.
>>>>>     
>>>>>
>>>>>          
>>>>>
>>>>I don't see how SCTP restart would affect M2PA. 
>>>>
>>>>        
>>>>
>>Should SCTP RESTART be a transition event to OUT OF SERVICE in Figure 7?
>>I think the following scenario is possible.
>>
>>If M2PA peer A detects the loss of association before M2PA peer B and A 
>>immediately reestablishes the association, then the SCTP implementation 
>>at B could see this as an association restart instead of a lost 
>>association.  In this case (RFC 2960, section 5.2.4):
>>
>>    -  Any SCTP DATA Chunks MAY be retained (this is an implementation
>>         specific option).
>>
>>      -  A notification of RESTART SHOULD be sent to the ULP instead of
>>         a "COMMUNICATION LOST" notification.
>>     
>>     All the congestion control parameters (e.g., cwnd, ssthresh)
>>      related to this peer MUST be reset to their initial values (see
>>      Section 6.2.1).
>>
>>      After this the endpoint shall enter the ESTABLISHED state.
>>
>>
>>Assuming that a RESTART indication is a transition event to OOS:
>>Note that if the SCTP implementation does retain data chunks then there 
>>could still be a fair amount of stale data to retransmitted.  I don't 
>>think this stale data should cause big problems, but it could slow down 
>>the alignment after a restart.
>>    
>>
>
>The peer that gets the restart indication doesn't have to do anthing:
>Look at both sides:
>
>The side getting the restart indication, if it supports the optional
>procedure of RFC 2960, it will allow the M2PA to continue and deliver
>a restart indication; otherwise, it will typically disconnect the ULP
>and provide another init indication.  The latter fall into the existing
>procedures for loss of an association and starting an association.
>
>The side restarting either is capable of continuing where it left off
>(unlikely, it has closed the association and is reopening it: there is
>no reason to do so unless the M2PA has crashed and restarted).  If the
>side restarting is capable of continuing, it continues where it left off.
>If it is not capable of continuiing, it follows the procedures for starting
>a new association and sends alignment.
>
>(IMO the first thing that an M2PA peer should do when forming the
>association is send LS OOS.  This is reflected in in the test spec
>draft-bidulock-sigtran-m2pa-test-02.txt.  That way, if MTP3 has not
>issued the Start signal, the other end will know the state of the link
>immediately.)
>
>So, no procedures are required outside of the normal procedures.
>
>I agree with Tom: I don't see how SCTP restart would affect M2PA.
>
>  
>
>>>>   
>>>>
>>>>        
>>>>
>>>>>Timer T7
>>>>>========
>>>>>In section 4.2.1 it sounds like a per message timer, or even an interval
>>>>>     
>>>>>
>>>>>          
>>>>>
>>>>It is intended to be a per-message timer. 
>>>>
>>>>
>>>>   
>>>>
>>>>        
>>>>
>>>>>e.g.
>>>>>  Timer T7 provides an indication of excessive delay of
>>>>>  acknowledgement. If the following conditions are true:
>>>>>
>>>>>  (a) There is at least one message in the M2PA retransmit buffer.
>>>>>
>>>>>  (b) The remote M2PA is not in a Busy condition (i.e., the local
>>>>>      timer T6 is not running).
>>>>>
>>>>>  (c) There is a message in the M2PA retransmit buffer that has not
>>>>>      received an acknowledgement in the span of T7 since its last
>>>>>      transmission.
>>>>>
>>>>>  then M2PA SHOULD take the link out of service.
>>>>>
>>>>>The interpretation of T7 as an interval (or just a condition on
>>>>>messages) is reinforced by the fact there are no references to starting T7.
>>>>>
>>>>>On the other hand in section 4.1.4 there is a reference to stopping T7
>>>>>during processor outage.  In 4.1.5 there is the comment
>>>>>  "The peer M2PA SHALL NOT fail the link due to expiration of timer T7
>>>>>  excessive delay of acknowledgement (see section 4.2.1 Sending and
>>>>>  receiving messages) while its T6 timer is running, i.e., while the
>>>>>  other end is Busy."
>>>>>There is no comment on how to handle T7 after Busy clears.  Is it
>>>>>appropriate to immediately check the conditions from 4.2.1?
>>>>>     
>>>>>
>>>>>          
>>>>>
>>>>I see the problem. After Busy clears, you may have a some unacked
>>>>messages that have been waiting for longer than T7. It doesn't seem like
>>>>you would want to fail the link for that. So it looks like T7 should be
>>>>restarted when Busy clears.
>>>>   
>>>>
>>>>        
>>>>
>>>T7 continues to run while busy (but not while processor outage).  This
>>>is the same as Q.703.  T7 expiry just does not fail the link while busy,
>>>T6 expiry does.
>>> 
>>>
>>>      
>>>
>>Is it then appropriate to check the condition in 4.2.1 when Busy abates, 
>>to see if T7 has expired for any unacknowledged messages?  If so, is it 
>>worth stating explicitly?
>>    
>>
>
>As stated in 4.2.1, if T7 has expired for any messages during the busy
>condition, it does not take the link out of service.  T7 does not need
>to be "checked" when the busy condition ceases: if T7 expires and there
>is no busy condition the link should be taken out of service as proscribed
>in section 4.2.1.
>
>--brian
>
>  
>

-- 
Andrew Booth, Software Developer
Signaling Systems Group, Performance Technologies
613-237-4344

Visit www.pt.com/mailing.html to subscribe to our Packets & Signals eLetter.


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 12:30:00 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA27031
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 12:30:00 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16HaXo17656
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 12:36:33 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16HaEp17596;
	Thu, 6 Feb 2003 12:36:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16HZMp17532
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 12:35:22 -0500
Received: from pt.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA26949
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 12:28:18 -0500 (EST)
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id MAA11584
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 12:28:11 -0500
Message-ID: <3E429BFD.70407@pt.com>
Date: Thu, 06 Feb 2003 12:31:41 -0500
From: Andrew Booth <abooth@pt.com>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.1) Gecko/20020826
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: sigtran@ietf.org
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] More M2PA 7 questions
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Hi All,

A few more questions on M2PA draft 7.

1) Does every User Data received from the peer increases the expected FSN
for the subsequent message, even in cases where the received message is 
later
discarded as part of a Flush Buffers or Changeover procedure?

2) Suppose that an M2PA implementation has received messages from a peer
and that some of the messages are buffered at the M2PA layer (perhaps it is
in processor outage).  Suppose now the link transitions to OOS and the
MTP3 sends a Retrieve BSNT to M2PA.

According to 4.2.6,

   "M2PA determines the Forward Sequence Number of the last User Data
   message received from the peer.  This is the BSNT."

In the M2PA situation it seems more appropriate to respond with the FSN
of the last message which was delivered to MTP3.
Am I missing something?

3) How should T7 be handled when remote processor outage abates?  Should 
we immediately check for excessive ack delay, or reset T7 somehow?

4) In section 5.1 we transition to proving without starting T3 and to 
aligned ready without receiving LS Proving.  Are these omissions 
intentional?

5) In section 5.8 there is some introductory text :

"Note that in this example, the sequence numbers and messages requested
by MTP3 are sent from SCTP to M2PA in the Communication Lost
primitive. In general, the retrieval of sequence numbers and messages
is implementation dependent."

Why are messages retrieved from SCTP?  Is this text indicating that the 
M2PA retransmit buffer is implemented in the SCTP?  Just a little confused.

6) In section 4.2.6, is the following correct or should it be rephrased 
in the context of a circular buffer?
"any transmitted User Data messages beginning with the first
       unacknowledged message with FSN greater than FSNC." 


Thanks for any info,
Andrew

-- 
Andrew Booth, Software Developer
Signaling Systems Group, Performance Technologies
613-237-4344

Visit www.pt.com/mailing.html to subscribe to our Packets & Signals eLetter.


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 17:04:20 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA06406
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 17:04:20 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h16MAw003499
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 17:10:58 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16MAXp03476;
	Thu, 6 Feb 2003 17:10:34 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h168jLp12260
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 03:45:21 -0500
Received: from mx1.etsi.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08552
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 03:38:28 -0500 (EST)
Received: from email10.etsihq.org ([172.27.1.62]) by mx1.etsi.org with Microsoft SMTPSVC(5.0.2195.5329);
	 Thu, 6 Feb 2003 09:41:41 +0100
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="US-ASCII"
Subject: RE: [Sigtran] M3UA - ASP database equalization
Date: Thu, 6 Feb 2003 09:41:41 +0100
Message-ID: <4091553999CBE4409CC2B562152B5A33EDDF58@email10.etsihq.org>
Thread-Topic: [Sigtran] M3UA - ASP database equalization
Thread-Index: AcLNuoa2un8p+B+2T4m/Onafq5XCNQAAD0Mw
From: "Richard Adams" <Richard.Adams@etsi.org>
To: <bidulock@openss7.org>, "Zannin, Francois" <Francois.Zannin@hp.com>
Cc: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 06 Feb 2003 08:41:41.0492 (UTC) FILETIME=[922EE340:01C2CDBB]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h168jMp12261
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Brian,
I agree with you. I think when the ASP becomes active, it should do an
audit to update its knowledge of the SS7 network. The ASP needs to take
the initiative to find out about the SS7 nodes etc. in which it is
interested. Any other method of working would lead to impossible loading
on the SG. 
Of course, I hope 199,999 of your inactive ASPs do not decide to become
active at the same time.....! 
Kind regards,
Dick Adams,
richard.adams@etsi.org

 

-----Original Message-----
From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
Sent: 06 February 2003 09:15
To: Zannin, Francois
Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization


Zannin,,

That is contrary to the conensus of the RFC.

Consider 1 ASP active, 200,000 ASPs inactive.

The RFC cannot require an SG to send SSNM to an inactive ASP.

--brian

Zannin, Francois wrote:
(Thu, 06 Feb 2003 09:02:41)
> Samuel,
> 
> IMO, it's a good idea to inform inactive (BUT UP) ASP of the state of
> the ss7 network especially in the override mode where an ASP can be
> ready to recover an AS as soon as it receive an AS-Pending
notification
> from the SGp. In this condition, the ASP is ready to operate quickly
and
> minimize the risk of buffered message lost during the pending timer.
> 
> If someone does not want to receive SSNM because it is not able to
> recover an AS, it just stays in an ASP-Down state.
> 
> Because SSNM reflects ss7 Point Code state that is not "SI" dependant,
> it affects all AS for a network and then, unless SG that plays
multiple
> networks, the ASP UP is sufficient to filter the undesired SSNM.
> 
> Francois.
> 
> -----Original Message-----
> From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> Sent: Thursday, February 06, 2003 8:35 AM
> To: Nagar, Ashish
> Cc: sigtran@ietf.org
> Subject: RE: [Sigtran] M3UA - ASP database equalization
> 
> 
> 
> Ashish,
> 
> But when inactive ASP changes state to active, first of all
> it does not know whether to send DAUD or not.
> 
> It will try to send traffic - but actually destination is unavailable.
> 
> The worst case is when SG triggers restart procedures this case all
the
> SS7 destinations are unavailable via restarted-SG.But this poor ASP
> (newly come-up ASP) keeps sending traffic via restarted-SG to reach
the
> SS7 destination and for every message ASP receives DUNA.
> 
> Are we not doing double-work here?
> 
> --Samuel
> 
> 
>   Alcatel USA, Inc.                  Internet:
> <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc.
> *******
> 
> On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> 
> > Hi Samuel
> > 
> > My point is why to do any processing if ASP is in Inactive state. 
> > Think about ASP is in Inactive state for longer period and receiving

> > lots of DUNA and DAVA, while what matter is the present only when
ASP 
> > state changes to ASP-ACTIVE.
> > 
> > Thanks & Regards
> > Ashish Nagar
> > 
> > 
> > -----Original Message-----
> > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > Sent: Thursday, February 06, 2003 12:30 PM
> > To: Nagar, Ashish
> > Cc: sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > Ashish,
> > 
> > My point is if SG is able to send DUNA to all inact and active ASPs,

> > we should not restrict it.
> > 
> > Sending DAUD is an alternative way ,If SG has direct way of sending 
> > DUNA then Why do you need alternative way?
> > 
> > --Samuel
> > 
> >   Alcatel USA, Inc.                  Internet:
> <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. 
> > *******
> > 
> > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > 
> > > You can generate DAUD procedure at any time, so whenever ASP
changes
> 
> > > state from Inactive to Active Just generate DAUD.
> > > 
> > > Thanks & Regards
> > > Ashish Nagar
> > > 
> > > 
> > > -----Original Message-----
> > > From: Samuel Dur D. Jeyaseelan
[mailto:sjeyasee@ssd.usa.alcatel.com]
> > > Sent: Thursday, February 06, 2003 11:27 AM
> > > To: sigtran@ietf.org
> > > Subject: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > Hi all,
> > > 
> > > Currently RFC 3332 says that SG sends SSNM messages to all 
> > > 'concerned ASPs'(means only ACTIVE ASPs).
> > > 
> > > +----+           +---+           +-----+
> > > |SEP | ----------|SG |-----------| ASP1|
> > > +----+           +---+           +-----+
> > >                    |             
> > >                    |             +-----+
> > >                    +-------------| ASP2|
> > >                                  +-----+
> > > 
> > > (ASP1 is active and ASP2 is incative - override mode)
> > > 
> > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > But SEP status is not changed in ASP2.There are two ways to solve 
> > > this problem
> > > 
> > > 1. ASP1 can equalize its database with ASP2
> > > or
> > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > 
> > > IMO, SG can send SSNM messages to ASP (either ASP is in active or 
> > > inact
> > > state)
> > > 
> > > Any comments?
> > > 
> > > -----
> > > 
> > > section 5.4.2.1
> > > 
> > > ...
> > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> > >         concerned Network Appearance.
> > > 
> > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER
request
> 
> > > is sent to MTP3.
> > > 
> > > But there is no primitive availabe between M3UA and MTP3.
> > > 
> > > I think it should be corrected.
> > > 
> > > ----
> > > Presently ASP is unaware of SG's MTP3 restart,
> > > ASP receives only DUNA for all SS7 point codes which can be
> > > reachable via that SG.But in case of peer to peer protocol
> > > SG and AS point codes are different, So why dont we send
> > > DUNA with SG's point code to ASP? So that just by getting
> > > single DUNA(with SG's PC) ASP will not route any message via that
> SG.
> > > This can avoid sending multiple DUNAs to ASP.
> > > 
> > > 
> > > Any comments?
> > > 
> > > -----
> > > section 1.4.1
> > > 
> > > ...
> > >    Where a particular AS can be reached via more than one SGP, the
> > >    corresponding Routing Keys in the SGPs should be identical.
> (Note:
> > >    It is possible for the SGP Routing Key configuration data to be
> > >    temporarily out-of-sync during configuration updates).
> > >  
> > >                               +--------+
> > >                               |        |
> > >                  +------------+  SG 1  +--------------+
> > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > >      |  SEP  +---+            +--------+              +---/      \
> > >      |   or  |                    |*                      | ASPs
|
> > >      |  STP  +---+            +--------+              +---\      /
> > >      +-------+   |            |        |              |     ----
> > >                  +------------+  SG 2  +--------------+
> > >                               | "STP"  |
> > >                               +--------+
> > >  
> > >                     Figure 1  Example with mated SGs
> > > 
> > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are 
> > > located geographically in different places.The ideas is when SG1 
> > > goes down in CITY1(due to earthquake...) , all the traffice still 
> > > can be handled by SG2 which is in CITY2.
> > > 
> > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will
be 
> > > routed from SGP1 to SGP2.
> > > 
> > > Does all these things mean to say that SGP1's mate 'should be' 
> > > present in SG2 and not in SG1 itself ?
> > >  
> > > 
> > > Comments are welcome.
> > > 
> > > --Samuel
> > > 
> > > 
> > > 
> > > 
> > >   Alcatel USA, Inc.                  Internet:
> > <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA,
Inc.
> > *******
> > > 
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> > 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 23:13:04 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA15758
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 23:13:04 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h174JnT24016
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 23:19:49 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h174JUp24005;
	Thu, 6 Feb 2003 23:19:31 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h174HNp23961
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 23:17:23 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA15707
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 23:10:07 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h174Dhqg021942;
	Thu, 6 Feb 2003 22:13:44 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h174DHM28920;
	Thu, 6 Feb 2003 22:13:17 -0600 (CST)
Date: Thu, 6 Feb 2003 22:13:17 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206074722.A21518@openss7.org>
Message-ID: <Pine.SOL.4.10.10302062148450.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,
(Sorry i did not see your last mail, just today morning saw it)

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 08:01:38)
> > 
> > Hi,
> > All i want to say - here i have precised.
> > 
> >     SG1
> >  +------+               +------+
> >  | SGP1 |---------------| ASP1 |
> >  +------+               +------+
> >      |
> >      |                  +------+
> >      +------------------| ASP2 |
> >                         +------+
> > Mode : over-ride
> > ASP1 - ACTIVE
> > ASP2 - INACTIVE
> > 
> > Say initially ss7 destination status - TABLE1 at ASP1 and ASP2.
> > 
> > PC   STATUS    DIRECT-ROUTE  ALT-ROUTE ...
> > --------------------------------------
> > 111 available   SG1           SG2      ...
> > 222 available   SG1           SG2
> > 333 available   SG1           SG2
> > 444 available   SG1           SG2
> > 555 available   SG1           SG2
> > 
> > Now ASP1 is handling traffice.
> > ASP1 gets DRST(222) and DUNA(333,444,555) from SGP1.
> > 
> > So the TABLE1 at ASP1 looks
> > 
> > PC   STATUS    ROUTE-ROUTE  ALT-ROUTE ...
> > -------------------------------------
> > 111 available   SG1          SG2      ...
> > 222 restrct     SG1          SG2
> > 333 unavail     SG1          SG2
> > 444 unavail     SG1          SG2
> > 555 restrct     SG1          SG2
> > 
> > But ASP2 keeps the old TABLE1.
> > 
> > Say ASP1 goes down so SGP1 sends NTFY to ASP2.
> > ASP2 becomes ACTIVE.
> > 
> > 1.ASP2 looks at TABLE1 which says all PCs are available So tries sending 
> > DATA(222) but actually PC 222 is restricted so it receives DRST(222).
> > 
> > 2.ASP2 tries sending DATA(333) but it receives DUNA(333)
> > 
> > 3.ASP2 tries sending DATA(444) but it receives DUNA(444)
> > 
> > 4.ASP2 tries sending DATA(555) but it receives DRST(555).
> > 
> > Here my point is If ASP2 already had received
> > DRST(222),DUNA(333),DUNA(444)
> > and DRST(555) from SGP1 when ASP2 was in INACT state, It would have
> > directly selected ALR-ROUTE SG2 to send the traffice. It does not waste its
> > precisous time by sending DATA or DAUD to find the dest node
> > status(according to the old TABLE1).
> 
> Answer me this simple question:
> 
>    What time is wasted sending DAUD?

1.First of all sending DAUD is unnecessary.
  (RFC says newly ACTIVE ASP MAY send DAUD and it never said "MUST")

2.In case if newly ACTIVE ASP sends DATA(222), definitely it will receive
DRST(222) - Here these two messages (DATA and DRST) are un-necessary.

ALL THESE UNNECESSARY MESSAGE COMMUNICATION CAN BE AVOIDED IF SGP HAD SENT
SSNM MESSAGES WHEN ASP2 WAS IN INACTIVE STATE.

SENDING SSNM MESSAGES TO INACTIVE ASP IS A DIRECT METHOD OF UPDATING
DESTINATION STATUS OF SS7 NODES AND SGP IS ALSO ABLE TO SEND SSNM TO 
INACTIVE ASP.

IMO IT WOULD BE GOOD IF WE ALLOW INACTIVE APS TO RECEIVE SSNM MESSAGES.

--Samuel

> 
> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 23:21:35 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA15931
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 23:21:35 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h174SLq24346
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 23:28:21 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h174S5p24317;
	Thu, 6 Feb 2003 23:28:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h174RRp24283
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 23:27:27 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA15902
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 23:20:10 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h174NhJT021315;
	Thu, 6 Feb 2003 22:23:43 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h174NCM02358;
	Thu, 6 Feb 2003 22:23:12 -0600 (CST)
Date: Thu, 6 Feb 2003 22:23:12 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: Richard Adams <Richard.Adams@etsi.org>
cc: bidulock@openss7.org, "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: RE: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <4091553999CBE4409CC2B562152B5A33EDDF58@email10.etsihq.org>
Message-ID: <Pine.SOL.4.10.10302062217200.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Richard,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Richard Adams wrote:

> Brian,
> I agree with you. I think when the ASP becomes active, it should do an
> audit to update its knowledge of the SS7 network. The ASP needs to take
> the initiative to find out about the SS7 nodes etc. in which it is
> interested. Any other method of working would lead to impossible loading
> on the SG.

Yes i agree but your suggestion is an indirect method.
I am proposing a direct method to send SSNM messages from SGP to INACTIVE
ASP. When SGP sends DUNA or DRST to ACTVIE ASP, let him also send a copy
of the same message to INACTIVE ASP. I dont find any difficulty here.

 
> Of course, I hope 199,999 of your inactive ASPs do not decide to become
> active at the same time.....! 

Theoritically your example sounds good but practically NOBODY will
configure 1 ASP ACTIVE and 199,999 ASPs INACTIVE.

--Samuel

> Kind regards,
> Dick Adams,
> richard.adams@etsi.org
> 
>  
> 
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: 06 February 2003 09:15
> To: Zannin, Francois
> Cc: Samuel Dur D. Jeyaseelan; Nagar, Ashish; sigtran@ietf.org
> Subject: Re: [Sigtran] M3UA - ASP database equalization
> 
> 
> Zannin,,
> 
> That is contrary to the conensus of the RFC.
> 
> Consider 1 ASP active, 200,000 ASPs inactive.
> 
> The RFC cannot require an SG to send SSNM to an inactive ASP.
> 
> --brian
> 
> Zannin, Francois wrote:
> (Thu, 06 Feb 2003 09:02:41)
> > Samuel,
> > 
> > IMO, it's a good idea to inform inactive (BUT UP) ASP of the state of
> > the ss7 network especially in the override mode where an ASP can be
> > ready to recover an AS as soon as it receive an AS-Pending
> notification
> > from the SGp. In this condition, the ASP is ready to operate quickly
> and
> > minimize the risk of buffered message lost during the pending timer.
> > 
> > If someone does not want to receive SSNM because it is not able to
> > recover an AS, it just stays in an ASP-Down state.
> > 
> > Because SSNM reflects ss7 Point Code state that is not "SI" dependant,
> > it affects all AS for a network and then, unless SG that plays
> multiple
> > networks, the ASP UP is sufficient to filter the undesired SSNM.
> > 
> > Francois.
> > 
> > -----Original Message-----
> > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> > Sent: Thursday, February 06, 2003 8:35 AM
> > To: Nagar, Ashish
> > Cc: sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > 
> > 
> > 
> > Ashish,
> > 
> > But when inactive ASP changes state to active, first of all
> > it does not know whether to send DAUD or not.
> > 
> > It will try to send traffic - but actually destination is unavailable.
> > 
> > The worst case is when SG triggers restart procedures this case all
> the
> > SS7 destinations are unavailable via restarted-SG.But this poor ASP
> > (newly come-up ASP) keeps sending traffic via restarted-SG to reach
> the
> > SS7 destination and for every message ASP receives DUNA.
> > 
> > Are we not doing double-work here?
> > 
> > --Samuel
> > 
> > 
> >   Alcatel USA, Inc.                  Internet:
> > <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > *******
> > 
> > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > 
> > > Hi Samuel
> > > 
> > > My point is why to do any processing if ASP is in Inactive state. 
> > > Think about ASP is in Inactive state for longer period and receiving
> 
> > > lots of DUNA and DAVA, while what matter is the present only when
> ASP 
> > > state changes to ASP-ACTIVE.
> > > 
> > > Thanks & Regards
> > > Ashish Nagar
> > > 
> > > 
> > > -----Original Message-----
> > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > Sent: Thursday, February 06, 2003 12:30 PM
> > > To: Nagar, Ashish
> > > Cc: sigtran@ietf.org
> > > Subject: RE: [Sigtran] M3UA - ASP database equalization
> > > 
> > > 
> > > Ashish,
> > > 
> > > My point is if SG is able to send DUNA to all inact and active ASPs,
> 
> > > we should not restrict it.
> > > 
> > > Sending DAUD is an alternative way ,If SG has direct way of sending 
> > > DUNA then Why do you need alternative way?
> > > 
> > > --Samuel
> > > 
> > >   Alcatel USA, Inc.                  Internet:
> > <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. 
> > > *******
> > > 
> > > On Wed, 5 Feb 2003, Nagar, Ashish wrote:
> > > 
> > > > You can generate DAUD procedure at any time, so whenever ASP
> changes
> > 
> > > > state from Inactive to Active Just generate DAUD.
> > > > 
> > > > Thanks & Regards
> > > > Ashish Nagar
> > > > 
> > > > 
> > > > -----Original Message-----
> > > > From: Samuel Dur D. Jeyaseelan
> [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > Sent: Thursday, February 06, 2003 11:27 AM
> > > > To: sigtran@ietf.org
> > > > Subject: [Sigtran] M3UA - ASP database equalization
> > > > 
> > > > 
> > > > Hi all,
> > > > 
> > > > Currently RFC 3332 says that SG sends SSNM messages to all 
> > > > 'concerned ASPs'(means only ACTIVE ASPs).
> > > > 
> > > > +----+           +---+           +-----+
> > > > |SEP | ----------|SG |-----------| ASP1|
> > > > +----+           +---+           +-----+
> > > >                    |             
> > > >                    |             +-----+
> > > >                    +-------------| ASP2|
> > > >                                  +-----+
> > > > 
> > > > (ASP1 is active and ASP2 is incative - override mode)
> > > > 
> > > > When SEP is getting restarted ,SG sends DUNA to ASP1.
> > > > But SEP status is not changed in ASP2.There are two ways to solve 
> > > > this problem
> > > > 
> > > > 1. ASP1 can equalize its database with ASP2
> > > > or
> > > > 2.SG itself can send DUNA... to ASP2 (eventhough ASP2 is inactive)
> > > > 
> > > > IMO, SG can send SSNM messages to ASP (either ASP is in active or 
> > > > inact
> > > > state)
> > > > 
> > > > Any comments?
> > > > 
> > > > -----
> > > > 
> > > > section 5.4.2.1
> > > > 
> > > > ...
> > > >       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
> > > >         concerned Network Appearance.
> > > > 
> > > > It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER
> request
> > 
> > > > is sent to MTP3.
> > > > 
> > > > But there is no primitive availabe between M3UA and MTP3.
> > > > 
> > > > I think it should be corrected.
> > > > 
> > > > ----
> > > > Presently ASP is unaware of SG's MTP3 restart,
> > > > ASP receives only DUNA for all SS7 point codes which can be
> > > > reachable via that SG.But in case of peer to peer protocol
> > > > SG and AS point codes are different, So why dont we send
> > > > DUNA with SG's point code to ASP? So that just by getting
> > > > single DUNA(with SG's PC) ASP will not route any message via that
> > SG.
> > > > This can avoid sending multiple DUNAs to ASP.
> > > > 
> > > > 
> > > > Any comments?
> > > > 
> > > > -----
> > > > section 1.4.1
> > > > 
> > > > ...
> > > >    Where a particular AS can be reached via more than one SGP, the
> > > >    corresponding Routing Keys in the SGPs should be identical.
> > (Note:
> > > >    It is possible for the SGP Routing Key configuration data to be
> > > >    temporarily out-of-sync during configuration updates).
> > > >  
> > > >                               +--------+
> > > >                               |        |
> > > >                  +------------+  SG 1  +--------------+
> > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > >      |  SEP  +---+            +--------+              +---/      \
> > > >      |   or  |                    |*                      | ASPs
> |
> > > >      |  STP  +---+            +--------+              +---\      /
> > > >      +-------+   |            |        |              |     ----
> > > >                  +------------+  SG 2  +--------------+
> > > >                               | "STP"  |
> > > >                               +--------+
> > > >  
> > > >                     Figure 1  Example with mated SGs
> > > > 
> > > > Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are 
> > > > located geographically in different places.The ideas is when SG1 
> > > > goes down in CITY1(due to earthquake...) , all the traffice still 
> > > > can be handled by SG2 which is in CITY2.
> > > > 
> > > > SGP2 is mate for SGP1, On failure of SGP1, all the messages will
> be 
> > > > routed from SGP1 to SGP2.
> > > > 
> > > > Does all these things mean to say that SGP1's mate 'should be' 
> > > > present in SG2 and not in SG1 itself ?
> > > >  
> > > > 
> > > > Comments are welcome.
> > > > 
> > > > --Samuel
> > > > 
> > > > 
> > > > 
> > > > 
> > > >   Alcatel USA, Inc.                  Internet:
> > > <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA,
> Inc.
> > > *******
> > > > 
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > 
> > > 
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 23:21:38 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA15945
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 23:21:38 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h174SO324359
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 23:28:24 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h174S7p24333;
	Thu, 6 Feb 2003 23:28:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h174Rep24296
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 23:27:40 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA15907
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 23:20:23 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:oUx1KRHrNAfgVDKek6W2M8RqnOzWjc8t@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h174O1D31394;
	Thu, 6 Feb 2003 21:24:02 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h174O1T05460;
	Thu, 6 Feb 2003 21:24:01 -0700
Date: Thu, 6 Feb 2003 21:24:01 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206212401.A5070@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206074722.A21518@openss7.org> <Pine.SOL.4.10.10302062148450.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302062148450.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 10:13:17PM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 22:13:17)
> > 
> > Answer me this simple question:
> > 
> >    What time is wasted sending DAUD?
> 
> 1.First of all sending DAUD is unnecessary.
>   (RFC says newly ACTIVE ASP MAY send DAUD and it never said "MUST")
> 
> 2.In case if newly ACTIVE ASP sends DATA(222), definitely it will receive
> DRST(222) - Here these two messages (DATA and DRST) are un-necessary.
> 
> ALL THESE UNNECESSARY MESSAGE COMMUNICATION CAN BE AVOIDED IF SGP HAD SENT
> SSNM MESSAGES WHEN ASP2 WAS IN INACTIVE STATE.
> 
> SENDING SSNM MESSAGES TO INACTIVE ASP IS A DIRECT METHOD OF UPDATING
> DESTINATION STATUS OF SS7 NODES AND SGP IS ALSO ABLE TO SEND SSNM TO 
> INACTIVE ASP.
> 
> IMO IT WOULD BE GOOD IF WE ALLOW INACTIVE APS TO RECEIVE SSNM MESSAGES.
> 

You didn't answer the question:

	What time is wasted sending DAUD?

I suppose that you cannot answer the question, because no time
is wasted sending DAUD.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 23:22:31 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA16018
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 23:22:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h174TG824456
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 23:29:16 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h174T6p24441;
	Thu, 6 Feb 2003 23:29:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h174Stp24396
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 23:28:55 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA15947
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 23:21:39 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:xu1L/6v0Y2PlWsbZ0PdYYxZ6WypZB2kG@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h174PHD31419;
	Thu, 6 Feb 2003 21:25:17 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h174PHM05643;
	Thu, 6 Feb 2003 21:25:17 -0700
Date: Thu, 6 Feb 2003 21:25:17 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: Richard Adams <Richard.Adams@etsi.org>,
        "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206212517.B5070@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	Richard Adams <Richard.Adams@etsi.org>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <4091553999CBE4409CC2B562152B5A33EDDF58@email10.etsihq.org> <Pine.SOL.4.10.10302062217200.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302062217200.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 10:23:12PM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

The RFC does not restrict the number of inactive ASPs.

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 22:23:12)
> 
> Richard,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Richard Adams wrote:
> 
> > Brian,
> > I agree with you. I think when the ASP becomes active, it should do an
> > audit to update its knowledge of the SS7 network. The ASP needs to take
> > the initiative to find out about the SS7 nodes etc. in which it is
> > interested. Any other method of working would lead to impossible loading
> > on the SG.
> 
> Yes i agree but your suggestion is an indirect method.
> I am proposing a direct method to send SSNM messages from SGP to INACTIVE
> ASP. When SGP sends DUNA or DRST to ACTVIE ASP, let him also send a copy
> of the same message to INACTIVE ASP. I dont find any difficulty here.
> 
>  
> > Of course, I hope 199,999 of your inactive ASPs do not decide to become
> > active at the same time.....! 
> 
> Theoritically your example sounds good but practically NOBODY will
> configure 1 ASP ACTIVE and 199,999 ASPs INACTIVE.
> 
> --Samuel
> 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 23:36:34 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA16356
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 23:36:34 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h174hK325585
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 23:43:20 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h174h8p25577;
	Thu, 6 Feb 2003 23:43:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h174gap25556
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 23:42:36 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA16324
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 23:35:19 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h174ctJV022327;
	Thu, 6 Feb 2003 22:38:57 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h174bwM07270;
	Thu, 6 Feb 2003 22:37:58 -0600 (CST)
Date: Thu, 6 Feb 2003 22:37:58 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206212401.A5070@openss7.org>
Message-ID: <Pine.SOL.4.10.10302062230540.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 22:13:17)
> > > 
> > > Answer me this simple question:
> > > 
> > >    What time is wasted sending DAUD?
> > 
> > 1.First of all sending DAUD is unnecessary.
> >   (RFC says newly ACTIVE ASP MAY send DAUD and it never said "MUST")
> > 
> > 2.In case if newly ACTIVE ASP sends DATA(222), definitely it will receive
> > DRST(222) - Here these two messages (DATA and DRST) are un-necessary.
> > 
> > ALL THESE UNNECESSARY MESSAGE COMMUNICATION CAN BE AVOIDED IF SGP HAD SENT
> > SSNM MESSAGES WHEN ASP2 WAS IN INACTIVE STATE.
> > 
> > SENDING SSNM MESSAGES TO INACTIVE ASP IS A DIRECT METHOD OF UPDATING
> > DESTINATION STATUS OF SS7 NODES AND SGP IS ALSO ABLE TO SEND SSNM TO 
> > INACTIVE ASP.
> > 
> > IMO IT WOULD BE GOOD IF WE ALLOW INACTIVE APS TO RECEIVE SSNM MESSAGES.
> > 
> 
> You didn't answer the question:
> 
> 	What time is wasted sending DAUD?
> 
> I suppose that you cannot answer the question, because no time
> is wasted sending DAUD.
> 
To prepare a DAUD message , ASP has to read BIG ss7 dest status table and
he has to send to SGP, RUN A TIMER AND WAIT FOR RESPONSE.

Another case if ASP has registered with a DPC and empty OPC list(means any
SS7 node can reach ASP via SGP). This case with what point code DAUD will
be sent from ASP?

--Samuel
 > --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 23:38:29 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA16419
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 23:38:29 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h174jEl25658
	for sigtran-archive@odin.ietf.org; Thu, 6 Feb 2003 23:45:14 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h174j3p25649;
	Thu, 6 Feb 2003 23:45:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h174icp25622
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 23:44:38 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA16387
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 23:37:21 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h174etJX022509;
	Thu, 6 Feb 2003 22:40:57 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h174emM08209;
	Thu, 6 Feb 2003 22:40:48 -0600 (CST)
Date: Thu, 6 Feb 2003 22:40:48 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: Richard Adams <Richard.Adams@etsi.org>,
        "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206212517.B5070@openss7.org>
Message-ID: <Pine.SOL.4.10.10302062238140.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

Ok, tell me, where you will configure 1 ASP ACT and 199,000 ASPs INACT?
I want to know which network does this poor configuration.

--Samuel


  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> The RFC does not restrict the number of inactive ASPs.
> 
> --brian
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 22:23:12)
> > 
> > Richard,
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Thu, 6 Feb 2003, Richard Adams wrote:
> > 
> > > Brian,
> > > I agree with you. I think when the ASP becomes active, it should do an
> > > audit to update its knowledge of the SS7 network. The ASP needs to take
> > > the initiative to find out about the SS7 nodes etc. in which it is
> > > interested. Any other method of working would lead to impossible loading
> > > on the SG.
> > 
> > Yes i agree but your suggestion is an indirect method.
> > I am proposing a direct method to send SSNM messages from SGP to INACTIVE
> > ASP. When SGP sends DUNA or DRST to ACTVIE ASP, let him also send a copy
> > of the same message to INACTIVE ASP. I dont find any difficulty here.
> > 
> >  
> > > Of course, I hope 199,999 of your inactive ASPs do not decide to become
> > > active at the same time.....! 
> > 
> > Theoritically your example sounds good but practically NOBODY will
> > configure 1 ASP ACTIVE and 199,999 ASPs INACTIVE.
> > 
> > --Samuel
> > 
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb  6 23:59:36 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA17068
	for <sigtran-archive@odin.ietf.org>; Thu, 6 Feb 2003 23:59:36 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1756N326541
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 00:06:23 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1756Bp26534;
	Fri, 7 Feb 2003 00:06:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1755gp26505
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 00:05:42 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA17035
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 23:58:25 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:Lsd2pAgNjJIM9iTZxqVORSHQGwuWtgyW@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h17523D32076;
	Thu, 6 Feb 2003 22:02:03 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h17522T10357;
	Thu, 6 Feb 2003 22:02:02 -0700
Date: Thu, 6 Feb 2003 22:02:02 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: Richard Adams <Richard.Adams@etsi.org>,
        "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206220202.A10248@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	Richard Adams <Richard.Adams@etsi.org>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206212517.B5070@openss7.org> <Pine.SOL.4.10.10302062238140.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302062238140.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 10:40:48PM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

The RFC does not restrict the number of inactive ASPs.

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 22:40:48)
> 
> Brian,
> 
> Ok, tell me, where you will configure 1 ASP ACT and 199,000 ASPs INACT?
> I want to know which network does this poor configuration.
> 
> --Samuel
> 
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > The RFC does not restrict the number of inactive ASPs.
> > 
> > --brian
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 22:23:12)
> > > 
> > > Richard,
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Thu, 6 Feb 2003, Richard Adams wrote:
> > > 
> > > > Brian,
> > > > I agree with you. I think when the ASP becomes active, it should do an
> > > > audit to update its knowledge of the SS7 network. The ASP needs to take
> > > > the initiative to find out about the SS7 nodes etc. in which it is
> > > > interested. Any other method of working would lead to impossible loading
> > > > on the SG.
> > > 
> > > Yes i agree but your suggestion is an indirect method.
> > > I am proposing a direct method to send SSNM messages from SGP to INACTIVE
> > > ASP. When SGP sends DUNA or DRST to ACTVIE ASP, let him also send a copy
> > > of the same message to INACTIVE ASP. I dont find any difficulty here.
> > > 
> > >  
> > > > Of course, I hope 199,999 of your inactive ASPs do not decide to become
> > > > active at the same time.....! 
> > > 
> > > Theoritically your example sounds good but practically NOBODY will
> > > configure 1 ASP ACTIVE and 199,999 ASPs INACTIVE.
> > > 
> > > --Samuel
> > > 
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 00:12:36 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA17356
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 00:12:36 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h175JNF27569
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 00:19:23 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h175JCp27542;
	Fri, 7 Feb 2003 00:19:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h175I6p27467
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 00:18:06 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA17316
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 00:10:47 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h175Dtf4001617;
	Thu, 6 Feb 2003 23:14:24 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h175DiM22235;
	Thu, 6 Feb 2003 23:13:44 -0600 (CST)
Date: Thu, 6 Feb 2003 23:13:44 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: Richard Adams <Richard.Adams@etsi.org>,
        "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206220202.A10248@openss7.org>
Message-ID: <Pine.SOL.4.10.10302062311170.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

Yes , The RFC does not restrict the number of inactive ASPs.
But network configuration will restrict.

--Samuel

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> The RFC does not restrict the number of inactive ASPs.
> 
> --brian
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 22:40:48)
> > 
> > Brian,
> > 
> > Ok, tell me, where you will configure 1 ASP ACT and 199,000 ASPs INACT?
> > I want to know which network does this poor configuration.
> > 
> > --Samuel
> > 
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > The RFC does not restrict the number of inactive ASPs.
> > > 
> > > --brian
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 22:23:12)
> > > > 
> > > > Richard,
> > > > 
> > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > 
> > > > On Thu, 6 Feb 2003, Richard Adams wrote:
> > > > 
> > > > > Brian,
> > > > > I agree with you. I think when the ASP becomes active, it should do an
> > > > > audit to update its knowledge of the SS7 network. The ASP needs to take
> > > > > the initiative to find out about the SS7 nodes etc. in which it is
> > > > > interested. Any other method of working would lead to impossible loading
> > > > > on the SG.
> > > > 
> > > > Yes i agree but your suggestion is an indirect method.
> > > > I am proposing a direct method to send SSNM messages from SGP to INACTIVE
> > > > ASP. When SGP sends DUNA or DRST to ACTVIE ASP, let him also send a copy
> > > > of the same message to INACTIVE ASP. I dont find any difficulty here.
> > > > 
> > > >  
> > > > > Of course, I hope 199,999 of your inactive ASPs do not decide to become
> > > > > active at the same time.....! 
> > > > 
> > > > Theoritically your example sounds good but practically NOBODY will
> > > > configure 1 ASP ACTIVE and 199,999 ASPs INACTIVE.
> > > > 
> > > > --Samuel
> > > > 
> > > 
> > > -- 
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 00:13:35 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA17386
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 00:13:35 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h175KMo27659
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 00:20:22 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h175K9p27647;
	Fri, 7 Feb 2003 00:20:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h175JDp27556
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 00:19:13 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA17340
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 00:11:55 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:bzfb6GtsAk5l36bhfCywwmMDyCaxAHM5@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h175FYD32221;
	Thu, 6 Feb 2003 22:15:34 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h175FYM11974;
	Thu, 6 Feb 2003 22:15:34 -0700
Date: Thu, 6 Feb 2003 22:15:33 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206221533.B10248@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206212401.A5070@openss7.org> <Pine.SOL.4.10.10302062230540.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302062230540.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 10:37:58PM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 22:37:58)
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 22:13:17)
> > > > 
> > > > Answer me this simple question:
> > > > 
> > > >    What time is wasted sending DAUD?
> > > 
> > > 1.First of all sending DAUD is unnecessary.
> > >   (RFC says newly ACTIVE ASP MAY send DAUD and it never said "MUST")
> > > 
> > > 2.In case if newly ACTIVE ASP sends DATA(222), definitely it will receive
> > > DRST(222) - Here these two messages (DATA and DRST) are un-necessary.
> > > 
> > > ALL THESE UNNECESSARY MESSAGE COMMUNICATION CAN BE AVOIDED IF SGP HAD SENT
> > > SSNM MESSAGES WHEN ASP2 WAS IN INACTIVE STATE.
> > > 
> > > SENDING SSNM MESSAGES TO INACTIVE ASP IS A DIRECT METHOD OF UPDATING
> > > DESTINATION STATUS OF SS7 NODES AND SGP IS ALSO ABLE TO SEND SSNM TO 
> > > INACTIVE ASP.
> > > 
> > > IMO IT WOULD BE GOOD IF WE ALLOW INACTIVE APS TO RECEIVE SSNM MESSAGES.
> > > 
> > 
> > You didn't answer the question:
> > 
> > 	What time is wasted sending DAUD?
> > 
> > I suppose that you cannot answer the question, because no time
> > is wasted sending DAUD.
> > 
> To prepare a DAUD message , ASP has to read BIG ss7 dest status table and
> he has to send to SGP, RUN A TIMER AND WAIT FOR RESPONSE.

It doesn't have to read any database, it can simply send DAUD(*).

It does not have to run a timer waiting for a response.  There is no
timer specified in the RFC for DAUD responses.  On the other hand, there
is a timer specified while waiting for ASPAC Ack, T(ack).

So the ASP simply tacks DAUD(*) onto the ASPAC Request and waits for
ASPAC Ack.  There is no delay.

> 
> Another case if ASP has registered with a DPC and empty OPC list(means any
> SS7 node can reach ASP via SGP). This case with what point code DAUD will
> be sent from ASP?

It sends DAUD(*) (where the '*" is a Mask value in the Affected Point Code
parameter greater than or equal to the number of bits in the point code,
indicating all routes available to the AS as indicated by the RC).

No time delay.

Any other absurd arguments?

Or, would you still have the SGs send hundreds or thousands of useless
messages to avoid you having to send a single DAUD(*) (only about 32
bytes long) immediately following the ASPAC?

Gee, if you have a decent SCTP you can even bundle the DAUD(*) with the
ASPAC.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 00:16:28 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA17441
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 00:16:28 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h175NF927770
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 00:23:15 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h175N3p27762;
	Fri, 7 Feb 2003 00:23:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h175Mup27738
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 00:22:56 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA17431
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 00:15:38 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:tSx8z88JeObq+dupjiApuxQNdlVSXza3@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h175JHD32270;
	Thu, 6 Feb 2003 22:19:17 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h175JHi12442;
	Thu, 6 Feb 2003 22:19:17 -0700
Date: Thu, 6 Feb 2003 22:19:16 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: Richard Adams <Richard.Adams@etsi.org>,
        "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206221916.C10248@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	Richard Adams <Richard.Adams@etsi.org>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206220202.A10248@openss7.org> <Pine.SOL.4.10.10302062311170.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302062311170.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 11:13:44PM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

That is implementation dependent.  Sending SSNM to inactive ASPs restricts
a dimension not restricted by the current and adequate procedure.  The
SG cannot be required to send SSNM to inactive ASPs.  The RFC should not
place a restriction on the number of inactive ASPs, because we cannot
forsee all future dimensioning.  Procedures should scale as well as possible.
The existing procedure scales.  Sending SSNM to inactive ASPs does not.

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 23:13:44)
> 
> Brian,
> 
> Yes , The RFC does not restrict the number of inactive ASPs.
> But network configuration will restrict.
> 
> --Samuel
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > The RFC does not restrict the number of inactive ASPs.
> > 
> > --brian
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 22:40:48)
> > > 
> > > Brian,
> > > 
> > > Ok, tell me, where you will configure 1 ASP ACT and 199,000 ASPs INACT?
> > > I want to know which network does this poor configuration.
> > > 
> > > --Samuel
> > > 
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Samuel,
> > > > 
> > > > The RFC does not restrict the number of inactive ASPs.
> > > > 
> > > > --brian
> > > > 
> > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 22:23:12)
> > > > > 
> > > > > Richard,
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > > 
> > > > > On Thu, 6 Feb 2003, Richard Adams wrote:
> > > > > 
> > > > > > Brian,
> > > > > > I agree with you. I think when the ASP becomes active, it should do an
> > > > > > audit to update its knowledge of the SS7 network. The ASP needs to take
> > > > > > the initiative to find out about the SS7 nodes etc. in which it is
> > > > > > interested. Any other method of working would lead to impossible loading
> > > > > > on the SG.
> > > > > 
> > > > > Yes i agree but your suggestion is an indirect method.
> > > > > I am proposing a direct method to send SSNM messages from SGP to INACTIVE
> > > > > ASP. When SGP sends DUNA or DRST to ACTVIE ASP, let him also send a copy
> > > > > of the same message to INACTIVE ASP. I dont find any difficulty here.
> > > > > 
> > > > >  
> > > > > > Of course, I hope 199,999 of your inactive ASPs do not decide to become
> > > > > > active at the same time.....! 
> > > > > 
> > > > > Theoritically your example sounds good but practically NOBODY will
> > > > > configure 1 ASP ACTIVE and 199,999 ASPs INACTIVE.
> > > > > 
> > > > > --Samuel
> > > > > 
> > > > 
> > > > -- 
> > > > Brian F. G. Bidulock
> > > > bidulock@openss7.org
> > > > http://www.openss7.org/
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > 
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 00:31:32 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA17851
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 00:31:32 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h175cJG28945
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 00:38:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h175c8p28938;
	Fri, 7 Feb 2003 00:38:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h175blp28920
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 00:37:47 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA17828
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 00:30:28 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h175Xtei003805;
	Thu, 6 Feb 2003 23:34:06 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h175XXM03382;
	Thu, 6 Feb 2003 23:33:33 -0600 (CST)
Date: Thu, 6 Feb 2003 23:33:33 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206221533.B10248@openss7.org>
Message-ID: <Pine.SOL.4.10.10302062319500.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 22:37:58)
> > 
> > Brian,
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 22:13:17)
> > > > > 
> > > > > Answer me this simple question:
> > > > > 
> > > > >    What time is wasted sending DAUD?
> > > > 
> > > > 1.First of all sending DAUD is unnecessary.
> > > >   (RFC says newly ACTIVE ASP MAY send DAUD and it never said "MUST")
> > > > 
> > > > 2.In case if newly ACTIVE ASP sends DATA(222), definitely it will receive
> > > > DRST(222) - Here these two messages (DATA and DRST) are un-necessary.
> > > > 
> > > > ALL THESE UNNECESSARY MESSAGE COMMUNICATION CAN BE AVOIDED IF SGP HAD SENT
> > > > SSNM MESSAGES WHEN ASP2 WAS IN INACTIVE STATE.
> > > > 
> > > > SENDING SSNM MESSAGES TO INACTIVE ASP IS A DIRECT METHOD OF UPDATING
> > > > DESTINATION STATUS OF SS7 NODES AND SGP IS ALSO ABLE TO SEND SSNM TO 
> > > > INACTIVE ASP.
> > > > 
> > > > IMO IT WOULD BE GOOD IF WE ALLOW INACTIVE APS TO RECEIVE SSNM MESSAGES.
> > > > 
> > > 
> > > You didn't answer the question:
> > > 
> > > 	What time is wasted sending DAUD?
> > > 
> > > I suppose that you cannot answer the question, because no time
> > > is wasted sending DAUD.
> > > 
> > To prepare a DAUD message , ASP has to read BIG ss7 dest status table and
> > he has to send to SGP, RUN A TIMER AND WAIT FOR RESPONSE.
> 
> It doesn't have to read any database, it can simply send DAUD(*).
> 
> It does not have to run a timer waiting for a response.  There is no
> timer specified in the RFC for DAUD responses.  On the other hand, there
> is a timer specified while waiting for ASPAC Ack, T(ack).

Again i am saying RFC never says "NEWLY ACTIVE ASP MUST HAVE TO SEND DAUD"
 your initial example was wrong so it leads into unnecesary things.

> 
> So the ASP simply tacks DAUD(*) onto the ASPAC Request and waits for
> ASPAC Ack.  There is no delay.
> 
> > 
> > Another case if ASP has registered with a DPC and empty OPC list(means any
> > SS7 node can reach ASP via SGP). This case with what point code DAUD will
> > be sent from ASP?
> 
> It sends DAUD(*) (where the '*" is a Mask value in the Affected Point Code
> parameter greater than or equal to the number of bits in the point code,
> indicating all routes available to the AS as indicated by the RC).
> 
> No time delay.
> 
> Any other absurd arguments?
> 
> Or, would you still have the SGs send hundreds or thousands of useless
> messages to avoid you having to send a single DAUD(*) (only about 32
> bytes long) immediately following the ASPAC?
> 
After sending DAUD(*), ASP will receive DRST(222,999,112,...), DUNA(333,876,...)
and SCON(555,666,...) all these useless messages, according to your absurd
arguements. And these useless messages i have to process at ASP then ASP
has to start its real work. ASP has to spend its precious time to process
all these useless messages. Thats all your argument.


> Gee, if you have a decent SCTP you can even bundle the DAUD(*) with the
> ASPAC.
> 
Indirect method and tomorrow your will have one more indirect method...

--Samuel


> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 00:49:38 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA18238
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 00:49:38 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h175uQa29721
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 00:56:26 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h175uEp29693;
	Fri, 7 Feb 2003 00:56:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h175tBp29564
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 00:55:11 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA18157
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 00:47:52 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:mDXVL/F54cmcQ6UnPFPTNKx2f84C/toL@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h175pUD32656;
	Thu, 6 Feb 2003 22:51:31 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h175pUZ16129;
	Thu, 6 Feb 2003 22:51:30 -0700
Date: Thu, 6 Feb 2003 22:51:30 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206225130.A15019@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206221533.B10248@openss7.org> <Pine.SOL.4.10.10302062319500.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302062319500.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Feb 06, 2003 at 11:33:33PM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 23:33:33)
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 22:37:58)
> > > 
> > > Brian,
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Samuel,
> > > > 
> > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 22:13:17)
> > > > > > 
> > > > > > Answer me this simple question:
> > > > > > 
> > > > > >    What time is wasted sending DAUD?
> > > > > 
> > > > > 1.First of all sending DAUD is unnecessary.
> > > > >   (RFC says newly ACTIVE ASP MAY send DAUD and it never said "MUST")
> > > > > 
> > > > > 2.In case if newly ACTIVE ASP sends DATA(222), definitely it will receive
> > > > > DRST(222) - Here these two messages (DATA and DRST) are un-necessary.
> > > > > 
> > > > > ALL THESE UNNECESSARY MESSAGE COMMUNICATION CAN BE AVOIDED IF SGP HAD SENT
> > > > > SSNM MESSAGES WHEN ASP2 WAS IN INACTIVE STATE.
> > > > > 
> > > > > SENDING SSNM MESSAGES TO INACTIVE ASP IS A DIRECT METHOD OF UPDATING
> > > > > DESTINATION STATUS OF SS7 NODES AND SGP IS ALSO ABLE TO SEND SSNM TO 
> > > > > INACTIVE ASP.
> > > > > 
> > > > > IMO IT WOULD BE GOOD IF WE ALLOW INACTIVE APS TO RECEIVE SSNM MESSAGES.
> > > > > 
> > > > 
> > > > You didn't answer the question:
> > > > 
> > > > 	What time is wasted sending DAUD?
> > > > 
> > > > I suppose that you cannot answer the question, because no time
> > > > is wasted sending DAUD.
> > > > 
> > > To prepare a DAUD message , ASP has to read BIG ss7 dest status table and
> > > he has to send to SGP, RUN A TIMER AND WAIT FOR RESPONSE.
> > 
> > It doesn't have to read any database, it can simply send DAUD(*).
> > 
> > It does not have to run a timer waiting for a response.  There is no
> > timer specified in the RFC for DAUD responses.  On the other hand, there
> > is a timer specified while waiting for ASPAC Ack, T(ack).
> 
> Again i am saying RFC never says "NEWLY ACTIVE ASP MUST HAVE TO SEND DAUD"
>  your initial example was wrong so it leads into unnecesary things.

So don't send DAUD.

The RFC is quite clear:

      - Isolation. The ASP is newly ASP-ACTIVE or has been
                   isolated from an SGP for an extended period.  The ASP
                   MAY request the availability/congestion status of one
                   or more SS7 destinations to which it expects to
                   communicate.

   The ASP MAY choose to audit the availability of unavailable
   destinations by sending DAUD messages.  This would be for example the
   case when an AS becomes active at an ASP and does not have current
   destination statuses.  If MTP restart is in progress at the SG, the
   SGP returns a DUNA message for that destination, even if it received
   an indication that the destination became available or restricted.

You said your ASP was coming ASP-ACTIVE and did not have current
destination statuses.  If you want to update them, send DAUD as
described in the RFC.  If you don't, don't.

And, there is no delay cause by the procedure in any event.


> > 
> > So the ASP simply tacks DAUD(*) onto the ASPAC Request and waits for
> > ASPAC Ack.  There is no delay.
> > 
> > > 
> > > Another case if ASP has registered with a DPC and empty OPC list(means any
> > > SS7 node can reach ASP via SGP). This case with what point code DAUD will
> > > be sent from ASP?
> > 
> > It sends DAUD(*) (where the '*" is a Mask value in the Affected Point Code
> > parameter greater than or equal to the number of bits in the point code,
> > indicating all routes available to the AS as indicated by the RC).
> > 
> > No time delay.
> > 
> > Any other absurd arguments?
> > 
> > Or, would you still have the SGs send hundreds or thousands of useless
> > messages to avoid you having to send a single DAUD(*) (only about 32
> > bytes long) immediately following the ASPAC?
> > 
> After sending DAUD(*), ASP will receive DRST(222,999,112,...), DUNA(333,876,...)
> and SCON(555,666,...) all these useless messages, according to your absurd
> arguements.

There are not useless, the ASP has asked for them because it does not
know the current status of any of the destinations.  If the ASP has
current status it does not need to send DAUD at all.  The ASP gets the
information requested.  If it doesn't need the information it should not
request it.

> And these useless messages i have to process at ASP then ASP
> has to start its real work. ASP has to spend its precious time to process
> all these useless messages. Thats all your argument.

If you don't want to receive them, don't request them with DAUD.

So which is it?  Does you database need to be updated at the ASP or not?
If it needs to be updated, it needs the status of all destinations
(including congestion and restriction).  If it doesn't need to be updated
don't send DAUD.

In any event, contrary to your claims, there is no delay.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 00:58:34 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA18383
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 00:58:34 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1765L230029
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 01:05:21 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1765Ap30018;
	Fri, 7 Feb 2003 01:05:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1764ip29982
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 01:04:44 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA18359
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 00:57:25 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1760tqk029463;
	Fri, 7 Feb 2003 00:01:01 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h1760YM18325;
	Fri, 7 Feb 2003 00:00:34 -0600 (CST)
Date: Fri, 7 Feb 2003 00:00:34 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: Richard Adams <Richard.Adams@etsi.org>,
        "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206221916.C10248@openss7.org>
Message-ID: <Pine.SOL.4.10.10302062352540.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

An efficent ASP will take up the load as soon as it comes up.
Inefficient ASP may send DAUD and will receive DRST(...),DUNA(...),
SCON(...).  So from the point of ASP following are the unnecessary
messages.
 .SENDING DAUD
 .RECEING DRST
 .RECEIVING DUNA
 .RECEIVING SCON

From the point of SGP these are the unnecessay messages.
 .Receiving DAUD
 .SENDING DRST
 .SENDING DUNA
 .SENDING SCON.

Thats what your argument leads.

--Samuel

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> That is implementation dependent.  Sending SSNM to inactive ASPs restricts
> a dimension not restricted by the current and adequate procedure.  The
> SG cannot be required to send SSNM to inactive ASPs.  The RFC should not
> place a restriction on the number of inactive ASPs, because we cannot
> forsee all future dimensioning.  Procedures should scale as well as possible.
> The existing procedure scales.  Sending SSNM to inactive ASPs does not.
> 
> --brian
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 23:13:44)
> > 
> > Brian,
> > 
> > Yes , The RFC does not restrict the number of inactive ASPs.
> > But network configuration will restrict.
> > 
> > --Samuel
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > The RFC does not restrict the number of inactive ASPs.
> > > 
> > > --brian
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 22:40:48)
> > > > 
> > > > Brian,
> > > > 
> > > > Ok, tell me, where you will configure 1 ASP ACT and 199,000 ASPs INACT?
> > > > I want to know which network does this poor configuration.
> > > > 
> > > > --Samuel
> > > > 
> > > > 
> > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > 
> > > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > > 
> > > > > Samuel,
> > > > > 
> > > > > The RFC does not restrict the number of inactive ASPs.
> > > > > 
> > > > > --brian
> > > > > 
> > > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 22:23:12)
> > > > > > 
> > > > > > Richard,
> > > > > > 
> > > > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > > >   1000 Coit Road, Plano, Texas 75075
> > > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > > > 
> > > > > > On Thu, 6 Feb 2003, Richard Adams wrote:
> > > > > > 
> > > > > > > Brian,
> > > > > > > I agree with you. I think when the ASP becomes active, it should do an
> > > > > > > audit to update its knowledge of the SS7 network. The ASP needs to take
> > > > > > > the initiative to find out about the SS7 nodes etc. in which it is
> > > > > > > interested. Any other method of working would lead to impossible loading
> > > > > > > on the SG.
> > > > > > 
> > > > > > Yes i agree but your suggestion is an indirect method.
> > > > > > I am proposing a direct method to send SSNM messages from SGP to INACTIVE
> > > > > > ASP. When SGP sends DUNA or DRST to ACTVIE ASP, let him also send a copy
> > > > > > of the same message to INACTIVE ASP. I dont find any difficulty here.
> > > > > > 
> > > > > >  
> > > > > > > Of course, I hope 199,999 of your inactive ASPs do not decide to become
> > > > > > > active at the same time.....! 
> > > > > > 
> > > > > > Theoritically your example sounds good but practically NOBODY will
> > > > > > configure 1 ASP ACTIVE and 199,999 ASPs INACTIVE.
> > > > > > 
> > > > > > --Samuel
> > > > > > 
> > > > > 
> > > > > -- 
> > > > > Brian F. G. Bidulock
> > > > > bidulock@openss7.org
> > > > > http://www.openss7.org/
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > 
> > > 
> > > -- 
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 01:16:43 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA18686
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 01:16:43 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h176NV931185
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 01:23:31 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h176NIp31156;
	Fri, 7 Feb 2003 01:23:18 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h176Mlp31125
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 01:22:47 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA18641
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 01:15:27 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:MJu/CwPogUZFKpr+iRqzCSZT3nvR4LRC@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h176J6D00533;
	Thu, 6 Feb 2003 23:19:06 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h176J5u19467;
	Thu, 6 Feb 2003 23:19:05 -0700
Date: Thu, 6 Feb 2003 23:19:04 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: Richard Adams <Richard.Adams@etsi.org>,
        "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206231904.A17419@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	Richard Adams <Richard.Adams@etsi.org>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206221916.C10248@openss7.org> <Pine.SOL.4.10.10302062352540.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302062352540.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Fri, Feb 07, 2003 at 12:00:34AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

If DAUD is unnecessary to your implementation, don't send it.
The RFC permits you to not send DAUD.

But, the SG cannot be required to send SSNM to inactive ASPs.

BTW it is not _my_ argument, it was the conensus of the WG as
captured in the RFC, reinforced by emails from Dick, Tolga,
and myself on this thread.

I was responsible for having the DAUD procedure reduced from
SHOULD to MAY on the "MTP Restart" thread from September 2001.
So you can thank me that you don't have to send DAUD if you
don't need the reponses.

A practical approach if your ASPs are stupid enough to not
share any state information in override mode, is to not
send DAUD and rely on responsive procedures.  This is because
having a destination other than accessible is a rare
circumstance in the SS7 network.

For MTP Restart, the smart SG will send DUNA(*) immediately
following ASPAC Ack and I don't see any problem even requiring
the SG to do so when ASPs are activating into an MTP Restart.

So its simple, if you think your ASP needs to update its
database, send DAUD.  If you think it doesn't, don't send DAUD.

Dumb ASPs that don't share state are not going to be able to
process any calls or transactions after the failover anyway,
so it really doesn't matter.

I think that your concerns have been adequately addressed on
this thread.

Unless you can come up with a different argument, further
discussion is pointless.

--brian

Samuel Dur D. Jeyaseelan wrote:              (Fri, 07 Feb 2003 00:00:34)
> 
> Brian,
> 
> An efficent ASP will take up the load as soon as it comes up.
> Inefficient ASP may send DAUD and will receive DRST(...),DUNA(...),
> SCON(...).  So from the point of ASP following are the unnecessary
> messages.
>  .SENDING DAUD
>  .RECEING DRST
>  .RECEIVING DUNA
>  .RECEIVING SCON
> 
> >From the point of SGP these are the unnecessay messages.
>  .Receiving DAUD
>  .SENDING DRST
>  .SENDING DUNA
>  .SENDING SCON.
> 
> Thats what your argument leads.
> 
> --Samuel
> 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 01:20:26 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA18786
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 01:20:26 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h176RFn31369
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 01:27:15 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h176R4p31356;
	Fri, 7 Feb 2003 01:27:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h176Q7p31333
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 01:26:07 -0500
Received: from rtp-msg-core-1.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA18758
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 01:18:48 -0500 (EST)
Received: from dingdong.cisco.com (localhost [127.0.0.1])
	by rtp-msg-core-1.cisco.com (8.12.2/8.12.6) with ESMTP id h176NLrf009714
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 01:23:21 -0500 (EST)
Received: from cisco.com (rtp-vpn1-263.cisco.com [10.82.225.7])
	by dingdong.cisco.com (Mirapoint Messaging Server MOS 3.2.1-GA)
	with ESMTP id ABJ55326;
	Fri, 7 Feb 2003 01:22:25 -0500 (EST)
Message-ID: <3E4350A2.8060B90@cisco.com>
Date: Fri, 07 Feb 2003 01:22:26 -0500
From: wayne n taylor <waynet@cisco.com>
X-Mailer: Mozilla 4.79 [en]C-CCK-MCD   (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: sigtran@ietf.org
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] M2PA draft 7 question
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Hi All,

i have a couple of comments/questions about the initial exchange of the
Link Status Alignment messages. there seems to be some inconsistency in
the text and finite state machine.

section 4.1.3 paragraph 4:

"Once the association is established and M2PA has received a Start
Request from MTP3, M2PA sends the Link Status Alignment message to its
peer. If M2PA has not already received the Link Status Alignment
message from its peer, then M2PA starts timer T2."

what state would the local M2PA be in to receive a LS Alignment message
from its peer prior to getting a MTP3 Start ?. i assume the OOS state,
but the finite state machine does not show LS Alignment as an input. 

section 4.1.3 paragraph 5:

"(Note that if the remote M2PA has not received a Start Request from
its MTP3, it will not send the Link Status Alignment message to the
local M2PA. Eventually timer T2 in the local M2PA will expire. If the
remote M2PA receives a Start Request from its MTP3 and sends Link
Status Alignment before the local M2PA's timer T2 expires, the
alignment procedure can continue.)"

assume the local M2PA receives a Start Request from MTP3, sends LS
Alignment and start T2, but the remote peer has not received a Start
Request and is in OOS. the current finite state machine does not show
what to do with the LS Alignment from the local peer so lets assume the
LS Alignment is dropped. the remote peer receives a Start Request from
MTP3, sends LS Alignment and start T2 (since the remote peer didnt get
the local peers LS Alignment). assume further that timer T2 is still
running on the local peer. the text from section 4.1.3 above states the
"alignment procedure can continue". in this case the local peer sent a
LS Alignment and received a LS Alignment. the remote peer only sent a LS
Alignment and did not receive an LS Alignment (since LS Alignment
messages are not retried). eventually T2 on the remote peer will expire,
T3 on the local peer will expire and the alignment procedure starts over
again. it seems possible for the peers to get caught in a loop trying to
exchange LS Alignment messages.

section 4.1.3 paragraph 9:

"When M2PA has both sent and received the Link Status Alignment
message, it has completed alignment.  M2PA starts the aligned timer T3
and moves to the proving state."

is it a requirement for both the local and remote peers to send/receive
LS Alignment for alignment to be complete ?. the only way to transition
from Initial Alignment to Proving is to receive LS Alignment. in the
scenario above, both peers do not send and receive LS Alignment
messages.


regards,

wayne
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 01:21:26 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA18822
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 01:21:26 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h176SEH31429
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 01:28:14 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h176S3p31422;
	Fri, 7 Feb 2003 01:28:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h176Rhp31402
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 01:27:43 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA18784
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 01:20:23 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h176O0qg001385;
	Fri, 7 Feb 2003 00:24:01 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h176NYM26884;
	Fri, 7 Feb 2003 00:23:34 -0600 (CST)
Date: Fri, 7 Feb 2003 00:23:34 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@SSD.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206225130.A15019@openss7.org>
Message-ID: <Pine.SOL.4.10.10302070010480.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 23:33:33)
> > 
> > Brian,
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 22:37:58)
> > > > 
> > > > Brian,
> > > > 
> > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > 
> > > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > > 
> > > > > Samuel,
> > > > > 
> > > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 06 Feb 2003 22:13:17)
> > > > > > > 
> > > > > > > Answer me this simple question:
> > > > > > > 
> > > > > > >    What time is wasted sending DAUD?
> > > > > > 
> > > > > > 1.First of all sending DAUD is unnecessary.
> > > > > >   (RFC says newly ACTIVE ASP MAY send DAUD and it never said "MUST")
> > > > > > 
> > > > > > 2.In case if newly ACTIVE ASP sends DATA(222), definitely it will receive
> > > > > > DRST(222) - Here these two messages (DATA and DRST) are un-necessary.
> > > > > > 
> > > > > > ALL THESE UNNECESSARY MESSAGE COMMUNICATION CAN BE AVOIDED IF SGP HAD SENT
> > > > > > SSNM MESSAGES WHEN ASP2 WAS IN INACTIVE STATE.
> > > > > > 
> > > > > > SENDING SSNM MESSAGES TO INACTIVE ASP IS A DIRECT METHOD OF UPDATING
> > > > > > DESTINATION STATUS OF SS7 NODES AND SGP IS ALSO ABLE TO SEND SSNM TO 
> > > > > > INACTIVE ASP.
> > > > > > 
> > > > > > IMO IT WOULD BE GOOD IF WE ALLOW INACTIVE APS TO RECEIVE SSNM MESSAGES.
> > > > > > 
> > > > > 
> > > > > You didn't answer the question:
> > > > > 
> > > > > 	What time is wasted sending DAUD?
> > > > > 
> > > > > I suppose that you cannot answer the question, because no time
> > > > > is wasted sending DAUD.
> > > > > 
> > > > To prepare a DAUD message , ASP has to read BIG ss7 dest status table and
> > > > he has to send to SGP, RUN A TIMER AND WAIT FOR RESPONSE.
> > > 
> > > It doesn't have to read any database, it can simply send DAUD(*).
> > > 
> > > It does not have to run a timer waiting for a response.  There is no
> > > timer specified in the RFC for DAUD responses.  On the other hand, there
> > > is a timer specified while waiting for ASPAC Ack, T(ack).
> > 
> > Again i am saying RFC never says "NEWLY ACTIVE ASP MUST HAVE TO SEND DAUD"
> >  your initial example was wrong so it leads into unnecesary things.
> 
> So don't send DAUD.
> 
> The RFC is quite clear:
> 
>       - Isolation. The ASP is newly ASP-ACTIVE or has been
>                    isolated from an SGP for an extended period.  The ASP
>                    MAY request the availability/congestion status of one
>                    or more SS7 destinations to which it expects to
>                    communicate.
> 
>    The ASP MAY choose to audit the availability of unavailable
>    destinations by sending DAUD messages.  This would be for example the
>    case when an AS becomes active at an ASP and does not have current
>    destination statuses.  If MTP restart is in progress at the SG, the
>    SGP returns a DUNA message for that destination, even if it received
>    an indication that the destination became available or restricted.

It says "ASP MAY..." but not "ASP MUST..."

> 
> You said your ASP was coming ASP-ACTIVE and did not have current
> destination statuses.  If you want to update them, send DAUD as
> described in the RFC.  If you don't, don't.
> 
> And, there is no delay cause by the procedure in any event.
> 

What do you mean? AT ASP Sending DAUD and PROCESSING DRST,DUNA, SOCN
will not consume any CPU time?If so please suggest me what CPU you are
using?
An efficent ASP will never spend its CPU time to process all these
unnecessary messages.

An efficent ASP would have received all ss7 destination status messages
when it was in INACT state.And it will start processing real traffic as
soon as comes up.

 

> 
> > > 
> > > So the ASP simply tacks DAUD(*) onto the ASPAC Request and waits for
> > > ASPAC Ack.  There is no delay.
> > > 
> > > > 
> > > > Another case if ASP has registered with a DPC and empty OPC list(means any
> > > > SS7 node can reach ASP via SGP). This case with what point code DAUD will
> > > > be sent from ASP?
> > > 
> > > It sends DAUD(*) (where the '*" is a Mask value in the Affected Point Code
> > > parameter greater than or equal to the number of bits in the point code,
> > > indicating all routes available to the AS as indicated by the RC).
> > > 
> > > No time delay.
> > > 
> > > Any other absurd arguments?
> > > 
> > > Or, would you still have the SGs send hundreds or thousands of useless
> > > messages to avoid you having to send a single DAUD(*) (only about 32
> > > bytes long) immediately following the ASPAC?
> > > 
> > After sending DAUD(*), ASP will receive DRST(222,999,112,...), DUNA(333,876,...)
> > and SCON(555,666,...) all these useless messages, according to your absurd
> > arguements.
> 
> There are not useless, the ASP has asked for them because it does not
> know the current status of any of the destinations.  If the ASP has
> current status it does not need to send DAUD at all.  The ASP gets the
> information requested.  If it doesn't need the information it should not
> request it.
They are absultely usefull for an INEFFICENT ASP but 100% useless for an
EFFICENT ASP.

> 
> > And these useless messages i have to process at ASP then ASP
> > has to start its real work. ASP has to spend its precious time to process
> > all these useless messages. Thats all your argument.
> 
> If you don't want to receive them, don't request them with DAUD.
> 
> So which is it?  Does you database need to be updated at the ASP or not?
> If it needs to be updated, it needs the status of all destinations
> (including congestion and restriction).  If it doesn't need to be updated
> don't send DAUD.
> 
> In any event, contrary to your claims, there is no delay.
see above.

--Samuel

> 
> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 01:26:27 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA18914
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 01:26:27 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h176XFK31592
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 01:33:15 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h176X4p31578;
	Fri, 7 Feb 2003 01:33:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h176Wep31543
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 01:32:40 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA18889
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 01:25:20 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:ngWsmRYt5/LC8qNvVqdZTt3tlnD5SQFp@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h176SwD00645;
	Thu, 6 Feb 2003 23:28:58 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h176Sw120676;
	Thu, 6 Feb 2003 23:28:58 -0700
Date: Thu, 6 Feb 2003 23:28:58 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@SSD.usa.alcatel.com>
Cc: "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206232858.A20208@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@SSD.usa.alcatel.com>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206225130.A15019@openss7.org> <Pine.SOL.4.10.10302070010480.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302070010480.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@SSD.usa.alcatel.com on Fri, Feb 07, 2003 at 12:23:34AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 07 Feb 2003 00:23:34)
> > 
> >    The ASP MAY choose to audit the availability of unavailable
> >    destinations by sending DAUD messages.  This would be for example the
> >    case when an AS becomes active at an ASP and does not have current
> >    destination statuses.  If MTP restart is in progress at the SG, the
> >    SGP returns a DUNA message for that destination, even if it received
> >    an indication that the destination became available or restricted.
> 
> It says "ASP MAY..." but not "ASP MUST..."

Thank you.  I had it changed to MAY.  I'm glad you like it.

> 
> > 
> > You said your ASP was coming ASP-ACTIVE and did not have current
> > destination statuses.  If you want to update them, send DAUD as
> > described in the RFC.  If you don't, don't.
> > 
> > And, there is no delay cause by the procedure in any event.
> > 
> 
> What do you mean? AT ASP Sending DAUD and PROCESSING DRST,DUNA, SOCN
> will not consume any CPU time?If so please suggest me what CPU you are
> using?
> An efficent ASP will never spend its CPU time to process all these
> unnecessary messages.
> 
> An efficent ASP would have received all ss7 destination status messages
> when it was in INACT state.And it will start processing real traffic as
> soon as comes up.

Don't send DAUD if you don't need the responses.
Start sending traffic immediately.

> 
>  
> 
> > 
> > > > 
> > > > So the ASP simply tacks DAUD(*) onto the ASPAC Request and waits for
> > > > ASPAC Ack.  There is no delay.
> > > > 
> > > > > 
> > > > > Another case if ASP has registered with a DPC and empty OPC list(means any
> > > > > SS7 node can reach ASP via SGP). This case with what point code DAUD will
> > > > > be sent from ASP?
> > > > 
> > > > It sends DAUD(*) (where the '*" is a Mask value in the Affected Point Code
> > > > parameter greater than or equal to the number of bits in the point code,
> > > > indicating all routes available to the AS as indicated by the RC).
> > > > 
> > > > No time delay.
> > > > 
> > > > Any other absurd arguments?
> > > > 
> > > > Or, would you still have the SGs send hundreds or thousands of useless
> > > > messages to avoid you having to send a single DAUD(*) (only about 32
> > > > bytes long) immediately following the ASPAC?
> > > > 
> > > After sending DAUD(*), ASP will receive DRST(222,999,112,...), DUNA(333,876,...)
> > > and SCON(555,666,...) all these useless messages, according to your absurd
> > > arguements.
> > 
> > There are not useless, the ASP has asked for them because it does not
> > know the current status of any of the destinations.  If the ASP has
> > current status it does not need to send DAUD at all.  The ASP gets the
> > information requested.  If it doesn't need the information it should not
> > request it.
> They are absultely usefull for an INEFFICENT ASP but 100% useless for an
> EFFICENT ASP.

So don't have your EFFICIENT ASP send DAUD.  But, sorry,  it cannot expect
to receive SSNM while inactive.  If you want database updates without DAUD,
get them from the active ASP using whatever means you want.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 01:35:29 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA19083
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 01:35:29 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h176gI332594
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 01:42:18 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h176g7p32586;
	Fri, 7 Feb 2003 01:42:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h176fhp32542
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 01:41:43 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA19043
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 01:34:23 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h176c0qe002368;
	Fri, 7 Feb 2003 00:38:00 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h176b4M01818;
	Fri, 7 Feb 2003 00:37:04 -0600 (CST)
Date: Fri, 7 Feb 2003 00:37:04 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: Richard Adams <Richard.Adams@etsi.org>,
        "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
In-Reply-To: <20030206231904.A17419@openss7.org>
Message-ID: <Pine.SOL.4.10.10302070033350.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

What is the risk at SGP to send SSNM messages to INACT ASP?

What factor restricts INACT ASP to receive SSNM messages?

(please dont give me IMPOSSIBLE and practically unimaginable examples).

--Samuel



  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> If DAUD is unnecessary to your implementation, don't send it.
> The RFC permits you to not send DAUD.
> 
> But, the SG cannot be required to send SSNM to inactive ASPs.
> 
> BTW it is not _my_ argument, it was the conensus of the WG as
> captured in the RFC, reinforced by emails from Dick, Tolga,
> and myself on this thread.
> 
> I was responsible for having the DAUD procedure reduced from
> SHOULD to MAY on the "MTP Restart" thread from September 2001.
> So you can thank me that you don't have to send DAUD if you
> don't need the reponses.
> 
> A practical approach if your ASPs are stupid enough to not
> share any state information in override mode, is to not
> send DAUD and rely on responsive procedures.  This is because
> having a destination other than accessible is a rare
> circumstance in the SS7 network.
> 
> For MTP Restart, the smart SG will send DUNA(*) immediately
> following ASPAC Ack and I don't see any problem even requiring
> the SG to do so when ASPs are activating into an MTP Restart.
> 
> So its simple, if you think your ASP needs to update its
> database, send DAUD.  If you think it doesn't, don't send DAUD.
> 
> Dumb ASPs that don't share state are not going to be able to
> process any calls or transactions after the failover anyway,
> so it really doesn't matter.
> 
> I think that your concerns have been adequately addressed on
> this thread.
> 
> Unless you can come up with a different argument, further
> discussion is pointless.
> 
> --brian
> 
> Samuel Dur D. Jeyaseelan wrote:              (Fri, 07 Feb 2003 00:00:34)
> > 
> > Brian,
> > 
> > An efficent ASP will take up the load as soon as it comes up.
> > Inefficient ASP may send DAUD and will receive DRST(...),DUNA(...),
> > SCON(...).  So from the point of ASP following are the unnecessary
> > messages.
> >  .SENDING DAUD
> >  .RECEING DRST
> >  .RECEIVING DUNA
> >  .RECEIVING SCON
> > 
> > >From the point of SGP these are the unnecessay messages.
> >  .Receiving DAUD
> >  .SENDING DRST
> >  .SENDING DUNA
> >  .SENDING SCON.
> > 
> > Thats what your argument leads.
> > 
> > --Samuel
> > 
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 01:44:28 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA19354
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 01:44:28 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h176pHE00412
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 01:51:17 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h176p6p00405;
	Fri, 7 Feb 2003 01:51:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h176oop00386
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 01:50:50 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA19340
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 01:43:30 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:hP3HebrWypoAXBPNDssCn4jZ6MZcrBEg@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h176lAD00891;
	Thu, 6 Feb 2003 23:47:10 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h176l9L22846;
	Thu, 6 Feb 2003 23:47:09 -0700
Date: Thu, 6 Feb 2003 23:47:09 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: Richard Adams <Richard.Adams@etsi.org>,
        "Zannin, Francois" <Francois.Zannin@hp.com>,
        "Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030206234709.A22731@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	Richard Adams <Richard.Adams@etsi.org>,
	"Zannin, Francois" <Francois.Zannin@hp.com>,
	"Nagar, Ashish" <ashish.nagar@intel.com>, sigtran@ietf.org
References: <20030206231904.A17419@openss7.org> <Pine.SOL.4.10.10302070033350.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302070033350.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Fri, Feb 07, 2003 at 12:37:04AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 07 Feb 2003 00:37:04)
> 
> Brian,
> 
> What is the risk at SGP to send SSNM messages to INACT ASP?
> 
> What factor restricts INACT ASP to receive SSNM messages?

Concensus.

--brian

> 
> (please dont give me IMPOSSIBLE and practically unimaginable examples).
> 
> --Samuel

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 03:03:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA00952
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 03:03:56 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h178Alp15762
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 03:10:47 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h178AUp15749;
	Fri, 7 Feb 2003 03:10:32 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1789Np15708
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 03:09:23 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA00907
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 03:02:01 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:tgVHRszSAN4lzyg5gl5ycsPgHnoSTh/w@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1785eD01991
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 01:05:40 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1785eI31816
	for sigtran@ietf.org; Fri, 7 Feb 2003 01:05:40 -0700
Date: Fri, 7 Feb 2003 01:05:40 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: sigtran@ietf.org
Message-ID: <20030207010540.A31552@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: sigtran@ietf.org
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Subject: [Sigtran] Packing M3UA messages into SCTP ULPDUs
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Sigtran,

Where does it say that only one M3UA message may be
placed into an SCTP ULPDU?

As messages contain the length in the header, can I
pack two or more M3UA messages in an SCTP ULPDU?

This is so basic, I wonder why it was not mentioned
anywhere...

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 04:08:08 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA02294
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 04:08:08 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h179ExP19499
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 04:14:59 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h179Ekp19490;
	Fri, 7 Feb 2003 04:14:46 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h179DLp19404
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 04:13:21 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA02240
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 04:05:54 -0500 (EST)
Subject: Re: [Sigtran] Packing M3UA messages into SCTP ULPDUs
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF767196AA.BBBAF488-ONC2256CC6.0030A76E@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Fri, 7 Feb 2003 11:05:52 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/07/2003
 11:06:10 AM
MIME-Version: 1.0
Content-type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h179DLp19405
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit


Brian,
> Sigtran,
>
> Where does it say that only one M3UA message may be
> placed into an SCTP ULPDU?
>
> As messages contain the length in the header, can I
> pack two or more M3UA messages in an SCTP ULPDU?
>
> This is so basic, I wonder why it was not mentioned
> anywhere...
If it is not forbidden, shouldn´t it be interpreted as allowed (although I
would guess that most of the implementations are assuming one M3UA message
in one SCTP ULPDU)?
BTW related with this topic:
from 3.1.4
  "The Message Length defines the length of the message in octets,
   including the Common Header.  The Message Length MUST include
   parameter padding bytes, if any.

   Note: A receiver SHOULD accept the message whether or not the final
   parameter padding is included in the message length."
The above paragraphs sound strange to me, including padding bytes in the
length is a MUST but still receiver should be prepared for the opposite.
But still based on that I assume a M3UA peer can receive messages, whose
length does not contain padding bytes (otherwise forget what stands below)
From 3.2
     "The total length of a parameter (including Tag, Parameter Length
      and Value fields) MUST be a multiple of 4 bytes.  If the length of
      the parameter is not a multiple of 4 bytes, the sender pads the
      Parameter at the end (i.e., after the Parameter Value field) with
      all zero bytes.  The length of the padding is NOT included in the
      parameter length field.  A sender SHOULD NOT pad with more than 3
      bytes.  The receiver MUST ignore the padding bytes."
Not to add more than 3 padding bytes is a SHOULD NOT, not a MUST.

For a scenario where message length does not contain padding bytes length
and more than 3 padding bytes are used, I would think receiver can´t detect
boundary between two M3UA messages (but I have to repeat, it sounds strange
to me that including padding bytes length in message length is a MUST and
still receiver SHOULD accept messages, which aren´t prepared in that way)

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 04:31:31 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA02850
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 04:31:30 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h179cM820881
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 04:38:22 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h179c8p20873;
	Fri, 7 Feb 2003 04:38:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h179bKp20758
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 04:37:20 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA02816
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 04:29:56 -0500 (EST)
Subject: Re: [Sigtran] Packing M3UA messages into SCTP ULPDUs
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OFF43AC2C0.37E2F51C-ONC2256CC6.003368BB@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Fri, 7 Feb 2003 11:30:06 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/07/2003
 11:30:11 AM
MIME-Version: 1.0
Content-type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h179bKp20759
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit


A comment about my statement that message boundary can´t be detected in
certain cases between multiple M3UA messages:
In fact it seems it can be: Padding is done with 0 and the first octet of
the message header (version) has a value of 1(and I assume it never will go
down to 0). So, just scan till 1 is found.

But I still find it strange that including padding bytes in message length
is a MUST and expecting the opposite is a SHOULD.
    Tolga



Brian,
> Sigtran,
>
> Where does it say that only one M3UA message may be
> placed into an SCTP ULPDU?
>
> As messages contain the length in the header, can I
> pack two or more M3UA messages in an SCTP ULPDU?
>
> This is so basic, I wonder why it was not mentioned
> anywhere...
If it is not forbidden, shouldn´t it be interpreted as allowed (although I
would guess that most of the implementations are assuming one M3UA message
in one SCTP ULPDU)?
BTW related with this topic:
from 3.1.4
  "The Message Length defines the length of the message in octets,
   including the Common Header.  The Message Length MUST include
   parameter padding bytes, if any.

   Note: A receiver SHOULD accept the message whether or not the final
   parameter padding is included in the message length."
The above paragraphs sound strange to me, including padding bytes in the
length is a MUST but still receiver should be prepared for the opposite.
But still based on that I assume a M3UA peer can receive messages, whose
length does not contain padding bytes (otherwise forget what stands below)
From 3.2
     "The total length of a parameter (including Tag, Parameter Length
      and Value fields) MUST be a multiple of 4 bytes.  If the length of
      the parameter is not a multiple of 4 bytes, the sender pads the
      Parameter at the end (i.e., after the Parameter Value field) with
      all zero bytes.  The length of the padding is NOT included in the
      parameter length field.  A sender SHOULD NOT pad with more than 3
      bytes.  The receiver MUST ignore the padding bytes."
Not to add more than 3 padding bytes is a SHOULD NOT, not a MUST.

For a scenario where message length does not contain padding bytes length
and more than 3 padding bytes are used, I would think receiver can´t detect
boundary between two M3UA messages (but I have to repeat, it sounds strange
to me that including padding bytes length in message length is a MUST and
still receiver SHOULD accept messages, which aren´t prepared in that way)

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 04:35:37 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA02932
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 04:35:37 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h179gSI21048
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 04:42:28 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h179g3p21026;
	Fri, 7 Feb 2003 04:42:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h179f6p21006
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 04:41:06 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA02892
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 04:33:43 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:CYh5tofJyytTJVUYR4pRZTebwTdzxvxy@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h179bLD03160;
	Fri, 7 Feb 2003 02:37:21 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h179bK010482;
	Fri, 7 Feb 2003 02:37:20 -0700
Date: Fri, 7 Feb 2003 02:37:20 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Packing M3UA messages into SCTP ULPDUs
Message-ID: <20030207023720.A9728@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OF767196AA.BBBAF488-ONC2256CC6.0030A76E@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF767196AA.BBBAF488-ONC2256CC6.0030A76E@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Fri, Feb 07, 2003 at 11:05:52AM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: 8bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Tolga,

It does not even forbid splitting an M3UA message across multiple
SCTP ULPDUs.

I think this should be clarified in the IGs for the sake of
interoperablity.  1 M3UA (M2UA, SUA) message to 1 SCTP ULPDU.

TOLGA ASVEREN wrote:                                                                 (Fri, 07 Feb 2003 11:05:52)
> 
> Brian,
> > Sigtran,
> >
> > Where does it say that only one M3UA message may be
> > placed into an SCTP ULPDU?
> >
> > As messages contain the length in the header, can I
> > pack two or more M3UA messages in an SCTP ULPDU?
> >
> > This is so basic, I wonder why it was not mentioned
> > anywhere...
> If it is not forbidden, shouldn´t it be interpreted as allowed (although I
> would guess that most of the implementations are assuming one M3UA message
> in one SCTP ULPDU)?
> BTW related with this topic:
> from 3.1.4
>   "The Message Length defines the length of the message in octets,
>    including the Common Header.  The Message Length MUST include
>    parameter padding bytes, if any.
> 
>    Note: A receiver SHOULD accept the message whether or not the final
>    parameter padding is included in the message length."
> The above paragraphs sound strange to me, including padding bytes in the
> length is a MUST but still receiver should be prepared for the opposite.
> But still based on that I assume a M3UA peer can receive messages, whose
> length does not contain padding bytes (otherwise forget what stands below)
> >From 3.2
>      "The total length of a parameter (including Tag, Parameter Length
>       and Value fields) MUST be a multiple of 4 bytes.  If the length of
>       the parameter is not a multiple of 4 bytes, the sender pads the
>       Parameter at the end (i.e., after the Parameter Value field) with
>       all zero bytes.  The length of the padding is NOT included in the
>       parameter length field.  A sender SHOULD NOT pad with more than 3
>       bytes.  The receiver MUST ignore the padding bytes."
> Not to add more than 3 padding bytes is a SHOULD NOT, not a MUST.
> 
> For a scenario where message length does not contain padding bytes length
> and more than 3 padding bytes are used, I would think receiver can´t detect
> boundary between two M3UA messages (but I have to repeat, it sounds strange
> to me that including padding bytes length in message length is a MUST and
> still receiver SHOULD accept messages, which aren´t prepared in that way)

One could still detect the next message by scanning for the next valid
M3UA message header...

Also, that wouldn't cause any problem for splitting an M3UA message across
several SCTP ULPDUs.

I think we need to state in the IG: 1 UA message, 1 SCTP ULPDU.

==brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 04:37:29 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA02971
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 04:37:29 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h179iK121123
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 04:44:20 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h179i5p21110;
	Fri, 7 Feb 2003 04:44:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h179hrp21093
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 04:43:53 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA02951
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 04:36:30 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:X8JnQIQcn8i1ldo1d/9qR/69WsOc22wN@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h179e9D03204;
	Fri, 7 Feb 2003 02:40:09 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h179e9D10841;
	Fri, 7 Feb 2003 02:40:09 -0700
Date: Fri, 7 Feb 2003 02:40:08 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Packing M3UA messages into SCTP ULPDUs
Message-ID: <20030207024008.B9728@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OFF43AC2C0.37E2F51C-ONC2256CC6.003368BB@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OFF43AC2C0.37E2F51C-ONC2256CC6.003368BB@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Fri, Feb 07, 2003 at 11:30:06AM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Tolga,

TOLGA ASVEREN wrote:                                                                 (Fri, 07 Feb 2003 11:30:06)
> 
> But I still find it strange that including padding bytes in message length
> is a MUST and expecting the opposite is a SHOULD.

It's just the be_strict_on_what_you_send_and_be_forgiving_on_what_you_receive
principle.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 05:11:03 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA03720
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 05:11:03 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17AHtp23007
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 05:17:55 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17AHbp22959;
	Fri, 7 Feb 2003 05:17:38 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17AGPp22912
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 05:16:25 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA03660
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 05:09:02 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h17ACcqe019925;
	Fri, 7 Feb 2003 04:12:39 -0600 (CST)
Received: from axes-mach01.usa.alcatel.com by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id OAA01973; Fri, 7 Feb 2003 14:41:48 -0500 (GMT)
Message-ID: <3E4386D3.83D2B8EF@axes-mach01.usa.alcatel.com>
Date: Fri, 07 Feb 2003 15:43:40 +0530
From: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: bidulock@openss7.org
CC: sigtran@ietf.org
Content-Type: text/plain; charset=x-user-defined
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] [M3UA] ASP Active Ack message
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Brian,

   In RFC 3332, Section 4.3.4.3 ASP Active Procedures, say the
following:

.....Multiple ASP Active Ack messages MAY be used in response to an ASP
Active message containing multiple Routing Contexts, allowing the SGP or
IPSP to independently acknowledge the ASP Active message for different
(sets of) Routing Contexts......

.....When the ASP sends an ASP Active message it starts timer T(ack).
If the ASP does not receive a response to an ASP Active message within
T(ack), the ASP MAY restart T(ack) and resend ASP Active messages until
it receives an ASP Active Ack message......

Can you please clarify my following doubt,

"As we can expect more than one ASP Active Ack messages for a single ASP
Active message", at what moment we can declare that we have received a
response and can cancel the T(ack) timer. (Is it after the first ASP
Active Ack or last ASP Active Ack?. Moreover unless we keep track of the
Routing Contexts we cannot say the last ASP Active Ack has arrived).

--Srini

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 05:58:31 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA04743
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 05:58:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17B5Oe25416
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 06:05:24 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17B5Cp25408;
	Fri, 7 Feb 2003 06:05:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17B44p25344
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 06:04:04 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA04708
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 05:56:39 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:7lnx2G/kydhe//WCnkS2zJ5YeUEf9hVP@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h17B0ID04362;
	Fri, 7 Feb 2003 04:00:18 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h17B0Hm20757;
	Fri, 7 Feb 2003 04:00:17 -0700
Date: Fri, 7 Feb 2003 04:00:17 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
Cc: sigtran@ietf.org
Message-ID: <20030207040017.A19578@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>,
	sigtran@ietf.org
References: <3E4386D3.83D2B8EF@axes-mach01.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E4386D3.83D2B8EF@axes-mach01.usa.alcatel.com>; from ksrini@axes-mach01.usa.alcatel.com on Fri, Feb 07, 2003 at 03:43:40PM +0530
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Subject: [Sigtran] Re: [M3UA] ASP Active Ack message
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Srini,

I see four obvious approaches:

1)  don't send multiple RC in ASPAC (you are not required to).

2)  start one T(ack) timer and resend an ASPAC with only those
    RC not acknowledged active at the time that T(ack) expires.
    Only stop timer when all RCs are acknowledged.

3)  start one T(ack) timer and resend the original ASPAC with
    all RCs if any RC not acknowledged.  (This will solicit
    ASPAC Ack for AS which are already active, but who cares?)
    Only stop timer when all RCs are acknowledged.

4)  start one T(ack) timer for each RC in the multiple RC
    ASPAC.  If any timer expires, resend indivitual ASPAC with
    only that RC.  Only stop each timer when each RC is
    acknowledged.

There are probably others.


I prefer (1).  It sounds like you will too.


--brian


ksrinivas wrote:                                                                 (Fri, 07 Feb 2003 15:43:40)
> Brian,
> 
>    In RFC 3332, Section 4.3.4.3 ASP Active Procedures, say the
> following:
> 
> .....Multiple ASP Active Ack messages MAY be used in response to an ASP
> Active message containing multiple Routing Contexts, allowing the SGP or
> IPSP to independently acknowledge the ASP Active message for different
> (sets of) Routing Contexts......
> 
> .....When the ASP sends an ASP Active message it starts timer T(ack).
> If the ASP does not receive a response to an ASP Active message within
> T(ack), the ASP MAY restart T(ack) and resend ASP Active messages until
> it receives an ASP Active Ack message......
> 
> Can you please clarify my following doubt,
> 
> "As we can expect more than one ASP Active Ack messages for a single ASP
> Active message", at what moment we can declare that we have received a
> response and can cancel the T(ack) timer. (Is it after the first ASP
> Active Ack or last ASP Active Ack?. Moreover unless we keep track of the
> Routing Contexts we cannot say the last ASP Active Ack has arrived).
> 
> --Srini

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 06:05:27 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA04929
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 06:05:27 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17BCKo26280
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 06:12:20 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17BC6p26271;
	Fri, 7 Feb 2003 06:12:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17BBVp26250
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 06:11:31 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA04844
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 06:04:06 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:GbFy0ddyC8oV2A3Laa3l1tRuzBFensTy@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h17B7iD04912;
	Fri, 7 Feb 2003 04:07:44 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h17B7fZ22243;
	Fri, 7 Feb 2003 04:07:41 -0700
Date: Fri, 7 Feb 2003 04:07:41 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] Re: [M3UA] ASP Active Ack message
Message-ID: <20030207040741.B19578@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>,
	sigtran@ietf.org
References: <3E4386D3.83D2B8EF@axes-mach01.usa.alcatel.com> <20030207040017.A19578@openss7.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20030207040017.A19578@openss7.org>; from bidulock@openss7.org on Fri, Feb 07, 2003 at 04:00:17AM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Srini,

I should mention that the other reason I prefer (1) is that
(per option in 4.2.1) I send ASPAC and ASPAC Ack ordered on
the stream that will carry the DATA messages to avoid message
loss during activation.  As I usually send DATA for one RC on
a different stream than other RC, the only way to place each
ASPAC/Ack message on its own stream is by sending only one RC
in the ASPAC/Ack.

Because IMO messages containing RC (DATA/ASPTM/SSNM) should
always be sent on the stream upon which the DATA is sent, I see
little use in sending multiple RCs in any of those messages.

--brian

Brian F. G. Bidulock wrote:                                                      (Fri, 07 Feb 2003 04:00:17)
> Srini,
> 
> I see four obvious approaches:
> 
> 1)  don't send multiple RC in ASPAC (you are not required to).
> 
> 2)  start one T(ack) timer and resend an ASPAC with only those
>     RC not acknowledged active at the time that T(ack) expires.
>     Only stop timer when all RCs are acknowledged.
> 
> 3)  start one T(ack) timer and resend the original ASPAC with
>     all RCs if any RC not acknowledged.  (This will solicit
>     ASPAC Ack for AS which are already active, but who cares?)
>     Only stop timer when all RCs are acknowledged.
> 
> 4)  start one T(ack) timer for each RC in the multiple RC
>     ASPAC.  If any timer expires, resend indivitual ASPAC with
>     only that RC.  Only stop each timer when each RC is
>     acknowledged.
> 
> There are probably others.
> 
> 
> I prefer (1).  It sounds like you will too.
> 
> 
> --brian
> 
> 
> ksrinivas wrote:                                                                 (Fri, 07 Feb 2003 15:43:40)
> > Brian,
> > 
> >    In RFC 3332, Section 4.3.4.3 ASP Active Procedures, say the
> > following:
> > 
> > .....Multiple ASP Active Ack messages MAY be used in response to an ASP
> > Active message containing multiple Routing Contexts, allowing the SGP or
> > IPSP to independently acknowledge the ASP Active message for different
> > (sets of) Routing Contexts......
> > 
> > .....When the ASP sends an ASP Active message it starts timer T(ack).
> > If the ASP does not receive a response to an ASP Active message within
> > T(ack), the ASP MAY restart T(ack) and resend ASP Active messages until
> > it receives an ASP Active Ack message......
> > 
> > Can you please clarify my following doubt,
> > 
> > "As we can expect more than one ASP Active Ack messages for a single ASP
> > Active message", at what moment we can declare that we have received a
> > response and can cancel the T(ack) timer. (Is it after the first ASP
> > Active Ack or last ASP Active Ack?. Moreover unless we keep track of the
> > Routing Contexts we cannot say the last ASP Active Ack has arrived).
> > 
> > --Srini
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 06:17:48 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA05182
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 06:17:48 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17BOg026691
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 06:24:42 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17BORp26675;
	Fri, 7 Feb 2003 06:24:29 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17BNap26649
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 06:23:36 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA05155
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 06:16:10 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h17BJkeW001320;
	Fri, 7 Feb 2003 05:19:48 -0600 (CST)
Received: from axsppc51 by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id PAA03650; Fri, 7 Feb 2003 15:48:44 -0500 (GMT)
Message-ID: <00b801c2ce9b$4806e7a0$6501ff0a@axsppc51>
From: "Manjunath" <manju@axes-mach01.usa.alcatel.com>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        <sigtran@ietf.org>
References: <Pine.SOL.4.10.10302052218190.22291-100000@sun3271.ssd.usa.alcatel.com>
Date: Fri, 7 Feb 2003 16:52:46 +0530
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 5.50.4133.2400
X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] ASP database equalization quaked off this question....
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

hi all,

> section 1.4.1
>
> ...
>    Where a particular AS can be reached via more than one SGP, the
>    corresponding Routing Keys in the SGPs should be identical.  (Note:
>    It is possible for the SGP Routing Key configuration data to be
>    temporarily out-of-sync during configuration updates).
>
>                               +--------+
>                               |        |
>                  +------------+  SG 1  +--------------+
>      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
>      |  SEP  +---+            +--------+              +---/      \
>      |   or  |                    |*                      | ASPs  |
>      |  STP  +---+            +--------+              +---\      /
>      +-------+   |            |        |              |     ----
>                  +------------+  SG 2  +--------------+
>                               | "STP"  |
>                               +--------+
>
>                     Figure 1  Example with mated SGs
>
> Here SGP1 is in SG1 and SGP2 is in SG2. Basically mated SGs are located
> geographically in different places.The ideas is when SG1 goes down in
> CITY1(due to earthquake...) , all the traffice still can be handled by SG2
> which is in CITY2.
>
> SGP2 is mate for SGP1, On failure of SGP1, all the messages will be
> routed from SGP1 to SGP2.
>
> Does all these things mean to say that SGP1's mate 'should be' present in
> SG2 and not in SG1 itself ?
>
>
> Comments are welcome.
>

please give  comments for this question ,so  my doubts
regarding the configuration of SG's , limiting redundant SGP's for an AS
to be configured only on different SG's  can be cleared ....
  Somebody throw some light regarding this!!

thanks
manju/sibi




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 06:48:30 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA05832
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 06:48:30 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17BtP228386
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 06:55:25 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17BtCp28362;
	Fri, 7 Feb 2003 06:55:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17Bt0p28341
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 06:55:00 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA05797
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 06:47:33 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:3OaggNSymQ11gIzWw9/lD0b8PEc1XVdY@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h17BpDD05523;
	Fri, 7 Feb 2003 04:51:13 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h17BpCm27362;
	Fri, 7 Feb 2003 04:51:12 -0700
Date: Fri, 7 Feb 2003 04:51:12 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Manjunath <manju@axes-mach01.usa.alcatel.com>
Cc: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        sigtran@ietf.org
Subject: Re: [Sigtran] ASP database equalization quaked off this question....
Message-ID: <20030207045112.A26702@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Manjunath <manju@axes-mach01.usa.alcatel.com>,
	"Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	sigtran@ietf.org
References: <Pine.SOL.4.10.10302052218190.22291-100000@sun3271.ssd.usa.alcatel.com> <00b801c2ce9b$4806e7a0$6501ff0a@axsppc51>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <00b801c2ce9b$4806e7a0$6501ff0a@axsppc51>; from manju@axes-mach01.usa.alcatel.com on Fri, Feb 07, 2003 at 04:52:46PM +0530
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Manjunath,

Manjunath wrote:                                                                 (Fri, 07 Feb 2003 16:52:46)
> >
> > Does all these things mean to say that SGP1's mate 'should be' present in
> > SG2 and not in SG1 itself ?
> >

No.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 06:53:19 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA05919
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 06:53:19 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17C0Ec28530
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 07:00:14 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17C03p28520;
	Fri, 7 Feb 2003 07:00:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17Bxtp28477
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 06:59:55 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA05901
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 06:52:29 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h17Bu5JT006359;
	Fri, 7 Feb 2003 05:56:07 -0600 (CST)
Received: from axes-mach01.usa.alcatel.com by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id QAA04714; Fri, 7 Feb 2003 16:25:13 -0500 (GMT)
Message-ID: <3E439F11.731D832A@axes-mach01.usa.alcatel.com>
Date: Fri, 07 Feb 2003 17:27:05 +0530
From: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: bidulock@openss7.org
CC: sigtran@ietf.org
References: <3E4386D3.83D2B8EF@axes-mach01.usa.alcatel.com> <20030207040017.A19578@openss7.org>
Content-Type: text/plain; charset=x-user-defined
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] Re: [M3UA] ASP Active Ack message
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Srini,

    I feel the second approach is the better one. As we cannot send ASPAC without multiple RCs always.

--Srini

"Brian F. G. Bidulock" wrote:

> Srini,
>
> I see four obvious approaches:
>
> 1)  don't send multiple RC in ASPAC (you are not required to).
>
> 2)  start one T(ack) timer and resend an ASPAC with only those
>     RC not acknowledged active at the time that T(ack) expires.
>     Only stop timer when all RCs are acknowledged.
>
> 3)  start one T(ack) timer and resend the original ASPAC with
>     all RCs if any RC not acknowledged.  (This will solicit
>     ASPAC Ack for AS which are already active, but who cares?)
>     Only stop timer when all RCs are acknowledged.
>
> 4)  start one T(ack) timer for each RC in the multiple RC
>     ASPAC.  If any timer expires, resend indivitual ASPAC with
>     only that RC.  Only stop each timer when each RC is
>     acknowledged.
>
> There are probably others.
>
> I prefer (1).  It sounds like you will too.
>
> --brian
>
> ksrinivas wrote:                                                                 (Fri, 07 Feb 2003 15:43:40)
> > Brian,
> >
> >    In RFC 3332, Section 4.3.4.3 ASP Active Procedures, say the
> > following:
> >
> > .....Multiple ASP Active Ack messages MAY be used in response to an ASP
> > Active message containing multiple Routing Contexts, allowing the SGP or
> > IPSP to independently acknowledge the ASP Active message for different
> > (sets of) Routing Contexts......
> >
> > .....When the ASP sends an ASP Active message it starts timer T(ack).
> > If the ASP does not receive a response to an ASP Active message within
> > T(ack), the ASP MAY restart T(ack) and resend ASP Active messages until
> > it receives an ASP Active Ack message......
> >
> > Can you please clarify my following doubt,
> >
> > "As we can expect more than one ASP Active Ack messages for a single ASP
> > Active message", at what moment we can declare that we have received a
> > response and can cancel the T(ack) timer. (Is it after the first ASP
> > Active Ack or last ASP Active Ack?. Moreover unless we keep track of the
> > Routing Contexts we cannot say the last ASP Active Ack has arrived).
> >
> > --Srini
>
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 06:57:21 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA06025
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 06:57:20 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17C4F428715
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 07:04:15 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17C44p28707;
	Fri, 7 Feb 2003 07:04:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17C3op28678
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 07:03:50 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA06015
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 06:56:24 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h17Bxlqg026837;
	Fri, 7 Feb 2003 05:59:49 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h17BwwM28473;
	Fri, 7 Feb 2003 05:58:58 -0600 (CST)
Date: Fri, 7 Feb 2003 05:58:58 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: Manjunath <manju@axes-mach01.usa.alcatel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] ASP database equalization quaked off this question....
In-Reply-To: <20030207045112.A26702@openss7.org>
Message-ID: <Pine.SOL.4.10.10302070558330.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,


  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Fri, 7 Feb 2003, Brian F. G. Bidulock wrote:

> Manjunath,
> 
> Manjunath wrote:                                                                 (Fri, 07 Feb 2003 16:52:46)
> > >
> > > Does all these things mean to say that SGP1's mate 'should be' present in
> > > SG2 and not in SG1 itself ?
> > >
> 
> No.
Why?

--Samuel
> 
> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 07:10:28 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA06439
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 07:10:28 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17CHN429795
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 07:17:23 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17CHBp29784;
	Fri, 7 Feb 2003 07:17:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17CGap29755
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 07:16:36 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA06296
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 07:09:09 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:bBhM3x0PCWc610JsvFzU2F778mz3lraK@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h17CCmD05837;
	Fri, 7 Feb 2003 05:12:48 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h17CCmf29967;
	Fri, 7 Feb 2003 05:12:48 -0700
Date: Fri, 7 Feb 2003 05:12:48 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: Manjunath <manju@axes-mach01.usa.alcatel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] ASP database equalization quaked off this question....
Message-ID: <20030207051248.A29762@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	Manjunath <manju@axes-mach01.usa.alcatel.com>, sigtran@ietf.org
References: <20030207045112.A26702@openss7.org> <Pine.SOL.4.10.10302070558330.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302070558330.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Fri, Feb 07, 2003 at 05:58:58AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 07 Feb 2003 05:58:58)
> 
> Brian,
> 
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Fri, 7 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Manjunath,
> > 
> > Manjunath wrote:                                                                 (Fri, 07 Feb 2003 16:52:46)
> > > >
> > > > Does all these things mean to say that SGP1's mate 'should be' present in
> > > > SG2 and not in SG1 itself ?
> > > >
> > 
> > No.
> Why?
SGP's don't have mates.

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 07:12:19 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA06483
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 07:12:19 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17CJFG29861
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 07:19:15 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17CJ4p29854;
	Fri, 7 Feb 2003 07:19:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17CIKp29829
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 07:18:20 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA06462
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 07:10:54 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:OYX9v3VnLI/xMs1HoE+59yLm2aQHfILe@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h17CEXD05859;
	Fri, 7 Feb 2003 05:14:33 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h17CEXU30176;
	Fri, 7 Feb 2003 05:14:33 -0700
Date: Fri, 7 Feb 2003 05:14:33 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
Cc: sigtran@ietf.org
Message-ID: <20030207051433.B29762@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>,
	sigtran@ietf.org
References: <3E4386D3.83D2B8EF@axes-mach01.usa.alcatel.com> <20030207040017.A19578@openss7.org> <3E439F11.731D832A@axes-mach01.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E439F11.731D832A@axes-mach01.usa.alcatel.com>; from ksrini@axes-mach01.usa.alcatel.com on Fri, Feb 07, 2003 at 05:27:05PM +0530
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Subject: [Sigtran] Re: [M3UA] ASP Active Ack message
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

ksrinivas,

Alrighty then.

--brian

ksrinivas wrote:                                                                 (Fri, 07 Feb 2003 17:27:05)
> Srini,
> 
>     I feel the second approach is the better one. As we cannot send ASPAC without multiple RCs always.
> 
> --Srini
> 
> "Brian F. G. Bidulock" wrote:
> 
> > Srini,
> >
> > I see four obvious approaches:
> >
> > 1)  don't send multiple RC in ASPAC (you are not required to).
> >
> > 2)  start one T(ack) timer and resend an ASPAC with only those
> >     RC not acknowledged active at the time that T(ack) expires.
> >     Only stop timer when all RCs are acknowledged.
> >
> > 3)  start one T(ack) timer and resend the original ASPAC with
> >     all RCs if any RC not acknowledged.  (This will solicit
> >     ASPAC Ack for AS which are already active, but who cares?)
> >     Only stop timer when all RCs are acknowledged.
> >
> > 4)  start one T(ack) timer for each RC in the multiple RC
> >     ASPAC.  If any timer expires, resend indivitual ASPAC with
> >     only that RC.  Only stop each timer when each RC is
> >     acknowledged.
> >
> > There are probably others.
> >
> > I prefer (1).  It sounds like you will too.
> >
> > --brian
> >
> > ksrinivas wrote:                                                                 (Fri, 07 Feb 2003 15:43:40)
> > > Brian,
> > >
> > >    In RFC 3332, Section 4.3.4.3 ASP Active Procedures, say the
> > > following:
> > >
> > > .....Multiple ASP Active Ack messages MAY be used in response to an ASP
> > > Active message containing multiple Routing Contexts, allowing the SGP or
> > > IPSP to independently acknowledge the ASP Active message for different
> > > (sets of) Routing Contexts......
> > >
> > > .....When the ASP sends an ASP Active message it starts timer T(ack).
> > > If the ASP does not receive a response to an ASP Active message within
> > > T(ack), the ASP MAY restart T(ack) and resend ASP Active messages until
> > > it receives an ASP Active Ack message......
> > >
> > > Can you please clarify my following doubt,
> > >
> > > "As we can expect more than one ASP Active Ack messages for a single ASP
> > > Active message", at what moment we can declare that we have received a
> > > response and can cancel the T(ack) timer. (Is it after the first ASP
> > > Active Ack or last ASP Active Ack?. Moreover unless we keep track of the
> > > Routing Contexts we cannot say the last ASP Active Ack has arrived).
> > >
> > > --Srini
> >
> > --
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 07:18:38 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA06643
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 07:18:38 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17CPYD30123
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 07:25:34 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17CPKp30098;
	Fri, 7 Feb 2003 07:25:20 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17COnp30073
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 07:24:49 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA06621
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 07:17:22 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h17CKlqe028826;
	Fri, 7 Feb 2003 06:20:47 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h17CKWM06440;
	Fri, 7 Feb 2003 06:20:32 -0600 (CST)
Date: Fri, 7 Feb 2003 06:20:32 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: Manjunath <manju@axes-mach01.usa.alcatel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] ASP database equalization quaked off this question....
In-Reply-To: <20030207051248.A29762@openss7.org>
Message-ID: <Pine.SOL.4.10.10302070615180.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Fri, 7 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 07 Feb 2003 05:58:58)
> > 
> > Brian,
> > 
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Fri, 7 Feb 2003, Brian F. G. Bidulock wrote:
> > 
> > > Manjunath,
> > > 
> > > Manjunath wrote:                                                                 (Fri, 07 Feb 2003 16:52:46)
> > > > >
> > > > > Does all these things mean to say that SGP1's mate 'should be' present in
> > > > > SG2 and not in SG1 itself ?
> > > > >
> > > 
> > > No.
> > Why?
> SGP's don't have mates.

Please See section 1.4.1  

...
   Where a particular AS can be reached via more than one SGP, the
   corresponding Routing Keys in the SGPs should be identical.  (Note:
   It is possible for the SGP Routing Key configuration data to be
   temporarily out-of-sync during configuration updates).
 
                              +--------+
                              |        |
                 +------------+  SG 1  +--------------+
     +-------+   |  SS7 links | "STP"  |  IP network  |     ----
     |  SEP  +---+            +--------+              +---/      \
     |   or  |                    |*                      | ASPs  |
     |  STP  +---+            +--------+              +---\      /
     +-------+   |            |        |              |     ----
                 +------------+  SG 2  +--------------+
                              | "STP"  |
                              +--------+
 
                    Figure 1  Example with mated SGs
 
   * Note:.  SG-to-SG communication (i.e., "C-links") is recommended for
   carrier grade networks, using an MTP3 linkset or an equivalent, to
   allow rerouting between the SGs in the event of route failures. Where
   SGPs are used, inter-SGP communication might be used.  Inter-SGP
   protocol is outside of the scope of this document.


Thanks,
Samuel

> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 07:22:24 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA06715
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 07:22:24 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17CTJf30237
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 07:29:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17CT7p30228;
	Fri, 7 Feb 2003 07:29:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17CSep30198
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 07:28:40 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA06690
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 07:21:13 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:1GoAhK+AudKHeHjUzxI2/w1yTj5zd9fA@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h17COqD05978;
	Fri, 7 Feb 2003 05:24:53 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h17COq231400;
	Fri, 7 Feb 2003 05:24:52 -0700
Date: Fri, 7 Feb 2003 05:24:52 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: Manjunath <manju@axes-mach01.usa.alcatel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] ASP database equalization quaked off this question....
Message-ID: <20030207052452.A31052@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	Manjunath <manju@axes-mach01.usa.alcatel.com>, sigtran@ietf.org
References: <20030207051248.A29762@openss7.org> <Pine.SOL.4.10.10302070615180.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302070615180.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Fri, Feb 07, 2003 at 06:20:32AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 07 Feb 2003 06:20:32)
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Fri, 7 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 07 Feb 2003 05:58:58)
> > > 
> > > Brian,
> > > 
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Fri, 7 Feb 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Manjunath,
> > > > 
> > > > Manjunath wrote:                                                                 (Fri, 07 Feb 2003 16:52:46)
> > > > > >
> > > > > > Does all these things mean to say that SGP1's mate 'should be' present in
> > > > > > SG2 and not in SG1 itself ?
> > > > > >
> > > > 
> > > > No.
> > > Why?
> > SGP's don't have mates.
> 
> Please See section 1.4.1  
> 
> ...
>    Where a particular AS can be reached via more than one SGP, the
>    corresponding Routing Keys in the SGPs should be identical.  (Note:
>    It is possible for the SGP Routing Key configuration data to be
>    temporarily out-of-sync during configuration updates).
>  
>                               +--------+
>                               |        |
>                  +------------+  SG 1  +--------------+
>      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
>      |  SEP  +---+            +--------+              +---/      \
>      |   or  |                    |*                      | ASPs  |
>      |  STP  +---+            +--------+              +---\      /
>      +-------+   |            |        |              |     ----
>                  +------------+  SG 2  +--------------+
>                               | "STP"  |
>                               +--------+
>  
>                     Figure 1  Example with mated SGs
>  
>    * Note:.  SG-to-SG communication (i.e., "C-links") is recommended for
>    carrier grade networks, using an MTP3 linkset or an equivalent, to
>    allow rerouting between the SGs in the event of route failures. Where
>    SGPs are used, inter-SGP communication might be used.  Inter-SGP
>    protocol is outside of the scope of this document.
> 

SGPs don't have mates.
^^^^

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 07:27:23 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA06824
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 07:27:23 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17CYJJ30425
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 07:34:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17CY5p30418;
	Fri, 7 Feb 2003 07:34:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17CXFp30380
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 07:33:15 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA06799
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 07:25:43 -0500 (EST)
Subject: Re: [Sigtran] Packing M3UA messages into SCTP ULPDUs
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF6BCC0126.30EBEE0A-ONC2256CC6.004440AC@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Fri, 7 Feb 2003 14:25:51 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/07/2003
 02:26:00 PM
MIME-Version: 1.0
Content-type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h17CXGp30381
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit


resending......


I also think that it is a good idea to clarify this in IG.

 Tolga
--- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
> Tolga,
>
> It does not even forbid splitting an M3UA message
> across multiple
> SCTP ULPDUs.
>
> I think this should be clarified in the IGs for the
> sake of
> interoperablity.  1 M3UA (M2UA, SUA) message to 1
> SCTP ULPDU.
>
> TOLGA ASVEREN wrote:
>                                 (Fri, 07 Feb 2003
> 11:05:52)
> >
> > Brian,
> > > Sigtran,
> > >
> > > Where does it say that only one M3UA message may
> be
> > > placed into an SCTP ULPDU?
> > >
> > > As messages contain the length in the header,
> can I
> > > pack two or more M3UA messages in an SCTP ULPDU?
> > >
> > > This is so basic, I wonder why it was not
> mentioned
> > > anywhere...
> > If it is not forbidden, shouldn´t it be
> interpreted as allowed (although I
> > would guess that most of the implementations are
> assuming one M3UA message
> > in one SCTP ULPDU)?
> > BTW related with this topic:
> > from 3.1.4
> >   "The Message Length defines the length of the
> message in octets,
> >    including the Common Header.  The Message
> Length MUST include
> >    parameter padding bytes, if any.
> >
> >    Note: A receiver SHOULD accept the message
> whether or not the final
> >    parameter padding is included in the message
> length."
> > The above paragraphs sound strange to me,
> including padding bytes in the
> > length is a MUST but still receiver should be
> prepared for the opposite.
> > But still based on that I assume a M3UA peer can
> receive messages, whose
> > length does not contain padding bytes (otherwise
> forget what stands below)
> > >From 3.2
> >      "The total length of a parameter (including
> Tag, Parameter Length
> >       and Value fields) MUST be a multiple of 4
> bytes.  If the length of
> >       the parameter is not a multiple of 4 bytes,
> the sender pads the
> >       Parameter at the end (i.e., after the
> Parameter Value field) with
> >       all zero bytes.  The length of the padding
> is NOT included in the
> >       parameter length field.  A sender SHOULD NOT
> pad with more than 3
> >       bytes.  The receiver MUST ignore the padding
> bytes."
> > Not to add more than 3 padding bytes is a SHOULD
> NOT, not a MUST.
> >
> > For a scenario where message length does not
> contain padding bytes length
> > and more than 3 padding bytes are used, I would
> think receiver can´t detect
> > boundary between two M3UA messages (but I have to
> repeat, it sounds strange
> > to me that including padding bytes length in
> message length is a MUST and
> > still receiver SHOULD accept messages, which
> aren´t prepared in that way)
>
> One could still detect the next message by scanning
> for the next valid
> M3UA message header...
>
> Also, that wouldn't cause any problem for splitting
> an M3UA message across
> several SCTP ULPDUs.
>
> I think we need to state in the IG: 1 UA message, 1
> SCTP ULPDU.



_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 07:40:22 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA07107
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 07:40:22 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17ClIn31542
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 07:47:18 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17Cl6p31531;
	Fri, 7 Feb 2003 07:47:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17Cknp31513
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 07:46:49 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA07065
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 07:39:21 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h17CglJT010304;
	Fri, 7 Feb 2003 06:42:47 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h17Cg4M13705;
	Fri, 7 Feb 2003 06:42:04 -0600 (CST)
Date: Fri, 7 Feb 2003 06:42:04 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: Manjunath <manju@axes-mach01.usa.alcatel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] ASP database equalization quaked off this question....
In-Reply-To: <20030207052452.A31052@openss7.org>
Message-ID: <Pine.SOL.4.10.10302070638200.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Fri, 7 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 07 Feb 2003 06:20:32)
> > 
> > Brian,
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Fri, 7 Feb 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 07 Feb 2003 05:58:58)
> > > > 
> > > > Brian,
> > > > 
> > > > 
> > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > 
> > > > On Fri, 7 Feb 2003, Brian F. G. Bidulock wrote:
> > > > 
> > > > > Manjunath,
> > > > > 
> > > > > Manjunath wrote:                                                                 (Fri, 07 Feb 2003 16:52:46)
> > > > > > >
> > > > > > > Does all these things mean to say that SGP1's mate 'should be' present in
> > > > > > > SG2 and not in SG1 itself ?
> > > > > > >
> > > > > 
> > > > > No.
> > > > Why?
> > > SGP's don't have mates.
> > 
> > Please See section 1.4.1  
> > 
> > ...
> >    Where a particular AS can be reached via more than one SGP, the
> >    corresponding Routing Keys in the SGPs should be identical.  (Note:
> >    It is possible for the SGP Routing Key configuration data to be
> >    temporarily out-of-sync during configuration updates).
> >  
> >                               +--------+
> >                               |        |
> >                  +------------+  SG 1  +--------------+
> >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> >      |  SEP  +---+            +--------+              +---/      \
> >      |   or  |                    |*                      | ASPs  |
> >      |  STP  +---+            +--------+              +---\      /
> >      +-------+   |            |        |              |     ----
> >                  +------------+  SG 2  +--------------+
> >                               | "STP"  |
> >                               +--------+
> >  
> >                     Figure 1  Example with mated SGs
> >  
> >    * Note:.  SG-to-SG communication (i.e., "C-links") is recommended for
> >    carrier grade networks, using an MTP3 linkset or an equivalent, to
> >    allow rerouting between the SGs in the event of route failures. Where
> >    SGPs are used, inter-SGP communication might be used.  Inter-SGP
> >    protocol is outside of the scope of this document.
> > 
> 
> SGPs don't have mates.
> ^^^^

"to allow rerouting between the SGs in the event of route failures, Where
SGPs are used, inter-SGP communication might be used."

Only in the event of failures, inter-SGP communication is used.

It means SGP1(SG1) and SGP1(SG2) are serving same RK.

--Samuel

> 
> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 07:41:17 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA07165
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 07:41:17 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17CmD031588
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 07:48:13 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17Cm2p31581;
	Fri, 7 Feb 2003 07:48:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17Cl6p31525
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 07:47:06 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA07068
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 07:39:39 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:6k7HUgXasoHdWVJ95DfRm0bTEm8SeXW+@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h17ChID06416;
	Fri, 7 Feb 2003 05:43:18 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h17ChIR01157;
	Fri, 7 Feb 2003 05:43:18 -0700
Date: Fri, 7 Feb 2003 05:43:18 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - ASP database equalization
Message-ID: <20030207054318.A31507@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	sigtran@ietf.org
References: <Pine.SOL.4.10.10302052218190.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302052218190.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Wed, Feb 05, 2003 at 11:57:10PM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Wed, 05 Feb 2003 23:57:10)

[snip]
> 
> section 5.4.2.1
> 
> ...
>       - Pass the MTP-TRANSFER request primitive to the MTP3 of the
>         concerned Network Appearance.
> 
> It conveys the meaning that M3UA layer (at SGP) MTP-TRANSFER request is
> sent to MTP3.
> 
> But there is no primitive availabe between M3UA and MTP3.
> 
> I think it should be corrected.

The statement in section 5.4.2.1 is correct.  M3UA exports
the MTP/MTP-User interface from the SG to the ASP.  As such,
MTP/MTP-User primitives are available to the NIF.

> 
> ----
> Presently ASP is unaware of SG's MTP3 restart, 
> ASP receives only DUNA for all SS7 point codes which can be
> reachable via that SG.But in case of peer to peer protocol
> SG and AS point codes are different, So why dont we send
> DUNA with SG's point code to ASP? So that just by getting
> single DUNA(with SG's PC) ASP will not route any message via that SG.
> This can avoid sending multiple DUNAs to ASP.
> 
> 
> Any comments?

In the case that the SG's point code is different from the ASP,
the ASP's MTP is not restarting when the SG's PC is in MTP Restart.
Where the SG's PC is different from the ASP, there is never a
time that the need SG sends DUNA for its own point code to the ASP.
Anytime that the ASP is active, the SG's point code can be accessible
to it.

--brian


-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 07:42:17 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA07201
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 07:42:17 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17CnDv31648
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 07:49:13 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17Cn2p31633;
	Fri, 7 Feb 2003 07:49:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17CmVp31608
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 07:48:31 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA07154
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 07:41:04 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:LYuajMyV+tMdsPk5qr7GBBIxfyBx8M8g@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h17CihD06435;
	Fri, 7 Feb 2003 05:44:43 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h17Cih601368;
	Fri, 7 Feb 2003 05:44:43 -0700
Date: Fri, 7 Feb 2003 05:44:43 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: Manjunath <manju@axes-mach01.usa.alcatel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] ASP database equalization quaked off this question....
Message-ID: <20030207054443.B31507@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	Manjunath <manju@axes-mach01.usa.alcatel.com>, sigtran@ietf.org
References: <20030207052452.A31052@openss7.org> <Pine.SOL.4.10.10302070638200.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302070638200.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Fri, Feb 07, 2003 at 06:42:04AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 07 Feb 2003 06:42:04)
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Fri, 7 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 07 Feb 2003 06:20:32)
> > > 
> > > Brian,
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Fri, 7 Feb 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Samuel,
> > > > 
> > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 07 Feb 2003 05:58:58)
> > > > > 
> > > > > Brian,
> > > > > 
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > > 
> > > > > On Fri, 7 Feb 2003, Brian F. G. Bidulock wrote:
> > > > > 
> > > > > > Manjunath,
> > > > > > 
> > > > > > Manjunath wrote:                                                                 (Fri, 07 Feb 2003 16:52:46)
> > > > > > > >
> > > > > > > > Does all these things mean to say that SGP1's mate 'should be' present in
> > > > > > > > SG2 and not in SG1 itself ?
> > > > > > > >
> > > > > > 
> > > > > > No.
> > > > > Why?
> > > > SGP's don't have mates.
> > > 
> > > Please See section 1.4.1  
> > > 
> > > ...
> > >    Where a particular AS can be reached via more than one SGP, the
> > >    corresponding Routing Keys in the SGPs should be identical.  (Note:
> > >    It is possible for the SGP Routing Key configuration data to be
> > >    temporarily out-of-sync during configuration updates).
> > >  
> > >                               +--------+
> > >                               |        |
> > >                  +------------+  SG 1  +--------------+
> > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > >      |  SEP  +---+            +--------+              +---/      \
> > >      |   or  |                    |*                      | ASPs  |
> > >      |  STP  +---+            +--------+              +---\      /
> > >      +-------+   |            |        |              |     ----
> > >                  +------------+  SG 2  +--------------+
> > >                               | "STP"  |
> > >                               +--------+
> > >  
> > >                     Figure 1  Example with mated SGs
> > >  
> > >    * Note:.  SG-to-SG communication (i.e., "C-links") is recommended for
> > >    carrier grade networks, using an MTP3 linkset or an equivalent, to
> > >    allow rerouting between the SGs in the event of route failures. Where
> > >    SGPs are used, inter-SGP communication might be used.  Inter-SGP
> > >    protocol is outside of the scope of this document.
> > > 
> > 
> > SGPs don't have mates.
> > ^^^^
> 
> "to allow rerouting between the SGs in the event of route failures, Where
> SGPs are used, inter-SGP communication might be used."
> 
> Only in the event of failures, inter-SGP communication is used.
> 
> It means SGP1(SG1) and SGP1(SG2) are serving same RK.

SGPs still don't have mates, so the answer to your question is: No.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 08:06:24 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA08524
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 08:06:24 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17DDK901881
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 08:13:20 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17DD8p01869;
	Fri, 7 Feb 2003 08:13:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17DCop01826
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 08:12:50 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA08483
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 08:05:22 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h17D8lJT012856;
	Fri, 7 Feb 2003 07:08:47 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h17D8GM22612;
	Fri, 7 Feb 2003 07:08:16 -0600 (CST)
Date: Fri, 7 Feb 2003 07:08:16 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: Manjunath <manju@axes-mach01.usa.alcatel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] ASP database equalization quaked off this question....
In-Reply-To: <20030207054443.B31507@openss7.org>
Message-ID: <Pine.SOL.4.10.10302070704210.22291-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Fri, 7 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 07 Feb 2003 06:42:04)
> > 
> > Brian,
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Fri, 7 Feb 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 07 Feb 2003 06:20:32)
> > > > 
> > > > Brian,
> > > > 
> > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > 
> > > > On Fri, 7 Feb 2003, Brian F. G. Bidulock wrote:
> > > > 
> > > > > Samuel,
> > > > > 
> > > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 07 Feb 2003 05:58:58)
> > > > > > 
> > > > > > Brian,
> > > > > > 
> > > > > > 
> > > > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > > >   1000 Coit Road, Plano, Texas 75075
> > > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > > > 
> > > > > > On Fri, 7 Feb 2003, Brian F. G. Bidulock wrote:
> > > > > > 
> > > > > > > Manjunath,
> > > > > > > 
> > > > > > > Manjunath wrote:                                                                 (Fri, 07 Feb 2003 16:52:46)
> > > > > > > > >
> > > > > > > > > Does all these things mean to say that SGP1's mate 'should be' present in
> > > > > > > > > SG2 and not in SG1 itself ?
> > > > > > > > >
> > > > > > > 
> > > > > > > No.
> > > > > > Why?
> > > > > SGP's don't have mates.
> > > > 
> > > > Please See section 1.4.1  
> > > > 
> > > > ...
> > > >    Where a particular AS can be reached via more than one SGP, the
> > > >    corresponding Routing Keys in the SGPs should be identical.  (Note:
> > > >    It is possible for the SGP Routing Key configuration data to be
> > > >    temporarily out-of-sync during configuration updates).
> > > >  
> > > >                               +--------+
> > > >                               |        |
> > > >                  +------------+  SG 1  +--------------+
> > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > >      |  SEP  +---+            +--------+              +---/      \
> > > >      |   or  |                    |*                      | ASPs  |
> > > >      |  STP  +---+            +--------+              +---\      /
> > > >      +-------+   |            |        |              |     ----
> > > >                  +------------+  SG 2  +--------------+
> > > >                               | "STP"  |
> > > >                               +--------+
> > > >  
> > > >                     Figure 1  Example with mated SGs
> > > >  
> > > >    * Note:.  SG-to-SG communication (i.e., "C-links") is recommended for
> > > >    carrier grade networks, using an MTP3 linkset or an equivalent, to
> > > >    allow rerouting between the SGs in the event of route failures. Where
> > > >    SGPs are used, inter-SGP communication might be used.  Inter-SGP
> > > >    protocol is outside of the scope of this document.
> > > > 
> > > 
> > > SGPs don't have mates.
> > > ^^^^
> > 
> > "to allow rerouting between the SGs in the event of route failures, Where
> > SGPs are used, inter-SGP communication might be used."
> > 
> > Only in the event of failures, inter-SGP communication is used.
> > 
> > It means SGP1(SG1) and SGP1(SG2) are serving same RK.
> 
> SGPs still don't have mates, so the answer to your question is: No.
> 

When SGP1(SG1) goes down, SGP1(SG2) takes up the load(using
inter-SGP communication).Thats all i want to say.

--Samuel

> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 08:34:05 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10466
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 08:34:05 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17Df2s04036
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 08:41:02 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17Deip03994;
	Fri, 7 Feb 2003 08:40:46 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17Dcvp03754
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 08:38:57 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10347
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 08:31:27 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:H4039SVdtSm0Ynzn2AhBkcWBO8S9Bgqn@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h17DZ6D07462;
	Fri, 7 Feb 2003 06:35:06 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h17DZ5J07350;
	Fri, 7 Feb 2003 06:35:05 -0700
Date: Fri, 7 Feb 2003 06:35:04 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: Manjunath <manju@axes-mach01.usa.alcatel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] ASP database equalization quaked off this question....
Message-ID: <20030207063504.A7125@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	Manjunath <manju@axes-mach01.usa.alcatel.com>, sigtran@ietf.org
References: <20030207054443.B31507@openss7.org> <Pine.SOL.4.10.10302070704210.22291-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302070704210.22291-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Fri, Feb 07, 2003 at 07:08:16AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 07 Feb 2003 07:08:16)
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Fri, 7 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 07 Feb 2003 06:42:04)
> > > 
> > > Brian,
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Fri, 7 Feb 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Samuel,
> > > > 
> > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 07 Feb 2003 06:20:32)
> > > > > 
> > > > > Brian,
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > > 
> > > > > On Fri, 7 Feb 2003, Brian F. G. Bidulock wrote:
> > > > > 
> > > > > > Samuel,
> > > > > > 
> > > > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 07 Feb 2003 05:58:58)
> > > > > > > 
> > > > > > > Brian,
> > > > > > > 
> > > > > > > 
> > > > > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > > > >   1000 Coit Road, Plano, Texas 75075
> > > > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > > > > 
> > > > > > > On Fri, 7 Feb 2003, Brian F. G. Bidulock wrote:
> > > > > > > 
> > > > > > > > Manjunath,
> > > > > > > > 
> > > > > > > > Manjunath wrote:                                                                 (Fri, 07 Feb 2003 16:52:46)
> > > > > > > > > >
> > > > > > > > > > Does all these things mean to say that SGP1's mate 'should be' present in
> > > > > > > > > > SG2 and not in SG1 itself ?
> > > > > > > > > >
> > > > > > > > 
> > > > > > > > No.
> > > > > > > Why?
> > > > > > SGP's don't have mates.
> > > > > 
> > > > > Please See section 1.4.1  
> > > > > 
> > > > > ...
> > > > >    Where a particular AS can be reached via more than one SGP, the
> > > > >    corresponding Routing Keys in the SGPs should be identical.  (Note:
> > > > >    It is possible for the SGP Routing Key configuration data to be
> > > > >    temporarily out-of-sync during configuration updates).
> > > > >  
> > > > >                               +--------+
> > > > >                               |        |
> > > > >                  +------------+  SG 1  +--------------+
> > > > >      +-------+   |  SS7 links | "STP"  |  IP network  |     ----
> > > > >      |  SEP  +---+            +--------+              +---/      \
> > > > >      |   or  |                    |*                      | ASPs  |
> > > > >      |  STP  +---+            +--------+              +---\      /
> > > > >      +-------+   |            |        |              |     ----
> > > > >                  +------------+  SG 2  +--------------+
> > > > >                               | "STP"  |
> > > > >                               +--------+
> > > > >  
> > > > >                     Figure 1  Example with mated SGs
> > > > >  
> > > > >    * Note:.  SG-to-SG communication (i.e., "C-links") is recommended for
> > > > >    carrier grade networks, using an MTP3 linkset or an equivalent, to
> > > > >    allow rerouting between the SGs in the event of route failures. Where
> > > > >    SGPs are used, inter-SGP communication might be used.  Inter-SGP
> > > > >    protocol is outside of the scope of this document.
> > > > > 
> > > > 
> > > > SGPs don't have mates.
> > > > ^^^^
> > > 
> > > "to allow rerouting between the SGs in the event of route failures, Where
> > > SGPs are used, inter-SGP communication might be used."
> > > 
> > > Only in the event of failures, inter-SGP communication is used.
> > > 
> > > It means SGP1(SG1) and SGP1(SG2) are serving same RK.
> > 
> > SGPs still don't have mates, so the answer to your question is: No.
> > 
> 
> When SGP1(SG1) goes down, SGP1(SG2) takes up the load(using
> inter-SGP communication).Thats all i want to say.

No.  C-links are used.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 10:39:17 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA13724
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 10:39:17 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17FkGW11389
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 10:46:16 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17Fjqp11375;
	Fri, 7 Feb 2003 10:45:53 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17Fipp11311
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 10:44:51 -0500
Received: from hotmail.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA13665
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 10:37:21 -0500 (EST)
Received: from mail pickup service by hotmail.com with Microsoft SMTPSVC;
	 Fri, 7 Feb 2003 07:39:51 -0800
Received: from 128.251.168.94 by lw15fd.law15.hotmail.msn.com with HTTP;
	Fri, 07 Feb 2003 15:39:51 GMT
X-Originating-IP: [128.251.168.94]
From: "denim john" <denim_john@hotmail.com>
To: sigtran@ietf.org
Cc: j.javier.pastor@ericsson.com, kmorneau@cisco.com
Subject: Re: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
Date: Fri, 07 Feb 2003 15:39:51 +0000
Mime-Version: 1.0
Content-Type: text/plain; format=flowed
Message-ID: <F1269WrXdMfCgROTw3X0001b785@hotmail.com>
X-OriginalArrivalTime: 07 Feb 2003 15:39:51.0375 (UTC) FILETIME=[2754F5F0:01C2CEBF]
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

I would request any one of the authors to conclude this thread.
samuel's idea seems to be good but this will result in changes of m3ua and 
sua.

Regards,
Denim.

An efficent ASP will take up the load as soon as it comes up.
Inefficient ASP may send DAUD and will receive DRST(...),DUNA(...),
SCON(...).  So from the point of ASP following are the unnecessary
messages.
.SENDING DAUD
.RECEING DRST
.RECEIVING DUNA
.RECEIVING SCON

From the point of SGP these are the unnecessay messages.
.Receiving DAUD
.SENDING DRST
.SENDING DUNA
.SENDING SCON.
Thats what your argument leads.

--Samuel

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:

>Samuel,
>
>That is implementation dependent.  Sending SSNM to inactive ASPs restricts
>a dimension not restricted by the current and adequate procedure.  The
>SG cannot be required to send SSNM to inactive ASPs.  The RFC should not
>place a restriction on the number of inactive ASPs, because we cannot
>forsee all future dimensioning.  Procedures should scale as well as 
>possible.
>The existing procedure scales.  Sending SSNM to inactive ASPs does not.
>

>--brian
>
>Samuel Dur D. Jeyaseelan wrote:
(Thu, 06 Feb 2003 23:13:44)
> > > Brian,
> > > Yes , The RFC does not restrict the number of inactive ASPs.
> > But network configuration will restrict.
> > > --Samuel
> > >   Alcatel USA, Inc.                  Internet: 
><userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. 
>*******
> > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > > Samuel,

> > > > > The RFC does not restrict the number of inactive ASPs.
> > > > > --brian
> > > > > Samuel Dur D. Jeyaseelan wrote:
(Thu, 06 Feb 2003 22:40:48)
> > > > > > > Brian,
> > > > > > > Ok, tell me, where you will configure 1 ASP ACT and 199,000 
>ASPs INACT?
> > > > I want to know which network does this poor configuration.
> > > > > > > --Samuel
> > > > > > > > > >   Alcatel USA, Inc.                  Internet:
<userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
*******
> > > > > > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > > > > > > Samuel,
> > > > > > > > > The RFC does not restrict the number of inactive ASPs.
> > > > > > > > > --brian
> > > > > > > > > Samuel Dur D. Jeyaseelan wrote:
(Thu, 06 Feb 2003 22:23:12)
> > > > > > > > > > > Richard,
> > > > > > > > > > >   Alcatel USA, Inc.                  Internet:
<userid>@ssd.usa.alcatel.com
> > > > > >   1000 Coit Road, Plano, Texas 75075
> > > > > >   ******* The opinions expressed are not those of Alcatel USA, 
>Inc.
*******
> > > > > > > > > > > On Thu, 6 Feb 2003, Richard Adams wrote:
> > > > > > > > > > > > Brian,
> > > > > > > I agree with you. I think when the ASP becomes active, it 
>should
do an
> > > > > > > audit to update its knowledge of the SS7 network. The ASP 
>needs
to take
> > > > > > > the initiative to find out about the SS7 nodes etc. in which 
>it
is
> > > > > > > interested. Any other method of working would lead to 
>impossible
loading
> > > > > > > on the SG.
> > > > > > > > > > > Yes i agree but your suggestion is an indirect method.
> > > > > > I am proposing a direct method to send SSNM messages from SGP to
INACTIVE
> > > > > > ASP. When SGP sends DUNA or DRST to ACTVIE ASP, let him also 
>send a
copy
> > > > > > of the same message to INACTIVE ASP. I dont find any difficulty
here.
> > > > > > > > > > >  > > > > > > Of course, I hope 199,999 of your 
>inactive ASPs do not decide to
become
> > > > > > > active at the same time.....! > > > > > > > > > > 
>Theoritically your example sounds good but practically NOBODY will
> > > > > > configure 1 ASP ACTIVE and 199,999 ASPs INACTIVE.
> > > > > > > > > > > --Samuel
> > > > > > > > > > > > > > > > > > -- > > > > Brian F. G. Bidulock
> > > > > bidulock@openss7.org
> > > > > http://www.openss7.org/
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > > > > > -- > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >
>
>--
>Brian F. G. Bidulock
>bidulock@openss7.org
>http://www.openss7.org/
>










_________________________________________________________________
Help STOP SPAM with the new MSN 8 and get 2 months FREE*  
http://join.msn.com/?page=features/junkmail

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 11:05:20 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA14267
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 11:05:20 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17GCKI12907
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 11:12:20 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17GC5p12896;
	Fri, 7 Feb 2003 11:12:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17GBep12859
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 11:11:40 -0500
Received: from rtp-msg-core-1.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA14251
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 11:04:09 -0500 (EST)
Received: from cia.cisco.com (localhost [127.0.0.1])
	by rtp-msg-core-1.cisco.com (8.12.2/8.12.6) with ESMTP id h17G8ci8028444;
	Fri, 7 Feb 2003 11:08:38 -0500 (EST)
Received: from kmorneau-w2k.cisco.com (rtp-vpn2-269.cisco.com [10.82.241.13])
	by cia.cisco.com (Mirapoint)
	with ESMTP id ABO69596;
	Fri, 7 Feb 2003 10:57:44 -0500 (EST)
Message-Id: <4.3.2.7.2.20030207110631.04c39a88@cia.cisco.com>
X-Sender: kmorneau@cia.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Fri, 07 Feb 2003 11:07:41 -0500
To: "denim john" <denim_john@hotmail.com>, sigtran@ietf.org
From: Ken Morneault <kmorneau@cisco.com>
Subject: Re: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
Cc: j.javier.pastor@ericsson.com
In-Reply-To: <F1269WrXdMfCgROTw3X0001b785@hotmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>



It would be useful if Samuel proposes text changes that could be 
reviewed and, then, we could see if there was consensus for making
this change.

Regards,

Ken

At 03:39 PM 2/7/2003 +0000, denim john wrote:

>Brian,
>
>I would request any one of the authors to conclude this thread.
>samuel's idea seems to be good but this will result in changes of m3ua and sua.
>
>Regards,
>Denim.
>
>An efficent ASP will take up the load as soon as it comes up.
>Inefficient ASP may send DAUD and will receive DRST(...),DUNA(...),
>SCON(...).  So from the point of ASP following are the unnecessary
>messages.
>.SENDING DAUD
>.RECEING DRST
>.RECEIVING DUNA
>.RECEIVING SCON
>
> From the point of SGP these are the unnecessay messages.
>.Receiving DAUD
>.SENDING DRST
>.SENDING DUNA
>.SENDING SCON.
>Thats what your argument leads.
>
>--Samuel
>
> Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> 1000 Coit Road, Plano, Texas 75075
> ******* The opinions expressed are not those of Alcatel USA, Inc. *******
>
>On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
>
>>Samuel,
>>
>>That is implementation dependent.  Sending SSNM to inactive ASPs restricts
>>a dimension not restricted by the current and adequate procedure.  The
>>SG cannot be required to send SSNM to inactive ASPs.  The RFC should not
>>place a restriction on the number of inactive ASPs, because we cannot
>>forsee all future dimensioning.  Procedures should scale as well as possible.
>>The existing procedure scales.  Sending SSNM to inactive ASPs does not.
>
>>--brian
>>
>>Samuel Dur D. Jeyaseelan wrote:
>(Thu, 06 Feb 2003 23:13:44)
>>> > Brian,
>>> > Yes , The RFC does not restrict the number of inactive ASPs.
>>> But network configuration will restrict.
>>> > --Samuel
>>> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>>>   1000 Coit Road, Plano, Texas 75075
>>>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
>>> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
>>> > > Samuel,
>
>>> > > > The RFC does not restrict the number of inactive ASPs.
>>> > > > --brian
>>> > > > Samuel Dur D. Jeyaseelan wrote:
>(Thu, 06 Feb 2003 22:40:48)
>>> > > > > > Brian,
>>> > > > > > Ok, tell me, where you will configure 1 ASP ACT and 199,000 ASPs INACT?
>>> > > I want to know which network does this poor configuration.
>>> > > > > > --Samuel
>>> > > > > > > > >   Alcatel USA, Inc.                  Internet:
><userid>@ssd.usa.alcatel.com
>>> > >   1000 Coit Road, Plano, Texas 75075
>>> > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
>*******
>>> > > > > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
>>> > > > > > > Samuel,
>>> > > > > > > > The RFC does not restrict the number of inactive ASPs.
>>> > > > > > > > --brian
>>> > > > > > > > Samuel Dur D. Jeyaseelan wrote:
>(Thu, 06 Feb 2003 22:23:12)
>>> > > > > > > > > > Richard,
>>> > > > > > > > > >   Alcatel USA, Inc.                  Internet:
><userid>@ssd.usa.alcatel.com
>>> > > > >   1000 Coit Road, Plano, Texas 75075
>>> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
>*******
>>> > > > > > > > > > On Thu, 6 Feb 2003, Richard Adams wrote:
>>> > > > > > > > > > > Brian,
>>> > > > > > I agree with you. I think when the ASP becomes active, it should
>do an
>>> > > > > > audit to update its knowledge of the SS7 network. The ASP needs
>to take
>>> > > > > > the initiative to find out about the SS7 nodes etc. in which it
>is
>>> > > > > > interested. Any other method of working would lead to impossible
>loading
>>> > > > > > on the SG.
>>> > > > > > > > > > Yes i agree but your suggestion is an indirect method.
>>> > > > > I am proposing a direct method to send SSNM messages from SGP to
>INACTIVE
>>> > > > > ASP. When SGP sends DUNA or DRST to ACTVIE ASP, let him also send a
>copy
>>> > > > > of the same message to INACTIVE ASP. I dont find any difficulty
>here.
>>> > > > > > > > > >  > > > > > > Of course, I hope 199,999 of your inactive ASPs do not decide to
>become
>>> > > > > > active at the same time.....! > > > > > > > > > > Theoritically your example sounds good but practically NOBODY will
>>> > > > > configure 1 ASP ACTIVE and 199,999 ASPs INACTIVE.
>>> > > > > > > > > > --Samuel
>>> > > > > > > > > > > > > > > > > -- > > > > Brian F. G. Bidulock
>>> > > > bidulock@openss7.org
>>> > > > http://www.openss7.org/
>>> > > > _______________________________________________
>>> > > > Sigtran mailing list
>>> > > > Sigtran@ietf.org
>>> > > > https://www1.ietf.org/mailman/listinfo/sigtran
>>> > > > > > > > -- > > Brian F. G. Bidulock
>>> > bidulock@openss7.org
>>> > http://www.openss7.org/
>>> > _______________________________________________
>>> > Sigtran mailing list
>>> > Sigtran@ietf.org
>>> > https://www1.ietf.org/mailman/listinfo/sigtran
>>> >
>>
>>--
>>Brian F. G. Bidulock
>>bidulock@openss7.org
>>http://www.openss7.org/
>
>
>
>
>
>
>
>
>
>
>_________________________________________________________________
>Help STOP SPAM with the new MSN 8 and get 2 months FREE*  
>http://join.msn.com/?page=features/junkmail

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 13:59:31 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA18214
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 13:59:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17J6YM22654
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 14:06:34 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17J6Kp22616;
	Fri, 7 Feb 2003 14:06:20 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17J5Mp22560
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 14:05:22 -0500
Received: from domino2.nc.tekelec.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA18150
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 13:57:48 -0500 (EST)
From: sherry.karl@tekelec.com
Received: from domino4.nc.tekelec.com ([198.89.40.9])
          by domino2.nc.tekelec.com (Lotus Domino Release 5.0.10)
          with ESMTP id 2003020713551282:24859 ;
          Fri, 7 Feb 2003 13:55:12 -0500 
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.7  March 21, 2001
Message-ID: <OFA41BCE2E.4B19292A-ON85256CC6.0067F555@nc.tekelec.com>
Date: Fri, 7 Feb 2003 13:56:56 -0500
MIME-Version: 1.0
X-MIMETrack: Serialize by Router on Domino4/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/07/2003 01:57:15 PM,
	Itemize by SMTP Server on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/07/2003 01:55:12 PM,
	Serialize by Router on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/07/2003 01:55:15 PM,
	Serialize complete at 02/07/2003 01:55:15 PM
Content-type: text/plain; charset=us-ascii
Subject: [Sigtran] SCTP Test Tool
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

I was wondering if anyone is aware of any 3rd party M3UA/SCTP Test Tools?

Thanks


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 14:51:24 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA19557
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 14:51:24 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17JwTx26021
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 14:58:29 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17JwEp26003;
	Fri, 7 Feb 2003 14:58:16 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17JvSp25985
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 14:57:28 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA19526
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 14:49:42 -0500 (EST)
Subject: Re: [Sigtran] Packing M3UA messages into SCTP ULPDUs
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF4C7EEDF0.E54FB924-ONC2256CC6.0035C4A8@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Fri, 7 Feb 2003 11:48:47 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/07/2003
 09:50:07 PM
MIME-Version: 1.0
Content-type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h17JvSp25986
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit


I also think that it is a good idea to clarify this in IG.

 Tolga
--- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
> Tolga,
>
> It does not even forbid splitting an M3UA message
> across multiple
> SCTP ULPDUs.
>
> I think this should be clarified in the IGs for the
> sake of
> interoperablity.  1 M3UA (M2UA, SUA) message to 1
> SCTP ULPDU.
>
> TOLGA ASVEREN wrote:
>                                 (Fri, 07 Feb 2003
> 11:05:52)
> >
> > Brian,
> > > Sigtran,
> > >
> > > Where does it say that only one M3UA message may
> be
> > > placed into an SCTP ULPDU?
> > >
> > > As messages contain the length in the header,
> can I
> > > pack two or more M3UA messages in an SCTP ULPDU?
> > >
> > > This is so basic, I wonder why it was not
> mentioned
> > > anywhere...
> > If it is not forbidden, shouldn´t it be
> interpreted as allowed (although I
> > would guess that most of the implementations are
> assuming one M3UA message
> > in one SCTP ULPDU)?
> > BTW related with this topic:
> > from 3.1.4
> >   "The Message Length defines the length of the
> message in octets,
> >    including the Common Header.  The Message
> Length MUST include
> >    parameter padding bytes, if any.
> >
> >    Note: A receiver SHOULD accept the message
> whether or not the final
> >    parameter padding is included in the message
> length."
> > The above paragraphs sound strange to me,
> including padding bytes in the
> > length is a MUST but still receiver should be
> prepared for the opposite.
> > But still based on that I assume a M3UA peer can
> receive messages, whose
> > length does not contain padding bytes (otherwise
> forget what stands below)
> > >From 3.2
> >      "The total length of a parameter (including
> Tag, Parameter Length
> >       and Value fields) MUST be a multiple of 4
> bytes.  If the length of
> >       the parameter is not a multiple of 4 bytes,
> the sender pads the
> >       Parameter at the end (i.e., after the
> Parameter Value field) with
> >       all zero bytes.  The length of the padding
> is NOT included in the
> >       parameter length field.  A sender SHOULD NOT
> pad with more than 3
> >       bytes.  The receiver MUST ignore the padding
> bytes."
> > Not to add more than 3 padding bytes is a SHOULD
> NOT, not a MUST.
> >
> > For a scenario where message length does not
> contain padding bytes length
> > and more than 3 padding bytes are used, I would
> think receiver can´t detect
> > boundary between two M3UA messages (but I have to
> repeat, it sounds strange
> > to me that including padding bytes length in
> message length is a MUST and
> > still receiver SHOULD accept messages, which
> aren´t prepared in that way)
>
> One could still detect the next message by scanning
> for the next valid
> M3UA message header...
>
> Also, that wouldn't cause any problem for splitting
> an M3UA message across
> several SCTP ULPDUs.
>
> I think we need to state in the IG: 1 UA message, 1
> SCTP ULPDU.

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 16:59:36 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA22784
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 16:59:36 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17M6gN01370
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 17:06:42 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17M6Rp01362;
	Fri, 7 Feb 2003 17:06:27 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17M5Lp01325
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 17:05:21 -0500
Received: from ilsa.franken.de (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA22744
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 16:57:40 -0500 (EST)
Received: from lurchi.franken.de (p5080103F.dip.t-dialin.net [80.128.16.63])
	by ilsa.franken.de (Postfix) with ESMTP
	id 04A9A245ED; Fri,  7 Feb 2003 22:59:52 +0100 (CET)
Date: Fri, 7 Feb 2003 22:56:44 +0100
Subject: Re: [Sigtran] Packing M3UA messages into SCTP ULPDUs
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Mime-Version: 1.0 (Apple Message framework v551)
Cc: sigtran@ietf.org
To: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
From: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>
In-Reply-To: <OF4C7EEDF0.E54FB924-ONC2256CC6.0035C4A8@LocalDomain>
Message-Id: <0BD08F48-3AE7-11D7-8A8F-0030654C1AB6@lurchi.franken.de>
X-Mailer: Apple Mail (2.551)
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h17M5Lp01326
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Dear all,

as far as I know all adaptation layers MAY run over TCP. That is
why all messages have a length field. But when run over SCTP
it does not make any sense to me not to put exactly one adaptation
layer packet in one DATA chunk.

I think it is not 'unclear' but never stated explicitly. No one
send multiple upper layer packets in one DATA chunk, or half of
it, or 1 and a half. I also have never received a bug report for
ethereal stating that such behaviour is not supported. Ethereal
expects exactly one upper layer packet in one DATA chunk.

This SHOULD be clarified for ALL SIGTRAN adaptation layers, but does
not change the current understanding.

Best regards
Michael

On Friday, Feb 7, 2003, at 10:48 Europe/Berlin, TOLGA ASVEREN wrote:

>
> I also think that it is a good idea to clarify this in IG.
>
>  Tolga
> --- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
>> Tolga,
>>
>> It does not even forbid splitting an M3UA message
>> across multiple
>> SCTP ULPDUs.
>>
>> I think this should be clarified in the IGs for the
>> sake of
>> interoperablity.  1 M3UA (M2UA, SUA) message to 1
>> SCTP ULPDU.
>>
>> TOLGA ASVEREN wrote:
>>                                 (Fri, 07 Feb 2003
>> 11:05:52)
>>>
>>> Brian,
>>>> Sigtran,
>>>>
>>>> Where does it say that only one M3UA message may
>> be
>>>> placed into an SCTP ULPDU?
>>>>
>>>> As messages contain the length in the header,
>> can I
>>>> pack two or more M3UA messages in an SCTP ULPDU?
>>>>
>>>> This is so basic, I wonder why it was not
>> mentioned
>>>> anywhere...
>>> If it is not forbidden, shouldn´t it be
>> interpreted as allowed (although I
>>> would guess that most of the implementations are
>> assuming one M3UA message
>>> in one SCTP ULPDU)?
>>> BTW related with this topic:
>>> from 3.1.4
>>>   "The Message Length defines the length of the
>> message in octets,
>>>    including the Common Header.  The Message
>> Length MUST include
>>>    parameter padding bytes, if any.
>>>
>>>    Note: A receiver SHOULD accept the message
>> whether or not the final
>>>    parameter padding is included in the message
>> length."
>>> The above paragraphs sound strange to me,
>> including padding bytes in the
>>> length is a MUST but still receiver should be
>> prepared for the opposite.
>>> But still based on that I assume a M3UA peer can
>> receive messages, whose
>>> length does not contain padding bytes (otherwise
>> forget what stands below)
>>>> From 3.2
>>>      "The total length of a parameter (including
>> Tag, Parameter Length
>>>       and Value fields) MUST be a multiple of 4
>> bytes.  If the length of
>>>       the parameter is not a multiple of 4 bytes,
>> the sender pads the
>>>       Parameter at the end (i.e., after the
>> Parameter Value field) with
>>>       all zero bytes.  The length of the padding
>> is NOT included in the
>>>       parameter length field.  A sender SHOULD NOT
>> pad with more than 3
>>>       bytes.  The receiver MUST ignore the padding
>> bytes."
>>> Not to add more than 3 padding bytes is a SHOULD
>> NOT, not a MUST.
>>>
>>> For a scenario where message length does not
>> contain padding bytes length
>>> and more than 3 padding bytes are used, I would
>> think receiver can´t detect
>>> boundary between two M3UA messages (but I have to
>> repeat, it sounds strange
>>> to me that including padding bytes length in
>> message length is a MUST and
>>> still receiver SHOULD accept messages, which
>> aren´t prepared in that way)
>>
>> One could still detect the next message by scanning
>> for the next valid
>> M3UA message header...
>>
>> Also, that wouldn't cause any problem for splitting
>> an M3UA message across
>> several SCTP ULPDUs.
>>
>> I think we need to state in the IG: 1 UA message, 1
>> SCTP ULPDU.
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 17:08:17 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA23059
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 17:08:17 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17MFOK02255
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 17:15:24 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17MF9p02246;
	Fri, 7 Feb 2003 17:15:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17MEcp02214
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 17:14:38 -0500
Received: from rtp-msg-core-1.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA23032
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 17:07:00 -0500 (EST)
Received: from cia.cisco.com (localhost [127.0.0.1])
	by rtp-msg-core-1.cisco.com (8.12.2/8.12.6) with ESMTP id h17MBM0x021765;
	Fri, 7 Feb 2003 17:11:22 -0500 (EST)
Received: from kmorneau-w2k.cisco.com (rtp-vpn2-269.cisco.com [10.82.241.13])
	by cia.cisco.com (Mirapoint)
	with ESMTP id ABO74215;
	Fri, 7 Feb 2003 17:00:27 -0500 (EST)
Message-Id: <4.3.2.7.2.20030207170852.01b15008@cia.cisco.com>
X-Sender: kmorneau@cia.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Fri, 07 Feb 2003 17:10:25 -0500
To: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>,
        "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
From: Ken Morneault <kmorneau@cisco.com>
Subject: Re: [Sigtran] Packing M3UA messages into SCTP ULPDUs
Cc: sigtran@ietf.org
In-Reply-To: <0BD08F48-3AE7-11D7-8A8F-0030654C1AB6@lurchi.franken.de>
References: <OF4C7EEDF0.E54FB924-ONC2256CC6.0035C4A8@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h17MEcp02215
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit



Another piece of information to add to Michael's points...

No implementation at any of the Interops (M3UA/SUA/M2UA/
IUA/M2PA) has (to my knowledge) ever sent more than on
UA message in one SCTP DATA chunk.

Regards,

Ken

At 10:56 PM 2/7/2003 +0100, Michael Tuexen wrote:
>Dear all,
>
>as far as I know all adaptation layers MAY run over TCP. That is
>why all messages have a length field. But when run over SCTP
>it does not make any sense to me not to put exactly one adaptation
>layer packet in one DATA chunk.
>
>I think it is not 'unclear' but never stated explicitly. No one
>send multiple upper layer packets in one DATA chunk, or half of
>it, or 1 and a half. I also have never received a bug report for
>ethereal stating that such behaviour is not supported. Ethereal
>expects exactly one upper layer packet in one DATA chunk.
>
>This SHOULD be clarified for ALL SIGTRAN adaptation layers, but does
>not change the current understanding.
>
>Best regards
>Michael
>
>On Friday, Feb 7, 2003, at 10:48 Europe/Berlin, TOLGA ASVEREN wrote:
>
>>
>>I also think that it is a good idea to clarify this in IG.
>>
>> Tolga
>>--- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
>>>Tolga,
>>>
>>>It does not even forbid splitting an M3UA message
>>>across multiple
>>>SCTP ULPDUs.
>>>
>>>I think this should be clarified in the IGs for the
>>>sake of
>>>interoperablity.  1 M3UA (M2UA, SUA) message to 1
>>>SCTP ULPDU.
>>>
>>>TOLGA ASVEREN wrote:
>>>                                (Fri, 07 Feb 2003
>>>11:05:52)
>>>>
>>>>Brian,
>>>>>Sigtran,
>>>>>
>>>>>Where does it say that only one M3UA message may
>>>be
>>>>>placed into an SCTP ULPDU?
>>>>>
>>>>>As messages contain the length in the header,
>>>can I
>>>>>pack two or more M3UA messages in an SCTP ULPDU?
>>>>>
>>>>>This is so basic, I wonder why it was not
>>>mentioned
>>>>>anywhere...
>>>>If it is not forbidden, shouldn´t it be
>>>interpreted as allowed (although I
>>>>would guess that most of the implementations are
>>>assuming one M3UA message
>>>>in one SCTP ULPDU)?
>>>>BTW related with this topic:
>>>>from 3.1.4
>>>>  "The Message Length defines the length of the
>>>message in octets,
>>>>   including the Common Header.  The Message
>>>Length MUST include
>>>>   parameter padding bytes, if any.
>>>>
>>>>   Note: A receiver SHOULD accept the message
>>>whether or not the final
>>>>   parameter padding is included in the message
>>>length."
>>>>The above paragraphs sound strange to me,
>>>including padding bytes in the
>>>>length is a MUST but still receiver should be
>>>prepared for the opposite.
>>>>But still based on that I assume a M3UA peer can
>>>receive messages, whose
>>>>length does not contain padding bytes (otherwise
>>>forget what stands below)
>>>>>From 3.2
>>>>     "The total length of a parameter (including
>>>Tag, Parameter Length
>>>>      and Value fields) MUST be a multiple of 4
>>>bytes.  If the length of
>>>>      the parameter is not a multiple of 4 bytes,
>>>the sender pads the
>>>>      Parameter at the end (i.e., after the
>>>Parameter Value field) with
>>>>      all zero bytes.  The length of the padding
>>>is NOT included in the
>>>>      parameter length field.  A sender SHOULD NOT
>>>pad with more than 3
>>>>      bytes.  The receiver MUST ignore the padding
>>>bytes."
>>>>Not to add more than 3 padding bytes is a SHOULD
>>>NOT, not a MUST.
>>>>
>>>>For a scenario where message length does not
>>>contain padding bytes length
>>>>and more than 3 padding bytes are used, I would
>>>think receiver can´t detect
>>>>boundary between two M3UA messages (but I have to
>>>repeat, it sounds strange
>>>>to me that including padding bytes length in
>>>message length is a MUST and
>>>>still receiver SHOULD accept messages, which
>>>aren´t prepared in that way)
>>>
>>>One could still detect the next message by scanning
>>>for the next valid
>>>M3UA message header...
>>>
>>>Also, that wouldn't cause any problem for splitting
>>>an M3UA message across
>>>several SCTP ULPDUs.
>>>
>>>I think we need to state in the IG: 1 UA message, 1
>>>SCTP ULPDU.
>>
>>_______________________________________________
>>Sigtran mailing list
>>Sigtran@ietf.org
>>https://www1.ietf.org/mailman/listinfo/sigtran
>
>_______________________________________________
>Sigtran mailing list
>Sigtran@ietf.org
>https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 17:21:14 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA23333
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 17:21:14 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17MSLt02654
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 17:28:21 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17MS7p02640;
	Fri, 7 Feb 2003 17:28:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17MRGp02603
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 17:27:17 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA23290
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 17:19:38 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:JO5inmOF1JXfA+MWCKdVnPWicT6/hau2@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h17MNHD15267;
	Fri, 7 Feb 2003 15:23:17 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h17MNG511967;
	Fri, 7 Feb 2003 15:23:16 -0700
Date: Fri, 7 Feb 2003 15:23:16 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>
Cc: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
Subject: Re: [Sigtran] Packing M3UA messages into SCTP ULPDUs
Message-ID: <20030207152316.A11102@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>,
	TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OF4C7EEDF0.E54FB924-ONC2256CC6.0035C4A8@LocalDomain> <0BD08F48-3AE7-11D7-8A8F-0030654C1AB6@lurchi.franken.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <0BD08F48-3AE7-11D7-8A8F-0030654C1AB6@lurchi.franken.de>; from Michael.Tuexen@lurchi.franken.de on Fri, Feb 07, 2003 at 10:56:44PM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: 8bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Michael,

It is not one message in one DATA chunk.  As you know,
a DATA chunk without both the B and E flags set does
not represent an entire SCTP ULPDU.

If it is stated, it should be state clearly as 1 msg
to 1 SCTP ULPDU (and define SCTP ULPDU if there is no
better term).

--brian

Michael Tuexen wrote:                                                              (Fri, 07 Feb 2003 22:56:44)
> Dear all,
> 
> as far as I know all adaptation layers MAY run over TCP. That is
> why all messages have a length field. But when run over SCTP
> it does not make any sense to me not to put exactly one adaptation
> layer packet in one DATA chunk.
> 
> I think it is not 'unclear' but never stated explicitly. No one
> send multiple upper layer packets in one DATA chunk, or half of
> it, or 1 and a half. I also have never received a bug report for
> ethereal stating that such behaviour is not supported. Ethereal
> expects exactly one upper layer packet in one DATA chunk.
> 
> This SHOULD be clarified for ALL SIGTRAN adaptation layers, but does
> not change the current understanding.
> 
> Best regards
> Michael
> 
> On Friday, Feb 7, 2003, at 10:48 Europe/Berlin, TOLGA ASVEREN wrote:
> 
> >
> > I also think that it is a good idea to clarify this in IG.
> >
> >  Tolga
> > --- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
> >> Tolga,
> >>
> >> It does not even forbid splitting an M3UA message
> >> across multiple
> >> SCTP ULPDUs.
> >>
> >> I think this should be clarified in the IGs for the
> >> sake of
> >> interoperablity.  1 M3UA (M2UA, SUA) message to 1
> >> SCTP ULPDU.
> >>
> >> TOLGA ASVEREN wrote:
> >>                                 (Fri, 07 Feb 2003
> >> 11:05:52)
> >>>
> >>> Brian,
> >>>> Sigtran,
> >>>>
> >>>> Where does it say that only one M3UA message may
> >> be
> >>>> placed into an SCTP ULPDU?
> >>>>
> >>>> As messages contain the length in the header,
> >> can I
> >>>> pack two or more M3UA messages in an SCTP ULPDU?
> >>>>
> >>>> This is so basic, I wonder why it was not
> >> mentioned
> >>>> anywhere...
> >>> If it is not forbidden, shouldn´t it be
> >> interpreted as allowed (although I
> >>> would guess that most of the implementations are
> >> assuming one M3UA message
> >>> in one SCTP ULPDU)?
> >>> BTW related with this topic:
> >>> from 3.1.4
> >>>   "The Message Length defines the length of the
> >> message in octets,
> >>>    including the Common Header.  The Message
> >> Length MUST include
> >>>    parameter padding bytes, if any.
> >>>
> >>>    Note: A receiver SHOULD accept the message
> >> whether or not the final
> >>>    parameter padding is included in the message
> >> length."
> >>> The above paragraphs sound strange to me,
> >> including padding bytes in the
> >>> length is a MUST but still receiver should be
> >> prepared for the opposite.
> >>> But still based on that I assume a M3UA peer can
> >> receive messages, whose
> >>> length does not contain padding bytes (otherwise
> >> forget what stands below)
> >>>> From 3.2
> >>>      "The total length of a parameter (including
> >> Tag, Parameter Length
> >>>       and Value fields) MUST be a multiple of 4
> >> bytes.  If the length of
> >>>       the parameter is not a multiple of 4 bytes,
> >> the sender pads the
> >>>       Parameter at the end (i.e., after the
> >> Parameter Value field) with
> >>>       all zero bytes.  The length of the padding
> >> is NOT included in the
> >>>       parameter length field.  A sender SHOULD NOT
> >> pad with more than 3
> >>>       bytes.  The receiver MUST ignore the padding
> >> bytes."
> >>> Not to add more than 3 padding bytes is a SHOULD
> >> NOT, not a MUST.
> >>>
> >>> For a scenario where message length does not
> >> contain padding bytes length
> >>> and more than 3 padding bytes are used, I would
> >> think receiver can´t detect
> >>> boundary between two M3UA messages (but I have to
> >> repeat, it sounds strange
> >>> to me that including padding bytes length in
> >> message length is a MUST and
> >>> still receiver SHOULD accept messages, which
> >> aren´t prepared in that way)
> >>
> >> One could still detect the next message by scanning
> >> for the next valid
> >> M3UA message header...
> >>
> >> Also, that wouldn't cause any problem for splitting
> >> an M3UA message across
> >> several SCTP ULPDUs.
> >>
> >> I think we need to state in the IG: 1 UA message, 1
> >> SCTP ULPDU.
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 18:26:46 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA24715
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 18:26:46 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17NXtJ05712
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 18:33:55 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17NXcp05686;
	Fri, 7 Feb 2003 18:33:40 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17NWAp05640
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 18:32:10 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA24618
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 18:24:30 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:973whpZP7MTib0DPnjeBKesb4dOrXkTc@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h17NSAD16136;
	Fri, 7 Feb 2003 16:28:10 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h17NS9h20229;
	Fri, 7 Feb 2003 16:28:09 -0700
Date: Fri, 7 Feb 2003 16:28:09 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>
Cc: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
Subject: Re: [Sigtran] Packing M3UA messages into SCTP ULPDUs
Message-ID: <20030207162809.B11102@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>,
	TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OF4C7EEDF0.E54FB924-ONC2256CC6.0035C4A8@LocalDomain> <0BD08F48-3AE7-11D7-8A8F-0030654C1AB6@lurchi.franken.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <0BD08F48-3AE7-11D7-8A8F-0030654C1AB6@lurchi.franken.de>; from Michael.Tuexen@lurchi.franken.de on Fri, Feb 07, 2003 at 10:56:44PM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Michael,

You do not think its unclear, but should be clarified?

--brian

Michael Tuexen wrote:                                                              (Fri, 07 Feb 2003 22:56:44)
> Dear all,
> 
> as far as I know all adaptation layers MAY run over TCP. That is
> why all messages have a length field. But when run over SCTP
> it does not make any sense to me not to put exactly one adaptation
> layer packet in one DATA chunk.
> 
> I think it is not 'unclear' but never stated explicitly. No one
> send multiple upper layer packets in one DATA chunk, or half of
> it, or 1 and a half. I also have never received a bug report for
> ethereal stating that such behaviour is not supported. Ethereal
> expects exactly one upper layer packet in one DATA chunk.
> 
> This SHOULD be clarified for ALL SIGTRAN adaptation layers, but does
> not change the current understanding.
> 
> Best regards
> Michael
> 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 18:27:06 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA24739
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 18:27:06 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17NYFK05758
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 18:34:15 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17NY3p05739;
	Fri, 7 Feb 2003 18:34:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17NXop05706
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 18:33:50 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA24699
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 18:26:10 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:OCP/sCypQpDYDxTTmlaHBJdkd37Hhahv@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h17NTbD16146;
	Fri, 7 Feb 2003 16:29:37 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h17NTa220387;
	Fri, 7 Feb 2003 16:29:36 -0700
Date: Fri, 7 Feb 2003 16:29:36 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: denim john <denim_john@hotmail.com>
Cc: sigtran@ietf.org, j.javier.pastor@ericsson.com, kmorneau@cisco.com
Subject: Re: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
Message-ID: <20030207162936.C11102@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: denim john <denim_john@hotmail.com>, sigtran@ietf.org,
	j.javier.pastor@ericsson.com, kmorneau@cisco.com
References: <F1269WrXdMfCgROTw3X0001b785@hotmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <F1269WrXdMfCgROTw3X0001b785@hotmail.com>; from denim_john@hotmail.com on Fri, Feb 07, 2003 at 03:39:51PM +0000
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

denim,

I thought it was concluded.

--brian

denim john wrote:                                                                    (Fri, 07 Feb 2003 15:39:51)
> 
> Brian,
> 
> I would request any one of the authors to conclude this thread.
> samuel's idea seems to be good but this will result in changes of m3ua and 
> sua.
> 
> Regards,
> Denim.
> 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb  7 18:34:10 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA24947
	for <sigtran-archive@odin.ietf.org>; Fri, 7 Feb 2003 18:34:10 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h17NfJC06666
	for sigtran-archive@odin.ietf.org; Fri, 7 Feb 2003 18:41:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17Nf6p06654;
	Fri, 7 Feb 2003 18:41:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17NeHp06576
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 18:40:17 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA24913
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 18:32:37 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:GuIm4ZbuGXihYfpW8JFFHsKFe6aSSPnB@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h17NaHD16267;
	Fri, 7 Feb 2003 16:36:17 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h17NaGo21187;
	Fri, 7 Feb 2003 16:36:16 -0700
Date: Fri, 7 Feb 2003 16:36:16 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: sherry.karl@tekelec.com
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] SCTP Test Tool
Message-ID: <20030207163616.D11102@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: sherry.karl@tekelec.com, sigtran@ietf.org
References: <OFA41BCE2E.4B19292A-ON85256CC6.0067F555@nc.tekelec.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OFA41BCE2E.4B19292A-ON85256CC6.0067F555@nc.tekelec.com>; from sherry.karl@tekelec.com on Fri, Feb 07, 2003 at 01:56:56PM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Sherry,

Ethereal is a valuable tool.

It is considered bad form to use IETF WG mail reflectors for
commercial purposes.  A web search would probably reveal the
information you need.  Try "M3UA SCTP Test Tool" as a search
string.

--brian

sherry.karl@tekelec.com wrote:                                                 (Fri, 07 Feb 2003 13:56:56)
> I was wondering if anyone is aware of any 3rd party M3UA/SCTP Test Tools?
> 
> Thanks
> 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Sat Feb  8 04:37:22 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA16721
	for <sigtran-archive@odin.ietf.org>; Sat, 8 Feb 2003 04:37:22 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h189iiH12099
	for sigtran-archive@odin.ietf.org; Sat, 8 Feb 2003 04:44:44 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h189iKp12092;
	Sat, 8 Feb 2003 04:44:22 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h189eDp11888
	for <sigtran@optimus.ietf.org>; Sat, 8 Feb 2003 04:40:13 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA16601
	for <sigtran@ietf.org>; Sat, 8 Feb 2003 04:32:18 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h189Znqe005089;
	Sat, 8 Feb 2003 03:35:49 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h189ZQM10862;
	Sat, 8 Feb 2003 03:35:26 -0600 (CST)
Date: Sat, 8 Feb 2003 03:35:26 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: Ken Morneault <kmorneau@cisco.com>
cc: denim john <denim_john@hotmail.com>, sigtran@ietf.org,
        j.javier.pastor@ericsson.com, Francois.Zannin@hp.com,
        bidulock@openss7.org
Subject: Re: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
In-Reply-To: <4.3.2.7.2.20030207110631.04c39a88@cia.cisco.com>
Message-ID: <Pine.SOL.4.10.10302080309550.21630-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Ken,

Find the text changes below.

======================start===================

3.17 Disallowing SSNM messages in ASP-INACTIVE state
 
3.17.1 Description of the problem
 
   SSNM messages are received and updated(destination node status) in
   ASP-ACTIVE state but the backup ASP in inactive state is unaware
   of SS7 node status changes. So when backup ASP becomes active, it
   may not take over the traffic as soon as it becomes active.In
   order to avoid this problem following additional changes are
   proposed.
 
       1.SGP can send SSNM messages to ASP-INACTIVE state.
   
       2.ASP in inactive state can receive SSNM messages.
 
     Implementing above two points will enhance the efficiency of ASP by
     taking the traffic as soon as it becomes active.
 
     Above two points are true only when dynamic registration is allowed
     between ASP and SGP.
 
3.17.2 Text changes to the document
 
   ---------
   Old text: (Section 4.6)
   ---------
 
   When the SG discovers SS7 network isolation, the SGPs send an
   indication to all concerned available ASPs (i.e., ASPs in the ASP-
   ACTIVE state) using DUNA messages for the concerned destinations.
 
   When the SG has completed the MTP Restart procedure, the M3UA layers
   at the SGPs inform all concerned ASPs in the ASP-ACTIVE state of any
   available/restricted SS7 destinations using the DAVA/DRST messages.
   No message is necessary for those destinations still unavailable
   after the restart procedure.
 
   ---------
   New text: (Section 4.6)
   ---------
 
   When the SG discovers SS7 network isolation, the SGPs send an
   indication to all concerned available ASPs (i.e., ASPs in the ASP-
   ACTIVE state or ASP-INACTIVE state) using DUNA messages for the
   concerned destinations.
 
   When the SG has completed the MTP Restart procedure, the M3UA layers
   at the SGPs inform all concerned ASPs in the ASP-ACTIVE state or ASP-
   INACTIVE state of any available/restricted SS7 destinations using the
   DAVA/DRST messages.No message is necessary for those destinations still
   unavailable after the restart procedure.
 
   ---------
   Old text: (Section 4.5.3)
   ---------
 
      - Isolation. The ASP is newly ASP-ACTIVE or has been
                   isolated from an SGP for an extended period.  The ASP
                   MAY request the availability/congestion status of one
                   or more SS7 destinations to which it expects to
                   communicate.
 
   ---------
   New text: (Section 4.5.3)
   ---------
 
      - Isolation. The ASP has been isolated from an SGP for an extended
                   period.  The ASP MAY request the
                   availability/congestion status of one or more SS7
                   destinations to which it expects to communicate.
 
3.17.3 Solution description
 
   Wherever "concerned ASP" is referred in the specification, it does
   mean that ASP in active state or ASP in inactive state.This is true
   only when dynamic registration is allowed between ASP and SGP.

=================END=========

--Samuel 



  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Fri, 7 Feb 2003, Ken Morneault wrote:

> 
> 
> It would be useful if Samuel proposes text changes that could be 
> reviewed and, then, we could see if there was consensus for making
> this change.
> 
> Regards,
> 
> Ken
> 
> At 03:39 PM 2/7/2003 +0000, denim john wrote:
> 
> >Brian,
> >
> >I would request any one of the authors to conclude this thread.
> >samuel's idea seems to be good but this will result in changes of m3ua and sua.
> >
> >Regards,
> >Denim.
> >
> >An efficent ASP will take up the load as soon as it comes up.
> >Inefficient ASP may send DAUD and will receive DRST(...),DUNA(...),
> >SCON(...).  So from the point of ASP following are the unnecessary
> >messages.
> >.SENDING DAUD
> >.RECEING DRST
> >.RECEIVING DUNA
> >.RECEIVING SCON
> >
> > From the point of SGP these are the unnecessay messages.
> >.Receiving DAUD
> >.SENDING DRST
> >.SENDING DUNA
> >.SENDING SCON.
> >Thats what your argument leads.
> >
> >--Samuel
> >
> > Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > 1000 Coit Road, Plano, Texas 75075
> > ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> >
> >On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> >
> >>Samuel,
> >>
> >>That is implementation dependent.  Sending SSNM to inactive ASPs restricts
> >>a dimension not restricted by the current and adequate procedure.  The
> >>SG cannot be required to send SSNM to inactive ASPs.  The RFC should not
> >>place a restriction on the number of inactive ASPs, because we cannot
> >>forsee all future dimensioning.  Procedures should scale as well as possible.
> >>The existing procedure scales.  Sending SSNM to inactive ASPs does not.
> >
> >>--brian
> >>
> >>Samuel Dur D. Jeyaseelan wrote:
> >(Thu, 06 Feb 2003 23:13:44)
> >>> > Brian,
> >>> > Yes , The RFC does not restrict the number of inactive ASPs.
> >>> But network configuration will restrict.
> >>> > --Samuel
> >>> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >>>   1000 Coit Road, Plano, Texas 75075
> >>>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> >>> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> >>> > > Samuel,
> >
> >>> > > > The RFC does not restrict the number of inactive ASPs.
> >>> > > > --brian
> >>> > > > Samuel Dur D. Jeyaseelan wrote:
> >(Thu, 06 Feb 2003 22:40:48)
> >>> > > > > > Brian,
> >>> > > > > > Ok, tell me, where you will configure 1 ASP ACT and 199,000 ASPs INACT?
> >>> > > I want to know which network does this poor configuration.
> >>> > > > > > --Samuel
> >>> > > > > > > > >   Alcatel USA, Inc.                  Internet:
> ><userid>@ssd.usa.alcatel.com
> >>> > >   1000 Coit Road, Plano, Texas 75075
> >>> > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> >*******
> >>> > > > > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> >>> > > > > > > Samuel,
> >>> > > > > > > > The RFC does not restrict the number of inactive ASPs.
> >>> > > > > > > > --brian
> >>> > > > > > > > Samuel Dur D. Jeyaseelan wrote:
> >(Thu, 06 Feb 2003 22:23:12)
> >>> > > > > > > > > > Richard,
> >>> > > > > > > > > >   Alcatel USA, Inc.                  Internet:
> ><userid>@ssd.usa.alcatel.com
> >>> > > > >   1000 Coit Road, Plano, Texas 75075
> >>> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> >*******
> >>> > > > > > > > > > On Thu, 6 Feb 2003, Richard Adams wrote:
> >>> > > > > > > > > > > Brian,
> >>> > > > > > I agree with you. I think when the ASP becomes active, it should
> >do an
> >>> > > > > > audit to update its knowledge of the SS7 network. The ASP needs
> >to take
> >>> > > > > > the initiative to find out about the SS7 nodes etc. in which it
> >is
> >>> > > > > > interested. Any other method of working would lead to impossible
> >loading
> >>> > > > > > on the SG.
> >>> > > > > > > > > > Yes i agree but your suggestion is an indirect method.
> >>> > > > > I am proposing a direct method to send SSNM messages from SGP to
> >INACTIVE
> >>> > > > > ASP. When SGP sends DUNA or DRST to ACTVIE ASP, let him also send a
> >copy
> >>> > > > > of the same message to INACTIVE ASP. I dont find any difficulty
> >here.
> >>> > > > > > > > > >  > > > > > > Of course, I hope 199,999 of your inactive ASPs do not decide to
> >become
> >>> > > > > > active at the same time.....! > > > > > > > > > > Theoritically your example sounds good but practically NOBODY will
> >>> > > > > configure 1 ASP ACTIVE and 199,999 ASPs INACTIVE.
> >>> > > > > > > > > > --Samuel
> >>> > > > > > > > > > > > > > > > > -- > > > > Brian F. G. Bidulock
> >>> > > > bidulock@openss7.org
> >>> > > > http://www.openss7.org/
> >>> > > > _______________________________________________
> >>> > > > Sigtran mailing list
> >>> > > > Sigtran@ietf.org
> >>> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> >>> > > > > > > > -- > > Brian F. G. Bidulock
> >>> > bidulock@openss7.org
> >>> > http://www.openss7.org/
> >>> > _______________________________________________
> >>> > Sigtran mailing list
> >>> > Sigtran@ietf.org
> >>> > https://www1.ietf.org/mailman/listinfo/sigtran
> >>> >
> >>
> >>--
> >>Brian F. G. Bidulock
> >>bidulock@openss7.org
> >>http://www.openss7.org/
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >_________________________________________________________________
> >Help STOP SPAM with the new MSN 8 and get 2 months FREE*  
> >http://join.msn.com/?page=features/junkmail
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Sat Feb  8 04:53:06 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA17025
	for <sigtran-archive@odin.ietf.org>; Sat, 8 Feb 2003 04:53:06 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h18A0TX12534
	for sigtran-archive@odin.ietf.org; Sat, 8 Feb 2003 05:00:29 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h18A0Gp12520;
	Sat, 8 Feb 2003 05:00:16 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h189xvp12465
	for <sigtran@optimus.ietf.org>; Sat, 8 Feb 2003 04:59:57 -0500
Received: from ilsa.franken.de (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA16991
	for <sigtran@ietf.org>; Sat, 8 Feb 2003 04:52:03 -0500 (EST)
Received: from lurchi.franken.de (p5080103F.dip.t-dialin.net [80.128.16.63])
	by ilsa.franken.de (Postfix) with ESMTP
	id 773A424612; Sat,  8 Feb 2003 10:54:06 +0100 (CET)
Date: Sat, 8 Feb 2003 10:51:00 +0100
Subject: Re: [Sigtran] Packing M3UA messages into SCTP ULPDUs
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Mime-Version: 1.0 (Apple Message framework v551)
Cc: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
To: bidulock@openss7.org
From: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>
In-Reply-To: <20030207152316.A11102@openss7.org>
Message-Id: <D43324CC-3B4A-11D7-8A8F-0030654C1AB6@lurchi.franken.de>
X-Mailer: Apple Mail (2.551)
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h189xwp12466
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Brian,

I was thinking of short SS7 messages, thus no fragmentation is needed.

But you are correct, I mean one adaptation layer PDU as one SCTP
user data.

Best regards
Michael

On Friday, Feb 7, 2003, at 23:23 Europe/Berlin, Brian F. G. Bidulock 
wrote:

> Michael,
>
> It is not one message in one DATA chunk.  As you know,
> a DATA chunk without both the B and E flags set does
> not represent an entire SCTP ULPDU.
>
> If it is stated, it should be state clearly as 1 msg
> to 1 SCTP ULPDU (and define SCTP ULPDU if there is no
> better term).
>
> --brian
>
> Michael Tuexen wrote:                                                  
>             (Fri, 07 Feb 2003 22:56:44)
>> Dear all,
>>
>> as far as I know all adaptation layers MAY run over TCP. That is
>> why all messages have a length field. But when run over SCTP
>> it does not make any sense to me not to put exactly one adaptation
>> layer packet in one DATA chunk.
>>
>> I think it is not 'unclear' but never stated explicitly. No one
>> send multiple upper layer packets in one DATA chunk, or half of
>> it, or 1 and a half. I also have never received a bug report for
>> ethereal stating that such behaviour is not supported. Ethereal
>> expects exactly one upper layer packet in one DATA chunk.
>>
>> This SHOULD be clarified for ALL SIGTRAN adaptation layers, but does
>> not change the current understanding.
>>
>> Best regards
>> Michael
>>
>> On Friday, Feb 7, 2003, at 10:48 Europe/Berlin, TOLGA ASVEREN wrote:
>>
>>>
>>> I also think that it is a good idea to clarify this in IG.
>>>
>>>  Tolga
>>> --- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
>>>> Tolga,
>>>>
>>>> It does not even forbid splitting an M3UA message
>>>> across multiple
>>>> SCTP ULPDUs.
>>>>
>>>> I think this should be clarified in the IGs for the
>>>> sake of
>>>> interoperablity.  1 M3UA (M2UA, SUA) message to 1
>>>> SCTP ULPDU.
>>>>
>>>> TOLGA ASVEREN wrote:
>>>>                                 (Fri, 07 Feb 2003
>>>> 11:05:52)
>>>>>
>>>>> Brian,
>>>>>> Sigtran,
>>>>>>
>>>>>> Where does it say that only one M3UA message may
>>>> be
>>>>>> placed into an SCTP ULPDU?
>>>>>>
>>>>>> As messages contain the length in the header,
>>>> can I
>>>>>> pack two or more M3UA messages in an SCTP ULPDU?
>>>>>>
>>>>>> This is so basic, I wonder why it was not
>>>> mentioned
>>>>>> anywhere...
>>>>> If it is not forbidden, shouldn´t it be
>>>> interpreted as allowed (although I
>>>>> would guess that most of the implementations are
>>>> assuming one M3UA message
>>>>> in one SCTP ULPDU)?
>>>>> BTW related with this topic:
>>>>> from 3.1.4
>>>>>   "The Message Length defines the length of the
>>>> message in octets,
>>>>>    including the Common Header.  The Message
>>>> Length MUST include
>>>>>    parameter padding bytes, if any.
>>>>>
>>>>>    Note: A receiver SHOULD accept the message
>>>> whether or not the final
>>>>>    parameter padding is included in the message
>>>> length."
>>>>> The above paragraphs sound strange to me,
>>>> including padding bytes in the
>>>>> length is a MUST but still receiver should be
>>>> prepared for the opposite.
>>>>> But still based on that I assume a M3UA peer can
>>>> receive messages, whose
>>>>> length does not contain padding bytes (otherwise
>>>> forget what stands below)
>>>>>> From 3.2
>>>>>      "The total length of a parameter (including
>>>> Tag, Parameter Length
>>>>>       and Value fields) MUST be a multiple of 4
>>>> bytes.  If the length of
>>>>>       the parameter is not a multiple of 4 bytes,
>>>> the sender pads the
>>>>>       Parameter at the end (i.e., after the
>>>> Parameter Value field) with
>>>>>       all zero bytes.  The length of the padding
>>>> is NOT included in the
>>>>>       parameter length field.  A sender SHOULD NOT
>>>> pad with more than 3
>>>>>       bytes.  The receiver MUST ignore the padding
>>>> bytes."
>>>>> Not to add more than 3 padding bytes is a SHOULD
>>>> NOT, not a MUST.
>>>>>
>>>>> For a scenario where message length does not
>>>> contain padding bytes length
>>>>> and more than 3 padding bytes are used, I would
>>>> think receiver can´t detect
>>>>> boundary between two M3UA messages (but I have to
>>>> repeat, it sounds strange
>>>>> to me that including padding bytes length in
>>>> message length is a MUST and
>>>>> still receiver SHOULD accept messages, which
>>>> aren´t prepared in that way)
>>>>
>>>> One could still detect the next message by scanning
>>>> for the next valid
>>>> M3UA message header...
>>>>
>>>> Also, that wouldn't cause any problem for splitting
>>>> an M3UA message across
>>>> several SCTP ULPDUs.
>>>>
>>>> I think we need to state in the IG: 1 UA message, 1
>>>> SCTP ULPDU.
>>>
>>> _______________________________________________
>>> Sigtran mailing list
>>> Sigtran@ietf.org
>>> https://www1.ietf.org/mailman/listinfo/sigtran
>>>
>>
>> _______________________________________________
>> Sigtran mailing list
>> Sigtran@ietf.org
>> https://www1.ietf.org/mailman/listinfo/sigtran
>
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
>

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Sat Feb  8 04:56:54 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA17130
	for <sigtran-archive@odin.ietf.org>; Sat, 8 Feb 2003 04:56:54 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h18A4HY12659
	for sigtran-archive@odin.ietf.org; Sat, 8 Feb 2003 05:04:17 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h18A44p12651;
	Sat, 8 Feb 2003 05:04:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h18A3Wp12629
	for <sigtran@optimus.ietf.org>; Sat, 8 Feb 2003 05:03:32 -0500
Received: from ilsa.franken.de (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA17090
	for <sigtran@ietf.org>; Sat, 8 Feb 2003 04:55:39 -0500 (EST)
Received: from lurchi.franken.de (p5080103F.dip.t-dialin.net [80.128.16.63])
	by ilsa.franken.de (Postfix) with ESMTP
	id 2C58C245D4; Sat,  8 Feb 2003 10:57:52 +0100 (CET)
Date: Sat, 8 Feb 2003 10:54:46 +0100
Subject: Re: [Sigtran] Packing M3UA messages into SCTP ULPDUs
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v551)
Cc: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
To: bidulock@openss7.org
From: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>
In-Reply-To: <20030207162809.B11102@openss7.org>
Message-Id: <5AE79342-3B4B-11D7-8A8F-0030654C1AB6@lurchi.franken.de>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.551)
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Brian,

I think it is clear to all people today, because of the reasons I gave 
and
the one Ken gave.

But I agree that it is not written somewhere and at least you thought 
about
it. A clarification does not hurt... This might be helpful for people 
staring to
work on SIGTRAN in the future without knowing the history of it.

We have such stuff also in the SCTP IG.

Best regards
Michael

On Saturday, Feb 8, 2003, at 00:28 Europe/Berlin, Brian F. G. Bidulock 
wrote:

> Michael,
>
> You do not think its unclear, but should be clarified?
>
> --brian
>
> Michael Tuexen wrote:                                                  
>             (Fri, 07 Feb 2003 22:56:44)
>> Dear all,
>>
>> as far as I know all adaptation layers MAY run over TCP. That is
>> why all messages have a length field. But when run over SCTP
>> it does not make any sense to me not to put exactly one adaptation
>> layer packet in one DATA chunk.
>>
>> I think it is not 'unclear' but never stated explicitly. No one
>> send multiple upper layer packets in one DATA chunk, or half of
>> it, or 1 and a half. I also have never received a bug report for
>> ethereal stating that such behaviour is not supported. Ethereal
>> expects exactly one upper layer packet in one DATA chunk.
>>
>> This SHOULD be clarified for ALL SIGTRAN adaptation layers, but does
>> not change the current understanding.
>>
>> Best regards
>> Michael
>>
>
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
>

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Sat Feb  8 05:29:24 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA17642
	for <sigtran-archive@odin.ietf.org>; Sat, 8 Feb 2003 05:29:24 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h18Aaji13956
	for sigtran-archive@odin.ietf.org; Sat, 8 Feb 2003 05:36:45 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h18AaXp13947;
	Sat, 8 Feb 2003 05:36:33 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h18AZ4p13909
	for <sigtran@optimus.ietf.org>; Sat, 8 Feb 2003 05:35:04 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA17611
	for <sigtran@ietf.org>; Sat, 8 Feb 2003 05:27:11 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:kYfdpViXvhW333nxKTHrR7BvH8vEereG@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h18AUoD25235;
	Sat, 8 Feb 2003 03:30:50 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h18AUlO02349;
	Sat, 8 Feb 2003 03:30:47 -0700
Date: Sat, 8 Feb 2003 03:30:47 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>
Cc: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
Subject: Re: [Sigtran] Packing M3UA messages into SCTP ULPDUs
Message-ID: <20030208033047.A2026@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>,
	TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <20030207162809.B11102@openss7.org> <5AE79342-3B4B-11D7-8A8F-0030654C1AB6@lurchi.franken.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <5AE79342-3B4B-11D7-8A8F-0030654C1AB6@lurchi.franken.de>; from Michael.Tuexen@lurchi.franken.de on Sat, Feb 08, 2003 at 10:54:46AM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Michael,

So you think it is clear, but a clarification wouldn't hurt? ;)

--brian

Michael Tuexen wrote:                                                              (Sat, 08 Feb 2003 10:54:46)
> Brian,
> 
> I think it is clear to all people today, because of the reasons I gave 
> and
> the one Ken gave.
> 
> But I agree that it is not written somewhere and at least you thought 
> about
> it. A clarification does not hurt... This might be helpful for people 
> staring to
> work on SIGTRAN in the future without knowing the history of it.
> 
> We have such stuff also in the SCTP IG.
> 
> Best regards
> Michael
> 
> On Saturday, Feb 8, 2003, at 00:28 Europe/Berlin, Brian F. G. Bidulock 
> wrote:
> 
> > Michael,
> >
> > You do not think its unclear, but should be clarified?
> >
> > --brian
> >
> > Michael Tuexen wrote:                                                  
> >             (Fri, 07 Feb 2003 22:56:44)
> >> Dear all,
> >>
> >> as far as I know all adaptation layers MAY run over TCP. That is
> >> why all messages have a length field. But when run over SCTP
> >> it does not make any sense to me not to put exactly one adaptation
> >> layer packet in one DATA chunk.
> >>
> >> I think it is not 'unclear' but never stated explicitly. No one
> >> send multiple upper layer packets in one DATA chunk, or half of
> >> it, or 1 and a half. I also have never received a bug report for
> >> ethereal stating that such behaviour is not supported. Ethereal
> >> expects exactly one upper layer packet in one DATA chunk.
> >>
> >> This SHOULD be clarified for ALL SIGTRAN adaptation layers, but does
> >> not change the current understanding.
> >>
> >> Best regards
> >> Michael
> >>
> >
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> >

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Sat Feb  8 06:01:12 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA18156
	for <sigtran-archive@odin.ietf.org>; Sat, 8 Feb 2003 06:01:12 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h18B8Y515907
	for sigtran-archive@odin.ietf.org; Sat, 8 Feb 2003 06:08:34 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h18B8Hp15896;
	Sat, 8 Feb 2003 06:08:18 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h18B7Kp15879
	for <sigtran@optimus.ietf.org>; Sat, 8 Feb 2003 06:07:20 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA18085
	for <sigtran@ietf.org>; Sat, 8 Feb 2003 05:59:24 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:IkMS2V1y23vxXR9/+EtECr7q7WWnSYEj@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h18B2iD25713;
	Sat, 8 Feb 2003 04:02:44 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h18B2ga06317;
	Sat, 8 Feb 2003 04:02:42 -0700
Date: Sat, 8 Feb 2003 04:02:42 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: Ken Morneault <kmorneau@cisco.com>, denim john <denim_john@hotmail.com>,
        sigtran@ietf.org, j.javier.pastor@ericsson.com, Francois.Zannin@hp.com
Subject: Re: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
Message-ID: <20030208040242.B2026@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	Ken Morneault <kmorneau@cisco.com>,
	denim john <denim_john@hotmail.com>, sigtran@ietf.org,
	j.javier.pastor@ericsson.com, Francois.Zannin@hp.com
References: <4.3.2.7.2.20030207110631.04c39a88@cia.cisco.com> <Pine.SOL.4.10.10302080309550.21630-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302080309550.21630-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Sat, Feb 08, 2003 at 03:35:26AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

The SG cannot be required to send SSNM to inactive ASPs.

Also, unregistered ASPs are also in the ASP-INACTIVE state, yet are
unable to know to which AS an RC corresponds.  An SG which supports both
dynamic and static configuration cannot know if and ASP is already
configured with the RC of the AS, or whether it must regsiter first.
(An AS does not have an "unregistered" state.)  Therefore, the SG cannot
know whether sending SSNM to ASP in the ASP-INACTIVE state for an AS
will induce an error (Invalid RC), and when an error is induced whether
it is indeed because the ASP needs to register for the AS, or whether
some other configuration problem has occured.

Also, in Section 4.6, and ASP in the ASP-INACTIVE state is not a
"concerned" ASP.  You should note that the SG is not required to send
SSNM to all of the ASP-ACTIVE ASPs either: only the concerned ASPs in
the ASP-ACTIVE state for the AS.  When sending SSNM in using reponsive
procedures, it is only the ASP to which the SSNM is a response that is
"concerned".

MTP does not indicate to its users whether an MTP-PAUSE, MTP-RESUME, or
MTP-STATUS primitive is responsive or broadcast.  Therefore, some
implementations cannot determine which of the ASP-ACTIVE ASPs are
"concerned".  But certainly, for both responsive indications, ASPs in
the ASP-INACTIVE state are not concerned.

Also, the ASP-INACTIVE state for an AS corresponds to no MTP-User being
associated with the MTP-SAP.  This means that when all ASPs are in the
ASP-INACTIVE state, there is no MTP User associated with the MTP at the
SG.  Because there is no MTP-User associated with the MTP at the SG, the
SG cannot deliver MTP-PAUSE, MTP-RESUME or MTP-STATUS primitives to M3UA
at the SG and cannot then issue SSNM to ASPs.

When all the ASPs are ASP-INACTIVE, the SG cannot be expected to send
SSNM to any of them.

The RFC cannot require an implementation at the SG to use facilities
which are not provided to the MTP-User by MTP.  (Things like DRST are
OPTIONAL).

ASPs which are not in the multiple SG configuration are not required to
maintain any destination state whatsoever: SSNM delivered to the ASP are
delivered directly to the user as the corresponding MTP-PAUSE,
MTP-RESUME and MTP-STATUS primitives.  When an ASP is in the
ASP-INACTIVE state for an AS, it has no MTP-User to delivere these
primitive to.  Therefore, it is unnecessary to send SSNM to ASPs in the
ASP-INACTIVE state in the single SG configuration, and the SG cannot be
required to send them.  Multiple SG arrangements are OPTIONAL.

So, as before, the SG cannot be required to send SSNM to inactives ASPs.

The current RFC says that the SG SHOULD NOT send DATA or SSNM to
inactive ASPs or it will risk data loss.  This does not forbit the SG
from sending SSNM to inactive ASPs, it just recommends that it doesn't.
Becase of the above, to recommend that SG send SSNM to inactive ASPs
with the text changes you propose is not workable.

I am in total disagreement with your proposal.

Three others on this thread have also expressed that the current RFC is
sufficient in this regard.

The only change that I see as necessary would be to add the following
statement:

	When an ASP becomes active for an AS and the SG is experiencing
	SS7 network isolation or is performing the MTP Restart procedure
	for the AS, the SG SHOULD send a DUNA(*) to the newly active ASP
	to prevent the ASP from sending traffic.

	When an ASP becomes active for an AS and destinations normally
	acessible to the ASP are inaccessible, restricted or congested
	for the AS, the SG SHOULD send DUNA, DRST or SCON AS for the
	inacessible, restricted or congested destinations to the ASP
	newly active for the AS to prevent the ASP from sending traffic
	for destinations that it might otherwise not know are
	inaccessible, restricted or congested.

Still, the SG cannot be required to send SSNM to inactive ASPs, and the
ASP can certainly not rely upon receiving SSNM while inactive.  So, the
only sure way for the ASP to obtain information is after it goes ASP-ACTIVE.

--brian


Samuel Dur D. Jeyaseelan wrote:                                                     (Sat, 08 Feb 2003 03:35:26)
> 
> Ken,
> 
> Find the text changes below.
> 
> ======================start===================
> 
> 3.17 Disallowing SSNM messages in ASP-INACTIVE state
>  
> 3.17.1 Description of the problem
>  
>    SSNM messages are received and updated(destination node status) in
>    ASP-ACTIVE state but the backup ASP in inactive state is unaware
>    of SS7 node status changes. So when backup ASP becomes active, it
>    may not take over the traffic as soon as it becomes active.In
>    order to avoid this problem following additional changes are
>    proposed.
>  
>        1.SGP can send SSNM messages to ASP-INACTIVE state.
>    
>        2.ASP in inactive state can receive SSNM messages.
>  
>      Implementing above two points will enhance the efficiency of ASP by
>      taking the traffic as soon as it becomes active.
>  
>      Above two points are true only when dynamic registration is allowed
>      between ASP and SGP.
>  
> 3.17.2 Text changes to the document
>  
>    ---------
>    Old text: (Section 4.6)
>    ---------
>  
>    When the SG discovers SS7 network isolation, the SGPs send an
>    indication to all concerned available ASPs (i.e., ASPs in the ASP-
>    ACTIVE state) using DUNA messages for the concerned destinations.
>  
>    When the SG has completed the MTP Restart procedure, the M3UA layers
>    at the SGPs inform all concerned ASPs in the ASP-ACTIVE state of any
>    available/restricted SS7 destinations using the DAVA/DRST messages.
>    No message is necessary for those destinations still unavailable
>    after the restart procedure.
>  
>    ---------
>    New text: (Section 4.6)
>    ---------
>  
>    When the SG discovers SS7 network isolation, the SGPs send an
>    indication to all concerned available ASPs (i.e., ASPs in the ASP-
>    ACTIVE state or ASP-INACTIVE state) using DUNA messages for the
>    concerned destinations.
>  
>    When the SG has completed the MTP Restart procedure, the M3UA layers
>    at the SGPs inform all concerned ASPs in the ASP-ACTIVE state or ASP-
>    INACTIVE state of any available/restricted SS7 destinations using the
>    DAVA/DRST messages.No message is necessary for those destinations still
>    unavailable after the restart procedure.
>  
>    ---------
>    Old text: (Section 4.5.3)
>    ---------
>  
>       - Isolation. The ASP is newly ASP-ACTIVE or has been
>                    isolated from an SGP for an extended period.  The ASP
>                    MAY request the availability/congestion status of one
>                    or more SS7 destinations to which it expects to
>                    communicate.
>  
>    ---------
>    New text: (Section 4.5.3)
>    ---------
>  
>       - Isolation. The ASP has been isolated from an SGP for an extended
>                    period.  The ASP MAY request the
>                    availability/congestion status of one or more SS7
>                    destinations to which it expects to communicate.
>  
> 3.17.3 Solution description
>  
>    Wherever "concerned ASP" is referred in the specification, it does
>    mean that ASP in active state or ASP in inactive state.This is true
>    only when dynamic registration is allowed between ASP and SGP.
> 
> =================END=========
> 
> --Samuel 
> 
> 
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Fri, 7 Feb 2003, Ken Morneault wrote:
> 
> > 
> > 
> > It would be useful if Samuel proposes text changes that could be 
> > reviewed and, then, we could see if there was consensus for making
> > this change.
> > 
> > Regards,
> > 
> > Ken
> > 
> > At 03:39 PM 2/7/2003 +0000, denim john wrote:
> > 
> > >Brian,
> > >
> > >I would request any one of the authors to conclude this thread.
> > >samuel's idea seems to be good but this will result in changes of m3ua and sua.
> > >
> > >Regards,
> > >Denim.
> > >
> > >An efficent ASP will take up the load as soon as it comes up.
> > >Inefficient ASP may send DAUD and will receive DRST(...),DUNA(...),
> > >SCON(...).  So from the point of ASP following are the unnecessary
> > >messages.
> > >.SENDING DAUD
> > >.RECEING DRST
> > >.RECEIVING DUNA
> > >.RECEIVING SCON
> > >
> > > From the point of SGP these are the unnecessay messages.
> > >.Receiving DAUD
> > >.SENDING DRST
> > >.SENDING DUNA
> > >.SENDING SCON.
> > >Thats what your argument leads.
> > >
> > >--Samuel
> > >
> > > Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > 1000 Coit Road, Plano, Texas 75075
> > > ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > >
> > >On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > >
> > >>Samuel,
> > >>
> > >>That is implementation dependent.  Sending SSNM to inactive ASPs restricts
> > >>a dimension not restricted by the current and adequate procedure.  The
> > >>SG cannot be required to send SSNM to inactive ASPs.  The RFC should not
> > >>place a restriction on the number of inactive ASPs, because we cannot
> > >>forsee all future dimensioning.  Procedures should scale as well as possible.
> > >>The existing procedure scales.  Sending SSNM to inactive ASPs does not.
> > >
> > >>--brian
> > >>
> > >>Samuel Dur D. Jeyaseelan wrote:
> > >(Thu, 06 Feb 2003 23:13:44)
> > >>> > Brian,
> > >>> > Yes , The RFC does not restrict the number of inactive ASPs.
> > >>> But network configuration will restrict.
> > >>> > --Samuel
> > >>> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >>>   1000 Coit Road, Plano, Texas 75075
> > >>>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > >>> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > >>> > > Samuel,
> > >
> > >>> > > > The RFC does not restrict the number of inactive ASPs.
> > >>> > > > --brian
> > >>> > > > Samuel Dur D. Jeyaseelan wrote:
> > >(Thu, 06 Feb 2003 22:40:48)
> > >>> > > > > > Brian,
> > >>> > > > > > Ok, tell me, where you will configure 1 ASP ACT and 199,000 ASPs INACT?
> > >>> > > I want to know which network does this poor configuration.
> > >>> > > > > > --Samuel
> > >>> > > > > > > > >   Alcatel USA, Inc.                  Internet:
> > ><userid>@ssd.usa.alcatel.com
> > >>> > >   1000 Coit Road, Plano, Texas 75075
> > >>> > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > >*******
> > >>> > > > > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > >>> > > > > > > Samuel,
> > >>> > > > > > > > The RFC does not restrict the number of inactive ASPs.
> > >>> > > > > > > > --brian
> > >>> > > > > > > > Samuel Dur D. Jeyaseelan wrote:
> > >(Thu, 06 Feb 2003 22:23:12)
> > >>> > > > > > > > > > Richard,
> > >>> > > > > > > > > >   Alcatel USA, Inc.                  Internet:
> > ><userid>@ssd.usa.alcatel.com
> > >>> > > > >   1000 Coit Road, Plano, Texas 75075
> > >>> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > >*******
> > >>> > > > > > > > > > On Thu, 6 Feb 2003, Richard Adams wrote:
> > >>> > > > > > > > > > > Brian,
> > >>> > > > > > I agree with you. I think when the ASP becomes active, it should
> > >do an
> > >>> > > > > > audit to update its knowledge of the SS7 network. The ASP needs
> > >to take
> > >>> > > > > > the initiative to find out about the SS7 nodes etc. in which it
> > >is
> > >>> > > > > > interested. Any other method of working would lead to impossible
> > >loading
> > >>> > > > > > on the SG.
> > >>> > > > > > > > > > Yes i agree but your suggestion is an indirect method.
> > >>> > > > > I am proposing a direct method to send SSNM messages from SGP to
> > >INACTIVE
> > >>> > > > > ASP. When SGP sends DUNA or DRST to ACTVIE ASP, let him also send a
> > >copy
> > >>> > > > > of the same message to INACTIVE ASP. I dont find any difficulty
> > >here.
> > >>> > > > > > > > > >  > > > > > > Of course, I hope 199,999 of your inactive ASPs do not decide to
> > >become
> > >>> > > > > > active at the same time.....! > > > > > > > > > > Theoritically your example sounds good but practically NOBODY will
> > >>> > > > > configure 1 ASP ACTIVE and 199,999 ASPs INACTIVE.
> > >>> > > > > > > > > > --Samuel
> > >>> > > > > > > > > > > > > > > > > -- > > > > Brian F. G. Bidulock
> > >>> > > > bidulock@openss7.org
> > >>> > > > http://www.openss7.org/
> > >>> > > > _______________________________________________
> > >>> > > > Sigtran mailing list
> > >>> > > > Sigtran@ietf.org
> > >>> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >>> > > > > > > > -- > > Brian F. G. Bidulock
> > >>> > bidulock@openss7.org
> > >>> > http://www.openss7.org/
> > >>> > _______________________________________________
> > >>> > Sigtran mailing list
> > >>> > Sigtran@ietf.org
> > >>> > https://www1.ietf.org/mailman/listinfo/sigtran
> > >>> >
> > >>
> > >>--
> > >>Brian F. G. Bidulock
> > >>bidulock@openss7.org
> > >>http://www.openss7.org/
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> > >_________________________________________________________________
> > >Help STOP SPAM with the new MSN 8 and get 2 months FREE*  
> > >http://join.msn.com/?page=features/junkmail
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Sat Feb  8 06:17:03 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA18440
	for <sigtran-archive@odin.ietf.org>; Sat, 8 Feb 2003 06:17:03 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h18BOQL16286
	for sigtran-archive@odin.ietf.org; Sat, 8 Feb 2003 06:24:26 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h18BO9p16274;
	Sat, 8 Feb 2003 06:24:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h18BNHp16248
	for <sigtran@optimus.ietf.org>; Sat, 8 Feb 2003 06:23:17 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA18421
	for <sigtran@ietf.org>; Sat, 8 Feb 2003 06:15:21 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:i8pMTwe9KOFhwG4ho/li4Oq7ULodQvXj@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h18BIvD26135;
	Sat, 8 Feb 2003 04:18:57 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h18BIuF08664;
	Sat, 8 Feb 2003 04:18:56 -0700
Date: Sat, 8 Feb 2003 04:18:55 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        Ken Morneault <kmorneau@cisco.com>,
        denim john <denim_john@hotmail.com>, sigtran@ietf.org,
        j.javier.pastor@ericsson.com, Francois.Zannin@hp.com
Subject: Re: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
Message-ID: <20030208041855.C2026@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	Ken Morneault <kmorneau@cisco.com>,
	denim john <denim_john@hotmail.com>, sigtran@ietf.org,
	j.javier.pastor@ericsson.com, Francois.Zannin@hp.com
References: <4.3.2.7.2.20030207110631.04c39a88@cia.cisco.com> <Pine.SOL.4.10.10302080309550.21630-100000@sun3271.ssd.usa.alcatel.com> <20030208040242.B2026@openss7.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20030208040242.B2026@openss7.org>; from bidulock@openss7.org on Sat, Feb 08, 2003 at 04:02:42AM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Sigtran,

I take it back: change SHOULD to MAY in the second clause:

 	When an ASP becomes active for an AS and the SG is experiencing
 	SS7 network isolation or is performing the MTP Restart procedure
 	for the AS, the SG SHOULD send a DUNA(*) to the newly active ASP
 	to prevent the ASP from sending traffic.
 
 	When an ASP becomes active for an AS and destinations normally
 	acessible to the ASP are inaccessible, restricted or congested
 	for the AS, the SG MAY send DUNA, DRST or SCON AS for the
		           ^^^
 	inacessible, restricted or congested destinations to the ASP
 	newly active for the AS to prevent the ASP from sending traffic
 	for destinations that it might otherwise not know are
 	inaccessible, restricted or congested.

Otherwise, the SG would be sending redundant information that an
single-SG ASP which does not maintain destination state will subsequently
obtain with DAUD using the currently recommended procedure.  This could
result in multiple MTP-PAUSE and MTP-STATUS primitives delivered to the
MTP User at the ASP which could confuse some MTP Users.

--brian


Brian F. G. Bidulock wrote:                              (Sat, 08 Feb 2003 04:02:42)
> Samuel,
> 
> The SG cannot be required to send SSNM to inactive ASPs.
> 
> Also, unregistered ASPs are also in the ASP-INACTIVE state, yet are
> unable to know to which AS an RC corresponds.  An SG which supports both
> dynamic and static configuration cannot know if and ASP is already
> configured with the RC of the AS, or whether it must regsiter first.
> (An AS does not have an "unregistered" state.)  Therefore, the SG cannot
> know whether sending SSNM to ASP in the ASP-INACTIVE state for an AS
> will induce an error (Invalid RC), and when an error is induced whether
> it is indeed because the ASP needs to register for the AS, or whether
> some other configuration problem has occured.
> 
> Also, in Section 4.6, and ASP in the ASP-INACTIVE state is not a
> "concerned" ASP.  You should note that the SG is not required to send
> SSNM to all of the ASP-ACTIVE ASPs either: only the concerned ASPs in
> the ASP-ACTIVE state for the AS.  When sending SSNM in using reponsive
> procedures, it is only the ASP to which the SSNM is a response that is
> "concerned".
> 
> MTP does not indicate to its users whether an MTP-PAUSE, MTP-RESUME, or
> MTP-STATUS primitive is responsive or broadcast.  Therefore, some
> implementations cannot determine which of the ASP-ACTIVE ASPs are
> "concerned".  But certainly, for both responsive indications, ASPs in
> the ASP-INACTIVE state are not concerned.
> 
> Also, the ASP-INACTIVE state for an AS corresponds to no MTP-User being
> associated with the MTP-SAP.  This means that when all ASPs are in the
> ASP-INACTIVE state, there is no MTP User associated with the MTP at the
> SG.  Because there is no MTP-User associated with the MTP at the SG, the
> SG cannot deliver MTP-PAUSE, MTP-RESUME or MTP-STATUS primitives to M3UA
> at the SG and cannot then issue SSNM to ASPs.
> 
> When all the ASPs are ASP-INACTIVE, the SG cannot be expected to send
> SSNM to any of them.
> 
> The RFC cannot require an implementation at the SG to use facilities
> which are not provided to the MTP-User by MTP.  (Things like DRST are
> OPTIONAL).
> 
> ASPs which are not in the multiple SG configuration are not required to
> maintain any destination state whatsoever: SSNM delivered to the ASP are
> delivered directly to the user as the corresponding MTP-PAUSE,
> MTP-RESUME and MTP-STATUS primitives.  When an ASP is in the
> ASP-INACTIVE state for an AS, it has no MTP-User to delivere these
> primitive to.  Therefore, it is unnecessary to send SSNM to ASPs in the
> ASP-INACTIVE state in the single SG configuration, and the SG cannot be
> required to send them.  Multiple SG arrangements are OPTIONAL.
> 
> So, as before, the SG cannot be required to send SSNM to inactives ASPs.
> 
> The current RFC says that the SG SHOULD NOT send DATA or SSNM to
> inactive ASPs or it will risk data loss.  This does not forbit the SG
> from sending SSNM to inactive ASPs, it just recommends that it doesn't.
> Becase of the above, to recommend that SG send SSNM to inactive ASPs
> with the text changes you propose is not workable.
> 
> I am in total disagreement with your proposal.
> 
> Three others on this thread have also expressed that the current RFC is
> sufficient in this regard.
> 
> The only change that I see as necessary would be to add the following
> statement:
> 
> 	When an ASP becomes active for an AS and the SG is experiencing
> 	SS7 network isolation or is performing the MTP Restart procedure
> 	for the AS, the SG SHOULD send a DUNA(*) to the newly active ASP
> 	to prevent the ASP from sending traffic.
> 
> 	When an ASP becomes active for an AS and destinations normally
> 	acessible to the ASP are inaccessible, restricted or congested
> 	for the AS, the SG SHOULD send DUNA, DRST or SCON AS for the
> 	inacessible, restricted or congested destinations to the ASP
> 	newly active for the AS to prevent the ASP from sending traffic
> 	for destinations that it might otherwise not know are
> 	inaccessible, restricted or congested.
> 
> Still, the SG cannot be required to send SSNM to inactive ASPs, and the
> ASP can certainly not rely upon receiving SSNM while inactive.  So, the
> only sure way for the ASP to obtain information is after it goes ASP-ACTIVE.
> 
> --brian
> 
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Sat, 08 Feb 2003 03:35:26)
> > 
> > Ken,
> > 
> > Find the text changes below.
> > 
> > ======================start===================
> > 
> > 3.17 Disallowing SSNM messages in ASP-INACTIVE state
> >  
> > 3.17.1 Description of the problem
> >  
> >    SSNM messages are received and updated(destination node status) in
> >    ASP-ACTIVE state but the backup ASP in inactive state is unaware
> >    of SS7 node status changes. So when backup ASP becomes active, it
> >    may not take over the traffic as soon as it becomes active.In
> >    order to avoid this problem following additional changes are
> >    proposed.
> >  
> >        1.SGP can send SSNM messages to ASP-INACTIVE state.
> >    
> >        2.ASP in inactive state can receive SSNM messages.
> >  
> >      Implementing above two points will enhance the efficiency of ASP by
> >      taking the traffic as soon as it becomes active.
> >  
> >      Above two points are true only when dynamic registration is allowed
> >      between ASP and SGP.
> >  
> > 3.17.2 Text changes to the document
> >  
> >    ---------
> >    Old text: (Section 4.6)
> >    ---------
> >  
> >    When the SG discovers SS7 network isolation, the SGPs send an
> >    indication to all concerned available ASPs (i.e., ASPs in the ASP-
> >    ACTIVE state) using DUNA messages for the concerned destinations.
> >  
> >    When the SG has completed the MTP Restart procedure, the M3UA layers
> >    at the SGPs inform all concerned ASPs in the ASP-ACTIVE state of any
> >    available/restricted SS7 destinations using the DAVA/DRST messages.
> >    No message is necessary for those destinations still unavailable
> >    after the restart procedure.
> >  
> >    ---------
> >    New text: (Section 4.6)
> >    ---------
> >  
> >    When the SG discovers SS7 network isolation, the SGPs send an
> >    indication to all concerned available ASPs (i.e., ASPs in the ASP-
> >    ACTIVE state or ASP-INACTIVE state) using DUNA messages for the
> >    concerned destinations.
> >  
> >    When the SG has completed the MTP Restart procedure, the M3UA layers
> >    at the SGPs inform all concerned ASPs in the ASP-ACTIVE state or ASP-
> >    INACTIVE state of any available/restricted SS7 destinations using the
> >    DAVA/DRST messages.No message is necessary for those destinations still
> >    unavailable after the restart procedure.
> >  
> >    ---------
> >    Old text: (Section 4.5.3)
> >    ---------
> >  
> >       - Isolation. The ASP is newly ASP-ACTIVE or has been
> >                    isolated from an SGP for an extended period.  The ASP
> >                    MAY request the availability/congestion status of one
> >                    or more SS7 destinations to which it expects to
> >                    communicate.
> >  
> >    ---------
> >    New text: (Section 4.5.3)
> >    ---------
> >  
> >       - Isolation. The ASP has been isolated from an SGP for an extended
> >                    period.  The ASP MAY request the
> >                    availability/congestion status of one or more SS7
> >                    destinations to which it expects to communicate.
> >  
> > 3.17.3 Solution description
> >  
> >    Wherever "concerned ASP" is referred in the specification, it does
> >    mean that ASP in active state or ASP in inactive state.This is true
> >    only when dynamic registration is allowed between ASP and SGP.
> > 
> > =================END=========
> > 
> > --Samuel 
> > 
> > 
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Fri, 7 Feb 2003, Ken Morneault wrote:
> > 
> > > 
> > > 
> > > It would be useful if Samuel proposes text changes that could be 
> > > reviewed and, then, we could see if there was consensus for making
> > > this change.
> > > 
> > > Regards,
> > > 
> > > Ken
> > > 
> > > At 03:39 PM 2/7/2003 +0000, denim john wrote:
> > > 
> > > >Brian,
> > > >
> > > >I would request any one of the authors to conclude this thread.
> > > >samuel's idea seems to be good but this will result in changes of m3ua and sua.
> > > >
> > > >Regards,
> > > >Denim.
> > > >
> > > >An efficent ASP will take up the load as soon as it comes up.
> > > >Inefficient ASP may send DAUD and will receive DRST(...),DUNA(...),
> > > >SCON(...).  So from the point of ASP following are the unnecessary
> > > >messages.
> > > >.SENDING DAUD
> > > >.RECEING DRST
> > > >.RECEIVING DUNA
> > > >.RECEIVING SCON
> > > >
> > > > From the point of SGP these are the unnecessay messages.
> > > >.Receiving DAUD
> > > >.SENDING DRST
> > > >.SENDING DUNA
> > > >.SENDING SCON.
> > > >Thats what your argument leads.
> > > >
> > > >--Samuel
> > > >
> > > > Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > 1000 Coit Road, Plano, Texas 75075
> > > > ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > >
> > > >On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > >
> > > >>Samuel,
> > > >>
> > > >>That is implementation dependent.  Sending SSNM to inactive ASPs restricts
> > > >>a dimension not restricted by the current and adequate procedure.  The
> > > >>SG cannot be required to send SSNM to inactive ASPs.  The RFC should not
> > > >>place a restriction on the number of inactive ASPs, because we cannot
> > > >>forsee all future dimensioning.  Procedures should scale as well as possible.
> > > >>The existing procedure scales.  Sending SSNM to inactive ASPs does not.
> > > >
> > > >>--brian
> > > >>
> > > >>Samuel Dur D. Jeyaseelan wrote:
> > > >(Thu, 06 Feb 2003 23:13:44)
> > > >>> > Brian,
> > > >>> > Yes , The RFC does not restrict the number of inactive ASPs.
> > > >>> But network configuration will restrict.
> > > >>> > --Samuel
> > > >>> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > >>>   1000 Coit Road, Plano, Texas 75075
> > > >>>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > >>> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > >>> > > Samuel,
> > > >
> > > >>> > > > The RFC does not restrict the number of inactive ASPs.
> > > >>> > > > --brian
> > > >>> > > > Samuel Dur D. Jeyaseelan wrote:
> > > >(Thu, 06 Feb 2003 22:40:48)
> > > >>> > > > > > Brian,
> > > >>> > > > > > Ok, tell me, where you will configure 1 ASP ACT and 199,000 ASPs INACT?
> > > >>> > > I want to know which network does this poor configuration.
> > > >>> > > > > > --Samuel
> > > >>> > > > > > > > >   Alcatel USA, Inc.                  Internet:
> > > ><userid>@ssd.usa.alcatel.com
> > > >>> > >   1000 Coit Road, Plano, Texas 75075
> > > >>> > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > >*******
> > > >>> > > > > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > >>> > > > > > > Samuel,
> > > >>> > > > > > > > The RFC does not restrict the number of inactive ASPs.
> > > >>> > > > > > > > --brian
> > > >>> > > > > > > > Samuel Dur D. Jeyaseelan wrote:
> > > >(Thu, 06 Feb 2003 22:23:12)
> > > >>> > > > > > > > > > Richard,
> > > >>> > > > > > > > > >   Alcatel USA, Inc.                  Internet:
> > > ><userid>@ssd.usa.alcatel.com
> > > >>> > > > >   1000 Coit Road, Plano, Texas 75075
> > > >>> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > >*******
> > > >>> > > > > > > > > > On Thu, 6 Feb 2003, Richard Adams wrote:
> > > >>> > > > > > > > > > > Brian,
> > > >>> > > > > > I agree with you. I think when the ASP becomes active, it should
> > > >do an
> > > >>> > > > > > audit to update its knowledge of the SS7 network. The ASP needs
> > > >to take
> > > >>> > > > > > the initiative to find out about the SS7 nodes etc. in which it
> > > >is
> > > >>> > > > > > interested. Any other method of working would lead to impossible
> > > >loading
> > > >>> > > > > > on the SG.
> > > >>> > > > > > > > > > Yes i agree but your suggestion is an indirect method.
> > > >>> > > > > I am proposing a direct method to send SSNM messages from SGP to
> > > >INACTIVE
> > > >>> > > > > ASP. When SGP sends DUNA or DRST to ACTVIE ASP, let him also send a
> > > >copy
> > > >>> > > > > of the same message to INACTIVE ASP. I dont find any difficulty
> > > >here.
> > > >>> > > > > > > > > >  > > > > > > Of course, I hope 199,999 of your inactive ASPs do not decide to
> > > >become
> > > >>> > > > > > active at the same time.....! > > > > > > > > > > Theoritically your example sounds good but practically NOBODY will
> > > >>> > > > > configure 1 ASP ACTIVE and 199,999 ASPs INACTIVE.
> > > >>> > > > > > > > > > --Samuel
> > > >>> > > > > > > > > > > > > > > > > -- > > > > Brian F. G. Bidulock
> > > >>> > > > bidulock@openss7.org
> > > >>> > > > http://www.openss7.org/
> > > >>> > > > _______________________________________________
> > > >>> > > > Sigtran mailing list
> > > >>> > > > Sigtran@ietf.org
> > > >>> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > >>> > > > > > > > -- > > Brian F. G. Bidulock
> > > >>> > bidulock@openss7.org
> > > >>> > http://www.openss7.org/
> > > >>> > _______________________________________________
> > > >>> > Sigtran mailing list
> > > >>> > Sigtran@ietf.org
> > > >>> > https://www1.ietf.org/mailman/listinfo/sigtran
> > > >>> >
> > > >>
> > > >>--
> > > >>Brian F. G. Bidulock
> > > >>bidulock@openss7.org
> > > >>http://www.openss7.org/
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >_________________________________________________________________
> > > >Help STOP SPAM with the new MSN 8 and get 2 months FREE*  
> > > >http://join.msn.com/?page=features/junkmail
> > > 
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Sun Feb  9 23:28:31 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA14751
	for <sigtran-archive@odin.ietf.org>; Sun, 9 Feb 2003 23:28:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1A4ajj08257
	for sigtran-archive@odin.ietf.org; Sun, 9 Feb 2003 23:36:45 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1A4aQp08239;
	Sun, 9 Feb 2003 23:36:27 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1A4XJp08165
	for <sigtran@optimus.ietf.org>; Sun, 9 Feb 2003 23:33:19 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA14696
	for <sigtran@ietf.org>; Sun, 9 Feb 2003 23:24:33 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1A4S0JT013953;
	Sun, 9 Feb 2003 22:28:01 -0600 (CST)
Received: from sun2594.ssd.usa.alcatel.com (sun2594.ssd.usa.alcatel.com [143.209.250.252])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h1A4REM27061;
	Sun, 9 Feb 2003 22:27:14 -0600 (CST)
Date: Sun, 9 Feb 2003 22:27:13 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: Ken Morneault <kmorneau@cisco.com>, denim john <denim_john@hotmail.com>,
        sigtran@ietf.org, j.javier.pastor@ericsson.com, Francois.Zannin@hp.com
Subject: Re: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
In-Reply-To: <20030208040242.B2026@openss7.org>
Message-ID: <Pine.SOL.4.10.10302092154080.1868-100000@sun2594.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Sat, 8 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> The SG cannot be required to send SSNM to inactive ASPs.
> 
> Also, unregistered ASPs are also in the ASP-INACTIVE state, yet are
> unable to know to which AS an RC corresponds.  An SG which supports both
> dynamic and static configuration cannot know if and ASP is already
> configured with the RC of the AS, or whether it must regsiter first.
> (An AS does not have an "unregistered" state.)  Therefore, the SG cannot
> know whether sending SSNM to ASP in the ASP-INACTIVE state for an AS
> will induce an error (Invalid RC), and when an error is induced whether
> it is indeed because the ASP needs to register for the AS, or whether
> some other configuration problem has occured.
> 
misunderstanding.See the proposed text changes,  it says that "SGP can
send SSNM messages to ASP", only when dynamic registration is allowed,
that means both SGP and ASP know RC.

> Also, in Section 4.6, and ASP in the ASP-INACTIVE state is not a
> "concerned" ASP.  You should note that the SG is not required to send
> SSNM to all of the ASP-ACTIVE ASPs either: only the concerned ASPs in
> the ASP-ACTIVE state for the AS.  When sending SSNM in using reponsive
> procedures, it is only the ASP to which the SSNM is a response that is
> "concerned".
> 
Again wrong. Proposed text changes will consider 'concerned ASP' as
ASP-ACTIVE or ASP-INACTIVE only when dynamic registration is allowed.
It means SGP knows AS and ASP knows RC.

> MTP does not indicate to its users whether an MTP-PAUSE, MTP-RESUME, or
> MTP-STATUS primitive is responsive or broadcast.  Therefore, some
> implementations cannot determine which of the ASP-ACTIVE ASPs are
> "concerned".  But certainly, for both responsive indications, ASPs in
> the ASP-INACTIVE state are not concerned.
> 
> Also, the ASP-INACTIVE state for an AS corresponds to no MTP-User being
> associated with the MTP-SAP.  This means that when all ASPs are in the
> ASP-INACTIVE state, there is no MTP User associated with the MTP at the
> SG.  Because there is no MTP-User associated with the MTP at the SG, the
> SG cannot deliver MTP-PAUSE, MTP-RESUME or MTP-STATUS primitives to M3UA
> at the SG and cannot then issue SSNM to ASPs.
> 
Wrong. LOGICALLY MTP's user part is available at SG.
Logically ASP-ACTIVE means ASP is ready to take over the traffic,
But still inactive ASP can receive SSNM messages but it SHOULD not 
take up the traffice as long as it is in inactive state.

> When all the ASPs are ASP-INACTIVE, the SG cannot be expected to send
> SSNM to any of them.
> 
No, Sending SSNM to inactive ASP will increase the efficiency of ASP.

> The RFC cannot require an implementation at the SG to use facilities
> which are not provided to the MTP-User by MTP.  (Things like DRST are
> OPTIONAL).
> 
> ASPs which are not in the multiple SG configuration are not required to
> maintain any destination state whatsoever: SSNM delivered to the ASP are
> delivered directly to the user as the corresponding MTP-PAUSE,
> MTP-RESUME and MTP-STATUS primitives.  When an ASP is in the
> ASP-INACTIVE state for an AS, it has no MTP-User to delivere these
> primitive to.  Therefore, it is unnecessary to send SSNM to ASPs in the
> ASP-INACTIVE state in the single SG configuration, and the SG cannot be
> required to send them.  Multiple SG arrangements are OPTIONAL.
> 

You have not read properly the 'proposed text changes'.

> So, as before, the SG cannot be required to send SSNM to inactives ASPs.
> 
> The current RFC says that the SG SHOULD NOT send DATA or SSNM to
> inactive ASPs or it will risk data loss.  This does not forbit the SG
> from sending SSNM to inactive ASPs, it just recommends that it doesn't.
> Becase of the above, to recommend that SG send SSNM to inactive ASPs
> with the text changes you propose is not workable.

Proposed text changes is only for DUNA,DRST,SCON and DAVA not for DATA.
Updating destination node status changes at inactive ASP will not have
any side effects.

> 
> I am in total disagreement with your proposal.
> 
> Three others on this thread have also expressed that the current RFC is
> sufficient in this regard.
> 
Three others also proposed "Sending SSNM to inactive ASP is good method".

> The only change that I see as necessary would be to add the following
> statement:
> 
> 	When an ASP becomes active for an AS and the SG is experiencing
> 	SS7 network isolation or is performing the MTP Restart procedure
> 	for the AS, the SG SHOULD send a DUNA(*) to the newly active ASP
> 	to prevent the ASP from sending traffic.
> 
> 	When an ASP becomes active for an AS and destinations normally
> 	acessible to the ASP are inaccessible, restricted or congested
> 	for the AS, the SG SHOULD send DUNA, DRST or SCON AS for the
> 	inacessible, restricted or congested destinations to the ASP
> 	newly active for the AS to prevent the ASP from sending traffic
> 	for destinations that it might otherwise not know are
> 	inaccessible, restricted or congested.
> 
> Still, the SG cannot be required to send SSNM to inactive ASPs, and the
> ASP can certainly not rely upon receiving SSNM while inactive.  So, the
> only sure way for the ASP to obtain information is after it goes ASP-ACTIVE.
> 

Once the dynamic registration is over, Both SGP and ASP know th RC.
Once RC is known to SGP and ASP , it means MTP user part is availabe at
SG.
ONLY DATA message should not get exchanged when ASP in inactive state.

--Samuel
> --brian
> 
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Sat, 08 Feb 2003 03:35:26)
> > 
> > Ken,
> > 
> > Find the text changes below.
> > 
> > ======================start===================
> > 
> > 3.17 Disallowing SSNM messages in ASP-INACTIVE state
> >  
> > 3.17.1 Description of the problem
> >  
> >    SSNM messages are received and updated(destination node status) in
> >    ASP-ACTIVE state but the backup ASP in inactive state is unaware
> >    of SS7 node status changes. So when backup ASP becomes active, it
> >    may not take over the traffic as soon as it becomes active.In
> >    order to avoid this problem following additional changes are
> >    proposed.
> >  
> >        1.SGP can send SSNM messages to ASP-INACTIVE state.
> >    
> >        2.ASP in inactive state can receive SSNM messages.
> >  
> >      Implementing above two points will enhance the efficiency of ASP by
> >      taking the traffic as soon as it becomes active.
> >  
> >      Above two points are true only when dynamic registration is allowed
> >      between ASP and SGP.
> >  
> > 3.17.2 Text changes to the document
> >  
> >    ---------
> >    Old text: (Section 4.6)
> >    ---------
> >  
> >    When the SG discovers SS7 network isolation, the SGPs send an
> >    indication to all concerned available ASPs (i.e., ASPs in the ASP-
> >    ACTIVE state) using DUNA messages for the concerned destinations.
> >  
> >    When the SG has completed the MTP Restart procedure, the M3UA layers
> >    at the SGPs inform all concerned ASPs in the ASP-ACTIVE state of any
> >    available/restricted SS7 destinations using the DAVA/DRST messages.
> >    No message is necessary for those destinations still unavailable
> >    after the restart procedure.
> >  
> >    ---------
> >    New text: (Section 4.6)
> >    ---------
> >  
> >    When the SG discovers SS7 network isolation, the SGPs send an
> >    indication to all concerned available ASPs (i.e., ASPs in the ASP-
> >    ACTIVE state or ASP-INACTIVE state) using DUNA messages for the
> >    concerned destinations.
> >  
> >    When the SG has completed the MTP Restart procedure, the M3UA layers
> >    at the SGPs inform all concerned ASPs in the ASP-ACTIVE state or ASP-
> >    INACTIVE state of any available/restricted SS7 destinations using the
> >    DAVA/DRST messages.No message is necessary for those destinations still
> >    unavailable after the restart procedure.
> >  
> >    ---------
> >    Old text: (Section 4.5.3)
> >    ---------
> >  
> >       - Isolation. The ASP is newly ASP-ACTIVE or has been
> >                    isolated from an SGP for an extended period.  The ASP
> >                    MAY request the availability/congestion status of one
> >                    or more SS7 destinations to which it expects to
> >                    communicate.
> >  
> >    ---------
> >    New text: (Section 4.5.3)
> >    ---------
> >  
> >       - Isolation. The ASP has been isolated from an SGP for an extended
> >                    period.  The ASP MAY request the
> >                    availability/congestion status of one or more SS7
> >                    destinations to which it expects to communicate.
> >  
> > 3.17.3 Solution description
> >  
> >    Wherever "concerned ASP" is referred in the specification, it does
> >    mean that ASP in active state or ASP in inactive state.This is true
> >    only when dynamic registration is allowed between ASP and SGP.
> > 
> > =================END=========
> > 
> > --Samuel 
> > 
> > 
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Fri, 7 Feb 2003, Ken Morneault wrote:
> > 
> > > 
> > > 
> > > It would be useful if Samuel proposes text changes that could be 
> > > reviewed and, then, we could see if there was consensus for making
> > > this change.
> > > 
> > > Regards,
> > > 
> > > Ken
> > > 
> > > At 03:39 PM 2/7/2003 +0000, denim john wrote:
> > > 
> > > >Brian,
> > > >
> > > >I would request any one of the authors to conclude this thread.
> > > >samuel's idea seems to be good but this will result in changes of m3ua and sua.
> > > >
> > > >Regards,
> > > >Denim.
> > > >
> > > >An efficent ASP will take up the load as soon as it comes up.
> > > >Inefficient ASP may send DAUD and will receive DRST(...),DUNA(...),
> > > >SCON(...).  So from the point of ASP following are the unnecessary
> > > >messages.
> > > >.SENDING DAUD
> > > >.RECEING DRST
> > > >.RECEIVING DUNA
> > > >.RECEIVING SCON
> > > >
> > > > From the point of SGP these are the unnecessay messages.
> > > >.Receiving DAUD
> > > >.SENDING DRST
> > > >.SENDING DUNA
> > > >.SENDING SCON.
> > > >Thats what your argument leads.
> > > >
> > > >--Samuel
> > > >
> > > > Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > 1000 Coit Road, Plano, Texas 75075
> > > > ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > >
> > > >On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > >
> > > >>Samuel,
> > > >>
> > > >>That is implementation dependent.  Sending SSNM to inactive ASPs restricts
> > > >>a dimension not restricted by the current and adequate procedure.  The
> > > >>SG cannot be required to send SSNM to inactive ASPs.  The RFC should not
> > > >>place a restriction on the number of inactive ASPs, because we cannot
> > > >>forsee all future dimensioning.  Procedures should scale as well as possible.
> > > >>The existing procedure scales.  Sending SSNM to inactive ASPs does not.
> > > >
> > > >>--brian
> > > >>
> > > >>Samuel Dur D. Jeyaseelan wrote:
> > > >(Thu, 06 Feb 2003 23:13:44)
> > > >>> > Brian,
> > > >>> > Yes , The RFC does not restrict the number of inactive ASPs.
> > > >>> But network configuration will restrict.
> > > >>> > --Samuel
> > > >>> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > >>>   1000 Coit Road, Plano, Texas 75075
> > > >>>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > >>> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > >>> > > Samuel,
> > > >
> > > >>> > > > The RFC does not restrict the number of inactive ASPs.
> > > >>> > > > --brian
> > > >>> > > > Samuel Dur D. Jeyaseelan wrote:
> > > >(Thu, 06 Feb 2003 22:40:48)
> > > >>> > > > > > Brian,
> > > >>> > > > > > Ok, tell me, where you will configure 1 ASP ACT and 199,000 ASPs INACT?
> > > >>> > > I want to know which network does this poor configuration.
> > > >>> > > > > > --Samuel
> > > >>> > > > > > > > >   Alcatel USA, Inc.                  Internet:
> > > ><userid>@ssd.usa.alcatel.com
> > > >>> > >   1000 Coit Road, Plano, Texas 75075
> > > >>> > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > >*******
> > > >>> > > > > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > >>> > > > > > > Samuel,
> > > >>> > > > > > > > The RFC does not restrict the number of inactive ASPs.
> > > >>> > > > > > > > --brian
> > > >>> > > > > > > > Samuel Dur D. Jeyaseelan wrote:
> > > >(Thu, 06 Feb 2003 22:23:12)
> > > >>> > > > > > > > > > Richard,
> > > >>> > > > > > > > > >   Alcatel USA, Inc.                  Internet:
> > > ><userid>@ssd.usa.alcatel.com
> > > >>> > > > >   1000 Coit Road, Plano, Texas 75075
> > > >>> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > >*******
> > > >>> > > > > > > > > > On Thu, 6 Feb 2003, Richard Adams wrote:
> > > >>> > > > > > > > > > > Brian,
> > > >>> > > > > > I agree with you. I think when the ASP becomes active, it should
> > > >do an
> > > >>> > > > > > audit to update its knowledge of the SS7 network. The ASP needs
> > > >to take
> > > >>> > > > > > the initiative to find out about the SS7 nodes etc. in which it
> > > >is
> > > >>> > > > > > interested. Any other method of working would lead to impossible
> > > >loading
> > > >>> > > > > > on the SG.
> > > >>> > > > > > > > > > Yes i agree but your suggestion is an indirect method.
> > > >>> > > > > I am proposing a direct method to send SSNM messages from SGP to
> > > >INACTIVE
> > > >>> > > > > ASP. When SGP sends DUNA or DRST to ACTVIE ASP, let him also send a
> > > >copy
> > > >>> > > > > of the same message to INACTIVE ASP. I dont find any difficulty
> > > >here.
> > > >>> > > > > > > > > >  > > > > > > Of course, I hope 199,999 of your inactive ASPs do not decide to
> > > >become
> > > >>> > > > > > active at the same time.....! > > > > > > > > > > Theoritically your example sounds good but practically NOBODY will
> > > >>> > > > > configure 1 ASP ACTIVE and 199,999 ASPs INACTIVE.
> > > >>> > > > > > > > > > --Samuel
> > > >>> > > > > > > > > > > > > > > > > -- > > > > Brian F. G. Bidulock
> > > >>> > > > bidulock@openss7.org
> > > >>> > > > http://www.openss7.org/
> > > >>> > > > _______________________________________________
> > > >>> > > > Sigtran mailing list
> > > >>> > > > Sigtran@ietf.org
> > > >>> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > >>> > > > > > > > -- > > Brian F. G. Bidulock
> > > >>> > bidulock@openss7.org
> > > >>> > http://www.openss7.org/
> > > >>> > _______________________________________________
> > > >>> > Sigtran mailing list
> > > >>> > Sigtran@ietf.org
> > > >>> > https://www1.ietf.org/mailman/listinfo/sigtran
> > > >>> >
> > > >>
> > > >>--
> > > >>Brian F. G. Bidulock
> > > >>bidulock@openss7.org
> > > >>http://www.openss7.org/
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >_________________________________________________________________
> > > >Help STOP SPAM with the new MSN 8 and get 2 months FREE*  
> > > >http://join.msn.com/?page=features/junkmail
> > > 
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 10 04:26:05 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA01459
	for <sigtran-archive@odin.ietf.org>; Mon, 10 Feb 2003 04:26:05 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1A9YPn01870
	for sigtran-archive@odin.ietf.org; Mon, 10 Feb 2003 04:34:25 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1A9Y5p01849;
	Mon, 10 Feb 2003 04:34:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1A9Upp01739
	for <sigtran@optimus.ietf.org>; Mon, 10 Feb 2003 04:30:51 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA01371
	for <sigtran@ietf.org>; Mon, 10 Feb 2003 04:21:58 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:ud4TgWQJhR59h2R9VBQYclI3dV5fxTI4@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1A9P5D01864;
	Mon, 10 Feb 2003 02:25:05 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1A9P3u06442;
	Mon, 10 Feb 2003 02:25:03 -0700
Date: Mon, 10 Feb 2003 02:25:03 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: Ken Morneault <kmorneau@cisco.com>, denim john <denim_john@hotmail.com>,
        sigtran@ietf.org, j.javier.pastor@ericsson.com, Francois.Zannin@hp.com
Subject: Re: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
Message-ID: <20030210022503.A4738@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	Ken Morneault <kmorneau@cisco.com>,
	denim john <denim_john@hotmail.com>, sigtran@ietf.org,
	j.javier.pastor@ericsson.com, Francois.Zannin@hp.com
References: <20030208040242.B2026@openss7.org> <Pine.SOL.4.10.10302092154080.1868-100000@sun2594.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302092154080.1868-100000@sun2594.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Sun, Feb 09, 2003 at 10:27:13PM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

I disagree with all your comments, and your proposal.

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Sun, 09 Feb 2003 22:27:13)
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Sat, 8 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > The SG cannot be required to send SSNM to inactive ASPs.
> > 
> > Also, unregistered ASPs are also in the ASP-INACTIVE state, yet are
> > unable to know to which AS an RC corresponds.  An SG which supports both
> > dynamic and static configuration cannot know if and ASP is already
> > configured with the RC of the AS, or whether it must regsiter first.
> > (An AS does not have an "unregistered" state.)  Therefore, the SG cannot
> > know whether sending SSNM to ASP in the ASP-INACTIVE state for an AS
> > will induce an error (Invalid RC), and when an error is induced whether
> > it is indeed because the ASP needs to register for the AS, or whether
> > some other configuration problem has occured.
> > 
> misunderstanding.See the proposed text changes,  it says that "SGP can
> send SSNM messages to ASP", only when dynamic registration is allowed,
> that means both SGP and ASP know RC.
> 
> > Also, in Section 4.6, and ASP in the ASP-INACTIVE state is not a
> > "concerned" ASP.  You should note that the SG is not required to send
> > SSNM to all of the ASP-ACTIVE ASPs either: only the concerned ASPs in
> > the ASP-ACTIVE state for the AS.  When sending SSNM in using reponsive
> > procedures, it is only the ASP to which the SSNM is a response that is
> > "concerned".
> > 
> Again wrong. Proposed text changes will consider 'concerned ASP' as
> ASP-ACTIVE or ASP-INACTIVE only when dynamic registration is allowed.
> It means SGP knows AS and ASP knows RC.
> 
> > MTP does not indicate to its users whether an MTP-PAUSE, MTP-RESUME, or
> > MTP-STATUS primitive is responsive or broadcast.  Therefore, some
> > implementations cannot determine which of the ASP-ACTIVE ASPs are
> > "concerned".  But certainly, for both responsive indications, ASPs in
> > the ASP-INACTIVE state are not concerned.
> > 
> > Also, the ASP-INACTIVE state for an AS corresponds to no MTP-User being
> > associated with the MTP-SAP.  This means that when all ASPs are in the
> > ASP-INACTIVE state, there is no MTP User associated with the MTP at the
> > SG.  Because there is no MTP-User associated with the MTP at the SG, the
> > SG cannot deliver MTP-PAUSE, MTP-RESUME or MTP-STATUS primitives to M3UA
> > at the SG and cannot then issue SSNM to ASPs.
> > 
> Wrong. LOGICALLY MTP's user part is available at SG.
> Logically ASP-ACTIVE means ASP is ready to take over the traffic,
> But still inactive ASP can receive SSNM messages but it SHOULD not 
> take up the traffice as long as it is in inactive state.
> 
> > When all the ASPs are ASP-INACTIVE, the SG cannot be expected to send
> > SSNM to any of them.
> > 
> No, Sending SSNM to inactive ASP will increase the efficiency of ASP.
> 
> > The RFC cannot require an implementation at the SG to use facilities
> > which are not provided to the MTP-User by MTP.  (Things like DRST are
> > OPTIONAL).
> > 
> > ASPs which are not in the multiple SG configuration are not required to
> > maintain any destination state whatsoever: SSNM delivered to the ASP are
> > delivered directly to the user as the corresponding MTP-PAUSE,
> > MTP-RESUME and MTP-STATUS primitives.  When an ASP is in the
> > ASP-INACTIVE state for an AS, it has no MTP-User to delivere these
> > primitive to.  Therefore, it is unnecessary to send SSNM to ASPs in the
> > ASP-INACTIVE state in the single SG configuration, and the SG cannot be
> > required to send them.  Multiple SG arrangements are OPTIONAL.
> > 
> 
> You have not read properly the 'proposed text changes'.
> 
> > So, as before, the SG cannot be required to send SSNM to inactives ASPs.
> > 
> > The current RFC says that the SG SHOULD NOT send DATA or SSNM to
> > inactive ASPs or it will risk data loss.  This does not forbit the SG
> > from sending SSNM to inactive ASPs, it just recommends that it doesn't.
> > Becase of the above, to recommend that SG send SSNM to inactive ASPs
> > with the text changes you propose is not workable.
> 
> Proposed text changes is only for DUNA,DRST,SCON and DAVA not for DATA.
> Updating destination node status changes at inactive ASP will not have
> any side effects.
> 
> > 
> > I am in total disagreement with your proposal.
> > 
> > Three others on this thread have also expressed that the current RFC is
> > sufficient in this regard.
> > 
> Three others also proposed "Sending SSNM to inactive ASP is good method".
> 
> > The only change that I see as necessary would be to add the following
> > statement:
> > 
> > 	When an ASP becomes active for an AS and the SG is experiencing
> > 	SS7 network isolation or is performing the MTP Restart procedure
> > 	for the AS, the SG SHOULD send a DUNA(*) to the newly active ASP
> > 	to prevent the ASP from sending traffic.
> > 
> > 	When an ASP becomes active for an AS and destinations normally
> > 	acessible to the ASP are inaccessible, restricted or congested
> > 	for the AS, the SG SHOULD send DUNA, DRST or SCON AS for the
> > 	inacessible, restricted or congested destinations to the ASP
> > 	newly active for the AS to prevent the ASP from sending traffic
> > 	for destinations that it might otherwise not know are
> > 	inaccessible, restricted or congested.
> > 
> > Still, the SG cannot be required to send SSNM to inactive ASPs, and the
> > ASP can certainly not rely upon receiving SSNM while inactive.  So, the
> > only sure way for the ASP to obtain information is after it goes ASP-ACTIVE.
> > 
> 
> Once the dynamic registration is over, Both SGP and ASP know th RC.
> Once RC is known to SGP and ASP , it means MTP user part is availabe at
> SG.
> ONLY DATA message should not get exchanged when ASP in inactive state.
> 
> --Samuel
> > --brian
> > 
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Sat, 08 Feb 2003 03:35:26)
> > > 
> > > Ken,
> > > 
> > > Find the text changes below.
> > > 
> > > ======================start===================
> > > 
> > > 3.17 Disallowing SSNM messages in ASP-INACTIVE state
> > >  
> > > 3.17.1 Description of the problem
> > >  
> > >    SSNM messages are received and updated(destination node status) in
> > >    ASP-ACTIVE state but the backup ASP in inactive state is unaware
> > >    of SS7 node status changes. So when backup ASP becomes active, it
> > >    may not take over the traffic as soon as it becomes active.In
> > >    order to avoid this problem following additional changes are
> > >    proposed.
> > >  
> > >        1.SGP can send SSNM messages to ASP-INACTIVE state.
> > >    
> > >        2.ASP in inactive state can receive SSNM messages.
> > >  
> > >      Implementing above two points will enhance the efficiency of ASP by
> > >      taking the traffic as soon as it becomes active.
> > >  
> > >      Above two points are true only when dynamic registration is allowed
> > >      between ASP and SGP.
> > >  
> > > 3.17.2 Text changes to the document
> > >  
> > >    ---------
> > >    Old text: (Section 4.6)
> > >    ---------
> > >  
> > >    When the SG discovers SS7 network isolation, the SGPs send an
> > >    indication to all concerned available ASPs (i.e., ASPs in the ASP-
> > >    ACTIVE state) using DUNA messages for the concerned destinations.
> > >  
> > >    When the SG has completed the MTP Restart procedure, the M3UA layers
> > >    at the SGPs inform all concerned ASPs in the ASP-ACTIVE state of any
> > >    available/restricted SS7 destinations using the DAVA/DRST messages.
> > >    No message is necessary for those destinations still unavailable
> > >    after the restart procedure.
> > >  
> > >    ---------
> > >    New text: (Section 4.6)
> > >    ---------
> > >  
> > >    When the SG discovers SS7 network isolation, the SGPs send an
> > >    indication to all concerned available ASPs (i.e., ASPs in the ASP-
> > >    ACTIVE state or ASP-INACTIVE state) using DUNA messages for the
> > >    concerned destinations.
> > >  
> > >    When the SG has completed the MTP Restart procedure, the M3UA layers
> > >    at the SGPs inform all concerned ASPs in the ASP-ACTIVE state or ASP-
> > >    INACTIVE state of any available/restricted SS7 destinations using the
> > >    DAVA/DRST messages.No message is necessary for those destinations still
> > >    unavailable after the restart procedure.
> > >  
> > >    ---------
> > >    Old text: (Section 4.5.3)
> > >    ---------
> > >  
> > >       - Isolation. The ASP is newly ASP-ACTIVE or has been
> > >                    isolated from an SGP for an extended period.  The ASP
> > >                    MAY request the availability/congestion status of one
> > >                    or more SS7 destinations to which it expects to
> > >                    communicate.
> > >  
> > >    ---------
> > >    New text: (Section 4.5.3)
> > >    ---------
> > >  
> > >       - Isolation. The ASP has been isolated from an SGP for an extended
> > >                    period.  The ASP MAY request the
> > >                    availability/congestion status of one or more SS7
> > >                    destinations to which it expects to communicate.
> > >  
> > > 3.17.3 Solution description
> > >  
> > >    Wherever "concerned ASP" is referred in the specification, it does
> > >    mean that ASP in active state or ASP in inactive state.This is true
> > >    only when dynamic registration is allowed between ASP and SGP.
> > > 
> > > =================END=========
> > > 
> > > --Samuel 
> > > 
> > > 
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Fri, 7 Feb 2003, Ken Morneault wrote:
> > > 
> > > > 
> > > > 
> > > > It would be useful if Samuel proposes text changes that could be 
> > > > reviewed and, then, we could see if there was consensus for making
> > > > this change.
> > > > 
> > > > Regards,
> > > > 
> > > > Ken
> > > > 
> > > > At 03:39 PM 2/7/2003 +0000, denim john wrote:
> > > > 
> > > > >Brian,
> > > > >
> > > > >I would request any one of the authors to conclude this thread.
> > > > >samuel's idea seems to be good but this will result in changes of m3ua and sua.
> > > > >
> > > > >Regards,
> > > > >Denim.
> > > > >
> > > > >An efficent ASP will take up the load as soon as it comes up.
> > > > >Inefficient ASP may send DAUD and will receive DRST(...),DUNA(...),
> > > > >SCON(...).  So from the point of ASP following are the unnecessary
> > > > >messages.
> > > > >.SENDING DAUD
> > > > >.RECEING DRST
> > > > >.RECEIVING DUNA
> > > > >.RECEIVING SCON
> > > > >
> > > > > From the point of SGP these are the unnecessay messages.
> > > > >.Receiving DAUD
> > > > >.SENDING DRST
> > > > >.SENDING DUNA
> > > > >.SENDING SCON.
> > > > >Thats what your argument leads.
> > > > >
> > > > >--Samuel
> > > > >
> > > > > Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > > 1000 Coit Road, Plano, Texas 75075
> > > > > ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > >
> > > > >On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > > >
> > > > >>Samuel,
> > > > >>
> > > > >>That is implementation dependent.  Sending SSNM to inactive ASPs restricts
> > > > >>a dimension not restricted by the current and adequate procedure.  The
> > > > >>SG cannot be required to send SSNM to inactive ASPs.  The RFC should not
> > > > >>place a restriction on the number of inactive ASPs, because we cannot
> > > > >>forsee all future dimensioning.  Procedures should scale as well as possible.
> > > > >>The existing procedure scales.  Sending SSNM to inactive ASPs does not.
> > > > >
> > > > >>--brian
> > > > >>
> > > > >>Samuel Dur D. Jeyaseelan wrote:
> > > > >(Thu, 06 Feb 2003 23:13:44)
> > > > >>> > Brian,
> > > > >>> > Yes , The RFC does not restrict the number of inactive ASPs.
> > > > >>> But network configuration will restrict.
> > > > >>> > --Samuel
> > > > >>> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > >>>   1000 Coit Road, Plano, Texas 75075
> > > > >>>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > >>> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > > >>> > > Samuel,
> > > > >
> > > > >>> > > > The RFC does not restrict the number of inactive ASPs.
> > > > >>> > > > --brian
> > > > >>> > > > Samuel Dur D. Jeyaseelan wrote:
> > > > >(Thu, 06 Feb 2003 22:40:48)
> > > > >>> > > > > > Brian,
> > > > >>> > > > > > Ok, tell me, where you will configure 1 ASP ACT and 199,000 ASPs INACT?
> > > > >>> > > I want to know which network does this poor configuration.
> > > > >>> > > > > > --Samuel
> > > > >>> > > > > > > > >   Alcatel USA, Inc.                  Internet:
> > > > ><userid>@ssd.usa.alcatel.com
> > > > >>> > >   1000 Coit Road, Plano, Texas 75075
> > > > >>> > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > > >*******
> > > > >>> > > > > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > > > >>> > > > > > > Samuel,
> > > > >>> > > > > > > > The RFC does not restrict the number of inactive ASPs.
> > > > >>> > > > > > > > --brian
> > > > >>> > > > > > > > Samuel Dur D. Jeyaseelan wrote:
> > > > >(Thu, 06 Feb 2003 22:23:12)
> > > > >>> > > > > > > > > > Richard,
> > > > >>> > > > > > > > > >   Alcatel USA, Inc.                  Internet:
> > > > ><userid>@ssd.usa.alcatel.com
> > > > >>> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >>> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > > > >*******
> > > > >>> > > > > > > > > > On Thu, 6 Feb 2003, Richard Adams wrote:
> > > > >>> > > > > > > > > > > Brian,
> > > > >>> > > > > > I agree with you. I think when the ASP becomes active, it should
> > > > >do an
> > > > >>> > > > > > audit to update its knowledge of the SS7 network. The ASP needs
> > > > >to take
> > > > >>> > > > > > the initiative to find out about the SS7 nodes etc. in which it
> > > > >is
> > > > >>> > > > > > interested. Any other method of working would lead to impossible
> > > > >loading
> > > > >>> > > > > > on the SG.
> > > > >>> > > > > > > > > > Yes i agree but your suggestion is an indirect method.
> > > > >>> > > > > I am proposing a direct method to send SSNM messages from SGP to
> > > > >INACTIVE
> > > > >>> > > > > ASP. When SGP sends DUNA or DRST to ACTVIE ASP, let him also send a
> > > > >copy
> > > > >>> > > > > of the same message to INACTIVE ASP. I dont find any difficulty
> > > > >here.
> > > > >>> > > > > > > > > >  > > > > > > Of course, I hope 199,999 of your inactive ASPs do not decide to
> > > > >become
> > > > >>> > > > > > active at the same time.....! > > > > > > > > > > Theoritically your example sounds good but practically NOBODY will
> > > > >>> > > > > configure 1 ASP ACTIVE and 199,999 ASPs INACTIVE.
> > > > >>> > > > > > > > > > --Samuel
> > > > >>> > > > > > > > > > > > > > > > > -- > > > > Brian F. G. Bidulock
> > > > >>> > > > bidulock@openss7.org
> > > > >>> > > > http://www.openss7.org/
> > > > >>> > > > _______________________________________________
> > > > >>> > > > Sigtran mailing list
> > > > >>> > > > Sigtran@ietf.org
> > > > >>> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > >>> > > > > > > > -- > > Brian F. G. Bidulock
> > > > >>> > bidulock@openss7.org
> > > > >>> > http://www.openss7.org/
> > > > >>> > _______________________________________________
> > > > >>> > Sigtran mailing list
> > > > >>> > Sigtran@ietf.org
> > > > >>> > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > >>> >
> > > > >>
> > > > >>--
> > > > >>Brian F. G. Bidulock
> > > > >>bidulock@openss7.org
> > > > >>http://www.openss7.org/
> > > > >
> > > > >
> > > > >
> > > > >
> > > > >
> > > > >
> > > > >
> > > > >
> > > > >
> > > > >
> > > > >_________________________________________________________________
> > > > >Help STOP SPAM with the new MSN 8 and get 2 months FREE*  
> > > > >http://join.msn.com/?page=features/junkmail
> > > > 
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > 
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 10 05:57:13 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA04030
	for <sigtran-archive@odin.ietf.org>; Mon, 10 Feb 2003 05:57:13 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1AB5Zj06931
	for sigtran-archive@odin.ietf.org; Mon, 10 Feb 2003 06:05:35 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1AB5Ip06918;
	Mon, 10 Feb 2003 06:05:18 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1AB4op06897
	for <sigtran@optimus.ietf.org>; Mon, 10 Feb 2003 06:04:50 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA04000
	for <sigtran@ietf.org>; Mon, 10 Feb 2003 05:55:56 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1AAx7eY006474;
	Mon, 10 Feb 2003 04:59:08 -0600 (CST)
Received: from sun2594.ssd.usa.alcatel.com (sun2594.ssd.usa.alcatel.com [143.209.250.252])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h1AAwhM20529;
	Mon, 10 Feb 2003 04:58:43 -0600 (CST)
Date: Mon, 10 Feb 2003 04:58:43 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: Ken Morneault <kmorneau@cisco.com>
cc: denim john <denim_john@hotmail.com>, sigtran@ietf.org,
        j.javier.pastor@ericsson.com, Francois.Zannin@hp.com,
        bidulock@openss7.org
Subject: Re: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
In-Reply-To: <Pine.SOL.4.10.10302080309550.21630-100000@sun3271.ssd.usa.alcatel.com>
Message-ID: <Pine.SOL.4.10.10302100458030.9486-100000@sun2594.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Any comment from authors?

--Samuel

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Sat, 8 Feb 2003, Samuel Dur D. Jeyaseelan wrote:

> 
> Ken,
> 
> Find the text changes below.
> 
> ======================start===================
> 
> 3.17 Disallowing SSNM messages in ASP-INACTIVE state
>  
> 3.17.1 Description of the problem
>  
>    SSNM messages are received and updated(destination node status) in
>    ASP-ACTIVE state but the backup ASP in inactive state is unaware
>    of SS7 node status changes. So when backup ASP becomes active, it
>    may not take over the traffic as soon as it becomes active.In
>    order to avoid this problem following additional changes are
>    proposed.
>  
>        1.SGP can send SSNM messages to ASP-INACTIVE state.
>    
>        2.ASP in inactive state can receive SSNM messages.
>  
>      Implementing above two points will enhance the efficiency of ASP by
>      taking the traffic as soon as it becomes active.
>  
>      Above two points are true only when dynamic registration is allowed
>      between ASP and SGP.
>  
> 3.17.2 Text changes to the document
>  
>    ---------
>    Old text: (Section 4.6)
>    ---------
>  
>    When the SG discovers SS7 network isolation, the SGPs send an
>    indication to all concerned available ASPs (i.e., ASPs in the ASP-
>    ACTIVE state) using DUNA messages for the concerned destinations.
>  
>    When the SG has completed the MTP Restart procedure, the M3UA layers
>    at the SGPs inform all concerned ASPs in the ASP-ACTIVE state of any
>    available/restricted SS7 destinations using the DAVA/DRST messages.
>    No message is necessary for those destinations still unavailable
>    after the restart procedure.
>  
>    ---------
>    New text: (Section 4.6)
>    ---------
>  
>    When the SG discovers SS7 network isolation, the SGPs send an
>    indication to all concerned available ASPs (i.e., ASPs in the ASP-
>    ACTIVE state or ASP-INACTIVE state) using DUNA messages for the
>    concerned destinations.
>  
>    When the SG has completed the MTP Restart procedure, the M3UA layers
>    at the SGPs inform all concerned ASPs in the ASP-ACTIVE state or ASP-
>    INACTIVE state of any available/restricted SS7 destinations using the
>    DAVA/DRST messages.No message is necessary for those destinations still
>    unavailable after the restart procedure.
>  
>    ---------
>    Old text: (Section 4.5.3)
>    ---------
>  
>       - Isolation. The ASP is newly ASP-ACTIVE or has been
>                    isolated from an SGP for an extended period.  The ASP
>                    MAY request the availability/congestion status of one
>                    or more SS7 destinations to which it expects to
>                    communicate.
>  
>    ---------
>    New text: (Section 4.5.3)
>    ---------
>  
>       - Isolation. The ASP has been isolated from an SGP for an extended
>                    period.  The ASP MAY request the
>                    availability/congestion status of one or more SS7
>                    destinations to which it expects to communicate.
>  
> 3.17.3 Solution description
>  
>    Wherever "concerned ASP" is referred in the specification, it does
>    mean that ASP in active state or ASP in inactive state.This is true
>    only when dynamic registration is allowed between ASP and SGP.
> 
> =================END=========
> 
> --Samuel 
> 
> 
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Fri, 7 Feb 2003, Ken Morneault wrote:
> 
> > 
> > 
> > It would be useful if Samuel proposes text changes that could be 
> > reviewed and, then, we could see if there was consensus for making
> > this change.
> > 
> > Regards,
> > 
> > Ken
> > 
> > At 03:39 PM 2/7/2003 +0000, denim john wrote:
> > 
> > >Brian,
> > >
> > >I would request any one of the authors to conclude this thread.
> > >samuel's idea seems to be good but this will result in changes of m3ua and sua.
> > >
> > >Regards,
> > >Denim.
> > >
> > >An efficent ASP will take up the load as soon as it comes up.
> > >Inefficient ASP may send DAUD and will receive DRST(...),DUNA(...),
> > >SCON(...).  So from the point of ASP following are the unnecessary
> > >messages.
> > >.SENDING DAUD
> > >.RECEING DRST
> > >.RECEIVING DUNA
> > >.RECEIVING SCON
> > >
> > > From the point of SGP these are the unnecessay messages.
> > >.Receiving DAUD
> > >.SENDING DRST
> > >.SENDING DUNA
> > >.SENDING SCON.
> > >Thats what your argument leads.
> > >
> > >--Samuel
> > >
> > > Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > 1000 Coit Road, Plano, Texas 75075
> > > ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > >
> > >On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > >
> > >>Samuel,
> > >>
> > >>That is implementation dependent.  Sending SSNM to inactive ASPs restricts
> > >>a dimension not restricted by the current and adequate procedure.  The
> > >>SG cannot be required to send SSNM to inactive ASPs.  The RFC should not
> > >>place a restriction on the number of inactive ASPs, because we cannot
> > >>forsee all future dimensioning.  Procedures should scale as well as possible.
> > >>The existing procedure scales.  Sending SSNM to inactive ASPs does not.
> > >
> > >>--brian
> > >>
> > >>Samuel Dur D. Jeyaseelan wrote:
> > >(Thu, 06 Feb 2003 23:13:44)
> > >>> > Brian,
> > >>> > Yes , The RFC does not restrict the number of inactive ASPs.
> > >>> But network configuration will restrict.
> > >>> > --Samuel
> > >>> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >>>   1000 Coit Road, Plano, Texas 75075
> > >>>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > >>> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > >>> > > Samuel,
> > >
> > >>> > > > The RFC does not restrict the number of inactive ASPs.
> > >>> > > > --brian
> > >>> > > > Samuel Dur D. Jeyaseelan wrote:
> > >(Thu, 06 Feb 2003 22:40:48)
> > >>> > > > > > Brian,
> > >>> > > > > > Ok, tell me, where you will configure 1 ASP ACT and 199,000 ASPs INACT?
> > >>> > > I want to know which network does this poor configuration.
> > >>> > > > > > --Samuel
> > >>> > > > > > > > >   Alcatel USA, Inc.                  Internet:
> > ><userid>@ssd.usa.alcatel.com
> > >>> > >   1000 Coit Road, Plano, Texas 75075
> > >>> > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > >*******
> > >>> > > > > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > >>> > > > > > > Samuel,
> > >>> > > > > > > > The RFC does not restrict the number of inactive ASPs.
> > >>> > > > > > > > --brian
> > >>> > > > > > > > Samuel Dur D. Jeyaseelan wrote:
> > >(Thu, 06 Feb 2003 22:23:12)
> > >>> > > > > > > > > > Richard,
> > >>> > > > > > > > > >   Alcatel USA, Inc.                  Internet:
> > ><userid>@ssd.usa.alcatel.com
> > >>> > > > >   1000 Coit Road, Plano, Texas 75075
> > >>> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > >*******
> > >>> > > > > > > > > > On Thu, 6 Feb 2003, Richard Adams wrote:
> > >>> > > > > > > > > > > Brian,
> > >>> > > > > > I agree with you. I think when the ASP becomes active, it should
> > >do an
> > >>> > > > > > audit to update its knowledge of the SS7 network. The ASP needs
> > >to take
> > >>> > > > > > the initiative to find out about the SS7 nodes etc. in which it
> > >is
> > >>> > > > > > interested. Any other method of working would lead to impossible
> > >loading
> > >>> > > > > > on the SG.
> > >>> > > > > > > > > > Yes i agree but your suggestion is an indirect method.
> > >>> > > > > I am proposing a direct method to send SSNM messages from SGP to
> > >INACTIVE
> > >>> > > > > ASP. When SGP sends DUNA or DRST to ACTVIE ASP, let him also send a
> > >copy
> > >>> > > > > of the same message to INACTIVE ASP. I dont find any difficulty
> > >here.
> > >>> > > > > > > > > >  > > > > > > Of course, I hope 199,999 of your inactive ASPs do not decide to
> > >become
> > >>> > > > > > active at the same time.....! > > > > > > > > > > Theoritically your example sounds good but practically NOBODY will
> > >>> > > > > configure 1 ASP ACTIVE and 199,999 ASPs INACTIVE.
> > >>> > > > > > > > > > --Samuel
> > >>> > > > > > > > > > > > > > > > > -- > > > > Brian F. G. Bidulock
> > >>> > > > bidulock@openss7.org
> > >>> > > > http://www.openss7.org/
> > >>> > > > _______________________________________________
> > >>> > > > Sigtran mailing list
> > >>> > > > Sigtran@ietf.org
> > >>> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >>> > > > > > > > -- > > Brian F. G. Bidulock
> > >>> > bidulock@openss7.org
> > >>> > http://www.openss7.org/
> > >>> > _______________________________________________
> > >>> > Sigtran mailing list
> > >>> > Sigtran@ietf.org
> > >>> > https://www1.ietf.org/mailman/listinfo/sigtran
> > >>> >
> > >>
> > >>--
> > >>Brian F. G. Bidulock
> > >>bidulock@openss7.org
> > >>http://www.openss7.org/
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> > >_________________________________________________________________
> > >Help STOP SPAM with the new MSN 8 and get 2 months FREE*  
> > >http://join.msn.com/?page=features/junkmail
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 10 07:37:24 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA06892
	for <sigtran-archive@odin.ietf.org>; Mon, 10 Feb 2003 07:37:24 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1ACjl712727
	for sigtran-archive@odin.ietf.org; Mon, 10 Feb 2003 07:45:47 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1ACjXp12715;
	Mon, 10 Feb 2003 07:45:34 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1ACimp12684
	for <sigtran@optimus.ietf.org>; Mon, 10 Feb 2003 07:44:48 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA06849
	for <sigtran@ietf.org>; Mon, 10 Feb 2003 07:35:51 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:Zk1nxgmkNIyo3/7GFV8N8nXlqnxa+lT0@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1ACdID04806;
	Mon, 10 Feb 2003 05:39:18 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1ACdHM00537;
	Mon, 10 Feb 2003 05:39:17 -0700
Date: Mon, 10 Feb 2003 05:39:17 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: Ken Morneault <kmorneau@cisco.com>, denim john <denim_john@hotmail.com>,
        sigtran@ietf.org, j.javier.pastor@ericsson.com, Francois.Zannin@hp.com
Subject: Re: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
Message-ID: <20030210053917.A32713@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	Ken Morneault <kmorneau@cisco.com>,
	denim john <denim_john@hotmail.com>, sigtran@ietf.org,
	j.javier.pastor@ericsson.com, Francois.Zannin@hp.com
References: <Pine.SOL.4.10.10302080309550.21630-100000@sun3271.ssd.usa.alcatel.com> <Pine.SOL.4.10.10302100458030.9486-100000@sun2594.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302100458030.9486-100000@sun2594.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Mon, Feb 10, 2003 at 04:58:43AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

I am an author.

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Mon, 10 Feb 2003 04:58:43)
> 
> Any comment from authors?
> 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 10 08:57:28 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA09014
	for <sigtran-archive@odin.ietf.org>; Mon, 10 Feb 2003 08:57:28 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1AE5po16619
	for sigtran-archive@odin.ietf.org; Mon, 10 Feb 2003 09:05:51 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1AE5Xp16601;
	Mon, 10 Feb 2003 09:05:34 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1AE4tp16575
	for <sigtran@optimus.ietf.org>; Mon, 10 Feb 2003 09:04:55 -0500
Received: from domino2.nc.tekelec.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA08954
	for <sigtran@ietf.org>; Mon, 10 Feb 2003 08:55:58 -0500 (EST)
From: jeffrey.craig@tekelec.com
Received: from domino4.nc.tekelec.com ([198.89.40.9])
          by domino2.nc.tekelec.com (Lotus Domino Release 5.0.10)
          with ESMTP id 2003021008532145:26379 ;
          Mon, 10 Feb 2003 08:53:21 -0500 
Subject: Re: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: Ken Morneault <kmorneau@cisco.com>, denim john <denim_john@hotmail.com>,
        sigtran@ietf.org, j.javier.pastor@ericsson.com, Francois.Zannin@hp.com,
        bidulock@openss7.org
X-Mailer: Lotus Notes Release 5.0.7  March 21, 2001
Message-ID: <OFB258F7C9.41E159E2-ON85256CC9.004BF508@nc.tekelec.com>
Date: Mon, 10 Feb 2003 08:55:24 -0500
MIME-Version: 1.0
X-MIMETrack: Serialize by Router on Domino4/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/10/2003 08:55:25 AM,
	Itemize by SMTP Server on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/10/2003 08:53:21 AM,
	Serialize by Router on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/10/2003 08:53:24 AM,
	Serialize complete at 02/10/2003 08:53:24 AM
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hello All,

>Any comment from authors?

I am not an author, but my thinking is that an SG SHOULD NOT send
unsolicited SSNMs to an INACTIVE ASP, and an SG MAY respond to an
audit issued by an INACTIVE ASP (i.e. send solicited SSNMs).

Regards,

Jeff







"Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
02/10/2003 05:58:43 AM

Sent by:    sigtran-admin@ietf.org


To:    Ken Morneault <kmorneau@cisco.com>
cc:    denim john <denim_john@hotmail.com>, sigtran@ietf.org,
       j.javier.pastor@ericsson.com, Francois.Zannin@hp.com,
       bidulock@openss7.org
Subject:    Re: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization



Any comment from authors?

--Samuel

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Sat, 8 Feb 2003, Samuel Dur D. Jeyaseelan wrote:

>
> Ken,
>
> Find the text changes below.
>
> ======================start===================
>
> 3.17 Disallowing SSNM messages in ASP-INACTIVE state
>
> 3.17.1 Description of the problem
>
>    SSNM messages are received and updated(destination node status) in
>    ASP-ACTIVE state but the backup ASP in inactive state is unaware
>    of SS7 node status changes. So when backup ASP becomes active, it
>    may not take over the traffic as soon as it becomes active.In
>    order to avoid this problem following additional changes are
>    proposed.
>
>        1.SGP can send SSNM messages to ASP-INACTIVE state.
>
>        2.ASP in inactive state can receive SSNM messages.
>
>      Implementing above two points will enhance the efficiency of ASP by
>      taking the traffic as soon as it becomes active.
>
>      Above two points are true only when dynamic registration is allowed
>      between ASP and SGP.
>
> 3.17.2 Text changes to the document
>
>    ---------
>    Old text: (Section 4.6)
>    ---------
>
>    When the SG discovers SS7 network isolation, the SGPs send an
>    indication to all concerned available ASPs (i.e., ASPs in the ASP-
>    ACTIVE state) using DUNA messages for the concerned destinations.
>
>    When the SG has completed the MTP Restart procedure, the M3UA layers
>    at the SGPs inform all concerned ASPs in the ASP-ACTIVE state of any
>    available/restricted SS7 destinations using the DAVA/DRST messages.
>    No message is necessary for those destinations still unavailable
>    after the restart procedure.
>
>    ---------
>    New text: (Section 4.6)
>    ---------
>
>    When the SG discovers SS7 network isolation, the SGPs send an
>    indication to all concerned available ASPs (i.e., ASPs in the ASP-
>    ACTIVE state or ASP-INACTIVE state) using DUNA messages for the
>    concerned destinations.
>
>    When the SG has completed the MTP Restart procedure, the M3UA layers
>    at the SGPs inform all concerned ASPs in the ASP-ACTIVE state or ASP-
>    INACTIVE state of any available/restricted SS7 destinations using the
>    DAVA/DRST messages.No message is necessary for those destinations
still
>    unavailable after the restart procedure.
>
>    ---------
>    Old text: (Section 4.5.3)
>    ---------
>
>       - Isolation. The ASP is newly ASP-ACTIVE or has been
>                    isolated from an SGP for an extended period.  The ASP
>                    MAY request the availability/congestion status of one
>                    or more SS7 destinations to which it expects to
>                    communicate.
>
>    ---------
>    New text: (Section 4.5.3)
>    ---------
>
>       - Isolation. The ASP has been isolated from an SGP for an extended
>                    period.  The ASP MAY request the
>                    availability/congestion status of one or more SS7
>                    destinations to which it expects to communicate.
>
> 3.17.3 Solution description
>
>    Wherever "concerned ASP" is referred in the specification, it does
>    mean that ASP in active state or ASP in inactive state.This is true
>    only when dynamic registration is allowed between ASP and SGP.
>
> =================END=========
>
> --Samuel
>
>
>
>   Alcatel USA, Inc.                  Internet: <userid>
@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc.
*******
>
> On Fri, 7 Feb 2003, Ken Morneault wrote:
>
> >
> >
> > It would be useful if Samuel proposes text changes that could be
> > reviewed and, then, we could see if there was consensus for making
> > this change.
> >
> > Regards,
> >
> > Ken
> >
> > At 03:39 PM 2/7/2003 +0000, denim john wrote:
> >
> > >Brian,
> > >
> > >I would request any one of the authors to conclude this thread.
> > >samuel's idea seems to be good but this will result in changes of m3ua
and sua.
> > >
> > >Regards,
> > >Denim.
> > >
> > >An efficent ASP will take up the load as soon as it comes up.
> > >Inefficient ASP may send DAUD and will receive DRST(...),DUNA(...),
> > >SCON(...).  So from the point of ASP following are the unnecessary
> > >messages.
> > >.SENDING DAUD
> > >.RECEING DRST
> > >.RECEIVING DUNA
> > >.RECEIVING SCON
> > >
> > > From the point of SGP these are the unnecessay messages.
> > >.Receiving DAUD
> > >.SENDING DRST
> > >.SENDING DUNA
> > >.SENDING SCON.
> > >Thats what your argument leads.
> > >
> > >--Samuel
> > >
> > > Alcatel USA, Inc.                  Internet: <userid>
@ssd.usa.alcatel.com
> > > 1000 Coit Road, Plano, Texas 75075
> > > ******* The opinions expressed are not those of Alcatel USA, Inc.
*******
> > >
> > >On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > >
> > >>Samuel,
> > >>
> > >>That is implementation dependent.  Sending SSNM to inactive ASPs
restricts
> > >>a dimension not restricted by the current and adequate procedure.
The
> > >>SG cannot be required to send SSNM to inactive ASPs.  The RFC should
not
> > >>place a restriction on the number of inactive ASPs, because we cannot
> > >>forsee all future dimensioning.  Procedures should scale as well as
possible.
> > >>The existing procedure scales.  Sending SSNM to inactive ASPs does
not.
> > >
> > >>--brian
> > >>
> > >>Samuel Dur D. Jeyaseelan wrote:
> > >(Thu, 06 Feb 2003 23:13:44)
> > >>> > Brian,
> > >>> > Yes , The RFC does not restrict the number of inactive ASPs.
> > >>> But network configuration will restrict.
> > >>> > --Samuel
> > >>> >   Alcatel USA, Inc.                  Internet: <userid>
@ssd.usa.alcatel.com
> > >>>   1000 Coit Road, Plano, Texas 75075
> > >>>   ******* The opinions expressed are not those of Alcatel USA, Inc.
*******
> > >>> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > >>> > > Samuel,
> > >
> > >>> > > > The RFC does not restrict the number of inactive ASPs.
> > >>> > > > --brian
> > >>> > > > Samuel Dur D. Jeyaseelan wrote:
> > >(Thu, 06 Feb 2003 22:40:48)
> > >>> > > > > > Brian,
> > >>> > > > > > Ok, tell me, where you will configure 1 ASP ACT and
199,000 ASPs INACT?
> > >>> > > I want to know which network does this poor configuration.
> > >>> > > > > > --Samuel
> > >>> > > > > > > > >   Alcatel USA, Inc.                  Internet:
> > ><userid>@ssd.usa.alcatel.com
> > >>> > >   1000 Coit Road, Plano, Texas 75075
> > >>> > >   ******* The opinions expressed are not those of Alcatel USA,
Inc.
> > >*******
> > >>> > > > > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
> > >>> > > > > > > Samuel,
> > >>> > > > > > > > The RFC does not restrict the number of inactive
ASPs.
> > >>> > > > > > > > --brian
> > >>> > > > > > > > Samuel Dur D. Jeyaseelan wrote:
> > >(Thu, 06 Feb 2003 22:23:12)
> > >>> > > > > > > > > > Richard,
> > >>> > > > > > > > > >   Alcatel USA, Inc.                  Internet:
> > ><userid>@ssd.usa.alcatel.com
> > >>> > > > >   1000 Coit Road, Plano, Texas 75075
> > >>> > > > >   ******* The opinions expressed are not those of Alcatel
USA, Inc.
> > >*******
> > >>> > > > > > > > > > On Thu, 6 Feb 2003, Richard Adams wrote:
> > >>> > > > > > > > > > > Brian,
> > >>> > > > > > I agree with you. I think when the ASP becomes active, it
should
> > >do an
> > >>> > > > > > audit to update its knowledge of the SS7 network. The ASP
needs
> > >to take
> > >>> > > > > > the initiative to find out about the SS7 nodes etc. in
which it
> > >is
> > >>> > > > > > interested. Any other method of working would lead to
impossible
> > >loading
> > >>> > > > > > on the SG.
> > >>> > > > > > > > > > Yes i agree but your suggestion is an indirect
method.
> > >>> > > > > I am proposing a direct method to send SSNM messages from
SGP to
> > >INACTIVE
> > >>> > > > > ASP. When SGP sends DUNA or DRST to ACTVIE ASP, let him
also send a
> > >copy
> > >>> > > > > of the same message to INACTIVE ASP. I dont find any
difficulty
> > >here.
> > >>> > > > > > > > > >  > > > > > > Of course, I hope 199,999 of your
inactive ASPs do not decide to
> > >become
> > >>> > > > > > active at the same time.....! > > > > > > > > > >
Theoritically your example sounds good but practically NOBODY will
> > >>> > > > > configure 1 ASP ACTIVE and 199,999 ASPs INACTIVE.
> > >>> > > > > > > > > > --Samuel
> > >>> > > > > > > > > > > > > > > > > -- > > > > Brian F. G. Bidulock
> > >>> > > > bidulock@openss7.org
> > >>> > > > http://www.openss7.org/
> > >>> > > > _______________________________________________
> > >>> > > > Sigtran mailing list
> > >>> > > > Sigtran@ietf.org
> > >>> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >>> > > > > > > > -- > > Brian F. G. Bidulock
> > >>> > bidulock@openss7.org
> > >>> > http://www.openss7.org/
> > >>> > _______________________________________________
> > >>> > Sigtran mailing list
> > >>> > Sigtran@ietf.org
> > >>> > https://www1.ietf.org/mailman/listinfo/sigtran
> > >>> >
> > >>
> > >>--
> > >>Brian F. G. Bidulock
> > >>bidulock@openss7.org
> > >>http://www.openss7.org/
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> > >_________________________________________________________________
> > >Help STOP SPAM with the new MSN 8 and get 2 months FREE*
> > >http://join.msn.com/?page=features/junkmail
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> >
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 10 09:04:54 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA09493
	for <sigtran-archive@odin.ietf.org>; Mon, 10 Feb 2003 09:04:54 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1AEDJE17580
	for sigtran-archive@odin.ietf.org; Mon, 10 Feb 2003 09:13:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1AED6p17570;
	Mon, 10 Feb 2003 09:13:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1AEChp17528
	for <sigtran@optimus.ietf.org>; Mon, 10 Feb 2003 09:12:43 -0500
Received: from rtp-core-1.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA09446
	for <sigtran@ietf.org>; Mon, 10 Feb 2003 09:03:46 -0500 (EST)
Received: from cia.cisco.com (IDENT:mirapoint@cia.cisco.com [161.44.85.200])
	by rtp-core-1.cisco.com (8.12.6/8.12.6) with ESMTP id h1AE7LJR000727;
	Mon, 10 Feb 2003 09:07:22 -0500 (EST)
Received: from kmorneau-w2k.cisco.com (rtp-vpn2-464.cisco.com [10.82.241.208])
	by cia.cisco.com (Mirapoint)
	with ESMTP id ABO82689;
	Mon, 10 Feb 2003 08:57:20 -0500 (EST)
Message-Id: <4.3.2.7.2.20030210090258.04b96008@cia.cisco.com>
X-Sender: kmorneau@cia.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Mon, 10 Feb 2003 09:07:15 -0500
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
From: Ken Morneault <kmorneau@cisco.com>
Subject: Re: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
Cc: denim john <denim_john@hotmail.com>, sigtran@ietf.org,
        j.javier.pastor@ericsson.com, Francois.Zannin@hp.com,
        bidulock@openss7.org
In-Reply-To: <Pine.SOL.4.10.10302100458030.9486-100000@sun2594.ssd.usa.a
 lcatel.com>
References: <Pine.SOL.4.10.10302080309550.21630-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>



Samuel,

It really doesn't matter what the authors think.  What matters is
the consensus of the group.

I'm not convinced that these text changes are needed.  Your
one argument is to increase the efficiency of the ASP during
failover.  However, these changes decrease the
efficiency of the SG for each inactive ASP.  This could lead
to unnecessary scaling problems for the SG.

If efficiency during failover is desired, it should be a goal
of the ASP design (which I think has been discussed previously
on this thread).  

Hopefully, more people will send in their comments/opinions.

Regards,

Ken

At 04:58 AM 2/10/2003 -0600, Samuel Dur D. Jeyaseelan wrote:

>Any comment from authors?
>
>--Samuel
>
>  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>  1000 Coit Road, Plano, Texas 75075
>  ******* The opinions expressed are not those of Alcatel USA, Inc. *******
>
>On Sat, 8 Feb 2003, Samuel Dur D. Jeyaseelan wrote:
>
>> 
>> Ken,
>> 
>> Find the text changes below.
>> 
>> ======================start===================
>> 
>> 3.17 Disallowing SSNM messages in ASP-INACTIVE state
>>  
>> 3.17.1 Description of the problem
>>  
>>    SSNM messages are received and updated(destination node status) in
>>    ASP-ACTIVE state but the backup ASP in inactive state is unaware
>>    of SS7 node status changes. So when backup ASP becomes active, it
>>    may not take over the traffic as soon as it becomes active.In
>>    order to avoid this problem following additional changes are
>>    proposed.
>>  
>>        1.SGP can send SSNM messages to ASP-INACTIVE state.
>>    
>>        2.ASP in inactive state can receive SSNM messages.
>>  
>>      Implementing above two points will enhance the efficiency of ASP by
>>      taking the traffic as soon as it becomes active.
>>  
>>      Above two points are true only when dynamic registration is allowed
>>      between ASP and SGP.
>>  
>> 3.17.2 Text changes to the document
>>  
>>    ---------
>>    Old text: (Section 4.6)
>>    ---------
>>  
>>    When the SG discovers SS7 network isolation, the SGPs send an
>>    indication to all concerned available ASPs (i.e., ASPs in the ASP-
>>    ACTIVE state) using DUNA messages for the concerned destinations.
>>  
>>    When the SG has completed the MTP Restart procedure, the M3UA layers
>>    at the SGPs inform all concerned ASPs in the ASP-ACTIVE state of any
>>    available/restricted SS7 destinations using the DAVA/DRST messages.
>>    No message is necessary for those destinations still unavailable
>>    after the restart procedure.
>>  
>>    ---------
>>    New text: (Section 4.6)
>>    ---------
>>  
>>    When the SG discovers SS7 network isolation, the SGPs send an
>>    indication to all concerned available ASPs (i.e., ASPs in the ASP-
>>    ACTIVE state or ASP-INACTIVE state) using DUNA messages for the
>>    concerned destinations.
>>  
>>    When the SG has completed the MTP Restart procedure, the M3UA layers
>>    at the SGPs inform all concerned ASPs in the ASP-ACTIVE state or ASP-
>>    INACTIVE state of any available/restricted SS7 destinations using the
>>    DAVA/DRST messages.No message is necessary for those destinations still
>>    unavailable after the restart procedure.
>>  
>>    ---------
>>    Old text: (Section 4.5.3)
>>    ---------
>>  
>>       - Isolation. The ASP is newly ASP-ACTIVE or has been
>>                    isolated from an SGP for an extended period.  The ASP
>>                    MAY request the availability/congestion status of one
>>                    or more SS7 destinations to which it expects to
>>                    communicate.
>>  
>>    ---------
>>    New text: (Section 4.5.3)
>>    ---------
>>  
>>       - Isolation. The ASP has been isolated from an SGP for an extended
>>                    period.  The ASP MAY request the
>>                    availability/congestion status of one or more SS7
>>                    destinations to which it expects to communicate.
>>  
>> 3.17.3 Solution description
>>  
>>    Wherever "concerned ASP" is referred in the specification, it does
>>    mean that ASP in active state or ASP in inactive state.This is true
>>    only when dynamic registration is allowed between ASP and SGP.
>> 
>> =================END=========
>> 
>> --Samuel 
>> 
>> 
>> 
>>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>>   1000 Coit Road, Plano, Texas 75075
>>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
>> 
>> On Fri, 7 Feb 2003, Ken Morneault wrote:
>> 
>> > 
>> > 
>> > It would be useful if Samuel proposes text changes that could be 
>> > reviewed and, then, we could see if there was consensus for making
>> > this change.
>> > 
>> > Regards,
>> > 
>> > Ken
>> > 
>> > At 03:39 PM 2/7/2003 +0000, denim john wrote:
>> > 
>> > >Brian,
>> > >
>> > >I would request any one of the authors to conclude this thread.
>> > >samuel's idea seems to be good but this will result in changes of m3ua and sua.
>> > >
>> > >Regards,
>> > >Denim.
>> > >
>> > >An efficent ASP will take up the load as soon as it comes up.
>> > >Inefficient ASP may send DAUD and will receive DRST(...),DUNA(...),
>> > >SCON(...).  So from the point of ASP following are the unnecessary
>> > >messages.
>> > >.SENDING DAUD
>> > >.RECEING DRST
>> > >.RECEIVING DUNA
>> > >.RECEIVING SCON
>> > >
>> > > From the point of SGP these are the unnecessay messages.
>> > >.Receiving DAUD
>> > >.SENDING DRST
>> > >.SENDING DUNA
>> > >.SENDING SCON.
>> > >Thats what your argument leads.
>> > >
>> > >--Samuel
>> > >
>> > > Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>> > > 1000 Coit Road, Plano, Texas 75075
>> > > ******* The opinions expressed are not those of Alcatel USA, Inc. *******
>> > >
>> > >On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
>> > >
>> > >>Samuel,
>> > >>
>> > >>That is implementation dependent.  Sending SSNM to inactive ASPs restricts
>> > >>a dimension not restricted by the current and adequate procedure.  The
>> > >>SG cannot be required to send SSNM to inactive ASPs.  The RFC should not
>> > >>place a restriction on the number of inactive ASPs, because we cannot
>> > >>forsee all future dimensioning.  Procedures should scale as well as possible.
>> > >>The existing procedure scales.  Sending SSNM to inactive ASPs does not.
>> > >
>> > >>--brian
>> > >>
>> > >>Samuel Dur D. Jeyaseelan wrote:
>> > >(Thu, 06 Feb 2003 23:13:44)
>> > >>> > Brian,
>> > >>> > Yes , The RFC does not restrict the number of inactive ASPs.
>> > >>> But network configuration will restrict.
>> > >>> > --Samuel
>> > >>> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>> > >>>   1000 Coit Road, Plano, Texas 75075
>> > >>>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
>> > >>> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
>> > >>> > > Samuel,
>> > >
>> > >>> > > > The RFC does not restrict the number of inactive ASPs.
>> > >>> > > > --brian
>> > >>> > > > Samuel Dur D. Jeyaseelan wrote:
>> > >(Thu, 06 Feb 2003 22:40:48)
>> > >>> > > > > > Brian,
>> > >>> > > > > > Ok, tell me, where you will configure 1 ASP ACT and 199,000 ASPs INACT?
>> > >>> > > I want to know which network does this poor configuration.
>> > >>> > > > > > --Samuel
>> > >>> > > > > > > > >   Alcatel USA, Inc.                  Internet:
>> > ><userid>@ssd.usa.alcatel.com
>> > >>> > >   1000 Coit Road, Plano, Texas 75075
>> > >>> > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
>> > >*******
>> > >>> > > > > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
>> > >>> > > > > > > Samuel,
>> > >>> > > > > > > > The RFC does not restrict the number of inactive ASPs.
>> > >>> > > > > > > > --brian
>> > >>> > > > > > > > Samuel Dur D. Jeyaseelan wrote:
>> > >(Thu, 06 Feb 2003 22:23:12)
>> > >>> > > > > > > > > > Richard,
>> > >>> > > > > > > > > >   Alcatel USA, Inc.                  Internet:
>> > ><userid>@ssd.usa.alcatel.com
>> > >>> > > > >   1000 Coit Road, Plano, Texas 75075
>> > >>> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
>> > >*******
>> > >>> > > > > > > > > > On Thu, 6 Feb 2003, Richard Adams wrote:
>> > >>> > > > > > > > > > > Brian,
>> > >>> > > > > > I agree with you. I think when the ASP becomes active, it should
>> > >do an
>> > >>> > > > > > audit to update its knowledge of the SS7 network. The ASP needs
>> > >to take
>> > >>> > > > > > the initiative to find out about the SS7 nodes etc. in which it
>> > >is
>> > >>> > > > > > interested. Any other method of working would lead to impossible
>> > >loading
>> > >>> > > > > > on the SG.
>> > >>> > > > > > > > > > Yes i agree but your suggestion is an indirect method.
>> > >>> > > > > I am proposing a direct method to send SSNM messages from SGP to
>> > >INACTIVE
>> > >>> > > > > ASP. When SGP sends DUNA or DRST to ACTVIE ASP, let him also send a
>> > >copy
>> > >>> > > > > of the same message to INACTIVE ASP. I dont find any difficulty
>> > >here.
>> > >>> > > > > > > > > >  > > > > > > Of course, I hope 199,999 of your inactive ASPs do not decide to
>> > >become
>> > >>> > > > > > active at the same time.....! > > > > > > > > > > Theoritically your example sounds good but practically NOBODY will
>> > >>> > > > > configure 1 ASP ACTIVE and 199,999 ASPs INACTIVE.
>> > >>> > > > > > > > > > --Samuel
>> > >>> > > > > > > > > > > > > > > > > -- > > > > Brian F. G. Bidulock
>> > >>> > > > bidulock@openss7.org
>> > >>> > > > http://www.openss7.org/
>> > >>> > > > _______________________________________________
>> > >>> > > > Sigtran mailing list
>> > >>> > > > Sigtran@ietf.org
>> > >>> > > > https://www1.ietf.org/mailman/listinfo/sigtran
>> > >>> > > > > > > > -- > > Brian F. G. Bidulock
>> > >>> > bidulock@openss7.org
>> > >>> > http://www.openss7.org/
>> > >>> > _______________________________________________
>> > >>> > Sigtran mailing list
>> > >>> > Sigtran@ietf.org
>> > >>> > https://www1.ietf.org/mailman/listinfo/sigtran
>> > >>> >
>> > >>
>> > >>--
>> > >>Brian F. G. Bidulock
>> > >>bidulock@openss7.org
>> > >>http://www.openss7.org/
>> > >
>> > >
>> > >
>> > >
>> > >
>> > >
>> > >
>> > >
>> > >
>> > >
>> > >_________________________________________________________________
>> > >Help STOP SPAM with the new MSN 8 and get 2 months FREE*  
>> > >http://join.msn.com/?page=features/junkmail
>> > 
>> > _______________________________________________
>> > Sigtran mailing list
>> > Sigtran@ietf.org
>> > https://www1.ietf.org/mailman/listinfo/sigtran
>> > 
>> 
>> _______________________________________________
>> Sigtran mailing list
>> Sigtran@ietf.org
>> https://www1.ietf.org/mailman/listinfo/sigtran
>> 
>
>_______________________________________________
>Sigtran mailing list
>Sigtran@ietf.org
>https://www1.ietf.org/mailman/listinfo/sigtran 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 10 09:05:52 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA09556
	for <sigtran-archive@odin.ietf.org>; Mon, 10 Feb 2003 09:05:51 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1AEEH417670
	for sigtran-archive@odin.ietf.org; Mon, 10 Feb 2003 09:14:17 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1AEE3p17661;
	Mon, 10 Feb 2003 09:14:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1AEDup17606
	for <sigtran@optimus.ietf.org>; Mon, 10 Feb 2003 09:13:56 -0500
Received: from rtp-core-1.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA09509
	for <sigtran@ietf.org>; Mon, 10 Feb 2003 09:04:58 -0500 (EST)
Received: from cia.cisco.com (IDENT:mirapoint@cia.cisco.com [161.44.85.200])
	by rtp-core-1.cisco.com (8.12.6/8.12.6) with ESMTP id h1AE8XJR000777;
	Mon, 10 Feb 2003 09:08:34 -0500 (EST)
Received: from kmorneau-w2k.cisco.com (rtp-vpn2-464.cisco.com [10.82.241.208])
	by cia.cisco.com (Mirapoint)
	with ESMTP id ABO82697;
	Mon, 10 Feb 2003 08:58:32 -0500 (EST)
Message-Id: <4.3.2.7.2.20030210090743.0497cd38@cia.cisco.com>
X-Sender: kmorneau@cia.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Mon, 10 Feb 2003 09:08:27 -0500
To: bidulock@openss7.org,
        "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
From: Ken Morneault <kmorneau@cisco.com>
Subject: Re: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
Cc: denim john <denim_john@hotmail.com>, sigtran@ietf.org,
        j.javier.pastor@ericsson.com, Francois.Zannin@hp.com
In-Reply-To: <20030210022503.A4738@openss7.org>
References: <Pine.SOL.4.10.10302092154080.1868-100000@sun2594.ssd.usa.alcatel.com>
 <20030208040242.B2026@openss7.org>
 <Pine.SOL.4.10.10302092154080.1868-100000@sun2594.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>



I think the recommended text Brian provides below would be useful
for the IG.

Ken

At 02:25 AM 2/10/2003 -0700, Brian F. G. Bidulock wrote:
>Samuel,
>
>I disagree with all your comments, and your proposal.
>
>--brian
>
>Samuel Dur D. Jeyaseelan wrote:                                                     (Sun, 09 Feb 2003 22:27:13)
>> 
>> Brian,
>> 
>>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>>   1000 Coit Road, Plano, Texas 75075
>>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
>> 
>> On Sat, 8 Feb 2003, Brian F. G. Bidulock wrote:
>> 
>> > Samuel,
>> > 
>> > The SG cannot be required to send SSNM to inactive ASPs.
>> > 
>> > Also, unregistered ASPs are also in the ASP-INACTIVE state, yet are
>> > unable to know to which AS an RC corresponds.  An SG which supports both
>> > dynamic and static configuration cannot know if and ASP is already
>> > configured with the RC of the AS, or whether it must regsiter first.
>> > (An AS does not have an "unregistered" state.)  Therefore, the SG cannot
>> > know whether sending SSNM to ASP in the ASP-INACTIVE state for an AS
>> > will induce an error (Invalid RC), and when an error is induced whether
>> > it is indeed because the ASP needs to register for the AS, or whether
>> > some other configuration problem has occured.
>> > 
>> misunderstanding.See the proposed text changes,  it says that "SGP can
>> send SSNM messages to ASP", only when dynamic registration is allowed,
>> that means both SGP and ASP know RC.
>> 
>> > Also, in Section 4.6, and ASP in the ASP-INACTIVE state is not a
>> > "concerned" ASP.  You should note that the SG is not required to send
>> > SSNM to all of the ASP-ACTIVE ASPs either: only the concerned ASPs in
>> > the ASP-ACTIVE state for the AS.  When sending SSNM in using reponsive
>> > procedures, it is only the ASP to which the SSNM is a response that is
>> > "concerned".
>> > 
>> Again wrong. Proposed text changes will consider 'concerned ASP' as
>> ASP-ACTIVE or ASP-INACTIVE only when dynamic registration is allowed.
>> It means SGP knows AS and ASP knows RC.
>> 
>> > MTP does not indicate to its users whether an MTP-PAUSE, MTP-RESUME, or
>> > MTP-STATUS primitive is responsive or broadcast.  Therefore, some
>> > implementations cannot determine which of the ASP-ACTIVE ASPs are
>> > "concerned".  But certainly, for both responsive indications, ASPs in
>> > the ASP-INACTIVE state are not concerned.
>> > 
>> > Also, the ASP-INACTIVE state for an AS corresponds to no MTP-User being
>> > associated with the MTP-SAP.  This means that when all ASPs are in the
>> > ASP-INACTIVE state, there is no MTP User associated with the MTP at the
>> > SG.  Because there is no MTP-User associated with the MTP at the SG, the
>> > SG cannot deliver MTP-PAUSE, MTP-RESUME or MTP-STATUS primitives to M3UA
>> > at the SG and cannot then issue SSNM to ASPs.
>> > 
>> Wrong. LOGICALLY MTP's user part is available at SG.
>> Logically ASP-ACTIVE means ASP is ready to take over the traffic,
>> But still inactive ASP can receive SSNM messages but it SHOULD not 
>> take up the traffice as long as it is in inactive state.
>> 
>> > When all the ASPs are ASP-INACTIVE, the SG cannot be expected to send
>> > SSNM to any of them.
>> > 
>> No, Sending SSNM to inactive ASP will increase the efficiency of ASP.
>> 
>> > The RFC cannot require an implementation at the SG to use facilities
>> > which are not provided to the MTP-User by MTP.  (Things like DRST are
>> > OPTIONAL).
>> > 
>> > ASPs which are not in the multiple SG configuration are not required to
>> > maintain any destination state whatsoever: SSNM delivered to the ASP are
>> > delivered directly to the user as the corresponding MTP-PAUSE,
>> > MTP-RESUME and MTP-STATUS primitives.  When an ASP is in the
>> > ASP-INACTIVE state for an AS, it has no MTP-User to delivere these
>> > primitive to.  Therefore, it is unnecessary to send SSNM to ASPs in the
>> > ASP-INACTIVE state in the single SG configuration, and the SG cannot be
>> > required to send them.  Multiple SG arrangements are OPTIONAL.
>> > 
>> 
>> You have not read properly the 'proposed text changes'.
>> 
>> > So, as before, the SG cannot be required to send SSNM to inactives ASPs.
>> > 
>> > The current RFC says that the SG SHOULD NOT send DATA or SSNM to
>> > inactive ASPs or it will risk data loss.  This does not forbit the SG
>> > from sending SSNM to inactive ASPs, it just recommends that it doesn't.
>> > Becase of the above, to recommend that SG send SSNM to inactive ASPs
>> > with the text changes you propose is not workable.
>> 
>> Proposed text changes is only for DUNA,DRST,SCON and DAVA not for DATA.
>> Updating destination node status changes at inactive ASP will not have
>> any side effects.
>> 
>> > 
>> > I am in total disagreement with your proposal.
>> > 
>> > Three others on this thread have also expressed that the current RFC is
>> > sufficient in this regard.
>> > 
>> Three others also proposed "Sending SSNM to inactive ASP is good method".
>> 
>> > The only change that I see as necessary would be to add the following
>> > statement:
>> > 
>> >     When an ASP becomes active for an AS and the SG is experiencing
>> >     SS7 network isolation or is performing the MTP Restart procedure
>> >     for the AS, the SG SHOULD send a DUNA(*) to the newly active ASP
>> >     to prevent the ASP from sending traffic.
>> > 
>> >     When an ASP becomes active for an AS and destinations normally
>> >     acessible to the ASP are inaccessible, restricted or congested
>> >     for the AS, the SG SHOULD send DUNA, DRST or SCON AS for the
>> >     inacessible, restricted or congested destinations to the ASP
>> >     newly active for the AS to prevent the ASP from sending traffic
>> >     for destinations that it might otherwise not know are
>> >     inaccessible, restricted or congested.
>> > 
>> > Still, the SG cannot be required to send SSNM to inactive ASPs, and the
>> > ASP can certainly not rely upon receiving SSNM while inactive.  So, the
>> > only sure way for the ASP to obtain information is after it goes ASP-ACTIVE.
>> > 
>> 
>> Once the dynamic registration is over, Both SGP and ASP know th RC.
>> Once RC is known to SGP and ASP , it means MTP user part is availabe at
>> SG.
>> ONLY DATA message should not get exchanged when ASP in inactive state.
>> 
>> --Samuel
>> > --brian
>> > 
>> > 
>> > Samuel Dur D. Jeyaseelan wrote:                                                     (Sat, 08 Feb 2003 03:35:26)
>> > > 
>> > > Ken,
>> > > 
>> > > Find the text changes below.
>> > > 
>> > > ======================start===================
>> > > 
>> > > 3.17 Disallowing SSNM messages in ASP-INACTIVE state
>> > >  
>> > > 3.17.1 Description of the problem
>> > >  
>> > >    SSNM messages are received and updated(destination node status) in
>> > >    ASP-ACTIVE state but the backup ASP in inactive state is unaware
>> > >    of SS7 node status changes. So when backup ASP becomes active, it
>> > >    may not take over the traffic as soon as it becomes active.In
>> > >    order to avoid this problem following additional changes are
>> > >    proposed.
>> > >  
>> > >        1.SGP can send SSNM messages to ASP-INACTIVE state.
>> > >    
>> > >        2.ASP in inactive state can receive SSNM messages.
>> > >  
>> > >      Implementing above two points will enhance the efficiency of ASP by
>> > >      taking the traffic as soon as it becomes active.
>> > >  
>> > >      Above two points are true only when dynamic registration is allowed
>> > >      between ASP and SGP.
>> > >  
>> > > 3.17.2 Text changes to the document
>> > >  
>> > >    ---------
>> > >    Old text: (Section 4.6)
>> > >    ---------
>> > >  
>> > >    When the SG discovers SS7 network isolation, the SGPs send an
>> > >    indication to all concerned available ASPs (i.e., ASPs in the ASP-
>> > >    ACTIVE state) using DUNA messages for the concerned destinations.
>> > >  
>> > >    When the SG has completed the MTP Restart procedure, the M3UA layers
>> > >    at the SGPs inform all concerned ASPs in the ASP-ACTIVE state of any
>> > >    available/restricted SS7 destinations using the DAVA/DRST messages.
>> > >    No message is necessary for those destinations still unavailable
>> > >    after the restart procedure.
>> > >  
>> > >    ---------
>> > >    New text: (Section 4.6)
>> > >    ---------
>> > >  
>> > >    When the SG discovers SS7 network isolation, the SGPs send an
>> > >    indication to all concerned available ASPs (i.e., ASPs in the ASP-
>> > >    ACTIVE state or ASP-INACTIVE state) using DUNA messages for the
>> > >    concerned destinations.
>> > >  
>> > >    When the SG has completed the MTP Restart procedure, the M3UA layers
>> > >    at the SGPs inform all concerned ASPs in the ASP-ACTIVE state or ASP-
>> > >    INACTIVE state of any available/restricted SS7 destinations using the
>> > >    DAVA/DRST messages.No message is necessary for those destinations still
>> > >    unavailable after the restart procedure.
>> > >  
>> > >    ---------
>> > >    Old text: (Section 4.5.3)
>> > >    ---------
>> > >  
>> > >       - Isolation. The ASP is newly ASP-ACTIVE or has been
>> > >                    isolated from an SGP for an extended period.  The ASP
>> > >                    MAY request the availability/congestion status of one
>> > >                    or more SS7 destinations to which it expects to
>> > >                    communicate.
>> > >  
>> > >    ---------
>> > >    New text: (Section 4.5.3)
>> > >    ---------
>> > >  
>> > >       - Isolation. The ASP has been isolated from an SGP for an extended
>> > >                    period.  The ASP MAY request the
>> > >                    availability/congestion status of one or more SS7
>> > >                    destinations to which it expects to communicate.
>> > >  
>> > > 3.17.3 Solution description
>> > >  
>> > >    Wherever "concerned ASP" is referred in the specification, it does
>> > >    mean that ASP in active state or ASP in inactive state.This is true
>> > >    only when dynamic registration is allowed between ASP and SGP.
>> > > 
>> > > =================END=========
>> > > 
>> > > --Samuel 
>> > > 
>> > > 
>> > > 
>> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>> > >   1000 Coit Road, Plano, Texas 75075
>> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
>> > > 
>> > > On Fri, 7 Feb 2003, Ken Morneault wrote:
>> > > 
>> > > > 
>> > > > 
>> > > > It would be useful if Samuel proposes text changes that could be 
>> > > > reviewed and, then, we could see if there was consensus for making
>> > > > this change.
>> > > > 
>> > > > Regards,
>> > > > 
>> > > > Ken
>> > > > 
>> > > > At 03:39 PM 2/7/2003 +0000, denim john wrote:
>> > > > 
>> > > > >Brian,
>> > > > >
>> > > > >I would request any one of the authors to conclude this thread.
>> > > > >samuel's idea seems to be good but this will result in changes of m3ua and sua.
>> > > > >
>> > > > >Regards,
>> > > > >Denim.
>> > > > >
>> > > > >An efficent ASP will take up the load as soon as it comes up.
>> > > > >Inefficient ASP may send DAUD and will receive DRST(...),DUNA(...),
>> > > > >SCON(...).  So from the point of ASP following are the unnecessary
>> > > > >messages.
>> > > > >.SENDING DAUD
>> > > > >.RECEING DRST
>> > > > >.RECEIVING DUNA
>> > > > >.RECEIVING SCON
>> > > > >
>> > > > > From the point of SGP these are the unnecessay messages.
>> > > > >.Receiving DAUD
>> > > > >.SENDING DRST
>> > > > >.SENDING DUNA
>> > > > >.SENDING SCON.
>> > > > >Thats what your argument leads.
>> > > > >
>> > > > >--Samuel
>> > > > >
>> > > > > Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>> > > > > 1000 Coit Road, Plano, Texas 75075
>> > > > > ******* The opinions expressed are not those of Alcatel USA, Inc. *******
>> > > > >
>> > > > >On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
>> > > > >
>> > > > >>Samuel,
>> > > > >>
>> > > > >>That is implementation dependent.  Sending SSNM to inactive ASPs restricts
>> > > > >>a dimension not restricted by the current and adequate procedure.  The
>> > > > >>SG cannot be required to send SSNM to inactive ASPs.  The RFC should not
>> > > > >>place a restriction on the number of inactive ASPs, because we cannot
>> > > > >>forsee all future dimensioning.  Procedures should scale as well as possible.
>> > > > >>The existing procedure scales.  Sending SSNM to inactive ASPs does not.
>> > > > >
>> > > > >>--brian
>> > > > >>
>> > > > >>Samuel Dur D. Jeyaseelan wrote:
>> > > > >(Thu, 06 Feb 2003 23:13:44)
>> > > > >>> > Brian,
>> > > > >>> > Yes , The RFC does not restrict the number of inactive ASPs.
>> > > > >>> But network configuration will restrict.
>> > > > >>> > --Samuel
>> > > > >>> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>> > > > >>>   1000 Coit Road, Plano, Texas 75075
>> > > > >>>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
>> > > > >>> > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
>> > > > >>> > > Samuel,
>> > > > >
>> > > > >>> > > > The RFC does not restrict the number of inactive ASPs.
>> > > > >>> > > > --brian
>> > > > >>> > > > Samuel Dur D. Jeyaseelan wrote:
>> > > > >(Thu, 06 Feb 2003 22:40:48)
>> > > > >>> > > > > > Brian,
>> > > > >>> > > > > > Ok, tell me, where you will configure 1 ASP ACT and 199,000 ASPs INACT?
>> > > > >>> > > I want to know which network does this poor configuration.
>> > > > >>> > > > > > --Samuel
>> > > > >>> > > > > > > > >   Alcatel USA, Inc.                  Internet:
>> > > > ><userid>@ssd.usa.alcatel.com
>> > > > >>> > >   1000 Coit Road, Plano, Texas 75075
>> > > > >>> > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
>> > > > >*******
>> > > > >>> > > > > > On Thu, 6 Feb 2003, Brian F. G. Bidulock wrote:
>> > > > >>> > > > > > > Samuel,
>> > > > >>> > > > > > > > The RFC does not restrict the number of inactive ASPs.
>> > > > >>> > > > > > > > --brian
>> > > > >>> > > > > > > > Samuel Dur D. Jeyaseelan wrote:
>> > > > >(Thu, 06 Feb 2003 22:23:12)
>> > > > >>> > > > > > > > > > Richard,
>> > > > >>> > > > > > > > > >   Alcatel USA, Inc.                  Internet:
>> > > > ><userid>@ssd.usa.alcatel.com
>> > > > >>> > > > >   1000 Coit Road, Plano, Texas 75075
>> > > > >>> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
>> > > > >*******
>> > > > >>> > > > > > > > > > On Thu, 6 Feb 2003, Richard Adams wrote:
>> > > > >>> > > > > > > > > > > Brian,
>> > > > >>> > > > > > I agree with you. I think when the ASP becomes active, it should
>> > > > >do an
>> > > > >>> > > > > > audit to update its knowledge of the SS7 network. The ASP needs
>> > > > >to take
>> > > > >>> > > > > > the initiative to find out about the SS7 nodes etc. in which it
>> > > > >is
>> > > > >>> > > > > > interested. Any other method of working would lead to impossible
>> > > > >loading
>> > > > >>> > > > > > on the SG.
>> > > > >>> > > > > > > > > > Yes i agree but your suggestion is an indirect method.
>> > > > >>> > > > > I am proposing a direct method to send SSNM messages from SGP to
>> > > > >INACTIVE
>> > > > >>> > > > > ASP. When SGP sends DUNA or DRST to ACTVIE ASP, let him also send a
>> > > > >copy
>> > > > >>> > > > > of the same message to INACTIVE ASP. I dont find any difficulty
>> > > > >here.
>> > > > >>> > > > > > > > > >  > > > > > > Of course, I hope 199,999 of your inactive ASPs do not decide to
>> > > > >become
>> > > > >>> > > > > > active at the same time.....! > > > > > > > > > > Theoritically your example sounds good but practically NOBODY will
>> > > > >>> > > > > configure 1 ASP ACTIVE and 199,999 ASPs INACTIVE.
>> > > > >>> > > > > > > > > > --Samuel
>> > > > >>> > > > > > > > > > > > > > > > > -- > > > > Brian F. G. Bidulock
>> > > > >>> > > > bidulock@openss7.org
>> > > > >>> > > > http://www.openss7.org/
>> > > > >>> > > > _______________________________________________
>> > > > >>> > > > Sigtran mailing list
>> > > > >>> > > > Sigtran@ietf.org
>> > > > >>> > > > https://www1.ietf.org/mailman/listinfo/sigtran
>> > > > >>> > > > > > > > -- > > Brian F. G. Bidulock
>> > > > >>> > bidulock@openss7.org
>> > > > >>> > http://www.openss7.org/
>> > > > >>> > _______________________________________________
>> > > > >>> > Sigtran mailing list
>> > > > >>> > Sigtran@ietf.org
>> > > > >>> > https://www1.ietf.org/mailman/listinfo/sigtran
>> > > > >>> >
>> > > > >>
>> > > > >>--
>> > > > >>Brian F. G. Bidulock
>> > > > >>bidulock@openss7.org
>> > > > >>http://www.openss7.org/
>> > > > >
>> > > > >
>> > > > >
>> > > > >
>> > > > >
>> > > > >
>> > > > >
>> > > > >
>> > > > >
>> > > > >
>> > > > >_________________________________________________________________
>> > > > >Help STOP SPAM with the new MSN 8 and get 2 months FREE*  
>> > > > >http://join.msn.com/?page=features/junkmail
>> > > > 
>> > > > _______________________________________________
>> > > > Sigtran mailing list
>> > > > Sigtran@ietf.org
>> > > > https://www1.ietf.org/mailman/listinfo/sigtran
>> > > > 
>> > 
>> > -- 
>> > Brian F. G. Bidulock
>> > bidulock@openss7.org
>> > http://www.openss7.org/
>> > 
>
>-- 
>Brian F. G. Bidulock
>bidulock@openss7.org
>http://www.openss7.org/
>_______________________________________________
>Sigtran mailing list
>Sigtran@ietf.org
>https://www1.ietf.org/mailman/listinfo/sigtran 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 10 10:41:17 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA12708
	for <sigtran-archive@odin.ietf.org>; Mon, 10 Feb 2003 10:41:17 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1AFnii23409
	for sigtran-archive@odin.ietf.org; Mon, 10 Feb 2003 10:49:44 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1AFnUp23400;
	Mon, 10 Feb 2003 10:49:32 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1AFmrp23359
	for <sigtran@optimus.ietf.org>; Mon, 10 Feb 2003 10:48:53 -0500
Received: from mx1.etsi.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA12656
	for <sigtran@ietf.org>; Mon, 10 Feb 2003 10:39:54 -0500 (EST)
Received: from email10.etsihq.org ([172.27.1.62]) by mx1.etsi.org with Microsoft SMTPSVC(5.0.2195.5329);
	 Mon, 10 Feb 2003 16:43:33 +0100
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="US-ASCII"
Subject: RE: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
Date: Mon, 10 Feb 2003 16:44:59 +0100
Message-ID: <4091553999CBE4409CC2B562152B5A33EDDF5C@email10.etsihq.org>
Thread-Topic: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
Thread-Index: AcLRDh87eH+tBg4oT3iGFBmzyS6HsAACUIbg
From: "Richard Adams" <Richard.Adams@etsi.org>
To: "Ken Morneault" <kmorneau@cisco.com>, <bidulock@openss7.org>,
        "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: "denim john" <denim_john@hotmail.com>, <sigtran@ietf.org>,
        <j.javier.pastor@ericsson.com>, <Francois.Zannin@hp.com>
X-OriginalArrivalTime: 10 Feb 2003 15:43:33.0963 (UTC) FILETIME=[2B3E71B0:01C2D11B]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h1AFmrp23360
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Ken,
Brian suggested a modification to his own text, and I agree with that : 
the second SHOULD should become a MAY.
So the proposed text reads:
"When an ASP becomes active for an AS and the SG is experiencing
SS7 network isolation or is performing the MTP Restart procedure
for the AS, the SG SHOULD send a DUNA(*) to the newly active ASP
to prevent the ASP from sending traffic.
 
When an ASP becomes active for an AS and destinations normally
acessible to the ASP are inaccessible, restricted or congested
for the AS, the SG MAY send DUNA, DRST or SCON AS for the
inacessible, restricted or congested destinations to the ASP
newly active for the AS to prevent the ASP from sending traffic
for destinations that it might otherwise not know are
inaccessible, restricted or congested."



What would concern me is that any other changes are starting to 
re-open AS/ASP management (and therefore any changes should
be consistent across M3UA, SUA, M2UA at least).
 Plugtests have shown the current procedures to work and
interwork between implementations. I fear we are opening 
a can of worms, which could ultimately lead to re-design to
build the ASP active/inactive procedures so that they integrated
into MTP restart. While this could be a laudable aim, I think
it might take a long time. MTP restart took ~5 to 7 years before
it was tightly-enough specified (and would work for a network
where a number of signalling points were re-starting at the same time).
 

Samuel's text did not appear to me to differentiate sufficiently
between actions when the SG allowed dynamic registration, and those when
it
didn't. And I would want to see the consequences, preferably
shown in a state transition diagram for the interaction between 
the SG and several ASPs. This would need to cover the sub-state changes
between INACTIVE_but_ready_to_become_active, INACTIVE_and_remaining_so,
and how the SG differentiated. And why would this be different
for static configurations? And in the transition from INACTIVE
to ACTIVE, does an ASP need to pass through the _ready_to_become_active
substate. Is it this transition that causes the SG to send out
SSNMs? If so, how does it know without signalling between the ASP and
SG ?  Or is this information pre-configured ?

Kind regards,
Dick Adams,
richard.adams@etsi.org

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 10 14:16:06 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA21312
	for <sigtran-archive@odin.ietf.org>; Mon, 10 Feb 2003 14:16:05 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1AJOam04937
	for sigtran-archive@odin.ietf.org; Mon, 10 Feb 2003 14:24:36 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1AJOOp04922;
	Mon, 10 Feb 2003 14:24:24 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1AJNmp04899
	for <sigtran@optimus.ietf.org>; Mon, 10 Feb 2003 14:23:48 -0500
Received: from web12803.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id OAA21277
	for <sigtran@ietf.org>; Mon, 10 Feb 2003 14:14:45 -0500 (EST)
Message-ID: <20030210191827.4590.qmail@web12803.mail.yahoo.com>
Received: from [62.29.76.221] by web12803.mail.yahoo.com via HTTP; Mon, 10 Feb 2003 11:18:27 PST
Date: Mon, 10 Feb 2003 11:18:27 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: RE: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
To: sigtran@ietf.org
In-Reply-To: <4091553999CBE4409CC2B562152B5A33EDDF5C@email10.etsihq.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

I also think the proposed text might be usefull in IG
and I have some commnets/questions below.

  Tolga
--- Richard Adams <Richard.Adams@etsi.org> wrote:
> Ken,
> Brian suggested a modification to his own text, and
> I agree with that : 
> the second SHOULD should become a MAY.
> So the proposed text reads:
> "When an ASP becomes active for an AS and the SG is
> experiencing
> SS7 network isolation or is performing the MTP
> Restart procedure
> for the AS, the SG SHOULD send a DUNA(*) to the
> newly active ASP
> to prevent the ASP from sending traffic.
>  
> When an ASP becomes active for an AS and
> destinations normally
> acessible to the ASP are inaccessible, restricted or
> congested
> for the AS, the SG MAY send DUNA, DRST or SCON AS
> for the
> inacessible, restricted or congested destinations to
> the ASP
> newly active for the AS to prevent the ASP from
> sending traffic
> for destinations that it might otherwise not know
> are
> inaccessible, restricted or congested."
[TOLGA]Because Q.704 does not specify actions when a
local user part becomes available, I think there is no
danger of divergence from Q.704. But how can ASP know,
when the last DUNA, DRST, SCON is received before
starting to send traffic? I would think, there is no
need that it waits for a specific message, i.e. it can
send DATA to any destination as soon as it becomes
ACTIVE and SSNM from SG is used just as additional
information. So, I would think, if an ASP really wants
to play safe, it should send DAUDs to SG for the
destinations it is interested in and wait replies from
them before sending DATA (or better synchronizes
destination table with its peer ASP)
And regarding sending SSNM while ASP inactive I  still
have the same idea, my personal opinion is to keep RFC
as it is. 


__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 10 16:24:24 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA26491
	for <sigtran-archive@odin.ietf.org>; Mon, 10 Feb 2003 16:24:24 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1ALWwb13355
	for sigtran-archive@odin.ietf.org; Mon, 10 Feb 2003 16:32:58 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1ALWcp13307;
	Mon, 10 Feb 2003 16:32:39 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1ALV7p12884
	for <sigtran@optimus.ietf.org>; Mon, 10 Feb 2003 16:31:07 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA26367
	for <sigtran@ietf.org>; Mon, 10 Feb 2003 16:22:01 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:5NDdcXZgxsm8QG6Bl2vijI0yDapL/xE4@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1ALPgD14226;
	Mon, 10 Feb 2003 14:25:42 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1ALPgm09283;
	Mon, 10 Feb 2003 14:25:42 -0700
Date: Mon, 10 Feb 2003 14:25:42 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tolga Asveren <tolga_asveren@yahoo.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
Message-ID: <20030210142542.A28140@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <4091553999CBE4409CC2B562152B5A33EDDF5C@email10.etsihq.org> <20030210191827.4590.qmail@web12803.mail.yahoo.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20030210191827.4590.qmail@web12803.mail.yahoo.com>; from tolga_asveren@yahoo.com on Mon, Feb 10, 2003 at 11:18:27AM -0800
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Tolga,

Tolga Asveren wrote:                                                                 (Mon, 10 Feb 2003 11:18:27)
> I also think the proposed text might be usefull in IG
> and I have some commnets/questions below.
> 
>   Tolga
> --- Richard Adams <Richard.Adams@etsi.org> wrote:
> > Ken,
> > Brian suggested a modification to his own text, and
> > I agree with that : 
> > the second SHOULD should become a MAY.
> > So the proposed text reads:
> > "When an ASP becomes active for an AS and the SG is
> > experiencing
> > SS7 network isolation or is performing the MTP
> > Restart procedure
> > for the AS, the SG SHOULD send a DUNA(*) to the
> > newly active ASP
> > to prevent the ASP from sending traffic.
> >  
> > When an ASP becomes active for an AS and
> > destinations normally
> > acessible to the ASP are inaccessible, restricted or
> > congested
> > for the AS, the SG MAY send DUNA, DRST or SCON AS
> > for the
> > inacessible, restricted or congested destinations to
> > the ASP
> > newly active for the AS to prevent the ASP from
> > sending traffic
> > for destinations that it might otherwise not know
> > are
> > inaccessible, restricted or congested."
> [TOLGA]Because Q.704 does not specify actions when a
> local user part becomes available, I think there is no
> danger of divergence from Q.704. But how can ASP know,
> when the last DUNA, DRST, SCON is received before
> starting to send traffic? I would think, there is no
> need that it waits for a specific message, i.e. it can
> send DATA to any destination as soon as it becomes
> ACTIVE and SSNM from SG is used just as additional
> information. So, I would think, if an ASP really wants
> to play safe, it should send DAUDs to SG for the
> destinations it is interested in and wait replies from
> them before sending DATA (or better synchronizes
> destination table with its peer ASP)
> And regarding sending SSNM while ASP inactive I  still
> have the same idea, my personal opinion is to keep RFC
> as it is. 

It could be possible for the SG to send the SSNM after
receiving ASPAC, but before sending ASPAC Ack.  That way
the ASPAC Ack acts like a TRA.

For example, an ASP inactive for the AS sends ASPAC Ack
and the SG has the ASP marked inactive at the SG for the
AS.  The SG sends DAVA(*) followed by any necessary
DUNA, SCON and DRST (which should be a short list), or,
if the SG is in MTP Restart, it sends DUNA(*).  The SG
could send these SSNM ordered in the AS stream.  Then
it could follow with the ASPAC Ack ordered in the same
stream.  The ASP should not start traffic until it
receives the ASPAC Ack, anyway.  Note that this would
only need to be performed for the first SGP that the ASP
becomes active for.  So it would look something like this:

   SGP                                 ASP
    |                                   |
    |                              /----|<-- (start)
    |                             /     |
    |      /-------ASPAC---------/      | MTP-User assumes
    |     /                             | all destinations
    |<---/                              | available but
    |----\                              | suppresses traffic.
    |----\\                             |
    |----\\\-------DAVA(*)-------\      |
    |     \\-------DUNA(1)-------\\     |
    |----\ \-------SCON(2)-------\\\--->| (reasserts above)
    |     \                       \\--->|--> MTP-PAUSE
    |      \------ASPAC Ack------\ \--->|--> MTP-STATUS
    |----\                        \     |
    |----\\                        \--->|--> MTP-RESTART-ENDS
    |     \\--------DATA---------\ /----|<-- MTP-TRANSFER
    |      \--------DATA---------\X/----|<-- MTP-TRANSFER
    |      /--------DATA---------/X\--->|--> MTP-TRANSFER
    |     //--------DATA---------/ \--->|--> MTP-TRANSFER
    |<---//                             |
    |<---/                              |
    |                                   |
    |                                   |

Now, that could be quite useful where the ASP Point Code
is different from the SG Point Code and the ASP Point Code
has routes to the entire SS7 network.  It also mimics the
SS7 MTP Restart procedure for this "detached" point code.

For the SG MTP Restart case, the SHOULD was intended for
something like this for the case where the ASP Point Code
is different from the SG point code:

   SGP                                 ASP
    |                                   |
    |                              /----|<-- (start)
    |                             /     |
    |      /-------ASPAC---------/      | MTP-User assumes
    |     /                             | all destinations
    |<---/                              | available but
    |----\                              | suppresses traffic.
    |----\\                             |
    |----\\\-------DUNA(*)-------\      |
    |     \\------DAVA(SG)-------\\     |
    |      \------ASPAC Ack------\\\--->|  (isolation)
    |                             \\--->|--> MTP-RESUME
    |----\                         \--->|--> MTP-RESTART-ENDS
    |----\\                             |
    |     \\--------DATA---------\ /----|<-- MTP-TRANSFER
    |      \--------DATA---------\X/----|<-- MTP-TRANSFER
    |      /--------DATA---------/X\--->|--> MTP-TRANSFER
    |     //--------DATA---------/ \--->|--> MTP-TRANSFER
    |<---//    (to and from SG only)    |
    |<---/                              |
    |                                   |
    |                                   |

If the ASP sends DAUD just before ASPAC, a similar scenario
can be acheived:

   SGP                                 ASP
    |                                   |
    |                              /----|<-- (start)
    |                             //----|
    |      /-------DAUD(*)-------//     |
    |     //-------ASPAC---------/      | MTP-User assumes
    |<---//                             | all destinations
    |<---/                              | available but
    |----\                              | suppresses traffic.
    |----\\                             |
    |----\\\-------DAVA(*)-------\      |
    |     \\-------DUNA(1)-------\\     |
    |----\ \-------SCON(2)-------\\\--->| (reasserts above)
    |     \                       \\--->|--> MTP-PAUSE
    |      \------ASPAC Ack------\ \--->|--> MTP-STATUS
    |----\                        \     |
    |----\\                        \--->|--> MTP-RESTART-ENDS
    |     \\--------DATA---------\ /----|<-- MTP-TRANSFER
    |      \--------DATA---------\X/----|<-- MTP-TRANSFER
    |      /--------DATA---------/X\--->|--> MTP-TRANSFER
    |     //--------DATA---------/ \--->|--> MTP-TRANSFER
    |<---//                             |
    |<---/                              |
    |                                   |
    |                                   |

And as long as the SG completes responding to the DAUD(*)
before responding to the ASPAC with ASPAC Ack, the same
scenario results.

I don't see a problem with sending the DAUD(*) before the
ASPAC.  DAUD is permitted to be sent unordered (and on
stream 0) and could be expedited ahead of the ASPAC
(which is sent ordered and normally on the same stream
as the AS DATA) by SCTP anyway.

I think that by saying that the inactive ASP MAY audit, and
the SG MAY respond to audits from inactive ASPs, this type
of smooth restart can be acheived.

Note that this would work equally well for the first active
ASP as it would for the subsequent active or overriding ASP.

So I think the following would suffice:

  If an SG is in MTP restart, the first SGP SHOULD send
  DUNA(*) before ASPAC Ack to the ASP newly activating for
  the AS.

  An SGP MAY send DAVA(*) followed by DUNA, DRST and SCON
  for affected destinations before sending ASPAC Ack to a
  newly activating AS.

  An SGP MAY respond to DAUD from an inactive ASP.

  An SGP SHOULD complete responding to a previous DAUD
  before responding to ASPAC with ASPAC Ack for an ASP newly
  activating for an AS.

  If an ASP wishes to update destination status when going
  active, it SHOULD send DAUD before sending ASPAC.

It would be nice to say somehow that in these scenarios the
SG can perform "abbreviations" such as above (DUNA(*) followed
by a couple of DAVAs, or DAVA(*) followed by a couple of
DUNAs).

Comments?

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 10 16:58:05 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA27586
	for <sigtran-archive@odin.ietf.org>; Mon, 10 Feb 2003 16:58:04 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1AM6eQ15715
	for sigtran-archive@odin.ietf.org; Mon, 10 Feb 2003 17:06:40 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1AM6Np15706;
	Mon, 10 Feb 2003 17:06:23 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h17NuSp07081
	for <sigtran@optimus.ietf.org>; Fri, 7 Feb 2003 18:56:28 -0500
Received: from catapult.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA25350
	for <sigtran@ietf.org>; Fri, 7 Feb 2003 18:48:48 -0500 (EST)
Received: from catapult.com ([192.168.0.136])
	by catapult.com (8.11.6+Sun/8.11.6) with ESMTP id h17Nn1G21276;
	Fri, 7 Feb 2003 15:49:01 -0800 (PST)
Message-ID: <3E4446C3.3000203@catapult.com>
Date: Fri, 07 Feb 2003 15:52:35 -0800
From: Nicholas Arraje <nick@catapult.com>
Reply-To: nick@catapult.com
Organization: Catapult Communications
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.0.1) Gecko/20020823 Netscape/7.0
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: sigtran@ietf.org
Subject: Re: FW: [Sigtran] SCTP Test Tool
References: <CIECLLOEEMIPEBGCMMKMOECECAAA.loren@catapult.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Catapult Communitations offers testing solutions for SCTP and M3UA.  Please feel 
free to contact us at http://www.catapult.com/.


> -----Original Message-----
> From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf Of
> sherry.karl@tekelec.com
> Sent: Friday, February 07, 2003 12:57 PM
> To: sigtran@ietf.org
> Subject: [Sigtran] SCTP Test Tool
> 
> I was wondering if anyone is aware of any 3rd party M3UA/SCTP Test Tools?
> 
> Thanks
> 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 
> 


-- 
Best Regards,

Nicholas Arraje
Regional Sales Manager
Catapult Communications
Email: nick@catapult.com

Phone: 650/960-1025 ext 149
Fax  : 650/960-1029
Cell : 415/336-4384

http://www.catapult.com/

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 10 17:04:44 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA27775
	for <sigtran-archive@odin.ietf.org>; Mon, 10 Feb 2003 17:04:44 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1AMDJh16673
	for sigtran-archive@odin.ietf.org; Mon, 10 Feb 2003 17:13:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1AMD5p16663;
	Mon, 10 Feb 2003 17:13:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1AMClp16647
	for <sigtran@optimus.ietf.org>; Mon, 10 Feb 2003 17:12:47 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA27750
	for <sigtran@ietf.org>; Mon, 10 Feb 2003 17:03:40 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1AM7LJT023681;
	Mon, 10 Feb 2003 16:07:21 -0600 (CST)
Message-ID: <3E482299.C3E86A0E@alcatel.com>
Date: Mon, 10 Feb 2003 16:07:21 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft 7 question
References: <3E4350A2.8060B90@cisco.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Wayne,

Yes, I want both sides to send and receive a LS Alignment message to
complete alignment. 

To fix the problem you describe, how about this" If M2PA has not
received MTP3 Start and receives LS Alignment, M2PA replies with LS OOS
(instead of not replying at all and counting on the other end's T2
expiring). 

tomG.

wayne n taylor wrote:
> 
> Hi All,
> 
> i have a couple of comments/questions about the initial exchange of the
> Link Status Alignment messages. there seems to be some inconsistency in
> the text and finite state machine.
> 
> section 4.1.3 paragraph 4:
> 
> "Once the association is established and M2PA has received a Start
> Request from MTP3, M2PA sends the Link Status Alignment message to its
> peer. If M2PA has not already received the Link Status Alignment
> message from its peer, then M2PA starts timer T2."
> 
> what state would the local M2PA be in to receive a LS Alignment message
> from its peer prior to getting a MTP3 Start ?. i assume the OOS state,
> but the finite state machine does not show LS Alignment as an input.
> 
> section 4.1.3 paragraph 5:
> 
> "(Note that if the remote M2PA has not received a Start Request from
> its MTP3, it will not send the Link Status Alignment message to the
> local M2PA. Eventually timer T2 in the local M2PA will expire. If the
> remote M2PA receives a Start Request from its MTP3 and sends Link
> Status Alignment before the local M2PA's timer T2 expires, the
> alignment procedure can continue.)"
> 
> assume the local M2PA receives a Start Request from MTP3, sends LS
> Alignment and start T2, but the remote peer has not received a Start
> Request and is in OOS. the current finite state machine does not show
> what to do with the LS Alignment from the local peer so lets assume the
> LS Alignment is dropped. the remote peer receives a Start Request from
> MTP3, sends LS Alignment and start T2 (since the remote peer didnt get
> the local peers LS Alignment). assume further that timer T2 is still
> running on the local peer. the text from section 4.1.3 above states the
> "alignment procedure can continue". in this case the local peer sent a
> LS Alignment and received a LS Alignment. the remote peer only sent a LS
> Alignment and did not receive an LS Alignment (since LS Alignment
> messages are not retried). eventually T2 on the remote peer will expire,
> T3 on the local peer will expire and the alignment procedure starts over
> again. it seems possible for the peers to get caught in a loop trying to
> exchange LS Alignment messages.
> 
> section 4.1.3 paragraph 9:
> 
> "When M2PA has both sent and received the Link Status Alignment
> message, it has completed alignment.  M2PA starts the aligned timer T3
> and moves to the proving state."
> 
> is it a requirement for both the local and remote peers to send/receive
> LS Alignment for alignment to be complete ?. the only way to transition
> from Initial Alignment to Proving is to receive LS Alignment. in the
> scenario above, both peers do not send and receive LS Alignment
> messages.
> 
> regards,
> 
> wayne

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 10 19:08:51 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA01352
	for <sigtran-archive@odin.ietf.org>; Mon, 10 Feb 2003 19:08:50 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1B0HTL24692
	for sigtran-archive@odin.ietf.org; Mon, 10 Feb 2003 19:17:29 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1B0HGp24679;
	Mon, 10 Feb 2003 19:17:16 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1B0F6p24594
	for <sigtran@optimus.ietf.org>; Mon, 10 Feb 2003 19:15:06 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA01297
	for <sigtran@ietf.org>; Mon, 10 Feb 2003 19:05:55 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:5yk6n8XUCmRGbDE7hVGK78EjzlPji9iN@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1B09cD16693;
	Mon, 10 Feb 2003 17:09:38 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1B09ci00572;
	Mon, 10 Feb 2003 17:09:38 -0700
Date: Mon, 10 Feb 2003 17:09:38 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tom George <Tom.George@alcatel.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft 7 question
Message-ID: <20030210170938.A11994@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tom George <Tom.George@alcatel.com>, sigtran@ietf.org
References: <3E4350A2.8060B90@cisco.com> <3E482299.C3E86A0E@alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E482299.C3E86A0E@alcatel.com>; from Tom.George@alcatel.com on Mon, Feb 10, 2003 at 04:07:21PM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Tom,

The first thing a link should send is LS OOS.

In the out of service state, it would be desireable for the out of
service side to always send LS OOS in response to other messages.

Also, the out of service side should be permitted to send LS OOS
periodically.

With regards to Q.703 state machines, LS OOS in response to alignment
will not abort alignment and T2 should still expire.  See test case
"3.1.2 Timer T2" in draft-bidulock-sigtran-m2pa-test-02.txt.

T2 is defined as the duration of time that the M2PA peer wil wait to
receive a status "Alignment" message after sending a status "Alignment"
message.

As status Alignment will abort an otherwise in-service link, other
link status received while running T2 should be ignored.

Any different behaviour could interfere with MTP Level 3's link
restoration timers and procedures.

More comments below...


Tom George wrote:                                                                      (Mon, 10 Feb 2003 16:07:21)
> Wayne,
> 
> Yes, I want both sides to send and receive a LS Alignment message to
> complete alignment. 
> 
> To fix the problem you describe, how about this" If M2PA has not
> received MTP3 Start and receives LS Alignment, M2PA replies with LS OOS
> (instead of not replying at all and counting on the other end's T2
> expiring). 
> 
> tomG.
> 
> wayne n taylor wrote:
> > 
> > Hi All,
> > 
> > i have a couple of comments/questions about the initial exchange of the
> > Link Status Alignment messages. there seems to be some inconsistency in
> > the text and finite state machine.
> > 
> > section 4.1.3 paragraph 4:
> > 
> > "Once the association is established and M2PA has received a Start
> > Request from MTP3, M2PA sends the Link Status Alignment message to its
> > peer. If M2PA has not already received the Link Status Alignment
> > message from its peer, then M2PA starts timer T2."
> > 
> > what state would the local M2PA be in to receive a LS Alignment message
> > from its peer prior to getting a MTP3 Start ?. i assume the OOS state,
> > but the finite state machine does not show LS Alignment as an input.
> > 
> > section 4.1.3 paragraph 5:
> > 
> > "(Note that if the remote M2PA has not received a Start Request from
> > its MTP3, it will not send the Link Status Alignment message to the
> > local M2PA. Eventually timer T2 in the local M2PA will expire. If the
> > remote M2PA receives a Start Request from its MTP3 and sends Link
> > Status Alignment before the local M2PA's timer T2 expires, the
> > alignment procedure can continue.)"
> > 
> > assume the local M2PA receives a Start Request from MTP3, sends LS
> > Alignment and start T2, but the remote peer has not received a Start
> > Request and is in OOS. the current finite state machine does not show
> > what to do with the LS Alignment from the local peer so lets assume the
> > LS Alignment is dropped. the remote peer receives a Start Request from
> > MTP3, sends LS Alignment and start T2 (since the remote peer didnt get
> > the local peers LS Alignment). assume further that timer T2 is still
> > running on the local peer. the text from section 4.1.3 above states the
> > "alignment procedure can continue". in this case the local peer sent a
> > LS Alignment and received a LS Alignment. the remote peer only sent a LS
> > Alignment and did not receive an LS Alignment (since LS Alignment
> > messages are not retried). eventually T2 on the remote peer will expire,
> > T3 on the local peer will expire and the alignment procedure starts over
> > again. it seems possible for the peers to get caught in a loop trying to
> > exchange LS Alignment messages.

See draft-bidulock-sigtran-m2pa-test-02.txt Section "3.1.3 Timer T3".  You
will see that what you describe does not happen if you follow Q.703 SDLs.
In Q.703 once SIO is sent SIN is sufficient to complete alignment.  So,
one side sends ALIGNMENT and starts T2, the other side sends ALIGNMENT and
starts T2.  The first side, receives the alignment, stops T2 and sends
PROVING-NORMAL or PROVING-EMERGENCY and starts T3.  When the second side
receives PROVING-NORMAL or PROVING-EMERGENCY it stops T2 and sends
PROVING-NORMAL or PROVING-EMERGENCY and starts T4.  When the first side
receives PROVING-NORMAL it stops T3 and starts T4.  When either T4 expires,
the side sends IN-SERVICE and starts T1.  When the side receives IN-SERVICE
it stops T1 and indicates in_service to MTP3.

This behavior can be seen on Figure 9/Q.703 (sheet 2 of 6) - Initial alignment
control.  When the IAC starts it sends SIO and starts T2.  If it receives
any of SIO, SIN or SIE it stops T2 and send SIN or SIE depending on emergency
status and starts T3.  Receipt of a subsequent SIN or SIE (sheet 3 of 6)
stops T3 and starts T4.  Note that if it receives SIO again, it ignores it
at that point and continues to send SIE or SIN.

These is no reason for M2PA not to follow the Q.703 SDLs in this regard.

A statement that might clarify is that the M2PA sends LS ALIGNMENT and if
it receives ALIGNMENT or PROVING it stops T2 and PROVING and starts T3.
Note that it takes two PROVING messages in this case to move the peer into
the proving state.  One PROVING message acts like an ALIGNMENT (cancels
T2 and starts T3), the next acts like the first PROVING message (cancels
T3 and starts T4).

--brian

> > 
> > section 4.1.3 paragraph 9:
> > 
> > "When M2PA has both sent and received the Link Status Alignment
> > message, it has completed alignment.  M2PA starts the aligned timer T3
> > and moves to the proving state."
> > 
> > is it a requirement for both the local and remote peers to send/receive
> > LS Alignment for alignment to be complete ?. the only way to transition
> > from Initial Alignment to Proving is to receive LS Alignment. in the
> > scenario above, both peers do not send and receive LS Alignment
> > messages.
> > 
> > regards,
> > 
> > wayne
> 
> -- 
> * Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
> * 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
> * Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
> * I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
> * to advertise or represent its position,  operations, or activities. *
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 11 02:35:26 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA22426
	for <sigtran-archive@odin.ietf.org>; Tue, 11 Feb 2003 02:35:26 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1B7iCj26300
	for sigtran-archive@odin.ietf.org; Tue, 11 Feb 2003 02:44:12 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1B7hpp26271;
	Tue, 11 Feb 2003 02:43:53 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1B7glp26238
	for <sigtran@optimus.ietf.org>; Tue, 11 Feb 2003 02:42:47 -0500
Received: from mx1.etsi.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA22337
	for <sigtran@ietf.org>; Tue, 11 Feb 2003 02:33:28 -0500 (EST)
Received: from email10.etsihq.org ([172.27.1.62]) by mx1.etsi.org with Microsoft SMTPSVC(5.0.2195.5329);
	 Tue, 11 Feb 2003 08:37:08 +0100
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="US-ASCII"
Subject: RE: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
Date: Tue, 11 Feb 2003 08:37:08 +0100
Message-ID: <4091553999CBE4409CC2B562152B5A33EDDF5D@email10.etsihq.org>
Thread-Topic: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
Thread-Index: AcLRS0xqNs80vk5iR4SEN3Iv06zCUgAUdAEg
From: "Richard Adams" <Richard.Adams@etsi.org>
To: <bidulock@openss7.org>, "Tolga Asveren" <tolga_asveren@yahoo.com>
Cc: <sigtran@ietf.org>
X-OriginalArrivalTime: 11 Feb 2003 07:37:08.0430 (UTC) FILETIME=[61B946E0:01C2D1A0]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h1B7glp26239
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Brian,
what you suggest does seem to work OK. So with your judicious use of
SHOULD and MAY, I like the proposal.
Incidentally, it would be nice to find some place in the documents to
put the sequence diagrams ! 
They are useful in explaining how the RFC is used.
I've additional comments below between [RAA] and {raa}



-----Original Message-----
From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
Sent: 10 February 2003 22:26
To: Tolga Asveren
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization


Tolga,

Tolga Asveren wrote:
(Mon, 10 Feb 2003 11:18:27)
> I also think the proposed text might be usefull in IG
> and I have some commnets/questions below.
> 
>   Tolga
> --- Richard Adams <Richard.Adams@etsi.org> wrote:
> > Ken,
> > Brian suggested a modification to his own text, and
> > I agree with that : 
> > the second SHOULD should become a MAY.
> > So the proposed text reads:
> > "When an ASP becomes active for an AS and the SG is
> > experiencing
> > SS7 network isolation or is performing the MTP
> > Restart procedure
> > for the AS, the SG SHOULD send a DUNA(*) to the
> > newly active ASP
> > to prevent the ASP from sending traffic.
> >  
> > When an ASP becomes active for an AS and
> > destinations normally
> > acessible to the ASP are inaccessible, restricted or
> > congested
> > for the AS, the SG MAY send DUNA, DRST or SCON AS
> > for the
> > inacessible, restricted or congested destinations to
> > the ASP
> > newly active for the AS to prevent the ASP from
> > sending traffic
> > for destinations that it might otherwise not know
> > are
> > inaccessible, restricted or congested."
> [TOLGA]Because Q.704 does not specify actions when a
> local user part becomes available, I think there is no
> danger of divergence from Q.704. But how can ASP know,
> when the last DUNA, DRST, SCON is received before
> starting to send traffic? I would think, there is no
> need that it waits for a specific message, i.e. it can
> send DATA to any destination as soon as it becomes
> ACTIVE and SSNM from SG is used just as additional
> information. So, I would think, if an ASP really wants
> to play safe, it should send DAUDs to SG for the
> destinations it is interested in and wait replies from
> them before sending DATA (or better synchronizes
> destination table with its peer ASP)
> And regarding sending SSNM while ASP inactive I  still
> have the same idea, my personal opinion is to keep RFC
> as it is. 

It could be possible for the SG to send the SSNM after
receiving ASPAC, but before sending ASPAC Ack.  That way
the ASPAC Ack acts like a TRA.

For example, an ASP inactive for the AS sends ASPAC Ack
and the SG has the ASP marked inactive at the SG for the
AS.  The SG sends DAVA(*) followed by any necessary
DUNA, SCON and DRST (which should be a short list), or,
if the SG is in MTP Restart, it sends DUNA(*).
[RAA] Could the restarting SG just ignore the ASPAC Ack for now, then
perhaps send the SSNMs later ? It depends in which part of its
restart it is in  - initially a restarting STP waits to
receive routeing info and constructs its routeings before
sending out any routeing data itself.
{raa}
  The SG
could send these SSNM ordered in the AS stream.  Then
it could follow with the ASPAC Ack ordered in the same
stream.  The ASP should not start traffic until it
receives the ASPAC Ack, anyway.  Note that this would
only need to be performed for the first SGP that the ASP
becomes active for.  So it would look something like this:

   SGP                                 ASP
    |                                   |
    |                              /----|<-- (start)
    |                             /     |
    |      /-------ASPAC---------/      | MTP-User assumes
    |     /                             | all destinations
    |<---/                              | available but
    |----\                              | suppresses traffic.
    |----\\                             |
    |----\\\-------DAVA(*)-------\      |
    |     \\-------DUNA(1)-------\\     |
    |----\ \-------SCON(2)-------\\\--->| (reasserts above)
    |     \                       \\--->|--> MTP-PAUSE
    |      \------ASPAC Ack------\ \--->|--> MTP-STATUS
    |----\                        \     |
    |----\\                        \--->|--> MTP-RESTART-ENDS
    |     \\--------DATA---------\ /----|<-- MTP-TRANSFER
    |      \--------DATA---------\X/----|<-- MTP-TRANSFER
    |      /--------DATA---------/X\--->|--> MTP-TRANSFER
    |     //--------DATA---------/ \--->|--> MTP-TRANSFER
    |<---//                             |
    |<---/                              |
    |                                   |
    |                                   |

Now, that could be quite useful where the ASP Point Code
is different from the SG Point Code and the ASP Point Code
has routes to the entire SS7 network.  It also mimics the
SS7 MTP Restart procedure for this "detached" point code.
[RAA] OK {raa}

For the SG MTP Restart case, the SHOULD was intended for
something like this for the case where the ASP Point Code
is different from the SG point code:

   SGP                                 ASP
    |                                   |
    |                              /----|<-- (start)
    |                             /     |
    |      /-------ASPAC---------/      | MTP-User assumes
    |     /                             | all destinations
    |<---/                              | available but
    |----\                              | suppresses traffic.
    |----\\                             |
    |----\\\-------DUNA(*)-------\      |
    |     \\------DAVA(SG)-------\\     |
    |      \------ASPAC Ack------\\\--->|  (isolation)
    |                             \\--->|--> MTP-RESUME
    |----\                         \--->|--> MTP-RESTART-ENDS
    |----\\                             |
    |     \\--------DATA---------\ /----|<-- MTP-TRANSFER
    |      \--------DATA---------\X/----|<-- MTP-TRANSFER
    |      /--------DATA---------/X\--->|--> MTP-TRANSFER
    |     //--------DATA---------/ \--->|--> MTP-TRANSFER
    |<---//    (to and from SG only)    |
    |<---/                              |
    |                                   |
    |                                   |

If the ASP sends DAUD just before ASPAC, a similar scenario
can be acheived:

   SGP                                 ASP
    |                                   |
    |                              /----|<-- (start)
    |                             //----|
    |      /-------DAUD(*)-------//     |
    |     //-------ASPAC---------/      | MTP-User assumes
    |<---//                             | all destinations
    |<---/                              | available but
    |----\                              | suppresses traffic.
    |----\\                             |
    |----\\\-------DAVA(*)-------\      |
    |     \\-------DUNA(1)-------\\     |
    |----\ \-------SCON(2)-------\\\--->| (reasserts above)
    |     \                       \\--->|--> MTP-PAUSE
    |      \------ASPAC Ack------\ \--->|--> MTP-STATUS
    |----\                        \     |
    |----\\                        \--->|--> MTP-RESTART-ENDS
    |     \\--------DATA---------\ /----|<-- MTP-TRANSFER
    |      \--------DATA---------\X/----|<-- MTP-TRANSFER
    |      /--------DATA---------/X\--->|--> MTP-TRANSFER
    |     //--------DATA---------/ \--->|--> MTP-TRANSFER
    |<---//                             |
    |<---/                              |
    |                                   |
    |                                   |

And as long as the SG completes responding to the DAUD(*)
before responding to the ASPAC with ASPAC Ack, the same
scenario results.

I don't see a problem with sending the DAUD(*) before the
ASPAC.  DAUD is permitted to be sent unordered (and on
stream 0) and could be expedited ahead of the ASPAC
(which is sent ordered and normally on the same stream
as the AS DATA) by SCTP anyway.

I think that by saying that the inactive ASP MAY audit, and
the SG MAY respond to audits from inactive ASPs, this type
of smooth restart can be acheived.

Note that this would work equally well for the first active
ASP as it would for the subsequent active or overriding ASP.

So I think the following would suffice:

  If an SG is in MTP restart, the first SGP SHOULD send
  DUNA(*) before ASPAC Ack to the ASP newly activating for
  the AS.

  An SGP MAY send DAVA(*) followed by DUNA, DRST and SCON
  for affected destinations before sending ASPAC Ack to a
  newly activating AS.

  An SGP MAY respond to DAUD from an inactive ASP.

  An SGP SHOULD complete responding to a previous DAUD
  before responding to ASPAC with ASPAC Ack for an ASP newly
  activating for an AS.

  If an ASP wishes to update destination status when going
  active, it SHOULD send DAUD before sending ASPAC.


It would be nice to say somehow that in these scenarios the
SG can perform "abbreviations" such as above (DUNA(*) followed
by a couple of DAVAs, or DAVA(*) followed by a couple of
DUNAs).

Comments?
[RAA] OK to the above. In the scenarios, perhaps put a note about 
the number of SSNM messages can be kept to a minimum by sending
an SSNM message which gives the status of the majority of the
concerned signalling points, followed by one or more SSNMs 
where necessary giving the status of the minority of SPs ( {raa}

Kind regards,
Dick Adams,
richard.adams@etsi.org
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 11 04:52:30 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA25191
	for <sigtran-archive@odin.ietf.org>; Tue, 11 Feb 2003 04:52:30 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1BA1K601138
	for sigtran-archive@odin.ietf.org; Tue, 11 Feb 2003 05:01:20 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1BA0rp01111;
	Tue, 11 Feb 2003 05:00:54 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1B9xbp01025
	for <sigtran@optimus.ietf.org>; Tue, 11 Feb 2003 04:59:37 -0500
Received: from web12801.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id EAA25132
	for <sigtran@ietf.org>; Tue, 11 Feb 2003 04:50:16 -0500 (EST)
Message-ID: <20030211095359.60925.qmail@web12801.mail.yahoo.com>
Received: from [62.29.65.21] by web12801.mail.yahoo.com via HTTP; Tue, 11 Feb 2003 01:53:59 PST
Date: Tue, 11 Feb 2003 01:53:59 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
To: sigtran@ietf.org
In-Reply-To: <20030210142542.A28140@openss7.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Brian,

Some more questions/comments.

  Tolga

[..snip...]
So I think the following would suffice:

  If an SG is in MTP restart, the first SGP SHOULD
send
  DUNA(*) before ASPAC Ack to the ASP newly activating
for
  the AS.

  An SGP MAY send DAVA(*) followed by DUNA, DRST and
SCON
  for affected destinations before sending ASPAC Ack
to a
  newly activating AS.

  An SGP MAY respond to DAUD from an inactive ASP.

  An SGP SHOULD complete responding to a previous DAUD
  before responding to ASPAC with ASPAC Ack for an ASP
newly
  activating for an AS.

  If an ASP wishes to update destination status when
going
  active, it SHOULD send DAUD before sending ASPAC.


[TOLGA] That SSNM and ASPAC/ASPAC-ACK should be sent
on the same stream is a key point in the proposed
procedure. So, I would think, this somehow should be
mentioned. 
In fact I would think, stream related issues deserve
to be explained in IG as a separate topic (like "AS
stream concept", synchronizing with HBEAT/HBEAT-ACK on
each stream to be used for the AS etc... IMO, still
some further discussions might be usefull about those
topics)

And I also would like to add -although not practically
important- even with the proposed procedures, I
wouldn't see ASPAC-ACK as corresponding to TRA but to
MTP-RestartEnd indication in case of MTP-Restart and
to the in Q.704 unspecified indication for other
cases(considering that this corresponds to becoming of
local user part available. In fact even when SG is in
MTP-Restart isn't ASPAC indicating that local user
part is available now?)





__________________________________________________
Do you Yahoo!?
Yahoo! Shopping - Send Flowers for Valentine's Day
http://shopping.yahoo.com
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 11 05:31:42 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA26026
	for <sigtran-archive@odin.ietf.org>; Tue, 11 Feb 2003 05:31:42 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1BAeWX03663
	for sigtran-archive@odin.ietf.org; Tue, 11 Feb 2003 05:40:32 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1BAeEp03655;
	Tue, 11 Feb 2003 05:40:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1BAdIp03608
	for <sigtran@optimus.ietf.org>; Tue, 11 Feb 2003 05:39:18 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA25986
	for <sigtran@ietf.org>; Tue, 11 Feb 2003 05:29:56 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:d20HEAII6NRYN7V9UmSykySAD9T7M5Rz@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1BAXcD25496;
	Tue, 11 Feb 2003 03:33:38 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1BAXbJ15519;
	Tue, 11 Feb 2003 03:33:37 -0700
Date: Tue, 11 Feb 2003 03:33:37 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tolga Asveren <tolga_asveren@yahoo.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
Message-ID: <20030211033337.A14614@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030210142542.A28140@openss7.org> <20030211095359.60925.qmail@web12801.mail.yahoo.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20030211095359.60925.qmail@web12801.mail.yahoo.com>; from tolga_asveren@yahoo.com on Tue, Feb 11, 2003 at 01:53:59AM -0800
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Tolga,

Tolga Asveren wrote:                                                                 (Tue, 11 Feb 2003 01:53:59)
> Brian,
> 
> Some more questions/comments.
> 
>   Tolga
> 
> [..snip...]
> So I think the following would suffice:
> 
>   If an SG is in MTP restart, the first SGP SHOULD
> send
>   DUNA(*) before ASPAC Ack to the ASP newly activating
> for
>   the AS.
> 
>   An SGP MAY send DAVA(*) followed by DUNA, DRST and
> SCON
>   for affected destinations before sending ASPAC Ack
> to a
>   newly activating AS.
> 
>   An SGP MAY respond to DAUD from an inactive ASP.
> 
>   An SGP SHOULD complete responding to a previous DAUD
>   before responding to ASPAC with ASPAC Ack for an ASP
> newly
>   activating for an AS.
> 
>   If an ASP wishes to update destination status when
> going
>   active, it SHOULD send DAUD before sending ASPAC.
> 
> 
> [TOLGA] That SSNM and ASPAC/ASPAC-ACK should be sent
> on the same stream is a key point in the proposed
> procedure. So, I would think, this somehow should be
> mentioned. 
> In fact I would think, stream related issues deserve
> to be explained in IG as a separate topic (like "AS
> stream concept", synchronizing with HBEAT/HBEAT-ACK on
> each stream to be used for the AS etc... IMO, still
> some further discussions might be usefull about those
> topics)
> 
> And I also would like to add -although not practically
> important- even with the proposed procedures, I
> wouldn't see ASPAC-ACK as corresponding to TRA but to
> MTP-RestartEnd indication in case of MTP-Restart and
> to the in Q.704 unspecified indication for other
> cases(considering that this corresponds to becoming of
> local user part available. In fact even when SG is in
> MTP-Restart isn't ASPAC indicating that local user
> part is available now?)
> 

Not when the ASP point code is separate from the SG
point code.  In that case, the ASP MTP is really restarting
and ASPAC is equivalent to TRA.  The ASP is not an MTP
user which has become available, but an entire point code
that has just appeared on the SS7 network.  But for the
pure backhaul cases, I agree, ASPAC corresponds to the
underspecified MTP-RESTART-ENDS (primitive?).

--brian


-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 11 08:20:50 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA03879
	for <sigtran-archive@odin.ietf.org>; Tue, 11 Feb 2003 08:20:50 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1BDTjI14292
	for sigtran-archive@odin.ietf.org; Tue, 11 Feb 2003 08:29:45 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1BDTWp14246;
	Tue, 11 Feb 2003 08:29:32 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1BDS2p14163
	for <sigtran@optimus.ietf.org>; Tue, 11 Feb 2003 08:28:02 -0500
Received: from web12806.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id IAA03753
	for <sigtran@ietf.org>; Tue, 11 Feb 2003 08:18:36 -0500 (EST)
Message-ID: <20030211132219.40026.qmail@web12806.mail.yahoo.com>
Received: from [62.29.76.49] by web12806.mail.yahoo.com via HTTP; Tue, 11 Feb 2003 05:22:19 PST
Date: Tue, 11 Feb 2003 05:22:19 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] M3UA - AUTHORS-FINAL-ASP database equalization
To: sigtran@ietf.org
In-Reply-To: <20030211033337.A14614@openss7.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Brian,
> > And I also would like to add -although not
> practically
> > important- even with the proposed procedures, I
> > wouldn't see ASPAC-ACK as corresponding to TRA but
> to
> > MTP-RestartEnd indication in case of MTP-Restart
> and
> > to the in Q.704 unspecified indication for other
> > cases(considering that this corresponds to
> becoming of
> > local user part available. In fact even when SG is
> in
> > MTP-Restart isn't ASPAC indicating that local user
> > part is available now?)
> > 
> 
> Not when the ASP point code is separate from the SG
> point code.  In that case, the ASP MTP is really
> restarting
> and ASPAC is equivalent to TRA.  The ASP is not an
> MTP
> user which has become available, but an entire point
> code
> that has just appeared on the SS7 network.  But for
> the
> pure backhaul cases, I agree, ASPAC corresponds to
> the
> underspecified MTP-RESTART-ENDS (primitive?).
[TOLGA]Even for STP SG case, it is quite discussible,
where MTP3 of PC for AS sits , or how much of it sits
where(for single SG case, for multiple SG case)(but no
need to discuss further in this thread IMO)


__________________________________________________
Do you Yahoo!?
Yahoo! Shopping - Send Flowers for Valentine's Day
http://shopping.yahoo.com
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 11 10:08:39 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA07566
	for <sigtran-archive@odin.ietf.org>; Tue, 11 Feb 2003 10:08:39 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1BFHZO21385
	for sigtran-archive@odin.ietf.org; Tue, 11 Feb 2003 10:17:35 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1BFHIp21362;
	Tue, 11 Feb 2003 10:17:20 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1BFG1p21313
	for <sigtran@optimus.ietf.org>; Tue, 11 Feb 2003 10:16:01 -0500
Received: from spf1.us.outblaze.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id KAA07512
	for <sigtran@ietf.org>; Tue, 11 Feb 2003 10:06:33 -0500 (EST)
Received: (qmail 27935 invoked from network); 11 Feb 2003 15:10:13 -0000
Received: from unknown (205.158.62.68)
  by spf1.us.outblaze.com with QMQP; 11 Feb 2003 15:10:13 -0000
Received: (qmail 41919 invoked from network); 11 Feb 2003 15:09:48 -0000
Received: from unknown (HELO ws1-6.us4.outblaze.com) (205.158.62.53)
  by 205-158-62-153.outblaze.com with SMTP; 11 Feb 2003 15:09:48 -0000
Received: (qmail 10242 invoked by uid 1001); 11 Feb 2003 15:09:47 -0000
Message-ID: <20030211150947.10241.qmail@mail.com>
Content-Type: text/plain; charset="iso-8859-1"
Content-Disposition: inline
Content-Transfer-Encoding: 7bit
MIME-Version: 1.0
X-Mailer: MIME-tools 5.41 (Entity 5.404)
Received: from [204.42.169.130] by ws1-6.us4.outblaze.com with http for
    jli@iname.com; Tue, 11 Feb 2003 10:09:47 -0500
From: "John Li" <jli@iname.com>
To: sigtran@ietf.org
Date: Tue, 11 Feb 2003 10:09:47 -0500
X-Originating-Ip: 204.42.169.130
X-Originating-Server: ws1-6.us4.outblaze.com
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] M2PA draft comments and questions.
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

All,
I sent the following comments and questions to the group a while ago, and have not got responses yet. here goes again. 
thanks.
John


1.	In section 1.5: the following statement is somewhat confusing.
“The SG is an IPSP equipped with both traditional SS7 and IP
network connections. In effect, the Signaling Gateway acts as a
Signal Transfer Point (STP)” 

Perhaps we can say "The SG interconnects SS7 and IP networks by
converting signaling traffic" or just refer to the definition in
section 1.2


2.	Section 2.1.5 Message length
Should we specify the valid value range of the length?

3.	Section 2.3.1 User Data

statement "M2PA does not add padding to the MTP3 message.".
I think we should make it a requirement: “M2PA should not add padding
to the MTP3 message.”


4.	In section 4.1.3 Link alignment

perhaps we can specify the recommended practice for choosing
emergency proving vs. normal proving during alignment procedure.


5.	Section 4.1.8: 
"If a Link Status Alignment message with an unsupported
received by the newer version, the receiving end's M2PA
SHALL NOT complete the alignment procedure."

can we clearly define the behavior of not completing the alignment
procedure?

6.	section 4.2.1 sending and receiving messages
When M2PA receives a User Data message with Backward Sequence Number
equal to 'n', it may remove all messages with Forward Sequence Number
<= n from its queue.

This is incorrect. Consider at the time SN is cycled to 0 from 127
(assume 128 buffers), there are 127, 126, etc are waiting for
acknowledgement. We should replace it with “The message n and all
outstanding (not yet acknowledged) messages sent before the message n
may be removed from its queue.

7.	section 4.2.1
If M2PA receives an empty User Data message, it SHALL NOT send an
acknowledgement of that message.

What is empty User Data message? FISUs? 

-- 
__________________________________________________________
Sign-up for your own FREE Personalized E-mail at Mail.com
http://www.mail.com/?sr=signup

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 11 13:27:09 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA14106
	for <sigtran-archive@odin.ietf.org>; Tue, 11 Feb 2003 13:27:09 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1BIaAC01615
	for sigtran-archive@odin.ietf.org; Tue, 11 Feb 2003 13:36:10 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1BIZip01547;
	Tue, 11 Feb 2003 13:35:46 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1BIYUp01465
	for <sigtran@optimus.ietf.org>; Tue, 11 Feb 2003 13:34:30 -0500
Received: from domino2.nc.tekelec.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA13984
	for <sigtran@ietf.org>; Tue, 11 Feb 2003 13:24:58 -0500 (EST)
From: jeffrey.craig@tekelec.com
Received: from domino4.nc.tekelec.com ([198.89.40.9])
          by domino2.nc.tekelec.com (Lotus Domino Release 5.0.10)
          with ESMTP id 2003021113222352:30259 ;
          Tue, 11 Feb 2003 13:22:23 -0500 
Subject: Re: [Sigtran] M2PA draft comments and questions.
To: "John Li" <jli@iname.com>
Cc: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.7  March 21, 2001
Message-ID: <OF8544DAD4.2B59ADE8-ON85256CCA.00634D4A@nc.tekelec.com>
Date: Tue, 11 Feb 2003 13:24:26 -0500
MIME-Version: 1.0
X-MIMETrack: Serialize by Router on Domino4/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/11/2003 01:24:27 PM,
	Itemize by SMTP Server on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/11/2003 01:22:23 PM,
	Serialize by Router on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/11/2003 01:22:26 PM,
	Serialize complete at 02/11/2003 01:22:26 PM
Content-type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h1BIYUp01466
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit


Hello John,

Since no one else is responding, I'll answer.

My responses are embedded below.

Regards,

Jeff

On 02/11/2003 10:09:47 AM sigtran-admin wrote:

>All,
>I sent the following comments and questions to the group a while ago, and
have
>not got responses yet. here goes again.
>thanks.
>John
>
>
>1.   In section 1.5: the following statement is somewhat confusing.
>"The SG is an IPSP equipped with both traditional SS7 and IP
>network connections. In effect, the Signaling Gateway acts as a
>Signal Transfer Point (STP)"
>
>Perhaps we can say "The SG interconnects SS7 and IP networks by
>converting signaling traffic" or just refer to the definition in
>section 1.2

I am not sure what defect this removes, and so I have no comment on this.

>
>
>2.   Section 2.1.5 Message length
>Should we specify the valid value range of the length?
>

No. Min/Max message length is dependent on message type,
and in the case of user data is governed by MTP3.

>3.   Section 2.3.1 User Data
>
>statement "M2PA does not add padding to the MTP3 message.".
>I think we should make it a requirement: "M2PA should not add padding
>to the MTP3 message."
>

Sounds good to me.

>
>4.   In section 4.1.3 Link alignment
>
>perhaps we can specify the recommended practice for choosing
>emergency proving vs. normal proving during alignment procedure.
>

This is governed by MTP3 procedures and is outside the scope
of the M2PA draft.

>
>5.   Section 4.1.8:
>"If a Link Status Alignment message with an unsupported
>received by the newer version, the receiving end's M2PA
>SHALL NOT complete the alignment procedure."
>
>can we clearly define the behavior of not completing the alignment
>procedure?

I believe that you are suggesting that the text be reworded to state
the following:

"If a Link Status Alignment message with an unsupported version
is received by the newer version, the receiving end's M2PA
SHOULD fail the link."

I think this was already understood, but the change improves clarity.

>
>6.   section 4.2.1 sending and receiving messages
>When M2PA receives a User Data message with Backward Sequence Number
>equal to 'n', it may remove all messages with Forward Sequence Number
><= n from its queue.
>
>This is incorrect. Consider at the time SN is cycled to 0 from 127
>(assume 128 buffers), there are 127, 126, etc are waiting for
>acknowledgement. We should replace it with "The message n and all
>outstanding (not yet acknowledged) messages sent before the message n
>may be removed from its queue.

I agree with the change.

>
>7.   section 4.2.1
>If M2PA receives an empty User Data message, it SHALL NOT send an
>acknowledgement of that message.
>
>What is empty User Data message? FISUs?
>

It is a M2PA user data message containing only the common header.
An 'empty'  User Data message is an M2PA receipt acknowledgement.
The statement in the draft is saying don't ACK an ACK.

>--
>__________________________________________________________
>Sign-up for your own FREE Personalized E-mail at Mail.com
>http://www.mail.com/?sr=signup
>
>_______________________________________________
>Sigtran mailing list
>Sigtran@ietf.org
>https://www1.ietf.org/mailman/listinfo/sigtran


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 11 15:30:37 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA17785
	for <sigtran-archive@odin.ietf.org>; Tue, 11 Feb 2003 15:30:37 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1BKddq10621
	for sigtran-archive@odin.ietf.org; Tue, 11 Feb 2003 15:39:39 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1BKdOp10602;
	Tue, 11 Feb 2003 15:39:25 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1BKcvp10555
	for <sigtran@optimus.ietf.org>; Tue, 11 Feb 2003 15:38:57 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA17761
	for <sigtran@ietf.org>; Tue, 11 Feb 2003 15:29:23 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1BKX4qe005538;
	Tue, 11 Feb 2003 14:33:04 -0600 (CST)
Message-ID: <3E495E00.8DE7C95E@alcatel.com>
Date: Tue, 11 Feb 2003 14:33:04 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA Link OOS (was: M2PA version 7 changes)
References: <3D72C18DF140D311A54F0008C7DBAC8D02917425@MCHH252E> <3E384D0E.AF500008@alcatel.com> <20030129152705.A4253@openss7.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Here is my idea for the new OOS section that I will reference throughout
the document. I thought I would make it section 4.1.6 and move the old
4.1.6ff down.

==========================
4.1.6  Link Out of Service

When M2PA places a link in the OUT OF SERVICE state, 

   (1) M2PA SHALL send a Link Status Out of Service message to its
       peer.

   (2) M2PA SHALL send an Out of Service indication to its upper layer
       MTP3. M2PA MAY send additional Link Status Out of Service
       messages to its peer while it remains in the OUT OF SERVICE
       state.

   (3) M2PA SHOULD NOT terminate the SCTP association.

M2PA waits for an MTP3 Start command to begin the alignment procedure.
==========================


"Brian F. G. Bidulock" wrote:
> 
> Tom,
> 
> I like the little OOS section idea.
> 
> --brian
> 
> Tom George wrote:                                                                      (Wed, 29 Jan 2003 15:52:14)
> > Franz and others,
> >
> > Yes, in that case M2PA should send LS OOS to its peer.
> >
> > There are several places in the document where the link goes OOS. As it
> > stands now, I do not always state all of the actions to be taken because
> > they are more-or-less understood:
> >
> > - Send LS OOS to the peer.
> > - Notify MTP3 of Out of Service condition.
> > - Leave the SCTP association up. (a SHOULD)
> > - Wait for MTP3 to begin alignment.
> >
> > What is the best way to explain this? Repeat it throughout the document,
> > or have a little OOS section (similar to 4.2.3  Link deactivation), or
> > just leave it as is and count on the reader to understand it?
> >
> > tomG.
> >
> > Gaiser Franz wrote:
> > >
> > > Dear All,
> > >
> > > some remarks to the new draft version 07:
> > >
> >   [...]
> > > 4.) 4.2.1 Paragraph "If M2PA receives a User Data message with
> > > an FSN that is out of order, M2PA SHALL take the link out of service."
> > > Does "take the link out of service"  mean also that the peer side is
> > > informed via sending LS OOS message.
> >   [...]
> >
> > --
> > * Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
> > * 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
> > * Alcatel USA, Inc.; 1000 Coit Rd, MS PB7-SWPRO; Plano TX 75075-5813  *
> > * I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
> > * to advertise or represent its position,  operations, or activities. *
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 11 16:04:33 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA18524
	for <sigtran-archive@odin.ietf.org>; Tue, 11 Feb 2003 16:04:33 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1BLDa512560
	for sigtran-archive@odin.ietf.org; Tue, 11 Feb 2003 16:13:36 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1BLDJp12547;
	Tue, 11 Feb 2003 16:13:19 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1BLCWp12513
	for <sigtran@optimus.ietf.org>; Tue, 11 Feb 2003 16:12:32 -0500
Received: from rtp-core-2.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA18487
	for <sigtran@ietf.org>; Tue, 11 Feb 2003 16:02:57 -0500 (EST)
Received: from dingdong.cisco.com (IDENT:mirapoint@dingdong.cisco.com [64.102.17.16])
	by rtp-core-2.cisco.com (8.12.6/8.12.6) with ESMTP id h1BL6aNh022757;
	Tue, 11 Feb 2003 16:06:37 -0500 (EST)
Received: from cisco.com (dhcp-64-102-86-148.cisco.com [64.102.86.148])
	by dingdong.cisco.com (Mirapoint Messaging Server MOS 3.2.1-GA)
	with ESMTP id ABK52155;
	Tue, 11 Feb 2003 16:06:34 -0500 (EST)
Message-ID: <3E4965DA.33245145@cisco.com>
Date: Tue, 11 Feb 2003 16:06:34 -0500
From: wayne n taylor <waynet@cisco.com>
X-Mailer: Mozilla 4.79 [en]C-CCK-MCD   (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Tom George <Tom.George@alcatel.com>
CC: bidulock@openss7.org, sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft 7 question
References: <3E4350A2.8060B90@cisco.com> <3E482299.C3E86A0E@alcatel.com> <20030210170938.A11994@openss7.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Tom,
as i described earlier, my concern is the potential loop that could
occur during the initial LS Alignment exchange based on the requirement
that both peers must send and receive LS Alignment messages. Brian has
shown in his response (and i agree) that the scenario i described will
not happen if the procedures defined in the SDLs of the Q.703 spec is
adopted for m2pa in regards to the initial "alignment" exchange.

wayne

"Brian F. G. Bidulock" wrote:
> 
> Tom,
> 
> The first thing a link should send is LS OOS.
> 
> In the out of service state, it would be desireable for the out of
> service side to always send LS OOS in response to other messages.
> 
> Also, the out of service side should be permitted to send LS OOS
> periodically.
> 
> With regards to Q.703 state machines, LS OOS in response to alignment
> will not abort alignment and T2 should still expire.  See test case
> "3.1.2 Timer T2" in draft-bidulock-sigtran-m2pa-test-02.txt.
> 
> T2 is defined as the duration of time that the M2PA peer wil wait to
> receive a status "Alignment" message after sending a status "Alignment"
> message.
> 
> As status Alignment will abort an otherwise in-service link, other
> link status received while running T2 should be ignored.
> 
> Any different behaviour could interfere with MTP Level 3's link
> restoration timers and procedures.
> 
> More comments below...
> 
> Tom George wrote:                                                                      (Mon, 10 Feb 2003 16:07:21)
> > Wayne,
> >
> > Yes, I want both sides to send and receive a LS Alignment message to
> > complete alignment.
> >
> > To fix the problem you describe, how about this" If M2PA has not
> > received MTP3 Start and receives LS Alignment, M2PA replies with LS OOS
> > (instead of not replying at all and counting on the other end's T2
> > expiring).
> >
> > tomG.
> >
> > wayne n taylor wrote:
> > >
> > > Hi All,
> > >
> > > i have a couple of comments/questions about the initial exchange of the
> > > Link Status Alignment messages. there seems to be some inconsistency in
> > > the text and finite state machine.
> > >
> > > section 4.1.3 paragraph 4:
> > >
> > > "Once the association is established and M2PA has received a Start
> > > Request from MTP3, M2PA sends the Link Status Alignment message to its
> > > peer. If M2PA has not already received the Link Status Alignment
> > > message from its peer, then M2PA starts timer T2."
> > >
> > > what state would the local M2PA be in to receive a LS Alignment message
> > > from its peer prior to getting a MTP3 Start ?. i assume the OOS state,
> > > but the finite state machine does not show LS Alignment as an input.
> > >
> > > section 4.1.3 paragraph 5:
> > >
> > > "(Note that if the remote M2PA has not received a Start Request from
> > > its MTP3, it will not send the Link Status Alignment message to the
> > > local M2PA. Eventually timer T2 in the local M2PA will expire. If the
> > > remote M2PA receives a Start Request from its MTP3 and sends Link
> > > Status Alignment before the local M2PA's timer T2 expires, the
> > > alignment procedure can continue.)"
> > >
> > > assume the local M2PA receives a Start Request from MTP3, sends LS
> > > Alignment and start T2, but the remote peer has not received a Start
> > > Request and is in OOS. the current finite state machine does not show
> > > what to do with the LS Alignment from the local peer so lets assume the
> > > LS Alignment is dropped. the remote peer receives a Start Request from
> > > MTP3, sends LS Alignment and start T2 (since the remote peer didnt get
> > > the local peers LS Alignment). assume further that timer T2 is still
> > > running on the local peer. the text from section 4.1.3 above states the
> > > "alignment procedure can continue". in this case the local peer sent a
> > > LS Alignment and received a LS Alignment. the remote peer only sent a LS
> > > Alignment and did not receive an LS Alignment (since LS Alignment
> > > messages are not retried). eventually T2 on the remote peer will expire,
> > > T3 on the local peer will expire and the alignment procedure starts over
> > > again. it seems possible for the peers to get caught in a loop trying to
> > > exchange LS Alignment messages.
> 
> See draft-bidulock-sigtran-m2pa-test-02.txt Section "3.1.3 Timer T3".  You
> will see that what you describe does not happen if you follow Q.703 SDLs.
> In Q.703 once SIO is sent SIN is sufficient to complete alignment.  So,
> one side sends ALIGNMENT and starts T2, the other side sends ALIGNMENT and
> starts T2.  The first side, receives the alignment, stops T2 and sends
> PROVING-NORMAL or PROVING-EMERGENCY and starts T3.  When the second side
> receives PROVING-NORMAL or PROVING-EMERGENCY it stops T2 and sends
> PROVING-NORMAL or PROVING-EMERGENCY and starts T4.  When the first side
> receives PROVING-NORMAL it stops T3 and starts T4.  When either T4 expires,
> the side sends IN-SERVICE and starts T1.  When the side receives IN-SERVICE
> it stops T1 and indicates in_service to MTP3.
> 
> This behavior can be seen on Figure 9/Q.703 (sheet 2 of 6) - Initial alignment
> control.  When the IAC starts it sends SIO and starts T2.  If it receives
> any of SIO, SIN or SIE it stops T2 and send SIN or SIE depending on emergency
> status and starts T3.  Receipt of a subsequent SIN or SIE (sheet 3 of 6)
> stops T3 and starts T4.  Note that if it receives SIO again, it ignores it
> at that point and continues to send SIE or SIN.
> 
> These is no reason for M2PA not to follow the Q.703 SDLs in this regard.
> 
> A statement that might clarify is that the M2PA sends LS ALIGNMENT and if
> it receives ALIGNMENT or PROVING it stops T2 and PROVING and starts T3.
> Note that it takes two PROVING messages in this case to move the peer into
> the proving state.  One PROVING message acts like an ALIGNMENT (cancels
> T2 and starts T3), the next acts like the first PROVING message (cancels
> T3 and starts T4).
> 
> --brian
> 
> > >
> > > section 4.1.3 paragraph 9:
> > >
> > > "When M2PA has both sent and received the Link Status Alignment
> > > message, it has completed alignment.  M2PA starts the aligned timer T3
> > > and moves to the proving state."
> > >
> > > is it a requirement for both the local and remote peers to send/receive
> > > LS Alignment for alignment to be complete ?. the only way to transition
> > > from Initial Alignment to Proving is to receive LS Alignment. in the
> > > scenario above, both peers do not send and receive LS Alignment
> > > messages.
> > >
> > > regards,
> > >
> > > wayne
> >
> > --
> > * Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
> > * 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
> > * Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
> > * I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
> > * to advertise or represent its position,  operations, or activities. *
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 11 16:57:07 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA19649
	for <sigtran-archive@odin.ietf.org>; Tue, 11 Feb 2003 16:57:07 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1BM6Bs15523
	for sigtran-archive@odin.ietf.org; Tue, 11 Feb 2003 17:06:11 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1BM5jp15486;
	Tue, 11 Feb 2003 17:05:47 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1BM41p15433
	for <sigtran@optimus.ietf.org>; Tue, 11 Feb 2003 17:04:01 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA19596
	for <sigtran@ietf.org>; Tue, 11 Feb 2003 16:54:26 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1BLw7JT001142;
	Tue, 11 Feb 2003 15:58:07 -0600 (CST)
Message-ID: <3E4971EF.BDD6AAE3@alcatel.com>
Date: Tue, 11 Feb 2003 15:58:07 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: "'sigtran@ietf.org'" <sigtran@ietf.org>
References: <2135200C183FD5119588009027DE572302836C0A@webdev-owa.oni.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] M2PA Security section (was: m2pa comment: author list)
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

On Jan 26, 2003 "Ong, Lyndon" wrote:
  [...]
> The security section probably needs to be updated based on the
> security framework document (which should come out this week).
  [...]

Here is my first stab at revising section 6 of the M2PA draft. Comments
are welcome, especially since I am not familiar with the security
protocols.

tomG.

=============================
6. Security

M2PA is designed to carry signaling messages for telephony
services. As such, M2PA MUST involve the security needs of several
parties: 

   - the end users of the services

   - the network providers

   - the applications involved

Additional requirements MAY come from local regulation.  

While these parties may have some overlapping security needs, their
needs may not be identical. Any security solution SHOULD fulfill all
of the different parties' needs.

6.1 Security Requirements

As a transport protocol, M2PA has the following security objectives:

   - Availability of reliable and timely user data transport.

   - Integrity of user data transport.

   - Confidentiality of user data.

M2PA runs on top of SCTP [RFC2960].  SCTP provides certain transport
related security features, as discussed in [Security]. However, these
security features MAY NOT be adequate for all of the involved parties'
requirements.

When M2PA is running in professionally managed corporate or service
provider network, it is reasonable to expect that this network
includes an appropriate security policy framework. The "Site Security
Handbook" [RFC2196] SHOULD be consulted for guidance.

When the network in which M2PA runs involves more than one party
(e.g., a non-private network), it MAY NOT be reasonable to expect that
all parties have implemented security in a sufficient manner.  In such
a case, it is recommended that IPSec or TLS be used to ensure
confidentiality of the user payload. Consult [Security] for more
guidance on the use of security protocols. 

6.2 Protecting Confidentiality

Particularly for mobile users, the requirement for confidentiality MAY
include the masking of IP addresses and ports. In this case
application-level encryption is not sufficient. IPSec ESP SHOULD be
used instead [RFC2401].  Regardless of which level performs the
encryption, the IPSec ISAKMP service SHOULD be used for key
management.

=============================

{Add an informative reference:}

  [Security]
       J. Loughney, M. Tuexen, and J. Pastor-Balbas, "Security
       Considerations for SIGTRAN Protocols,"
       draft-ietf-sigtran-security-01.txt (January 2003).

=============================

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 11 18:30:27 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA21754
	for <sigtran-archive@odin.ietf.org>; Tue, 11 Feb 2003 18:30:27 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1BNdYA21630
	for sigtran-archive@odin.ietf.org; Tue, 11 Feb 2003 18:39:34 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1BNdFp21614;
	Tue, 11 Feb 2003 18:39:16 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1BNcBp21564
	for <sigtran@optimus.ietf.org>; Tue, 11 Feb 2003 18:38:11 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA21642
	for <sigtran@ietf.org>; Tue, 11 Feb 2003 18:28:33 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1BNWFJT017392;
	Tue, 11 Feb 2003 17:32:16 -0600 (CST)
Message-ID: <3E4987FF.9744C944@alcatel.com>
Date: Tue, 11 Feb 2003 17:32:15 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
To: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA Link OOS (was: M2PA version 7 changes)
References: <3D72C18DF140D311A54F0008C7DBAC8D02917425@MCHH252E> <3E384D0E.AF500008@alcatel.com> <20030129152705.A4253@openss7.org> <3E495E00.8DE7C95E@alcatel.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Oops. Of course, the 2nd sentence in (2) should be moved to (1). 

tomG.

Tom George wrote:
> 
> Here is my idea for the new OOS section that I will reference throughout
> the document. I thought I would make it section 4.1.6 and move the old
> 4.1.6ff down.
> 
> ==========================
> 4.1.6  Link Out of Service
> 
> When M2PA places a link in the OUT OF SERVICE state,
> 
>    (1) M2PA SHALL send a Link Status Out of Service message to its
>        peer.
> 
>    (2) M2PA SHALL send an Out of Service indication to its upper layer
>        MTP3. M2PA MAY send additional Link Status Out of Service
>        messages to its peer while it remains in the OUT OF SERVICE
>        state.
> 
>    (3) M2PA SHOULD NOT terminate the SCTP association.
> 
> M2PA waits for an MTP3 Start command to begin the alignment procedure.
> ==========================
> 
> "Brian F. G. Bidulock" wrote:
> >
> > Tom,
> >
> > I like the little OOS section idea.
> >
> > --brian
> >
> > Tom George wrote:                                                                      (Wed, 29 Jan 2003 15:52:14)
> > > Franz and others,
> > >
> > > Yes, in that case M2PA should send LS OOS to its peer.
> > >
> > > There are several places in the document where the link goes OOS. As it
> > > stands now, I do not always state all of the actions to be taken because
> > > they are more-or-less understood:
> > >
> > > - Send LS OOS to the peer.
> > > - Notify MTP3 of Out of Service condition.
> > > - Leave the SCTP association up. (a SHOULD)
> > > - Wait for MTP3 to begin alignment.
> > >
> > > What is the best way to explain this? Repeat it throughout the document,
> > > or have a little OOS section (similar to 4.2.3  Link deactivation), or
> > > just leave it as is and count on the reader to understand it?
> > >
> > > tomG.
> > >
> > > Gaiser Franz wrote:
> > > >
> > > > Dear All,
> > > >
> > > > some remarks to the new draft version 07:
> > > >
> > >   [...]
> > > > 4.) 4.2.1 Paragraph "If M2PA receives a User Data message with
> > > > an FSN that is out of order, M2PA SHALL take the link out of service."
> > > > Does "take the link out of service"  mean also that the peer side is
> > > > informed via sending LS OOS message.
> > >   [...]
> > >
> > > --
> > > * Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
> > > * 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
> > > * Alcatel USA, Inc.; 1000 Coit Rd, MS PB7-SWPRO; Plano TX 75075-5813  *
> > > * I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
> > > * to advertise or represent its position,  operations, or activities. *
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> > --
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> --
> * Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
> * 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
> * Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
> * I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
> * to advertise or represent its position,  operations, or activities. *

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 11 21:14:16 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA24982
	for <sigtran-archive@odin.ietf.org>; Tue, 11 Feb 2003 21:14:16 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1C2NQ330626
	for sigtran-archive@odin.ietf.org; Tue, 11 Feb 2003 21:23:26 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1C2NCp30608;
	Tue, 11 Feb 2003 21:23:13 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1C2Lhp30538
	for <sigtran@optimus.ietf.org>; Tue, 11 Feb 2003 21:21:43 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA24936;
	Tue, 11 Feb 2003 21:12:00 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:nkteuQ4IYqcrgn/w5At224D1D1Drd8Mo@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1C2FiD07777;
	Tue, 11 Feb 2003 19:15:44 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1C2FhU16862;
	Tue, 11 Feb 2003 19:15:43 -0700
Date: Tue, 11 Feb 2003 19:15:43 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Randall Stewart <randall@stewart.chicago.il.us>
Cc: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>,
        Caitlin Bestler <cait@asomi.com>, tsvwg@ietf.org,
        Tuexen Michael <michael.tuexen@siemens.com>, sigtran@ietf.org
Message-ID: <20030211191543.A14481@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Randall Stewart <randall@stewart.chicago.il.us>,
	Michael Tuexen <Michael.Tuexen@lurchi.franken.de>,
	Caitlin Bestler <cait@asomi.com>, tsvwg@ietf.org,
	Tuexen Michael <michael.tuexen@siemens.com>, sigtran@ietf.org
References: <A2AA39E1-3DFB-11D7-90B4-0030654C1AB6@lurchi.franken.de> <3E49899A.6010809@stewart.chicago.il.us>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E49899A.6010809@stewart.chicago.il.us>; from randall@stewart.chicago.il.us on Tue, Feb 11, 2003 at 05:39:06PM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: 8bit
Subject: [Sigtran] Re: [Tsvwg] A question of Un-Ordered SCTP
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Randall,

I agree to having SCTP set the sequence number and, as you said in
your first note, SCTP passing the sequence number to the ULP on
the receiving side is already done for ordered traffic.  Some even
pass the zero now.

I suppose that SCTP would set this sequence number when it is given
the data for delivery at the sending ULP and the sequence would represent
the order that the data was presented to SCTP by the ULP at the sending
side.

I think it would be good to see the requirement that the sequence
number begin at 1 and be monotonically increasing in the saem fashion
as ordered streams.

That way RDDP and other applications could perform their own reordering
at the receiver.  I think it would be a good thing to have SCTP set
this value.

In fact, I can see another application for this feature.  M2PA
(SIGTRAN protocol) needs in some circumstances to send data unordered
with the hope of expedited delivery (does not want control information
stuck behind large amounts of data in an ordered stream), however, the
order of processing of the expedited control information is important.

This facility would permit M2PA to send such information unordered
to expedite it and the receiver could perform the reordering itself.
Up until now, the only solution was to place a sequence number in with
the control information.

Could we perhaps add a section to the SCTP IG.  "Problem statement:
the sender wishes to send unordered data but the receive needs to know
the original order of sending..." or some such.

It should be quite simple to add the feature to existing SCTP
implementations.

--brian


On Tue, 11 Feb 2003, Randall Stewart wrote:

> Michael:
> 
> This seems like a valid method.. it gets around Brian's objection
> (which was a good point) and yet still provides the application with
> a hint about what the assigned sequence order was on the unordered
> messages...
> 
> R
> 
> Michael Tuexen wrote:
> > Caitlin,
> > 
> > thank you very much for the clarification.
> > 
> > I do not like the idea to pass the SSN transparently from the ULP to the 
> > ULP.
> > But having two sequence spaces is OK for me. This means that the SSN at the
> > sending side is controlled by the SCTP and given to the ULP at the 
> > receiving side.
> > 
> > This would be a clean solution. So I voting for have two separate number 
> > spaces
> > for user data. Since the one for ordered delivery exists already in RFC 
> > 2960, this
> > only means that a new one is introduced for the unordered delivery user 
> > data.
> > 
> > Best regards
> > Michael
> > 
> > On Tuesday, Feb 11, 2003, at 17:37 Europe/Berlin, Caitlin Bestler wrote:
> > 
> >> On 2/11/03, Tuexen Michael wrote:
> >>
> >>> Dear all,
> >>>
> >>> does this mean that you want two sequence spaces per
> >>> stream: one for the 'ordered' user data and one for the
> >>> unordered user data?
> >>>
> >> The RDDP/SCTP adaptation only uses unordered delivery, so
> >> that question did not really arise.
> >>
> >> But if you were to generalize the problem, I believe the
> >> answer would be yes.
> >>
> >> The SSN for ordered delivery would be unchanged. The
> >> addition would be providing "advisory ordering" for
> >> unordered delivery. The general characteristic is that
> >> the receiving ULP wants the data in whatever order it
> >> arrives, but wants to know what the original order was.
> >>
> >> In addition to RDDP, RTP over UDP is another example of
> >> an application that wants to be aware of the original
> >> ordering, even while accepting the data out-of-order.
> >>
> >> Caitlin Bestler
> >> http://asomi.com/CaitlinBestler/
> >> _______________________________________________
> >> tsvwg mailing list
> >> tsvwg@ietf.org
> >> https://www1.ietf.org/mailman/listinfo/tsvwg
> >>
> > 
> > _______________________________________________
> > tsvwg mailing list
> > tsvwg@ietf.org
> > https://www1.ietf.org/mailman/listinfo/tsvwg
> > 
> > 
> 
> 
> -- 
> Randall R. Stewart
> randall@stewart.chicago.il.us 815-342-5222 (cell phone)
> 
> _______________________________________________
> tsvwg mailing list
> tsvwg@ietf.org
> https://www1.ietf.org/mailman/listinfo/tsvwg

-- 
Brian F. G. Bidulock    ¦ The reasonable man adapts himself to the ¦
bidulock@openss7.org    ¦ world; the unreasonable one persists in  ¦
http://www.openss7.org/ ¦ trying  to adapt the  world  to himself. ¦
                        ¦ Therefore  all  progress  depends on the ¦
                        ¦ unreasonable man. -- George Bernard Shaw ¦
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 11 21:23:09 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA25185
	for <sigtran-archive@odin.ietf.org>; Tue, 11 Feb 2003 21:23:09 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1C2WK730937
	for sigtran-archive@odin.ietf.org; Tue, 11 Feb 2003 21:32:20 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1C2W8p30929;
	Tue, 11 Feb 2003 21:32:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1C2V8p30898
	for <sigtran@optimus.ietf.org>; Tue, 11 Feb 2003 21:31:08 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA25163
	for <sigtran@ietf.org>; Tue, 11 Feb 2003 21:21:25 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:nmOC7OeI8hPlwYPiMD/p6ooFE8PxuQY6@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1C2P9D07894;
	Tue, 11 Feb 2003 19:25:09 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1C2P9O18069;
	Tue, 11 Feb 2003 19:25:09 -0700
Date: Tue, 11 Feb 2003 19:25:09 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: jeffrey.craig@tekelec.com
Cc: John Li <jli@iname.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft comments and questions.
Message-ID: <20030211192509.B14481@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: jeffrey.craig@tekelec.com, John Li <jli@iname.com>,
	sigtran@ietf.org
References: <OF8544DAD4.2B59ADE8-ON85256CCA.00634D4A@nc.tekelec.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF8544DAD4.2B59ADE8-ON85256CCA.00634D4A@nc.tekelec.com>; from jeffrey.craig@tekelec.com on Tue, Feb 11, 2003 at 01:24:26PM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

jeffrey.craig,

jeffrey.craig@tekelec.com wrote:                                             (Tue, 11 Feb 2003 13:24:26)
> 
> Hello John,
> 
> Since no one else is responding, I'll answer.
> 
> My responses are embedded below.
> 
> Regards,
> 
> Jeff
> 
> On 02/11/2003 10:09:47 AM sigtran-admin wrote:
> 
> >All,
> >I sent the following comments and questions to the group a while ago, and
> have
> >not got responses yet. here goes again.
> >thanks.
> >John
> >
> >
> >1.   In section 1.5: the following statement is somewhat confusing.
> >"The SG is an IPSP equipped with both traditional SS7 and IP
> >network connections. In effect, the Signaling Gateway acts as a
> >Signal Transfer Point (STP)"
> >
> >Perhaps we can say "The SG interconnects SS7 and IP networks by
> >converting signaling traffic" or just refer to the definition in
> >section 1.2
> 
> I am not sure what defect this removes, and so I have no comment on this.
> 
> >
> >
> >2.   Section 2.1.5 Message length
> >Should we specify the valid value range of the length?
> >
> 
> No. Min/Max message length is dependent on message type,
> and in the case of user data is governed by MTP3.

Agree.  No link restriction on message length.

> 
> >3.   Section 2.3.1 User Data
> >
> >statement "M2PA does not add padding to the MTP3 message.".
> >I think we should make it a requirement: "M2PA should not add padding
> >to the MTP3 message."
> >
> 
> Sounds good to me.

M2PA MUST not add padding to the MTP3 message.  (The receiver cannot
detect it, so it needs to be a MUST).

> 
> >
> >4.   In section 4.1.3 Link alignment
> >
> >perhaps we can specify the recommended practice for choosing
> >emergency proving vs. normal proving during alignment procedure.
> >
> 
> This is governed by MTP3 procedures and is outside the scope
> of the M2PA draft.

Agree.  MTP3 decides this.

> 
> >
> >5.   Section 4.1.8:
> >"If a Link Status Alignment message with an unsupported
> >received by the newer version, the receiving end's M2PA
> >SHALL NOT complete the alignment procedure."
> >
> >can we clearly define the behavior of not completing the alignment
> >procedure?
> 
> I believe that you are suggesting that the text be reworded to state
> the following:
> 
> "If a Link Status Alignment message with an unsupported version
> is received by the newer version, the receiving end's M2PA
> SHOULD fail the link."
> 
> I think this was already understood, but the change improves clarity.

The link has not align so it cannot be "failed."  I agree with the current
wording: "SHALL NOT complete the alignment procedure", because the link
is already in the OOS state.

> 
> >
> >6.   section 4.2.1 sending and receiving messages
> >When M2PA receives a User Data message with Backward Sequence Number
> >equal to 'n', it may remove all messages with Forward Sequence Number
> ><= n from its queue.
> >
> >This is incorrect. Consider at the time SN is cycled to 0 from 127
> >(assume 128 buffers), there are 127, 126, etc are waiting for
> >acknowledgement. We should replace it with "The message n and all
> >outstanding (not yet acknowledged) messages sent before the message n
> >may be removed from its queue.
> 
> I agree with the change.

I think it is understood that the <= is modulo arithmetic.

--brian

> 
> >
> >7.   section 4.2.1
> >If M2PA receives an empty User Data message, it SHALL NOT send an
> >acknowledgement of that message.
> >
> >What is empty User Data message? FISUs?
> >
> 
> It is a M2PA user data message containing only the common header.
> An 'empty'  User Data message is an M2PA receipt acknowledgement.
> The statement in the draft is saying don't ACK an ACK.
> 
> >--
> >__________________________________________________________
> >Sign-up for your own FREE Personalized E-mail at Mail.com
> >http://www.mail.com/?sr=signup
> >
> >_______________________________________________
> >Sigtran mailing list
> >Sigtran@ietf.org
> >https://www1.ietf.org/mailman/listinfo/sigtran
> 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 11 21:39:18 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA25795
	for <sigtran-archive@odin.ietf.org>; Tue, 11 Feb 2003 21:39:18 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1C2mRs32146
	for sigtran-archive@odin.ietf.org; Tue, 11 Feb 2003 21:48:27 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1C2mFp32138;
	Tue, 11 Feb 2003 21:48:15 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1C2lXp32105
	for <sigtran@optimus.ietf.org>; Tue, 11 Feb 2003 21:47:33 -0500
Received: from spf1.us.outblaze.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id VAA25713
	for <sigtran@ietf.org>; Tue, 11 Feb 2003 21:37:52 -0500 (EST)
Received: (qmail 1689 invoked from network); 12 Feb 2003 02:41:08 -0000
Received: from unknown (205.158.62.68)
  by spf1.us.outblaze.com with QMQP; 12 Feb 2003 02:41:08 -0000
Received: (qmail 14490 invoked from network); 12 Feb 2003 02:41:08 -0000
Received: from unknown (HELO ws1-9.us4.outblaze.com) (205.158.62.37)
  by 205-158-62-153.outblaze.com with SMTP; 12 Feb 2003 02:41:08 -0000
Received: (qmail 26376 invoked by uid 1001); 12 Feb 2003 02:41:07 -0000
Message-ID: <20030212024107.26375.qmail@mail.com>
Content-Type: text/plain; charset="iso-8859-1"
Content-Disposition: inline
Content-Transfer-Encoding: 7bit
MIME-Version: 1.0
X-Mailer: MIME-tools 5.41 (Entity 5.404)
Received: from [216.118.69.91] by ws1-9.us4.outblaze.com with http for
    jli@iname.com; Tue, 11 Feb 2003 21:41:07 -0500
From: "John Li" <jli@iname.com>
To: jeffrey.craig@tekelec.com
Cc: sigtran@ietf.org
Date: Tue, 11 Feb 2003 21:41:07 -0500
Subject: Re: [Sigtran] M2PA draft comments and questions.
X-Originating-Ip: 216.118.69.91
X-Originating-Server: ws1-9.us4.outblaze.com
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit


Hi, Jeff, 
Thanks for the responses. Further comments are inline. 
Thanks, 
John 

-----Original Message----- 
From: jeffrey.craig@tekelec.com [mailto:jeffrey.craig@tekelec.com] 
Sent: Tuesday, February 11, 2003 1:24 PM 
To: John Li 
Cc: sigtran@ietf.org 
Subject: Re: [Sigtran] M2PA draft comments and questions. 



Hello John, 

Since no one else is responding, I'll answer. 

My responses are embedded below. 

Regards, 

Jeff 

On 02/11/2003 10:09:47 AM sigtran-admin wrote: 

>All, 
>I sent the following comments and questions to the group a while ago, and 
have 
>not got responses yet. here goes again. 
>thanks. 
>John 
> 
> 
>1.   In section 1.5: the following statement is somewhat confusing. 
>"The SG is an IPSP equipped with both traditional SS7 and IP 
>network connections. In effect, the Signaling Gateway acts as a 
>Signal Transfer Point (STP)" 
> 
>Perhaps we can say "The SG interconnects SS7 and IP networks by 
>converting signaling traffic" or just refer to the definition in 
>section 1.2 

I am not sure what defect this removes, and so I have no comment on this. 

> 
> 
>2.   Section 2.1.5 Message length 
>Should we specify the valid value range of the length? 
> 

No. Min/Max message length is dependent on message type, 
and in the case of user data is governed by MTP3. 

[John] the max payload MTP2 can carry is 272 octets. So by leaving that unspecified, it would be fine if M2PA used in a pure IP environment, but would have issues when used in a SG, who will be doing the segmentation and reassembling when receiving oversize(>272) messages from peer M2PA before it can be passed to the ss7 network?  

>3.   Section 2.3.1 User Data 
> 
>statement "M2PA does not add padding to the MTP3 message.". 
>I think we should make it a requirement: "M2PA should not add padding 
>to the MTP3 message." 
> 

Sounds good to me. 

> 
>4.   In section 4.1.3 Link alignment 
> 
>perhaps we can specify the recommended practice for choosing 
>emergency proving vs. normal proving during alignment procedure. 
> 

This is governed by MTP3 procedures and is outside the scope 
of the M2PA draft. 

[John] I thought about the scope issue as well. But since the draft covers MTP3 changeover in 2.4.6, by the same token, it would be very useful to add a paragraph to explain the two different alignment procedures in section 2.4: Procedures to Support the MTP3/MTP2 Interface.

> 
>5.   Section 4.1.8: 
>"If a Link Status Alignment message with an unsupported 
>received by the newer version, the receiving end's M2PA 
>SHALL NOT complete the alignment procedure." 
> 
>can we clearly define the behavior of not completing the alignment 
>procedure? 

I believe that you are suggesting that the text be reworded to state 
the following: 

"If a Link Status Alignment message with an unsupported version 
is received by the newer version, the receiving end's M2PA 
SHOULD fail the link." 

I think this was already understood, but the change improves clarity. 

[John] Further, I am thinking in this case, what is the procedure to prevent further alignment when the M2PA knows the incompatible versions: for example, should M2PA notify its upper user about the incompatible so that the upper layer will not initiate alignment again. what is the status to the M2PA user to indicate it should not try to re-align if so, etc.

> 
>6.   section 4.2.1 sending and receiving messages 
>When M2PA receives a User Data message with Backward Sequence Number 
>equal to 'n', it may remove all messages with Forward Sequence Number 
><= n from its queue. 
> 
>This is incorrect. Consider at the time SN is cycled to 0 from 127 
>(assume 128 buffers), there are 127, 126, etc are waiting for 
>acknowledgement. We should replace it with "The message n and all 
>outstanding (not yet acknowledged) messages sent before the message n 
>may be removed from its queue. 

I agree with the change. 

> 
>7.   section 4.2.1 
>If M2PA receives an empty User Data message, it SHALL NOT send an 
>acknowledgement of that message. 
> 
>What is empty User Data message? FISUs? 
> 

It is a M2PA user data message containing only the common header. 
An 'empty'  User Data message is an M2PA receipt acknowledgement. 
The statement in the draft is saying don't ACK an ACK. 

>-- 
>__________________________________________________________ 
>Sign-up for your own FREE Personalized E-mail at Mail.com 
>http://www.mail.com/?sr=signup 
> 
>_______________________________________________ 
>Sigtran mailing list 
>Sigtran@ietf.org 
>https://www1.ietf.org/mailman/listinfo/sigtran 



_______________________________________________ 
Sigtran mailing list 
Sigtran@ietf.org 
https://www1.ietf.org/mailman/listinfo/sigtran 

-- 
__________________________________________________________
Sign-up for your own FREE Personalized E-mail at Mail.com
http://www.mail.com/?sr=signup

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 11 22:31:46 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA26904
	for <sigtran-archive@odin.ietf.org>; Tue, 11 Feb 2003 22:31:46 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1C3evb02585
	for sigtran-archive@odin.ietf.org; Tue, 11 Feb 2003 22:40:57 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1C3egp02571;
	Tue, 11 Feb 2003 22:40:43 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1C3dCp02503
	for <sigtran@optimus.ietf.org>; Tue, 11 Feb 2003 22:39:12 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA26665
	for <sigtran@ietf.org>; Tue, 11 Feb 2003 22:29:29 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:F3Z0Zce6WZ0Hf06kZEPefftPKqWKC3PN@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1C3XCD08817;
	Tue, 11 Feb 2003 20:33:12 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1C3XBr27270;
	Tue, 11 Feb 2003 20:33:11 -0700
Date: Tue, 11 Feb 2003 20:33:11 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: John Li <jli@iname.com>
Cc: jeffrey.craig@tekelec.com, sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft comments and questions.
Message-ID: <20030211203311.A26317@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: John Li <jli@iname.com>, jeffrey.craig@tekelec.com,
	sigtran@ietf.org
References: <20030212024107.26375.qmail@mail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20030212024107.26375.qmail@mail.com>; from jli@iname.com on Tue, Feb 11, 2003 at 09:41:07PM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

John,

John Li wrote:                                                                        (Tue, 11 Feb 2003 21:41:07)
> 
> Hi, Jeff, 
> Thanks for the responses. Further comments are inline. 
> Thanks, 
> John 
> 
> -----Original Message----- 
> From: jeffrey.craig@tekelec.com [mailto:jeffrey.craig@tekelec.com] 
> Sent: Tuesday, February 11, 2003 1:24 PM 
> To: John Li 
> Cc: sigtran@ietf.org 
> Subject: Re: [Sigtran] M2PA draft comments and questions. 
> 
> 
> 
> Hello John, 
> 
> Since no one else is responding, I'll answer. 
> 
> My responses are embedded below. 
> 
> Regards, 
> 
> Jeff 
> 
> On 02/11/2003 10:09:47 AM sigtran-admin wrote: 
> 
> >All, 
> >I sent the following comments and questions to the group a while ago, and 
> have 
> >not got responses yet. here goes again. 
> >thanks. 
> >John 
> > 
> > 
> >1.   In section 1.5: the following statement is somewhat confusing. 
> >"The SG is an IPSP equipped with both traditional SS7 and IP 
> >network connections. In effect, the Signaling Gateway acts as a 
> >Signal Transfer Point (STP)" 
> > 
> >Perhaps we can say "The SG interconnects SS7 and IP networks by 
> >converting signaling traffic" or just refer to the definition in 
> >section 1.2 
> 
> I am not sure what defect this removes, and so I have no comment on this. 
> 
> > 
> > 
> >2.   Section 2.1.5 Message length 
> >Should we specify the valid value range of the length? 
> > 
> 
> No. Min/Max message length is dependent on message type, 
> and in the case of user data is governed by MTP3. 
> 
> [John] the max payload MTP2 can carry is 272 octets. So by leaving that
> unspecified, it would be fine if M2PA used in a pure IP environment, but
> would have issues when used in a SG, who will be doing the segmentation and
> reassembling when receiving oversize(>272) messages from peer M2PA before it
> can be passed to the ss7 network?  

MTP3b does not limit to 272 octets.  Neither does Q.703 Annex A.

This is a network restriction understood by MTP3.  M2PA imposes no
restriction on message size.

> 
> >3.   Section 2.3.1 User Data 
> > 
> >statement "M2PA does not add padding to the MTP3 message.". 
> >I think we should make it a requirement: "M2PA should not add padding 
> >to the MTP3 message." 
> > 
> 
> Sounds good to me. 

It is necessary to say MUST rather than SHOULD.

> 
> > 
> >4.   In section 4.1.3 Link alignment 
> > 
> >perhaps we can specify the recommended practice for choosing 
> >emergency proving vs. normal proving during alignment procedure. 
> > 
> 
> This is governed by MTP3 procedures and is outside the scope 
> of the M2PA draft. 
> 
> [John] I thought about the scope issue as well. But since the draft covers
> MTP3 changeover in 2.4.6, by the same token, it would be very useful to add
> a paragraph to explain the two different alignment procedures in section
> 2.4: Procedures to Support the MTP3/MTP2 Interface.


Changeover is discussed because the chnageover message contains
the MTP2 sequence number.  XCO/XCA need to be used due to M2PA's
larger sequence number.

Emergency vs. Normal proving is adequately addressed by MTP3 which
is cited as a reference.

> 
> > 
> >5.   Section 4.1.8: 
> >"If a Link Status Alignment message with an unsupported 
> >received by the newer version, the receiving end's M2PA 
> >SHALL NOT complete the alignment procedure." 
> > 
> >can we clearly define the behavior of not completing the alignment 
> >procedure? 
> 
> I believe that you are suggesting that the text be reworded to state 
> the following: 
> 
> "If a Link Status Alignment message with an unsupported version 
> is received by the newer version, the receiving end's M2PA 
> SHOULD fail the link." 
> 
> I think this was already understood, but the change improves clarity. 
> 
> [John] Further, I am thinking in this case, what is the procedure to prevent
> further alignment when the M2PA knows the incompatible versions: for
> example, should M2PA notify its upper user about the incompatible so that
> the upper layer will not initiate alignment again. what is the status to the
> M2PA user to indicate it should not try to re-align if so, etc.

Management procedures are implementation dependent.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 03:10:07 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA25658
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 03:10:07 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1C8JPh27782
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 03:19:25 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1C8JCp27775;
	Wed, 12 Feb 2003 03:19:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1C8Htp27695
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 03:17:55 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA25608
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 03:08:07 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1C8BmJT004498;
	Wed, 12 Feb 2003 02:11:49 -0600 (CST)
Received: from axes-mach01.usa.alcatel.com by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id MAA07428; Wed, 12 Feb 2003 12:40:54 -0500 (GMT)
Message-ID: <3E4A01C7.CF2892B@axes-mach01.usa.alcatel.com>
Date: Wed, 12 Feb 2003 13:42:16 +0530
From: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: brian@openss7.org
CC: sigtran@ietf.org
Content-Type: text/plain; charset=x-user-defined
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] M3UA Notify message
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Brian,

     Can you tell me when an SGP should send Notify(AS-PENDING) and
Notify(Insufficient ASP Resources Active in AS)?. I felt RFC and IG of
M3UA is not clear on these two.

--Srini

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 04:39:18 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA27388
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 04:39:18 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1C9maZ00883
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 04:48:36 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1C9mOp00850;
	Wed, 12 Feb 2003 04:48:24 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1C9kBp00768
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 04:46:11 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA27326
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 04:36:21 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:dv5HP4baiGmpQsH4GMgm9KXpmqKsf5M3@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1C9e3D14020;
	Wed, 12 Feb 2003 02:40:03 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1C9e3f08823;
	Wed, 12 Feb 2003 02:40:03 -0700
Date: Wed, 12 Feb 2003 02:40:02 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
Cc: sigtran@ietf.org
Message-ID: <20030212024002.A6927@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>,
	sigtran@ietf.org
References: <3E4A01C7.CF2892B@axes-mach01.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E4A01C7.CF2892B@axes-mach01.usa.alcatel.com>; from ksrini@axes-mach01.usa.alcatel.com on Wed, Feb 12, 2003 at 01:42:16PM +0530
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Subject: [Sigtran] Re: M3UA Notify message
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Srini,

ksrinivas wrote:                                                                 (Wed, 12 Feb 2003 13:42:16)
> Brian,
> 
>      Can you tell me when an SGP should send Notify(AS-PENDING) and
> Notify(Insufficient ASP Resources Active in AS)?. I felt RFC and IG of
> M3UA is not clear on these two.
> 
> --Srini

It think that it is pretty straight-forward.

NTFY(Insufficient ASPs) is sent for an AS by an SGP when the number of
ASPs in the ASP-ACTIVE state for the AS falls below "n", but not to
zero, (as a result of ASP failures or deactivations); where "n" is the
minimum number of active ASPs defined for the AS.  So this notification
is only sent if "n" is greater than 1.

NTFY(AS-PENDING) is sent for an AS by an SGP when the number of ASPs in
the ASP-ACTIVE state for the AS falls to zero (as a result of ASP
failures or deactivations).

Although there is the notification to indicate when there is insufficent
ASPs available for an AS, there is no notification that there are
sufficient ASPs for an AS.  This causes problems for ASP wishing to use
the first notification as in indication to go active.  There should be a
NTFY("Sufficient ASPs active") to notify that the opposite of
insufficient ASPs active is now true.

I also send these two notifications to any ASP that transitions from the
ASP-DOWN state to the ASP-INACTIVE state for the AS when the numer of
ASPs is below "n" or is zero and the T(pend) timer has not yet expired.
I also send other state notifications for the AS in this case as well
(i.e, NTFY(AS-ACTIVE), NTFY(AS-INACTIVE)).

As there is no procedure for a newly inactive ASP to audit the state of
the AS, this helps ASPs which are just coming up for an AS to
immediately determine the state of the AS and their next best course of
action for the AS.

In addition to notifications resulting from AS state changes, I think we
should say in the IG:

   The SGP SHOULD send a NTFY message indicating the state of the AS to
   all ASPs newly becoming ASP-INACTIVE from ASP-DOWN for each AS.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 05:38:24 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA28419
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 05:38:24 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CAli004805
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 05:47:44 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CAlNp04791;
	Wed, 12 Feb 2003 05:47:23 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CAkmp04761
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 05:46:48 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA28404
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 05:36:57 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1CAeZqe009376;
	Wed, 12 Feb 2003 04:40:37 -0600 (CST)
Received: from axes-mach01.usa.alcatel.com by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id PAA10963; Wed, 12 Feb 2003 15:09:25 -0500 (GMT)
Message-ID: <3E4A243F.D9B240D3@axes-mach01.usa.alcatel.com>
Date: Wed, 12 Feb 2003 16:10:48 +0530
From: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: bidulock@openss7.org
CC: sigtran@ietf.org
References: <3E4A01C7.CF2892B@axes-mach01.usa.alcatel.com> <20030212024002.A6927@openss7.org>
Content-Type: text/plain; charset=x-user-defined
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] Re: M3UA Notify message
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Brian,

"Brian F. G. Bidulock" wrote:

> Srini,
>
> ksrinivas wrote:                                                                 (Wed, 12 Feb 2003 13:42:16)
> > Brian,
> >
> >      Can you tell me when an SGP should send Notify(AS-PENDING) and
> > Notify(Insufficient ASP Resources Active in AS)?. I felt RFC and IG of
> > M3UA is not clear on these two.
> >
> > --Srini
>
> It think that it is pretty straight-forward.
>
> NTFY(Insufficient ASPs) is sent for an AS by an SGP when the number of
> ASPs in the ASP-ACTIVE state for the AS falls below "n", but not to
> zero, (as a result of ASP failures or deactivations); where "n" is the
> minimum number of active ASPs defined for the AS.  So this notification
> is only sent if "n" is greater than 1.
>
> NTFY(AS-PENDING) is sent for an AS by an SGP when the number of ASPs in
> the ASP-ACTIVE state for the AS falls to zero (as a result of ASP
> failures or deactivations).
>

I understood the same way initially. But the section 3.15.2 of M3UA IG-02 has confused me. The following is the
text from this section.

"AC2PN: one ASP in ASP-ACTIVE state moves to ASP-INACTIVE or ASP-DOWN states, causing the number of ASPs in
ASP-ACTIVE drop below n."

So the AS moves to AS-PENDING state when ASPs in ASP-ACTIVE drop below 'n' (but not necessarily zero). Do we not
have to send Notify(AS-PENDING) when AS has moved to AS-PENDING state?.

>
> Although there is the notification to indicate when there is insufficent
> ASPs available for an AS, there is no notification that there are
> sufficient ASPs for an AS.  This causes problems for ASP wishing to use
> the first notification as in indication to go active.  There should be a
> NTFY("Sufficient ASPs active") to notify that the opposite of
> insufficient ASPs active is now true.
>
> I also send these two notifications to any ASP that transitions from the
> ASP-DOWN state to the ASP-INACTIVE state for the AS when the numer of
> ASPs is below "n" or is zero and the T(pend) timer has not yet expired.
> I also send other state notifications for the AS in this case as well
> (i.e, NTFY(AS-ACTIVE), NTFY(AS-INACTIVE)).
>
> As there is no procedure for a newly inactive ASP to audit the state of
> the AS, this helps ASPs which are just coming up for an AS to
> immediately determine the state of the AS and their next best course of
> action for the AS.
>
> In addition to notifications resulting from AS state changes, I think we
> should say in the IG:
>
>    The SGP SHOULD send a NTFY message indicating the state of the AS to
>    all ASPs newly becoming ASP-INACTIVE from ASP-DOWN for each AS.
>
> --brian
>
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/

--Srini


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 06:25:30 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA29249
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 06:25:30 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CBYp106786
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 06:34:51 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CBYYp06769;
	Wed, 12 Feb 2003 06:34:36 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CBX1p06713
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 06:33:01 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA29213
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 06:23:09 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:XKhAacAK6bF5C4kXE9ZEk0PCz0GBTtHH@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1CBQqD16021;
	Wed, 12 Feb 2003 04:26:52 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1CBQp023250;
	Wed, 12 Feb 2003 04:26:51 -0700
Date: Wed, 12 Feb 2003 04:26:51 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
Cc: sigtran@ietf.org
Message-ID: <20030212042651.A16495@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>,
	sigtran@ietf.org
References: <3E4A01C7.CF2892B@axes-mach01.usa.alcatel.com> <20030212024002.A6927@openss7.org> <3E4A243F.D9B240D3@axes-mach01.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E4A243F.D9B240D3@axes-mach01.usa.alcatel.com>; from ksrini@axes-mach01.usa.alcatel.com on Wed, Feb 12, 2003 at 04:10:48PM +0530
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Subject: [Sigtran] Re: M3UA Notify message
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

ksrinivas,

ksrinivas wrote:                                                                 (Wed, 12 Feb 2003 16:10:48)
> Brian,
> 
> "Brian F. G. Bidulock" wrote:
> 
> > Srini,
> >
> > ksrinivas wrote:                                                                 (Wed, 12 Feb 2003 13:42:16)
> > > Brian,
> > >
> > >      Can you tell me when an SGP should send Notify(AS-PENDING) and
> > > Notify(Insufficient ASP Resources Active in AS)?. I felt RFC and IG of
> > > M3UA is not clear on these two.
> > >
> > > --Srini
> >
> > It think that it is pretty straight-forward.
> >
> > NTFY(Insufficient ASPs) is sent for an AS by an SGP when the number of
> > ASPs in the ASP-ACTIVE state for the AS falls below "n", but not to
> > zero, (as a result of ASP failures or deactivations); where "n" is the
> > minimum number of active ASPs defined for the AS.  So this notification
> > is only sent if "n" is greater than 1.
> >
> > NTFY(AS-PENDING) is sent for an AS by an SGP when the number of ASPs in
> > the ASP-ACTIVE state for the AS falls to zero (as a result of ASP
> > failures or deactivations).
> >
> 
> I understood the same way initially. But the section 3.15.2 of M3UA IG-02
> has confused me. The following is the text from this section.
> 
> "AC2PN: one ASP in ASP-ACTIVE state moves to ASP-INACTIVE or ASP-DOWN
> states, causing the number of ASPs in ASP-ACTIVE drop below n."
> 
> So the AS moves to AS-PENDING state when ASPs in ASP-ACTIVE drop below 'n'
> (but not necessarily zero). Do we not have to send Notify(AS-PENDING) when
> AS has moved to AS-PENDING state?.

I see.

I largely disagree with the IG section 3.15.2.  Consider that the new
description of the AS-PENDING state is:

    AS-PENDING: An active ASP has transitioned from ASP-ACTIVE to ASP-
    INACTIVE or ASP-DOWN and it was the number n active ASP in the AS.  A
    recovery timer T(r) SHOULD be started and all incoming signalling messages
    SHOULD be queued by the SGP. If an ASP becomes ASP-ACTIVE before T(r)
    expires, the AS is moved to the AS-ACTIVE state and all the queued
    messages will be sent to the ASP.

I think that it would be undesireable for an SGP to queue "incoming"
signalling messages when the number of active ASPs is less than "n" for a
number of reasons stated on the "n+k redundancy" thread.

For example, at quiet periods, when there is not much traffic, is there any
reason to stop service because one ASP happens to be inactive?  (Remember that
this only counts for load-sharing and broadcast modes.)

I believe that the number of active ASPs less than "n" corresponds to the
AS-ACTIVE state (as in the RFC).

I disagree with the modified state machine as arranged between Tolga and
Javier with little input from others.

As far as I'm concerned the AC2PN transition is caused by the last active ASP
moving to the ASP-INACTIVE or ASP-DOWN state or only if the SGP follows the
"MAY" (and withholds traffic for m < n).

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 06:59:10 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA29872
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 06:59:10 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CC8VQ09244
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 07:08:31 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CC8Dp09192;
	Wed, 12 Feb 2003 07:08:13 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CC5Qp08423
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 07:05:26 -0500
Received: from stewart.chicago.il.us (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA29823;
	Wed, 12 Feb 2003 06:55:32 -0500 (EST)
Received: from stewart.chicago.il.us (stewart.chicago.il.us [127.0.0.1])
	by stewart.chicago.il.us (8.12.6/8.12.3) with ESMTP id h1CBwtu6003525;
	Wed, 12 Feb 2003 05:58:56 -0600 (CST)
	(envelope-from randall@stewart.chicago.il.us)
Message-ID: <3E4A36FF.8050205@stewart.chicago.il.us>
Date: Wed, 12 Feb 2003 05:58:55 -0600
From: Randall Stewart <randall@stewart.chicago.il.us>
User-Agent: Mozilla/5.0 (X11; U; FreeBSD i386; en-US; rv:1.1) Gecko/20021005
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: bidulock@openss7.org, Michael Tuexen <Michael.Tuexen@lurchi.franken.de>
CC: Caitlin Bestler <cait@asomi.com>, tsvwg@ietf.org,
        Tuexen Michael <michael.tuexen@siemens.com>, sigtran@ietf.org
References: <A2AA39E1-3DFB-11D7-90B4-0030654C1AB6@lurchi.franken.de> <3E49899A.6010809@stewart.chicago.il.us> <20030211191543.A14481@openss7.org>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] Re: [Tsvwg] A question of Un-Ordered SCTP
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Brian/Michael:

I do like the idea of keeping things the same (aka same
methodology)... this is good.

It is also neat to be able to tell the difference between
an implementation that supports this feature and one that
does not.  But there are positives and negatives, I think,
to each idea...

A) For using a TLV, (as Michael suggests)
    we really need to put this into a seperate
    draft.. when I do think one could easily fit
    this into the IG as Brian as suggested.

B) montonically increasing numbers would eventually
    wrap when the SSN hit FFFF back to 0... thus
    the application would need to have some sort
    of state to track if the peer supported the
    feature.

It seems to me it is only the first message that
arrives that benefits from the start at 0. If the
application assumes a "not supported", then by
starting the SSN at 1 the app can flip the flag
from "not supported" to "supported" on the arrival
of the first message.

If the application instead set its flag to "supported"
the first message would not tell this information and
instead it would only be upon the second arriving message
that the app could make the flag determination... causing
either a delay in delivery of the first message (yuck) or
just toss it up and wait.

More importantly, I am wondering, if the implementation does
NOT support this, what will the app do? Change its
message structure? If so it will need to have some clues
(as the initiator) that its implementation does not
support the feature .. possibly a HAVE_SCTP_UNORDERED_SSN
in sctp.h... but the problem is you would not know about
the other guy until its first unordered message arrives.
So if you have to send the first message what do you do?

if HAVE_SCTP_UNORDERED_SSN is 0.. no problem.. you
send a message that has the imbedded sequence number.. but
what do you send when HAVE_SCTP_UNRODERED_SSN is set to 1?

If you send the imbedded sequence in the message anyway whats
the point of this little trick? The second message (which you
may have to send before knowing receiving a unordered
msg from the peer) would still have this, and maybe the third
etc...

Hmm.. the more I think on this the more it becomes clear that
you must know ahead of time i.e. when the association comes
up to "do the right thing".... which goes back to Michaels
idea of a TLV.

I have been wondering for quite some time now if we don't
just need a TLV in the INIT/INIT-ACK that indicate features
in general... something like a bit mask or byte array indicating
each feature...

addip
delip
prsctp
setprimary
unordered-assigns-sequences

etc..

Maybe this is a better way to go.. and we could write a short
draft dictating this bit/byte TLV... code as a 0x80 class.. that
way old implementations will throw it away... (we really don't
care if they don't understand it)...

I have been avoiding this.. but it might be the best approach..

Then again, I guess it could also be put in addip or prsctp since
it would effect them as well :> Then one could put the
other change in the IG and thus have a very easy change that
is in the IG and a minor update to either addip or prsctp...


What do you all think?

R

Brian F. G. Bidulock wrote:
> Randall,
> 
> I agree to having SCTP set the sequence number and, as you said in
> your first note, SCTP passing the sequence number to the ULP on
> the receiving side is already done for ordered traffic.  Some even
> pass the zero now.
> 
> I suppose that SCTP would set this sequence number when it is given
> the data for delivery at the sending ULP and the sequence would represent
> the order that the data was presented to SCTP by the ULP at the sending
> side.
> 
> I think it would be good to see the requirement that the sequence
> number begin at 1 and be monotonically increasing in the saem fashion
> as ordered streams.
> 
> That way RDDP and other applications could perform their own reordering
> at the receiver.  I think it would be a good thing to have SCTP set
> this value.
> 
> In fact, I can see another application for this feature.  M2PA
> (SIGTRAN protocol) needs in some circumstances to send data unordered
> with the hope of expedited delivery (does not want control information
> stuck behind large amounts of data in an ordered stream), however, the
> order of processing of the expedited control information is important.
> 
> This facility would permit M2PA to send such information unordered
> to expedite it and the receiver could perform the reordering itself.
> Up until now, the only solution was to place a sequence number in with
> the control information.
> 
> Could we perhaps add a section to the SCTP IG.  "Problem statement:
> the sender wishes to send unordered data but the receive needs to know
> the original order of sending..." or some such.
> 
> It should be quite simple to add the feature to existing SCTP
> implementations.
> 
> --brian
> 
> 
> On Tue, 11 Feb 2003, Randall Stewart wrote:
> 
> 
>>Michael:
>>
>>This seems like a valid method.. it gets around Brian's objection
>>(which was a good point) and yet still provides the application with
>>a hint about what the assigned sequence order was on the unordered
>>messages...
>>
>>R
>>
>>Michael Tuexen wrote:
>>
>>>Caitlin,
>>>
>>>thank you very much for the clarification.
>>>
>>>I do not like the idea to pass the SSN transparently from the ULP to the 
>>>ULP.
>>>But having two sequence spaces is OK for me. This means that the SSN at the
>>>sending side is controlled by the SCTP and given to the ULP at the 
>>>receiving side.
>>>
>>>This would be a clean solution. So I voting for have two separate number 
>>>spaces
>>>for user data. Since the one for ordered delivery exists already in RFC 
>>>2960, this
>>>only means that a new one is introduced for the unordered delivery user 
>>>data.
>>>
>>>Best regards
>>>Michael
>>>
>>>On Tuesday, Feb 11, 2003, at 17:37 Europe/Berlin, Caitlin Bestler wrote:
>>>
>>>
>>>>On 2/11/03, Tuexen Michael wrote:
>>>>
>>>>
>>>>>Dear all,
>>>>>
>>>>>does this mean that you want two sequence spaces per
>>>>>stream: one for the 'ordered' user data and one for the
>>>>>unordered user data?
>>>>>
>>>>
>>>>The RDDP/SCTP adaptation only uses unordered delivery, so
>>>>that question did not really arise.
>>>>
>>>>But if you were to generalize the problem, I believe the
>>>>answer would be yes.
>>>>
>>>>The SSN for ordered delivery would be unchanged. The
>>>>addition would be providing "advisory ordering" for
>>>>unordered delivery. The general characteristic is that
>>>>the receiving ULP wants the data in whatever order it
>>>>arrives, but wants to know what the original order was.
>>>>
>>>>In addition to RDDP, RTP over UDP is another example of
>>>>an application that wants to be aware of the original
>>>>ordering, even while accepting the data out-of-order.
>>>>
>>>>Caitlin Bestler
>>>>http://asomi.com/CaitlinBestler/
>>>>_______________________________________________
>>>>tsvwg mailing list
>>>>tsvwg@ietf.org
>>>>https://www1.ietf.org/mailman/listinfo/tsvwg
>>>>
>>>
>>>_______________________________________________
>>>tsvwg mailing list
>>>tsvwg@ietf.org
>>>https://www1.ietf.org/mailman/listinfo/tsvwg
>>>
>>>
>>
>>
>>-- 
>>Randall R. Stewart
>>randall@stewart.chicago.il.us 815-342-5222 (cell phone)
>>
>>_______________________________________________
>>tsvwg mailing list
>>tsvwg@ietf.org
>>https://www1.ietf.org/mailman/listinfo/tsvwg
> 
> 


-- 
Randall R. Stewart
randall@stewart.chicago.il.us 815-342-5222 (cell phone)

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 07:15:12 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA00183
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 07:15:12 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CCOYT09954
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 07:24:34 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CCOHp09930;
	Wed, 12 Feb 2003 07:24:17 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CCNHp09871
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 07:23:17 -0500
Received: from gorilla.mchh.siemens.de (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA00144;
	Wed, 12 Feb 2003 07:13:21 -0500 (EST)
Received: from moody.mchh.siemens.de ([139.21.205.85])
	by gorilla.mchh.siemens.de (8.9.3/8.9.3) with ESMTP id NAA15542;
	Wed, 12 Feb 2003 13:17:02 +0100 (MET)
Received: from mchh248e.mchh.siemens.de (mchh248e.mchh.siemens.de [139.21.200.58])
	by moody.mchh.siemens.de (8.9.1/8.9.1) with ESMTP id NAA10951;
	Wed, 12 Feb 2003 13:17:03 +0100 (MET)
Received: by MCHH248E with Internet Mail Service (5.5.2653.19)
	id <D1N52V46>; Wed, 12 Feb 2003 13:16:13 +0100
Message-ID: <D73559B38D51D4119CB800508B9527B505AC15D4@mchh266e.mchh.siemens.de>
From: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>
To: "'Randall Stewart'" <randall@stewart.chicago.il.us>
Cc: tsvwg@ietf.org, sigtran@ietf.org
Subject: AW: [Sigtran] Re: [Tsvwg] A question of Un-Ordered SCTP
Date: Wed, 12 Feb 2003 13:15:42 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h1CCNHp09874
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Randy,

I definitely would appreciate such a possibility, but not based on a bit map.
Using only one bit per feature would narrow limits on the number of features that could be supported.

greetings

HannsJuergen

> -----Ursprüngliche Nachricht-----
> Von: Randall Stewart [mailto:randall@stewart.chicago.il.us]
> Gesendet: Mittwoch, 12. Februar 2003 12:59
> An: bidulock@openss7.org; Michael Tuexen
> Cc: Caitlin Bestler; tsvwg@ietf.org; Tuexen Michael; sigtran@ietf.org
> Betreff: [Sigtran] Re: [Tsvwg] A question of Un-Ordered SCTP
> 
> 
> Brian/Michael:
> 
> I do like the idea of keeping things the same (aka same
> methodology)... this is good.
> 
> It is also neat to be able to tell the difference between
> an implementation that supports this feature and one that
> does not.  But there are positives and negatives, I think,
> to each idea...
> 
> A) For using a TLV, (as Michael suggests)
>     we really need to put this into a seperate
>     draft.. when I do think one could easily fit
>     this into the IG as Brian as suggested.
> 
> B) montonically increasing numbers would eventually
>     wrap when the SSN hit FFFF back to 0... thus
>     the application would need to have some sort
>     of state to track if the peer supported the
>     feature.
> 
> It seems to me it is only the first message that
> arrives that benefits from the start at 0. If the
> application assumes a "not supported", then by
> starting the SSN at 1 the app can flip the flag
> from "not supported" to "supported" on the arrival
> of the first message.
> 
> If the application instead set its flag to "supported"
> the first message would not tell this information and
> instead it would only be upon the second arriving message
> that the app could make the flag determination... causing
> either a delay in delivery of the first message (yuck) or
> just toss it up and wait.
> 
> More importantly, I am wondering, if the implementation does
> NOT support this, what will the app do? Change its
> message structure? If so it will need to have some clues
> (as the initiator) that its implementation does not
> support the feature .. possibly a HAVE_SCTP_UNORDERED_SSN
> in sctp.h... but the problem is you would not know about
> the other guy until its first unordered message arrives.
> So if you have to send the first message what do you do?
> 
> if HAVE_SCTP_UNORDERED_SSN is 0.. no problem.. you
> send a message that has the imbedded sequence number.. but
> what do you send when HAVE_SCTP_UNRODERED_SSN is set to 1?
> 
> If you send the imbedded sequence in the message anyway whats
> the point of this little trick? The second message (which you
> may have to send before knowing receiving a unordered
> msg from the peer) would still have this, and maybe the third
> etc...
> 
> Hmm.. the more I think on this the more it becomes clear that
> you must know ahead of time i.e. when the association comes
> up to "do the right thing".... which goes back to Michaels
> idea of a TLV.
> 
> I have been wondering for quite some time now if we don't
> just need a TLV in the INIT/INIT-ACK that indicate features
> in general... something like a bit mask or byte array indicating
> each feature...
> 
> addip
> delip
> prsctp
> setprimary
> unordered-assigns-sequences
> 
> etc..
> 
> Maybe this is a better way to go.. and we could write a short
> draft dictating this bit/byte TLV... code as a 0x80 class.. that
> way old implementations will throw it away... (we really don't
> care if they don't understand it)...
> 
> I have been avoiding this.. but it might be the best approach..
> 
> Then again, I guess it could also be put in addip or prsctp since
> it would effect them as well :> Then one could put the
> other change in the IG and thus have a very easy change that
> is in the IG and a minor update to either addip or prsctp...
> 
> 
> What do you all think?
> 
> R
> 
> Brian F. G. Bidulock wrote:
> > Randall,
> > 
> > I agree to having SCTP set the sequence number and, as you said in
> > your first note, SCTP passing the sequence number to the ULP on
> > the receiving side is already done for ordered traffic.  Some even
> > pass the zero now.
> > 
> > I suppose that SCTP would set this sequence number when it is given
> > the data for delivery at the sending ULP and the sequence 
> would represent
> > the order that the data was presented to SCTP by the ULP at 
> the sending
> > side.
> > 
> > I think it would be good to see the requirement that the sequence
> > number begin at 1 and be monotonically increasing in the 
> saem fashion
> > as ordered streams.
> > 
> > That way RDDP and other applications could perform their 
> own reordering
> > at the receiver.  I think it would be a good thing to have SCTP set
> > this value.
> > 
> > In fact, I can see another application for this feature.  M2PA
> > (SIGTRAN protocol) needs in some circumstances to send data 
> unordered
> > with the hope of expedited delivery (does not want control 
> information
> > stuck behind large amounts of data in an ordered stream), 
> however, the
> > order of processing of the expedited control information is 
> important.
> > 
> > This facility would permit M2PA to send such information unordered
> > to expedite it and the receiver could perform the reordering itself.
> > Up until now, the only solution was to place a sequence 
> number in with
> > the control information.
> > 
> > Could we perhaps add a section to the SCTP IG.  "Problem statement:
> > the sender wishes to send unordered data but the receive 
> needs to know
> > the original order of sending..." or some such.
> > 
> > It should be quite simple to add the feature to existing SCTP
> > implementations.
> > 
> > --brian
> > 
> > 
> > On Tue, 11 Feb 2003, Randall Stewart wrote:
> > 
> > 
> >>Michael:
> >>
> >>This seems like a valid method.. it gets around Brian's objection
> >>(which was a good point) and yet still provides the application with
> >>a hint about what the assigned sequence order was on the unordered
> >>messages...
> >>
> >>R
> >>
> >>Michael Tuexen wrote:
> >>
> >>>Caitlin,
> >>>
> >>>thank you very much for the clarification.
> >>>
> >>>I do not like the idea to pass the SSN transparently from 
> the ULP to the 
> >>>ULP.
> >>>But having two sequence spaces is OK for me. This means 
> that the SSN at the
> >>>sending side is controlled by the SCTP and given to the ULP at the 
> >>>receiving side.
> >>>
> >>>This would be a clean solution. So I voting for have two 
> separate number 
> >>>spaces
> >>>for user data. Since the one for ordered delivery exists 
> already in RFC 
> >>>2960, this
> >>>only means that a new one is introduced for the unordered 
> delivery user 
> >>>data.
> >>>
> >>>Best regards
> >>>Michael
> >>>
> >>>On Tuesday, Feb 11, 2003, at 17:37 Europe/Berlin, Caitlin 
> Bestler wrote:
> >>>
> >>>
> >>>>On 2/11/03, Tuexen Michael wrote:
> >>>>
> >>>>
> >>>>>Dear all,
> >>>>>
> >>>>>does this mean that you want two sequence spaces per
> >>>>>stream: one for the 'ordered' user data and one for the
> >>>>>unordered user data?
> >>>>>
> >>>>
> >>>>The RDDP/SCTP adaptation only uses unordered delivery, so
> >>>>that question did not really arise.
> >>>>
> >>>>But if you were to generalize the problem, I believe the
> >>>>answer would be yes.
> >>>>
> >>>>The SSN for ordered delivery would be unchanged. The
> >>>>addition would be providing "advisory ordering" for
> >>>>unordered delivery. The general characteristic is that
> >>>>the receiving ULP wants the data in whatever order it
> >>>>arrives, but wants to know what the original order was.
> >>>>
> >>>>In addition to RDDP, RTP over UDP is another example of
> >>>>an application that wants to be aware of the original
> >>>>ordering, even while accepting the data out-of-order.
> >>>>
> >>>>Caitlin Bestler
> >>>>http://asomi.com/CaitlinBestler/
> >>>>_______________________________________________
> >>>>tsvwg mailing list
> >>>>tsvwg@ietf.org
> >>>>https://www1.ietf.org/mailman/listinfo/tsvwg
> >>>>
> >>>
> >>>_______________________________________________
> >>>tsvwg mailing list
> >>>tsvwg@ietf.org
> >>>https://www1.ietf.org/mailman/listinfo/tsvwg
> >>>
> >>>
> >>
> >>
> >>-- 
> >>Randall R. Stewart
> >>randall@stewart.chicago.il.us 815-342-5222 (cell phone)
> >>
> >>_______________________________________________
> >>tsvwg mailing list
> >>tsvwg@ietf.org
> >>https://www1.ietf.org/mailman/listinfo/tsvwg
> > 
> > 
> 
> 
> -- 
> Randall R. Stewart
> randall@stewart.chicago.il.us 815-342-5222 (cell phone)
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 07:20:02 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA00356
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 07:20:02 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CCTOF10268
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 07:29:24 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CCT9p10238;
	Wed, 12 Feb 2003 07:29:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CCSmp10185
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 07:28:48 -0500
Received: from stewart.chicago.il.us (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA00335;
	Wed, 12 Feb 2003 07:18:53 -0500 (EST)
Received: from stewart.chicago.il.us (stewart.chicago.il.us [127.0.0.1])
	by stewart.chicago.il.us (8.12.6/8.12.3) with ESMTP id h1CCMau6003573;
	Wed, 12 Feb 2003 06:22:37 -0600 (CST)
	(envelope-from randall@stewart.chicago.il.us)
Message-ID: <3E4A3C8C.6030805@stewart.chicago.il.us>
Date: Wed, 12 Feb 2003 06:22:36 -0600
From: Randall Stewart <randall@stewart.chicago.il.us>
User-Agent: Mozilla/5.0 (X11; U; FreeBSD i386; en-US; rv:1.1) Gecko/20021005
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>
CC: tsvwg@ietf.org, sigtran@ietf.org
Subject: Re: AW: [Sigtran] Re: [Tsvwg] A question of Un-Ordered SCTP
References: <D73559B38D51D4119CB800508B9527B505AC15D4@mchh266e.mchh.siemens.de>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Hanns:

Yes.. I knew you would feel that way and thats why I said
bit mask or byte array :>

I do like Michaels proposal..

R

Schwarzbauer Hanns Juergen wrote:
> Randy,
> 
> I definitely would appreciate such a possibility, but not based on a bit map.
> Using only one bit per feature would narrow limits on the number of features that could be supported.
> 
> greetings
> 
> HannsJuergen
> 
> 
>>-----Ursprüngliche Nachricht-----
>>Von: Randall Stewart [mailto:randall@stewart.chicago.il.us]
>>Gesendet: Mittwoch, 12. Februar 2003 12:59
>>An: bidulock@openss7.org; Michael Tuexen
>>Cc: Caitlin Bestler; tsvwg@ietf.org; Tuexen Michael; sigtran@ietf.org
>>Betreff: [Sigtran] Re: [Tsvwg] A question of Un-Ordered SCTP
>>
>>
>>Brian/Michael:
>>
>>I do like the idea of keeping things the same (aka same
>>methodology)... this is good.
>>
>>It is also neat to be able to tell the difference between
>>an implementation that supports this feature and one that
>>does not.  But there are positives and negatives, I think,
>>to each idea...
>>
>>A) For using a TLV, (as Michael suggests)
>>    we really need to put this into a seperate
>>    draft.. when I do think one could easily fit
>>    this into the IG as Brian as suggested.
>>
>>B) montonically increasing numbers would eventually
>>    wrap when the SSN hit FFFF back to 0... thus
>>    the application would need to have some sort
>>    of state to track if the peer supported the
>>    feature.
>>
>>It seems to me it is only the first message that
>>arrives that benefits from the start at 0. If the
>>application assumes a "not supported", then by
>>starting the SSN at 1 the app can flip the flag
>>from "not supported" to "supported" on the arrival
>>of the first message.
>>
>>If the application instead set its flag to "supported"
>>the first message would not tell this information and
>>instead it would only be upon the second arriving message
>>that the app could make the flag determination... causing
>>either a delay in delivery of the first message (yuck) or
>>just toss it up and wait.
>>
>>More importantly, I am wondering, if the implementation does
>>NOT support this, what will the app do? Change its
>>message structure? If so it will need to have some clues
>>(as the initiator) that its implementation does not
>>support the feature .. possibly a HAVE_SCTP_UNORDERED_SSN
>>in sctp.h... but the problem is you would not know about
>>the other guy until its first unordered message arrives.
>>So if you have to send the first message what do you do?
>>
>>if HAVE_SCTP_UNORDERED_SSN is 0.. no problem.. you
>>send a message that has the imbedded sequence number.. but
>>what do you send when HAVE_SCTP_UNRODERED_SSN is set to 1?
>>
>>If you send the imbedded sequence in the message anyway whats
>>the point of this little trick? The second message (which you
>>may have to send before knowing receiving a unordered
>>msg from the peer) would still have this, and maybe the third
>>etc...
>>
>>Hmm.. the more I think on this the more it becomes clear that
>>you must know ahead of time i.e. when the association comes
>>up to "do the right thing".... which goes back to Michaels
>>idea of a TLV.
>>
>>I have been wondering for quite some time now if we don't
>>just need a TLV in the INIT/INIT-ACK that indicate features
>>in general... something like a bit mask or byte array indicating
>>each feature...
>>
>>addip
>>delip
>>prsctp
>>setprimary
>>unordered-assigns-sequences
>>
>>etc..
>>
>>Maybe this is a better way to go.. and we could write a short
>>draft dictating this bit/byte TLV... code as a 0x80 class.. that
>>way old implementations will throw it away... (we really don't
>>care if they don't understand it)...
>>
>>I have been avoiding this.. but it might be the best approach..
>>
>>Then again, I guess it could also be put in addip or prsctp since
>>it would effect them as well :> Then one could put the
>>other change in the IG and thus have a very easy change that
>>is in the IG and a minor update to either addip or prsctp...
>>
>>
>>What do you all think?
>>
>>R
>>
>>Brian F. G. Bidulock wrote:
>>
>>>Randall,
>>>
>>>I agree to having SCTP set the sequence number and, as you said in
>>>your first note, SCTP passing the sequence number to the ULP on
>>>the receiving side is already done for ordered traffic.  Some even
>>>pass the zero now.
>>>
>>>I suppose that SCTP would set this sequence number when it is given
>>>the data for delivery at the sending ULP and the sequence 
>>
>>would represent
>>
>>>the order that the data was presented to SCTP by the ULP at 
>>
>>the sending
>>
>>>side.
>>>
>>>I think it would be good to see the requirement that the sequence
>>>number begin at 1 and be monotonically increasing in the 
>>
>>saem fashion
>>
>>>as ordered streams.
>>>
>>>That way RDDP and other applications could perform their 
>>
>>own reordering
>>
>>>at the receiver.  I think it would be a good thing to have SCTP set
>>>this value.
>>>
>>>In fact, I can see another application for this feature.  M2PA
>>>(SIGTRAN protocol) needs in some circumstances to send data 
>>
>>unordered
>>
>>>with the hope of expedited delivery (does not want control 
>>
>>information
>>
>>>stuck behind large amounts of data in an ordered stream), 
>>
>>however, the
>>
>>>order of processing of the expedited control information is 
>>
>>important.
>>
>>>This facility would permit M2PA to send such information unordered
>>>to expedite it and the receiver could perform the reordering itself.
>>>Up until now, the only solution was to place a sequence 
>>
>>number in with
>>
>>>the control information.
>>>
>>>Could we perhaps add a section to the SCTP IG.  "Problem statement:
>>>the sender wishes to send unordered data but the receive 
>>
>>needs to know
>>
>>>the original order of sending..." or some such.
>>>
>>>It should be quite simple to add the feature to existing SCTP
>>>implementations.
>>>
>>>--brian
>>>
>>>
>>>On Tue, 11 Feb 2003, Randall Stewart wrote:
>>>
>>>
>>>
>>>>Michael:
>>>>
>>>>This seems like a valid method.. it gets around Brian's objection
>>>>(which was a good point) and yet still provides the application with
>>>>a hint about what the assigned sequence order was on the unordered
>>>>messages...
>>>>
>>>>R
>>>>
>>>>Michael Tuexen wrote:
>>>>
>>>>
>>>>>Caitlin,
>>>>>
>>>>>thank you very much for the clarification.
>>>>>
>>>>>I do not like the idea to pass the SSN transparently from 
>>>>
>>the ULP to the 
>>
>>>>>ULP.
>>>>>But having two sequence spaces is OK for me. This means 
>>>>
>>that the SSN at the
>>
>>>>>sending side is controlled by the SCTP and given to the ULP at the 
>>>>>receiving side.
>>>>>
>>>>>This would be a clean solution. So I voting for have two 
>>>>
>>separate number 
>>
>>>>>spaces
>>>>>for user data. Since the one for ordered delivery exists 
>>>>
>>already in RFC 
>>
>>>>>2960, this
>>>>>only means that a new one is introduced for the unordered 
>>>>
>>delivery user 
>>
>>>>>data.
>>>>>
>>>>>Best regards
>>>>>Michael
>>>>>
>>>>>On Tuesday, Feb 11, 2003, at 17:37 Europe/Berlin, Caitlin 
>>>>
>>Bestler wrote:
>>
>>>>>
>>>>>>On 2/11/03, Tuexen Michael wrote:
>>>>>>
>>>>>>
>>>>>>
>>>>>>>Dear all,
>>>>>>>
>>>>>>>does this mean that you want two sequence spaces per
>>>>>>>stream: one for the 'ordered' user data and one for the
>>>>>>>unordered user data?
>>>>>>>
>>>>>>
>>>>>>The RDDP/SCTP adaptation only uses unordered delivery, so
>>>>>>that question did not really arise.
>>>>>>
>>>>>>But if you were to generalize the problem, I believe the
>>>>>>answer would be yes.
>>>>>>
>>>>>>The SSN for ordered delivery would be unchanged. The
>>>>>>addition would be providing "advisory ordering" for
>>>>>>unordered delivery. The general characteristic is that
>>>>>>the receiving ULP wants the data in whatever order it
>>>>>>arrives, but wants to know what the original order was.
>>>>>>
>>>>>>In addition to RDDP, RTP over UDP is another example of
>>>>>>an application that wants to be aware of the original
>>>>>>ordering, even while accepting the data out-of-order.
>>>>>>
>>>>>>Caitlin Bestler
>>>>>>http://asomi.com/CaitlinBestler/
>>>>>>_______________________________________________
>>>>>>tsvwg mailing list
>>>>>>tsvwg@ietf.org
>>>>>>https://www1.ietf.org/mailman/listinfo/tsvwg
>>>>>>
>>>>>
>>>>>_______________________________________________
>>>>>tsvwg mailing list
>>>>>tsvwg@ietf.org
>>>>>https://www1.ietf.org/mailman/listinfo/tsvwg
>>>>>
>>>>>
>>>>
>>>>
>>>>-- 
>>>>Randall R. Stewart
>>>>randall@stewart.chicago.il.us 815-342-5222 (cell phone)
>>>>
>>>>_______________________________________________
>>>>tsvwg mailing list
>>>>tsvwg@ietf.org
>>>>https://www1.ietf.org/mailman/listinfo/tsvwg
>>>
>>>
>>
>>-- 
>>Randall R. Stewart
>>randall@stewart.chicago.il.us 815-342-5222 (cell phone)
>>
>>_______________________________________________
>>Sigtran mailing list
>>Sigtran@ietf.org
>>https://www1.ietf.org/mailman/listinfo/sigtran
>>
> 
> _______________________________________________
> tsvwg mailing list
> tsvwg@ietf.org
> https://www1.ietf.org/mailman/listinfo/tsvwg
> 
> 


-- 
Randall R. Stewart
randall@stewart.chicago.il.us 815-342-5222 (cell phone)

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 07:22:56 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA00471
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 07:22:56 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CCWI210509
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 07:32:18 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CCW5p10478;
	Wed, 12 Feb 2003 07:32:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CCVdp10440
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 07:31:39 -0500
Received: from stewart.chicago.il.us (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA00437;
	Wed, 12 Feb 2003 07:21:44 -0500 (EST)
Received: from stewart.chicago.il.us (stewart.chicago.il.us [127.0.0.1])
	by stewart.chicago.il.us (8.12.6/8.12.3) with ESMTP id h1CCPQu6003585;
	Wed, 12 Feb 2003 06:25:27 -0600 (CST)
	(envelope-from randall@stewart.chicago.il.us)
Message-ID: <3E4A3D36.4030401@stewart.chicago.il.us>
Date: Wed, 12 Feb 2003 06:25:26 -0600
From: Randall Stewart <randall@stewart.chicago.il.us>
User-Agent: Mozilla/5.0 (X11; U; FreeBSD i386; en-US; rv:1.1) Gecko/20021005
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Tuexen Michael <michael.tuexen@siemens.com>
CC: bidulock@openss7.org, Michael Tuexen <Michael.Tuexen@lurchi.franken.de>,
        Caitlin Bestler <cait@asomi.com>, tsvwg@ietf.org, sigtran@ietf.org
References: <F92978223B01D311ACFF0008C71EE19D015DFAA8@MCHH225E>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] Re: [Tsvwg] A question of Un-Ordered SCTP
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

This looks good.. but would it be
better to make the FEATURE_ID 16 bits..

.. my reasoning is that cookies are getting
bigger and it might be best to control there
length to some degree... I could go either way
but I think 65,535 features is more than enough
to choose from :>

R

Tuexen Michael wrote:
> Randy,
> 
> I think having a FEATURES parameter would be a good idea. I do not
> like the idea of bitmask. Instead of this I would propose something
> like a FEATURES parameter which looks like:
> 
>        0                   1                   2                   3
>        0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
>       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>       |   FEATURES_PARAMETER_ID       |       Parameter Length        |
>       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>       |                         FEATURE_ID 1                          |
>       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>       |                         FEATURE_ID 2                          |
>       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>       |                         FEATURE_ID 3                          |
>       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>       |                         FEATURE_ID 4                          |
>       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
> 
> Each feature will be an IANA assiged 32 bit uint. This is extendable.
> 
> Best regards
> Michael
> 
> Michael Tuexen   Tel.:   +49 89 722 47210
> Siemens AG       Fax:    +49 89 722 48212
> ICN CP SE C 7    e-mail: Michael.Tuexen@siemens.com
> 
> 
>>-----Original Message-----
>>From: Randall Stewart [mailto:randall@stewart.chicago.il.us]
>>Sent: Wednesday, February 12, 2003 12:59 PM
>>To: bidulock@openss7.org; Michael Tuexen
>>Cc: Caitlin Bestler; tsvwg@ietf.org; Tuexen Michael; sigtran@ietf.org
>>Subject: Re: [Tsvwg] A question of Un-Ordered SCTP
>>
>>
>>Brian/Michael:
>>
>>I do like the idea of keeping things the same (aka same
>>methodology)... this is good.
>>
>>It is also neat to be able to tell the difference between
>>an implementation that supports this feature and one that
>>does not.  But there are positives and negatives, I think,
>>to each idea...
>>
>>A) For using a TLV, (as Michael suggests)
>>    we really need to put this into a seperate
>>    draft.. when I do think one could easily fit
>>    this into the IG as Brian as suggested.
>>
>>B) montonically increasing numbers would eventually
>>    wrap when the SSN hit FFFF back to 0... thus
>>    the application would need to have some sort
>>    of state to track if the peer supported the
>>    feature.
>>
>>It seems to me it is only the first message that
>>arrives that benefits from the start at 0. If the
>>application assumes a "not supported", then by
>>starting the SSN at 1 the app can flip the flag
>>from "not supported" to "supported" on the arrival
>>of the first message.
>>
>>If the application instead set its flag to "supported"
>>the first message would not tell this information and
>>instead it would only be upon the second arriving message
>>that the app could make the flag determination... causing
>>either a delay in delivery of the first message (yuck) or
>>just toss it up and wait.
>>
>>More importantly, I am wondering, if the implementation does
>>NOT support this, what will the app do? Change its
>>message structure? If so it will need to have some clues
>>(as the initiator) that its implementation does not
>>support the feature .. possibly a HAVE_SCTP_UNORDERED_SSN
>>in sctp.h... but the problem is you would not know about
>>the other guy until its first unordered message arrives.
>>So if you have to send the first message what do you do?
>>
>>if HAVE_SCTP_UNORDERED_SSN is 0.. no problem.. you
>>send a message that has the imbedded sequence number.. but
>>what do you send when HAVE_SCTP_UNRODERED_SSN is set to 1?
>>
>>If you send the imbedded sequence in the message anyway whats
>>the point of this little trick? The second message (which you
>>may have to send before knowing receiving a unordered
>>msg from the peer) would still have this, and maybe the third
>>etc...
>>
>>Hmm.. the more I think on this the more it becomes clear that
>>you must know ahead of time i.e. when the association comes
>>up to "do the right thing".... which goes back to Michaels
>>idea of a TLV.
>>
>>I have been wondering for quite some time now if we don't
>>just need a TLV in the INIT/INIT-ACK that indicate features
>>in general... something like a bit mask or byte array indicating
>>each feature...
>>
>>addip
>>delip
>>prsctp
>>setprimary
>>unordered-assigns-sequences
>>
>>etc..
>>
>>Maybe this is a better way to go.. and we could write a short
>>draft dictating this bit/byte TLV... code as a 0x80 class.. that
>>way old implementations will throw it away... (we really don't
>>care if they don't understand it)...
>>
>>I have been avoiding this.. but it might be the best approach..
>>
>>Then again, I guess it could also be put in addip or prsctp since
>>it would effect them as well :> Then one could put the
>>other change in the IG and thus have a very easy change that
>>is in the IG and a minor update to either addip or prsctp...
>>
>>
>>What do you all think?
>>
>>R
>>
>>Brian F. G. Bidulock wrote:
>>
>>>Randall,
>>>
>>>I agree to having SCTP set the sequence number and, as you said in
>>>your first note, SCTP passing the sequence number to the ULP on
>>>the receiving side is already done for ordered traffic.  Some even
>>>pass the zero now.
>>>
>>>I suppose that SCTP would set this sequence number when it is given
>>>the data for delivery at the sending ULP and the sequence 
>>
>>would represent
>>
>>>the order that the data was presented to SCTP by the ULP at 
>>
>>the sending
>>
>>>side.
>>>
>>>I think it would be good to see the requirement that the sequence
>>>number begin at 1 and be monotonically increasing in the 
>>
>>saem fashion
>>
>>>as ordered streams.
>>>
>>>That way RDDP and other applications could perform their 
>>
>>own reordering
>>
>>>at the receiver.  I think it would be a good thing to have SCTP set
>>>this value.
>>>
>>>In fact, I can see another application for this feature.  M2PA
>>>(SIGTRAN protocol) needs in some circumstances to send data 
>>
>>unordered
>>
>>>with the hope of expedited delivery (does not want control 
>>
>>information
>>
>>>stuck behind large amounts of data in an ordered stream), 
>>
>>however, the
>>
>>>order of processing of the expedited control information is 
>>
>>important.
>>
>>>This facility would permit M2PA to send such information unordered
>>>to expedite it and the receiver could perform the reordering itself.
>>>Up until now, the only solution was to place a sequence 
>>
>>number in with
>>
>>>the control information.
>>>
>>>Could we perhaps add a section to the SCTP IG.  "Problem statement:
>>>the sender wishes to send unordered data but the receive 
>>
>>needs to know
>>
>>>the original order of sending..." or some such.
>>>
>>>It should be quite simple to add the feature to existing SCTP
>>>implementations.
>>>
>>>--brian
>>>
>>>
>>>On Tue, 11 Feb 2003, Randall Stewart wrote:
>>>
>>>
>>>
>>>>Michael:
>>>>
>>>>This seems like a valid method.. it gets around Brian's objection
>>>>(which was a good point) and yet still provides the application with
>>>>a hint about what the assigned sequence order was on the unordered
>>>>messages...
>>>>
>>>>R
>>>>
>>>>Michael Tuexen wrote:
>>>>
>>>>
>>>>>Caitlin,
>>>>>
>>>>>thank you very much for the clarification.
>>>>>
>>>>>I do not like the idea to pass the SSN transparently from 
>>>>
>>the ULP to the 
>>
>>>>>ULP.
>>>>>But having two sequence spaces is OK for me. This means 
>>>>
>>that the SSN at the
>>
>>>>>sending side is controlled by the SCTP and given to the ULP at the 
>>>>>receiving side.
>>>>>
>>>>>This would be a clean solution. So I voting for have two 
>>>>
>>separate number 
>>
>>>>>spaces
>>>>>for user data. Since the one for ordered delivery exists 
>>>>
>>already in RFC 
>>
>>>>>2960, this
>>>>>only means that a new one is introduced for the unordered 
>>>>
>>delivery user 
>>
>>>>>data.
>>>>>
>>>>>Best regards
>>>>>Michael
>>>>>
>>>>>On Tuesday, Feb 11, 2003, at 17:37 Europe/Berlin, Caitlin 
>>>>
>>Bestler wrote:
>>
>>>>>
>>>>>>On 2/11/03, Tuexen Michael wrote:
>>>>>>
>>>>>>
>>>>>>
>>>>>>>Dear all,
>>>>>>>
>>>>>>>does this mean that you want two sequence spaces per
>>>>>>>stream: one for the 'ordered' user data and one for the
>>>>>>>unordered user data?
>>>>>>>
>>>>>>
>>>>>>The RDDP/SCTP adaptation only uses unordered delivery, so
>>>>>>that question did not really arise.
>>>>>>
>>>>>>But if you were to generalize the problem, I believe the
>>>>>>answer would be yes.
>>>>>>
>>>>>>The SSN for ordered delivery would be unchanged. The
>>>>>>addition would be providing "advisory ordering" for
>>>>>>unordered delivery. The general characteristic is that
>>>>>>the receiving ULP wants the data in whatever order it
>>>>>>arrives, but wants to know what the original order was.
>>>>>>
>>>>>>In addition to RDDP, RTP over UDP is another example of
>>>>>>an application that wants to be aware of the original
>>>>>>ordering, even while accepting the data out-of-order.
>>>>>>
>>>>>>Caitlin Bestler
>>>>>>http://asomi.com/CaitlinBestler/
>>>>>>_______________________________________________
>>>>>>tsvwg mailing list
>>>>>>tsvwg@ietf.org
>>>>>>https://www1.ietf.org/mailman/listinfo/tsvwg
>>>>>>
>>>>>
>>>>>_______________________________________________
>>>>>tsvwg mailing list
>>>>>tsvwg@ietf.org
>>>>>https://www1.ietf.org/mailman/listinfo/tsvwg
>>>>>
>>>>>
>>>>
>>>>
>>>>-- 
>>>>Randall R. Stewart
>>>>randall@stewart.chicago.il.us 815-342-5222 (cell phone)
>>>>
>>>>_______________________________________________
>>>>tsvwg mailing list
>>>>tsvwg@ietf.org
>>>>https://www1.ietf.org/mailman/listinfo/tsvwg
>>>
>>>
>>
>>-- 
>>Randall R. Stewart
>>randall@stewart.chicago.il.us 815-342-5222 (cell phone)
>>
> 
> 
> 


-- 
Randall R. Stewart
randall@stewart.chicago.il.us 815-342-5222 (cell phone)

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 07:28:24 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA00709
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 07:28:23 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CCbkC11410
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 07:37:46 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CCbTp11299;
	Wed, 12 Feb 2003 07:37:30 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CCahp10692
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 07:36:43 -0500
Received: from stewart.chicago.il.us (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA00627;
	Wed, 12 Feb 2003 07:26:48 -0500 (EST)
Received: from stewart.chicago.il.us (stewart.chicago.il.us [127.0.0.1])
	by stewart.chicago.il.us (8.12.6/8.12.3) with ESMTP id h1CCUQu6003598;
	Wed, 12 Feb 2003 06:30:26 -0600 (CST)
	(envelope-from randall@stewart.chicago.il.us)
Message-ID: <3E4A3E62.4050207@stewart.chicago.il.us>
Date: Wed, 12 Feb 2003 06:30:26 -0600
From: Randall Stewart <randall@stewart.chicago.il.us>
User-Agent: Mozilla/5.0 (X11; U; FreeBSD i386; en-US; rv:1.1) Gecko/20021005
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: bidulock@openss7.org
CC: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>,
        Caitlin Bestler <cait@asomi.com>, tsvwg@ietf.org,
        Tuexen Michael <michael.tuexen@siemens.com>, sigtran@ietf.org
References: <A2AA39E1-3DFB-11D7-90B4-0030654C1AB6@lurchi.franken.de> <3E49899A.6010809@stewart.chicago.il.us> <20030211191543.A14481@openss7.org> <3E4A36FF.8050205@stewart.chicago.il.us> <20030212052223.A29464@openss7.org>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] Re: [Tsvwg] A question of Un-Ordered SCTP
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Brian F. G. Bidulock wrote:
> Randall,
> 
> For the M2PA application, I think that reserving
> '0' for unordered messages for which the order of
> sending is unimportant is still a good idea.
> So the number space from 1 to 0xffff indicates
> importance of order.  '0' indicates no importance
> of order.  As in M2PA's case, a sender could mix
> both on the same stream.
> 
> There then becomes three different types of messages:
> 
> 1) Unordered messages for which the order of sending is
>    unimportant (U-bit SSN=0),
> 
> 2) unordered messages for which the order of sending is
>    important (U-bit SSN!=0), and
> 
> 3) ordered messages for which the order of sending is
>    important (~U-bit).
> 
> And this also handles compatibility because previous
> unordered messages fall into category (1) by nature
> of the fact that SSN==0.
> 
> But starting counting at 1, you get rid of the ability
> to indicate that the order of sending is unimportant and
> force the receiving ULP to handle ordering of all received
> unordered data.
> 
> So, in short, start counting at 1, regardless of whether
> there is an indicator in the INIT/INIT-ACK.
> 

That might be ok, but in the above does not that mean
when you reach 0xffff you end up sending a type
1) message... unless of course you want to have
a counting scheme where you always skip 0... which
is a bit more code on every sequence increase...

And I do still think you will need to "know" ahead of
time if both sides support 2) otherwise you would not
know to embed your own SSN (in cases where you must
emulate 2)

R

> --brian
> 
> Randall Stewart wrote:                                                             (Wed, 12 Feb 2003 05:58:55)
> 
>>Brian/Michael:
>>
>>I do like the idea of keeping things the same (aka same
>>methodology)... this is good.
>>
>>It is also neat to be able to tell the difference between
>>an implementation that supports this feature and one that
>>does not.  But there are positives and negatives, I think,
>>to each idea...
>>
>>A) For using a TLV, (as Michael suggests)
>>    we really need to put this into a seperate
>>    draft.. when I do think one could easily fit
>>    this into the IG as Brian as suggested.
>>
>>B) montonically increasing numbers would eventually
>>    wrap when the SSN hit FFFF back to 0... thus
>>    the application would need to have some sort
>>    of state to track if the peer supported the
>>    feature.
>>
>>It seems to me it is only the first message that
>>arrives that benefits from the start at 0. If the
>>application assumes a "not supported", then by
>>starting the SSN at 1 the app can flip the flag
>>from "not supported" to "supported" on the arrival
>>of the first message.
>>
>>If the application instead set its flag to "supported"
>>the first message would not tell this information and
>>instead it would only be upon the second arriving message
>>that the app could make the flag determination... causing
>>either a delay in delivery of the first message (yuck) or
>>just toss it up and wait.
>>
>>More importantly, I am wondering, if the implementation does
>>NOT support this, what will the app do? Change its
>>message structure? If so it will need to have some clues
>>(as the initiator) that its implementation does not
>>support the feature .. possibly a HAVE_SCTP_UNORDERED_SSN
>>in sctp.h... but the problem is you would not know about
>>the other guy until its first unordered message arrives.
>>So if you have to send the first message what do you do?
>>
>>if HAVE_SCTP_UNORDERED_SSN is 0.. no problem.. you
>>send a message that has the imbedded sequence number.. but
>>what do you send when HAVE_SCTP_UNRODERED_SSN is set to 1?
>>
>>If you send the imbedded sequence in the message anyway whats
>>the point of this little trick? The second message (which you
>>may have to send before knowing receiving a unordered
>>msg from the peer) would still have this, and maybe the third
>>etc...
>>
>>Hmm.. the more I think on this the more it becomes clear that
>>you must know ahead of time i.e. when the association comes
>>up to "do the right thing".... which goes back to Michaels
>>idea of a TLV.
>>
>>I have been wondering for quite some time now if we don't
>>just need a TLV in the INIT/INIT-ACK that indicate features
>>in general... something like a bit mask or byte array indicating
>>each feature...
>>
>>addip
>>delip
>>prsctp
>>setprimary
>>unordered-assigns-sequences
>>
>>etc..
>>
>>Maybe this is a better way to go.. and we could write a short
>>draft dictating this bit/byte TLV... code as a 0x80 class.. that
>>way old implementations will throw it away... (we really don't
>>care if they don't understand it)...
>>
>>I have been avoiding this.. but it might be the best approach..
>>
>>Then again, I guess it could also be put in addip or prsctp since
>>it would effect them as well :> Then one could put the
>>other change in the IG and thus have a very easy change that
>>is in the IG and a minor update to either addip or prsctp...
>>
>>
>>What do you all think?
>>
>>R
>>
>>Brian F. G. Bidulock wrote:
>>
>>>Randall,
>>>
>>>I agree to having SCTP set the sequence number and, as you said in
>>>your first note, SCTP passing the sequence number to the ULP on
>>>the receiving side is already done for ordered traffic.  Some even
>>>pass the zero now.
>>>
>>>I suppose that SCTP would set this sequence number when it is given
>>>the data for delivery at the sending ULP and the sequence would represent
>>>the order that the data was presented to SCTP by the ULP at the sending
>>>side.
>>>
>>>I think it would be good to see the requirement that the sequence
>>>number begin at 1 and be monotonically increasing in the saem fashion
>>>as ordered streams.
>>>
>>>That way RDDP and other applications could perform their own reordering
>>>at the receiver.  I think it would be a good thing to have SCTP set
>>>this value.
>>>
>>>In fact, I can see another application for this feature.  M2PA
>>>(SIGTRAN protocol) needs in some circumstances to send data unordered
>>>with the hope of expedited delivery (does not want control information
>>>stuck behind large amounts of data in an ordered stream), however, the
>>>order of processing of the expedited control information is important.
>>>
>>>This facility would permit M2PA to send such information unordered
>>>to expedite it and the receiver could perform the reordering itself.
>>>Up until now, the only solution was to place a sequence number in with
>>>the control information.
>>>
>>>Could we perhaps add a section to the SCTP IG.  "Problem statement:
>>>the sender wishes to send unordered data but the receive needs to know
>>>the original order of sending..." or some such.
>>>
>>>It should be quite simple to add the feature to existing SCTP
>>>implementations.
>>>
>>>--brian
>>>
>>>
>>>On Tue, 11 Feb 2003, Randall Stewart wrote:
>>>
>>>
>>>
>>>>Michael:
>>>>
>>>>This seems like a valid method.. it gets around Brian's objection
>>>>(which was a good point) and yet still provides the application with
>>>>a hint about what the assigned sequence order was on the unordered
>>>>messages...
>>>>
>>>>R
>>>>
>>>>Michael Tuexen wrote:
>>>>
>>>>
>>>>>Caitlin,
>>>>>
>>>>>thank you very much for the clarification.
>>>>>
>>>>>I do not like the idea to pass the SSN transparently from the ULP to the 
>>>>>ULP.
>>>>>But having two sequence spaces is OK for me. This means that the SSN at the
>>>>>sending side is controlled by the SCTP and given to the ULP at the 
>>>>>receiving side.
>>>>>
>>>>>This would be a clean solution. So I voting for have two separate number 
>>>>>spaces
>>>>>for user data. Since the one for ordered delivery exists already in RFC 
>>>>>2960, this
>>>>>only means that a new one is introduced for the unordered delivery user 
>>>>>data.
>>>>>
>>>>>Best regards
>>>>>Michael
>>>>>
>>>>>On Tuesday, Feb 11, 2003, at 17:37 Europe/Berlin, Caitlin Bestler wrote:
>>>>>
>>>>>
>>>>>
>>>>>>On 2/11/03, Tuexen Michael wrote:
>>>>>>
>>>>>>
>>>>>>
>>>>>>>Dear all,
>>>>>>>
>>>>>>>does this mean that you want two sequence spaces per
>>>>>>>stream: one for the 'ordered' user data and one for the
>>>>>>>unordered user data?
>>>>>>>
>>>>>>
>>>>>>The RDDP/SCTP adaptation only uses unordered delivery, so
>>>>>>that question did not really arise.
>>>>>>
>>>>>>But if you were to generalize the problem, I believe the
>>>>>>answer would be yes.
>>>>>>
>>>>>>The SSN for ordered delivery would be unchanged. The
>>>>>>addition would be providing "advisory ordering" for
>>>>>>unordered delivery. The general characteristic is that
>>>>>>the receiving ULP wants the data in whatever order it
>>>>>>arrives, but wants to know what the original order was.
>>>>>>
>>>>>>In addition to RDDP, RTP over UDP is another example of
>>>>>>an application that wants to be aware of the original
>>>>>>ordering, even while accepting the data out-of-order.
>>>>>>
>>>>>>Caitlin Bestler
>>>>>>http://asomi.com/CaitlinBestler/
>>>>>>_______________________________________________
>>>>>>tsvwg mailing list
>>>>>>tsvwg@ietf.org
>>>>>>https://www1.ietf.org/mailman/listinfo/tsvwg
>>>>>>
>>>>>
>>>>>_______________________________________________
>>>>>tsvwg mailing list
>>>>>tsvwg@ietf.org
>>>>>https://www1.ietf.org/mailman/listinfo/tsvwg
>>>>>
>>>>>
>>>>
>>>>
>>>>-- 
>>>>Randall R. Stewart
>>>>randall@stewart.chicago.il.us 815-342-5222 (cell phone)
>>>>
>>>>_______________________________________________
>>>>tsvwg mailing list
>>>>tsvwg@ietf.org
>>>>https://www1.ietf.org/mailman/listinfo/tsvwg
>>>
>>>
>>
>>-- 
>>Randall R. Stewart
>>randall@stewart.chicago.il.us 815-342-5222 (cell phone)
> 
> 


-- 
Randall R. Stewart
randall@stewart.chicago.il.us 815-342-5222 (cell phone)

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 07:36:24 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA00986
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 07:36:24 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CCjk711937
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 07:45:46 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CCjSp11912;
	Wed, 12 Feb 2003 07:45:28 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CCj0p11792
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 07:45:00 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA00918;
	Wed, 12 Feb 2003 07:35:02 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:tq1NvNhxlGYmkolmgEAF1dwn8oeWHWhu@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1CCchD17247;
	Wed, 12 Feb 2003 05:38:43 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1CCchg00685;
	Wed, 12 Feb 2003 05:38:43 -0700
Date: Wed, 12 Feb 2003 05:38:43 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Randall Stewart <randall@stewart.chicago.il.us>
Cc: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>,
        Caitlin Bestler <cait@asomi.com>, tsvwg@ietf.org,
        Tuexen Michael <michael.tuexen@siemens.com>, sigtran@ietf.org
Message-ID: <20030212053843.C29464@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Randall Stewart <randall@stewart.chicago.il.us>,
	Michael Tuexen <Michael.Tuexen@lurchi.franken.de>,
	Caitlin Bestler <cait@asomi.com>, tsvwg@ietf.org,
	Tuexen Michael <michael.tuexen@siemens.com>, sigtran@ietf.org
References: <A2AA39E1-3DFB-11D7-90B4-0030654C1AB6@lurchi.franken.de> <3E49899A.6010809@stewart.chicago.il.us> <20030211191543.A14481@openss7.org> <3E4A36FF.8050205@stewart.chicago.il.us> <20030212052223.A29464@openss7.org> <3E4A3E62.4050207@stewart.chicago.il.us>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E4A3E62.4050207@stewart.chicago.il.us>; from randall@stewart.chicago.il.us on Wed, Feb 12, 2003 at 06:30:26AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Subject: [Sigtran] Re: [Tsvwg] A question of Un-Ordered SCTP
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Randall,

So why not introduce another bit into the data chunk header?

Say, the O-bit which indicates that data is ordered at the
sender but not at the receiver?

An SCTP which does not know about the O-bit will assume that
ordered delivery is required.  Receivers which know about
the O-bit will know that unordered deliver is allowed.

That way you can start counting at 0 and leave unordered
data chunks alone.

While your at it, why not place the offset of the data chunk
in the final message in the DATA chunk header so that the
payload is identical for RDDP and non-RDDP SCTPs.  That way,
the non-RDDP receiver will be able to ignore the O-bit and
offset, and deliver the data ordered.  The RDDP receiver
will be able to observe the O-bit and offset and direct place
the data chunk.

--brian


Randall Stewart wrote:                                                             (Wed, 12 Feb 2003 06:30:26)
> Brian F. G. Bidulock wrote:
> > Randall,
> > 
> > For the M2PA application, I think that reserving
> > '0' for unordered messages for which the order of
> > sending is unimportant is still a good idea.
> > So the number space from 1 to 0xffff indicates
> > importance of order.  '0' indicates no importance
> > of order.  As in M2PA's case, a sender could mix
> > both on the same stream.
> > 
> > There then becomes three different types of messages:
> > 
> > 1) Unordered messages for which the order of sending is
> >    unimportant (U-bit SSN=0),
> > 
> > 2) unordered messages for which the order of sending is
> >    important (U-bit SSN!=0), and
> > 
> > 3) ordered messages for which the order of sending is
> >    important (~U-bit).
> > 
> > And this also handles compatibility because previous
> > unordered messages fall into category (1) by nature
> > of the fact that SSN==0.
> > 
> > But starting counting at 1, you get rid of the ability
> > to indicate that the order of sending is unimportant and
> > force the receiving ULP to handle ordering of all received
> > unordered data.
> > 
> > So, in short, start counting at 1, regardless of whether
> > there is an indicator in the INIT/INIT-ACK.
> > 
> 
> That might be ok, but in the above does not that mean
> when you reach 0xffff you end up sending a type
> 1) message... unless of course you want to have
> a counting scheme where you always skip 0... which
> is a bit more code on every sequence increase...
> 
> And I do still think you will need to "know" ahead of
> time if both sides support 2) otherwise you would not
> know to embed your own SSN (in cases where you must
> emulate 2)
> 
> R
> 
> > --brian
> > 
> > Randall Stewart wrote:                                                             (Wed, 12 Feb 2003 05:58:55)
> > 
> >>Brian/Michael:
> >>
> >>I do like the idea of keeping things the same (aka same
> >>methodology)... this is good.
> >>
> >>It is also neat to be able to tell the difference between
> >>an implementation that supports this feature and one that
> >>does not.  But there are positives and negatives, I think,
> >>to each idea...
> >>
> >>A) For using a TLV, (as Michael suggests)
> >>    we really need to put this into a seperate
> >>    draft.. when I do think one could easily fit
> >>    this into the IG as Brian as suggested.
> >>
> >>B) montonically increasing numbers would eventually
> >>    wrap when the SSN hit FFFF back to 0... thus
> >>    the application would need to have some sort
> >>    of state to track if the peer supported the
> >>    feature.
> >>
> >>It seems to me it is only the first message that
> >>arrives that benefits from the start at 0. If the
> >>application assumes a "not supported", then by
> >>starting the SSN at 1 the app can flip the flag
> >>from "not supported" to "supported" on the arrival
> >>of the first message.
> >>
> >>If the application instead set its flag to "supported"
> >>the first message would not tell this information and
> >>instead it would only be upon the second arriving message
> >>that the app could make the flag determination... causing
> >>either a delay in delivery of the first message (yuck) or
> >>just toss it up and wait.
> >>
> >>More importantly, I am wondering, if the implementation does
> >>NOT support this, what will the app do? Change its
> >>message structure? If so it will need to have some clues
> >>(as the initiator) that its implementation does not
> >>support the feature .. possibly a HAVE_SCTP_UNORDERED_SSN
> >>in sctp.h... but the problem is you would not know about
> >>the other guy until its first unordered message arrives.
> >>So if you have to send the first message what do you do?
> >>
> >>if HAVE_SCTP_UNORDERED_SSN is 0.. no problem.. you
> >>send a message that has the imbedded sequence number.. but
> >>what do you send when HAVE_SCTP_UNRODERED_SSN is set to 1?
> >>
> >>If you send the imbedded sequence in the message anyway whats
> >>the point of this little trick? The second message (which you
> >>may have to send before knowing receiving a unordered
> >>msg from the peer) would still have this, and maybe the third
> >>etc...
> >>
> >>Hmm.. the more I think on this the more it becomes clear that
> >>you must know ahead of time i.e. when the association comes
> >>up to "do the right thing".... which goes back to Michaels
> >>idea of a TLV.
> >>
> >>I have been wondering for quite some time now if we don't
> >>just need a TLV in the INIT/INIT-ACK that indicate features
> >>in general... something like a bit mask or byte array indicating
> >>each feature...
> >>
> >>addip
> >>delip
> >>prsctp
> >>setprimary
> >>unordered-assigns-sequences
> >>
> >>etc..
> >>
> >>Maybe this is a better way to go.. and we could write a short
> >>draft dictating this bit/byte TLV... code as a 0x80 class.. that
> >>way old implementations will throw it away... (we really don't
> >>care if they don't understand it)...
> >>
> >>I have been avoiding this.. but it might be the best approach..
> >>
> >>Then again, I guess it could also be put in addip or prsctp since
> >>it would effect them as well :> Then one could put the
> >>other change in the IG and thus have a very easy change that
> >>is in the IG and a minor update to either addip or prsctp...
> >>
> >>
> >>What do you all think?
> >>
> >>R
> >>
> >>Brian F. G. Bidulock wrote:
> >>
> >>>Randall,
> >>>
> >>>I agree to having SCTP set the sequence number and, as you said in
> >>>your first note, SCTP passing the sequence number to the ULP on
> >>>the receiving side is already done for ordered traffic.  Some even
> >>>pass the zero now.
> >>>
> >>>I suppose that SCTP would set this sequence number when it is given
> >>>the data for delivery at the sending ULP and the sequence would represent
> >>>the order that the data was presented to SCTP by the ULP at the sending
> >>>side.
> >>>
> >>>I think it would be good to see the requirement that the sequence
> >>>number begin at 1 and be monotonically increasing in the saem fashion
> >>>as ordered streams.
> >>>
> >>>That way RDDP and other applications could perform their own reordering
> >>>at the receiver.  I think it would be a good thing to have SCTP set
> >>>this value.
> >>>
> >>>In fact, I can see another application for this feature.  M2PA
> >>>(SIGTRAN protocol) needs in some circumstances to send data unordered
> >>>with the hope of expedited delivery (does not want control information
> >>>stuck behind large amounts of data in an ordered stream), however, the
> >>>order of processing of the expedited control information is important.
> >>>
> >>>This facility would permit M2PA to send such information unordered
> >>>to expedite it and the receiver could perform the reordering itself.
> >>>Up until now, the only solution was to place a sequence number in with
> >>>the control information.
> >>>
> >>>Could we perhaps add a section to the SCTP IG.  "Problem statement:
> >>>the sender wishes to send unordered data but the receive needs to know
> >>>the original order of sending..." or some such.
> >>>
> >>>It should be quite simple to add the feature to existing SCTP
> >>>implementations.
> >>>
> >>>--brian
> >>>
> >>>
> >>>On Tue, 11 Feb 2003, Randall Stewart wrote:
> >>>
> >>>
> >>>
> >>>>Michael:
> >>>>
> >>>>This seems like a valid method.. it gets around Brian's objection
> >>>>(which was a good point) and yet still provides the application with
> >>>>a hint about what the assigned sequence order was on the unordered
> >>>>messages...
> >>>>
> >>>>R
> >>>>
> >>>>Michael Tuexen wrote:
> >>>>
> >>>>
> >>>>>Caitlin,
> >>>>>
> >>>>>thank you very much for the clarification.
> >>>>>
> >>>>>I do not like the idea to pass the SSN transparently from the ULP to the 
> >>>>>ULP.
> >>>>>But having two sequence spaces is OK for me. This means that the SSN at the
> >>>>>sending side is controlled by the SCTP and given to the ULP at the 
> >>>>>receiving side.
> >>>>>
> >>>>>This would be a clean solution. So I voting for have two separate number 
> >>>>>spaces
> >>>>>for user data. Since the one for ordered delivery exists already in RFC 
> >>>>>2960, this
> >>>>>only means that a new one is introduced for the unordered delivery user 
> >>>>>data.
> >>>>>
> >>>>>Best regards
> >>>>>Michael
> >>>>>
> >>>>>On Tuesday, Feb 11, 2003, at 17:37 Europe/Berlin, Caitlin Bestler wrote:
> >>>>>
> >>>>>
> >>>>>
> >>>>>>On 2/11/03, Tuexen Michael wrote:
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>>>Dear all,
> >>>>>>>
> >>>>>>>does this mean that you want two sequence spaces per
> >>>>>>>stream: one for the 'ordered' user data and one for the
> >>>>>>>unordered user data?
> >>>>>>>
> >>>>>>
> >>>>>>The RDDP/SCTP adaptation only uses unordered delivery, so
> >>>>>>that question did not really arise.
> >>>>>>
> >>>>>>But if you were to generalize the problem, I believe the
> >>>>>>answer would be yes.
> >>>>>>
> >>>>>>The SSN for ordered delivery would be unchanged. The
> >>>>>>addition would be providing "advisory ordering" for
> >>>>>>unordered delivery. The general characteristic is that
> >>>>>>the receiving ULP wants the data in whatever order it
> >>>>>>arrives, but wants to know what the original order was.
> >>>>>>
> >>>>>>In addition to RDDP, RTP over UDP is another example of
> >>>>>>an application that wants to be aware of the original
> >>>>>>ordering, even while accepting the data out-of-order.
> >>>>>>
> >>>>>>Caitlin Bestler
> >>>>>>http://asomi.com/CaitlinBestler/
> >>>>>>_______________________________________________
> >>>>>>tsvwg mailing list
> >>>>>>tsvwg@ietf.org
> >>>>>>https://www1.ietf.org/mailman/listinfo/tsvwg
> >>>>>>
> >>>>>
> >>>>>_______________________________________________
> >>>>>tsvwg mailing list
> >>>>>tsvwg@ietf.org
> >>>>>https://www1.ietf.org/mailman/listinfo/tsvwg
> >>>>>
> >>>>>
> >>>>
> >>>>
> >>>>-- 
> >>>>Randall R. Stewart
> >>>>randall@stewart.chicago.il.us 815-342-5222 (cell phone)
> >>>>
> >>>>_______________________________________________
> >>>>tsvwg mailing list
> >>>>tsvwg@ietf.org
> >>>>https://www1.ietf.org/mailman/listinfo/tsvwg
> >>>
> >>>
> >>
> >>-- 
> >>Randall R. Stewart
> >>randall@stewart.chicago.il.us 815-342-5222 (cell phone)
> > 
> > 
> 
> 
> -- 
> Randall R. Stewart
> randall@stewart.chicago.il.us 815-342-5222 (cell phone)

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 07:57:12 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA01441
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 07:57:12 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CD6Zd13084
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 08:06:35 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CD6Lp13062;
	Wed, 12 Feb 2003 08:06:21 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CD5Bp12997
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 08:05:11 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA01382
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 07:55:16 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1CCwvqe020381;
	Wed, 12 Feb 2003 06:58:58 -0600 (CST)
Received: from axes-mach01.usa.alcatel.com by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id RAA13993; Wed, 12 Feb 2003 17:28:03 -0500 (GMT)
Message-ID: <3E4A452A.85F1083@axes-mach01.usa.alcatel.com>
Date: Wed, 12 Feb 2003 18:29:23 +0530
From: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: bidulock@openss7.org
CC: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
References: <3E4A01C7.CF2892B@axes-mach01.usa.alcatel.com> <20030212024002.A6927@openss7.org> <3E4A243F.D9B240D3@axes-mach01.usa.alcatel.com> <20030212042651.A16495@openss7.org>
Content-Type: text/plain; charset=x-user-defined
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Brian,

>>I believe that the number of active ASPs less than "n" corresponds to the AS-ACTIVE state (as in >>the RFC).

I agree with this.

--Srini

"Brian F. G. Bidulock" wrote:

> ksrinivas,
>
> ksrinivas wrote:                                                                 (Wed, 12 Feb 2003 16:10:48)
> > Brian,
> >
> > "Brian F. G. Bidulock" wrote:
> >
> > > Srini,
> > >
> > > ksrinivas wrote:                                                                 (Wed, 12 Feb 2003 13:42:16)
> > > > Brian,
> > > >
> > > >      Can you tell me when an SGP should send Notify(AS-PENDING) and
> > > > Notify(Insufficient ASP Resources Active in AS)?. I felt RFC and IG of
> > > > M3UA is not clear on these two.
> > > >
> > > > --Srini
> > >
> > > It think that it is pretty straight-forward.
> > >
> > > NTFY(Insufficient ASPs) is sent for an AS by an SGP when the number of
> > > ASPs in the ASP-ACTIVE state for the AS falls below "n", but not to
> > > zero, (as a result of ASP failures or deactivations); where "n" is the
> > > minimum number of active ASPs defined for the AS.  So this notification
> > > is only sent if "n" is greater than 1.
> > >
> > > NTFY(AS-PENDING) is sent for an AS by an SGP when the number of ASPs in
> > > the ASP-ACTIVE state for the AS falls to zero (as a result of ASP
> > > failures or deactivations).
> > >
> >
> > I understood the same way initially. But the section 3.15.2 of M3UA IG-02
> > has confused me. The following is the text from this section.
> >
> > "AC2PN: one ASP in ASP-ACTIVE state moves to ASP-INACTIVE or ASP-DOWN
> > states, causing the number of ASPs in ASP-ACTIVE drop below n."
> >
> > So the AS moves to AS-PENDING state when ASPs in ASP-ACTIVE drop below 'n'
> > (but not necessarily zero). Do we not have to send Notify(AS-PENDING) when
> > AS has moved to AS-PENDING state?.
>
> I see.
>
> I largely disagree with the IG section 3.15.2.  Consider that the new
> description of the AS-PENDING state is:
>
>     AS-PENDING: An active ASP has transitioned from ASP-ACTIVE to ASP-
>     INACTIVE or ASP-DOWN and it was the number n active ASP in the AS.  A
>     recovery timer T(r) SHOULD be started and all incoming signalling messages
>     SHOULD be queued by the SGP. If an ASP becomes ASP-ACTIVE before T(r)
>     expires, the AS is moved to the AS-ACTIVE state and all the queued
>     messages will be sent to the ASP.
>
> I think that it would be undesireable for an SGP to queue "incoming"
> signalling messages when the number of active ASPs is less than "n" for a
> number of reasons stated on the "n+k redundancy" thread.
>
> For example, at quiet periods, when there is not much traffic, is there any
> reason to stop service because one ASP happens to be inactive?  (Remember that
> this only counts for load-sharing and broadcast modes.)
>
> I believe that the number of active ASPs less than "n" corresponds to the
> AS-ACTIVE state (as in the RFC).
>
> I disagree with the modified state machine as arranged between Tolga and
> Javier with little input from others.
>
> As far as I'm concerned the AC2PN transition is caused by the last active ASP
> moving to the ASP-INACTIVE or ASP-DOWN state or only if the SGP follows the
> "MAY" (and withholds traffic for m < n).
>
> --brian
>
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 08:06:08 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA01723
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 08:06:08 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CDFWf14217
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 08:15:32 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CDFEp14194;
	Wed, 12 Feb 2003 08:15:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CDEQp14166
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 08:14:26 -0500
Received: from web12807.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id IAA01663
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 08:04:31 -0500 (EST)
Message-ID: <20030212130814.41239.qmail@web12807.mail.yahoo.com>
Received: from [62.29.64.191] by web12807.mail.yahoo.com via HTTP; Wed, 12 Feb 2003 05:08:14 PST
Date: Wed, 12 Feb 2003 05:08:14 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
In-Reply-To: <20030212042651.A16495@openss7.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Brian,
--- "Brian F. G. Bidulock" <bidulock@openss7.org>
wrote:
> ksrinivas,
> 
> ksrinivas wrote:                                    
>                             (Wed, 12 Feb 2003
> 16:10:48)
> > Brian,
> > 
> > "Brian F. G. Bidulock" wrote:
> > 
> > > Srini,
> > >
> > > ksrinivas wrote:                                
>                                 (Wed, 12 Feb 2003
> 13:42:16)
> > > > Brian,
> > > >
> > > >      Can you tell me when an SGP should send
> Notify(AS-PENDING) and
> > > > Notify(Insufficient ASP Resources Active in
> AS)?. I felt RFC and IG of
> > > > M3UA is not clear on these two.
> > > >
> > > > --Srini
> > >
> > > It think that it is pretty straight-forward.
> > >
> > > NTFY(Insufficient ASPs) is sent for an AS by an
> SGP when the number of
> > > ASPs in the ASP-ACTIVE state for the AS falls
> below "n", but not to
> > > zero, (as a result of ASP failures or
> deactivations); where "n" is the
> > > minimum number of active ASPs defined for the
> AS.  So this notification
> > > is only sent if "n" is greater than 1.
> > >
> > > NTFY(AS-PENDING) is sent for an AS by an SGP
> when the number of ASPs in
> > > the ASP-ACTIVE state for the AS falls to zero
> (as a result of ASP
> > > failures or deactivations).
> > >
> > 
> > I understood the same way initially. But the
> section 3.15.2 of M3UA IG-02
> > has confused me. The following is the text from
> this section.
> > 
> > "AC2PN: one ASP in ASP-ACTIVE state moves to
> ASP-INACTIVE or ASP-DOWN
> > states, causing the number of ASPs in ASP-ACTIVE
> drop below n."
> > 
> > So the AS moves to AS-PENDING state when ASPs in
> ASP-ACTIVE drop below 'n'
> > (but not necessarily zero). Do we not have to send
> Notify(AS-PENDING) when
> > AS has moved to AS-PENDING state?.
> 
> I see.
> 
> I largely disagree with the IG section 3.15.2. 
> Consider that the new
> description of the AS-PENDING state is:
> 
>     AS-PENDING: An active ASP has transitioned from
> ASP-ACTIVE to ASP-
>     INACTIVE or ASP-DOWN and it was the number n
> active ASP in the AS.  A
>     recovery timer T(r) SHOULD be started and all
> incoming signalling messages
>     SHOULD be queued by the SGP. If an ASP becomes
> ASP-ACTIVE before T(r)
>     expires, the AS is moved to the AS-ACTIVE state
> and all the queued
>     messages will be sent to the ASP.
> 
> I think that it would be undesireable for an SGP to
> queue "incoming"
> signalling messages when the number of active ASPs
> is less than "n" for a
> number of reasons stated on the "n+k redundancy"
> thread.
> 
> For example, at quiet periods, when there is not
> much traffic, is there any
> reason to stop service because one ASP happens to be
> inactive?  (Remember that
> this only counts for load-sharing and broadcast
> modes.)
[TOLGA] If this is the desired behavior then the
redundancy model to be choosen is 1+n . IMO, n+k
redundancy means that n active ASPs are required and I
think there is a difference between required and
preferred. IMO NTFY(Insuff. ASP resources) is used for
the case where more active ASPs are preferred. So, for
n+k redundancy case, when number of active ASPs fall
below n, AS state changes but not for 1+k case.
> 
> I believe that the number of active ASPs less than
> "n" corresponds to the
> AS-ACTIVE state (as in the RFC).
> 
> I disagree with the modified state machine as
> arranged between Tolga and
> Javier with little input from others.
[TOLGA]For n=1, the new n+k state machine is exactly
the same like the old one. But it just makes the state
transitions clear for the case where n>1 is
required.(everything related with this change was
discussed on the list, just I think you were busy at
that time ;) )
> 
> As far as I'm concerned the AC2PN transition is
> caused by the last active ASP
> moving to the ASP-INACTIVE or ASP-DOWN state or only
> if the SGP follows the
> "MAY" (and withholds traffic for m < n).
[TOLGA]I think, that "MAY" is a potential source for
problems, when ASP and SGP interprete it differently.
By letting the state of AS change for n+k case when
number of active ASPs fall below n, the goal was to
get rid of this ambiguity (but of course for the
"preferred" case, it is still there)
> 
> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran


__________________________________________________
Do you Yahoo!?
Yahoo! Shopping - Send Flowers for Valentine's Day
http://shopping.yahoo.com
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 08:10:46 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA00357
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 07:20:02 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CCTOA10281
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 07:29:24 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CCT7p10223;
	Wed, 12 Feb 2003 07:29:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CCSZp10175
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 07:28:35 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA00328;
	Wed, 12 Feb 2003 07:18:39 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:nrPNu/O+or2jPCcjvFc2Il/njb1X0mo3@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1CCMND16850;
	Wed, 12 Feb 2003 05:22:23 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1CCMNv30963;
	Wed, 12 Feb 2003 05:22:23 -0700
Date: Wed, 12 Feb 2003 05:22:23 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Randall Stewart <randall@stewart.chicago.il.us>
Cc: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>,
        Caitlin Bestler <cait@asomi.com>, tsvwg@ietf.org,
        Tuexen Michael <michael.tuexen@siemens.com>, sigtran@ietf.org
Message-ID: <20030212052223.A29464@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Randall Stewart <randall@stewart.chicago.il.us>,
	Michael Tuexen <Michael.Tuexen@lurchi.franken.de>,
	Caitlin Bestler <cait@asomi.com>, tsvwg@ietf.org,
	Tuexen Michael <michael.tuexen@siemens.com>, sigtran@ietf.org
References: <A2AA39E1-3DFB-11D7-90B4-0030654C1AB6@lurchi.franken.de> <3E49899A.6010809@stewart.chicago.il.us> <20030211191543.A14481@openss7.org> <3E4A36FF.8050205@stewart.chicago.il.us>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E4A36FF.8050205@stewart.chicago.il.us>; from randall@stewart.chicago.il.us on Wed, Feb 12, 2003 at 05:58:55AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Subject: [Sigtran] Re: [Tsvwg] A question of Un-Ordered SCTP
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Randall,

For the M2PA application, I think that reserving
'0' for unordered messages for which the order of
sending is unimportant is still a good idea.
So the number space from 1 to 0xffff indicates
importance of order.  '0' indicates no importance
of order.  As in M2PA's case, a sender could mix
both on the same stream.

There then becomes three different types of messages:

1) Unordered messages for which the order of sending is
   unimportant (U-bit SSN=0),

2) unordered messages for which the order of sending is
   important (U-bit SSN!=0), and

3) ordered messages for which the order of sending is
   important (~U-bit).

And this also handles compatibility because previous
unordered messages fall into category (1) by nature
of the fact that SSN==0.

But starting counting at 1, you get rid of the ability
to indicate that the order of sending is unimportant and
force the receiving ULP to handle ordering of all received
unordered data.

So, in short, start counting at 1, regardless of whether
there is an indicator in the INIT/INIT-ACK.

--brian

Randall Stewart wrote:                                                             (Wed, 12 Feb 2003 05:58:55)
> Brian/Michael:
> 
> I do like the idea of keeping things the same (aka same
> methodology)... this is good.
> 
> It is also neat to be able to tell the difference between
> an implementation that supports this feature and one that
> does not.  But there are positives and negatives, I think,
> to each idea...
> 
> A) For using a TLV, (as Michael suggests)
>     we really need to put this into a seperate
>     draft.. when I do think one could easily fit
>     this into the IG as Brian as suggested.
> 
> B) montonically increasing numbers would eventually
>     wrap when the SSN hit FFFF back to 0... thus
>     the application would need to have some sort
>     of state to track if the peer supported the
>     feature.
> 
> It seems to me it is only the first message that
> arrives that benefits from the start at 0. If the
> application assumes a "not supported", then by
> starting the SSN at 1 the app can flip the flag
> from "not supported" to "supported" on the arrival
> of the first message.
> 
> If the application instead set its flag to "supported"
> the first message would not tell this information and
> instead it would only be upon the second arriving message
> that the app could make the flag determination... causing
> either a delay in delivery of the first message (yuck) or
> just toss it up and wait.
> 
> More importantly, I am wondering, if the implementation does
> NOT support this, what will the app do? Change its
> message structure? If so it will need to have some clues
> (as the initiator) that its implementation does not
> support the feature .. possibly a HAVE_SCTP_UNORDERED_SSN
> in sctp.h... but the problem is you would not know about
> the other guy until its first unordered message arrives.
> So if you have to send the first message what do you do?
> 
> if HAVE_SCTP_UNORDERED_SSN is 0.. no problem.. you
> send a message that has the imbedded sequence number.. but
> what do you send when HAVE_SCTP_UNRODERED_SSN is set to 1?
> 
> If you send the imbedded sequence in the message anyway whats
> the point of this little trick? The second message (which you
> may have to send before knowing receiving a unordered
> msg from the peer) would still have this, and maybe the third
> etc...
> 
> Hmm.. the more I think on this the more it becomes clear that
> you must know ahead of time i.e. when the association comes
> up to "do the right thing".... which goes back to Michaels
> idea of a TLV.
> 
> I have been wondering for quite some time now if we don't
> just need a TLV in the INIT/INIT-ACK that indicate features
> in general... something like a bit mask or byte array indicating
> each feature...
> 
> addip
> delip
> prsctp
> setprimary
> unordered-assigns-sequences
> 
> etc..
> 
> Maybe this is a better way to go.. and we could write a short
> draft dictating this bit/byte TLV... code as a 0x80 class.. that
> way old implementations will throw it away... (we really don't
> care if they don't understand it)...
> 
> I have been avoiding this.. but it might be the best approach..
> 
> Then again, I guess it could also be put in addip or prsctp since
> it would effect them as well :> Then one could put the
> other change in the IG and thus have a very easy change that
> is in the IG and a minor update to either addip or prsctp...
> 
> 
> What do you all think?
> 
> R
> 
> Brian F. G. Bidulock wrote:
> > Randall,
> > 
> > I agree to having SCTP set the sequence number and, as you said in
> > your first note, SCTP passing the sequence number to the ULP on
> > the receiving side is already done for ordered traffic.  Some even
> > pass the zero now.
> > 
> > I suppose that SCTP would set this sequence number when it is given
> > the data for delivery at the sending ULP and the sequence would represent
> > the order that the data was presented to SCTP by the ULP at the sending
> > side.
> > 
> > I think it would be good to see the requirement that the sequence
> > number begin at 1 and be monotonically increasing in the saem fashion
> > as ordered streams.
> > 
> > That way RDDP and other applications could perform their own reordering
> > at the receiver.  I think it would be a good thing to have SCTP set
> > this value.
> > 
> > In fact, I can see another application for this feature.  M2PA
> > (SIGTRAN protocol) needs in some circumstances to send data unordered
> > with the hope of expedited delivery (does not want control information
> > stuck behind large amounts of data in an ordered stream), however, the
> > order of processing of the expedited control information is important.
> > 
> > This facility would permit M2PA to send such information unordered
> > to expedite it and the receiver could perform the reordering itself.
> > Up until now, the only solution was to place a sequence number in with
> > the control information.
> > 
> > Could we perhaps add a section to the SCTP IG.  "Problem statement:
> > the sender wishes to send unordered data but the receive needs to know
> > the original order of sending..." or some such.
> > 
> > It should be quite simple to add the feature to existing SCTP
> > implementations.
> > 
> > --brian
> > 
> > 
> > On Tue, 11 Feb 2003, Randall Stewart wrote:
> > 
> > 
> >>Michael:
> >>
> >>This seems like a valid method.. it gets around Brian's objection
> >>(which was a good point) and yet still provides the application with
> >>a hint about what the assigned sequence order was on the unordered
> >>messages...
> >>
> >>R
> >>
> >>Michael Tuexen wrote:
> >>
> >>>Caitlin,
> >>>
> >>>thank you very much for the clarification.
> >>>
> >>>I do not like the idea to pass the SSN transparently from the ULP to the 
> >>>ULP.
> >>>But having two sequence spaces is OK for me. This means that the SSN at the
> >>>sending side is controlled by the SCTP and given to the ULP at the 
> >>>receiving side.
> >>>
> >>>This would be a clean solution. So I voting for have two separate number 
> >>>spaces
> >>>for user data. Since the one for ordered delivery exists already in RFC 
> >>>2960, this
> >>>only means that a new one is introduced for the unordered delivery user 
> >>>data.
> >>>
> >>>Best regards
> >>>Michael
> >>>
> >>>On Tuesday, Feb 11, 2003, at 17:37 Europe/Berlin, Caitlin Bestler wrote:
> >>>
> >>>
> >>>>On 2/11/03, Tuexen Michael wrote:
> >>>>
> >>>>
> >>>>>Dear all,
> >>>>>
> >>>>>does this mean that you want two sequence spaces per
> >>>>>stream: one for the 'ordered' user data and one for the
> >>>>>unordered user data?
> >>>>>
> >>>>
> >>>>The RDDP/SCTP adaptation only uses unordered delivery, so
> >>>>that question did not really arise.
> >>>>
> >>>>But if you were to generalize the problem, I believe the
> >>>>answer would be yes.
> >>>>
> >>>>The SSN for ordered delivery would be unchanged. The
> >>>>addition would be providing "advisory ordering" for
> >>>>unordered delivery. The general characteristic is that
> >>>>the receiving ULP wants the data in whatever order it
> >>>>arrives, but wants to know what the original order was.
> >>>>
> >>>>In addition to RDDP, RTP over UDP is another example of
> >>>>an application that wants to be aware of the original
> >>>>ordering, even while accepting the data out-of-order.
> >>>>
> >>>>Caitlin Bestler
> >>>>http://asomi.com/CaitlinBestler/
> >>>>_______________________________________________
> >>>>tsvwg mailing list
> >>>>tsvwg@ietf.org
> >>>>https://www1.ietf.org/mailman/listinfo/tsvwg
> >>>>
> >>>
> >>>_______________________________________________
> >>>tsvwg mailing list
> >>>tsvwg@ietf.org
> >>>https://www1.ietf.org/mailman/listinfo/tsvwg
> >>>
> >>>
> >>
> >>
> >>-- 
> >>Randall R. Stewart
> >>randall@stewart.chicago.il.us 815-342-5222 (cell phone)
> >>
> >>_______________________________________________
> >>tsvwg mailing list
> >>tsvwg@ietf.org
> >>https://www1.ietf.org/mailman/listinfo/tsvwg
> > 
> > 
> 
> 
> -- 
> Randall R. Stewart
> randall@stewart.chicago.il.us 815-342-5222 (cell phone)

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 08:13:54 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA01856
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 08:13:54 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CDNI714561
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 08:23:18 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CDN6p14550;
	Wed, 12 Feb 2003 08:23:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CDM9p14517
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 08:22:09 -0500
Received: from web12801.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id IAA01833
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 08:12:14 -0500 (EST)
Message-ID: <20030212131558.41891.qmail@web12801.mail.yahoo.com>
Received: from [62.29.64.191] by web12801.mail.yahoo.com via HTTP; Wed, 12 Feb 2003 05:15:58 PST
Date: Wed, 12 Feb 2003 05:15:58 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
In-Reply-To: <3E4A452A.85F1083@axes-mach01.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

 If the traffic mode is 1+k I agree with that
statement but if it is n+k, I disagree.

   Tolga
--- ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
wrote:
> Brian,
> 
> >>I believe that the number of active ASPs less than
> "n" corresponds to the AS-ACTIVE state (as in >>the
> RFC).
> 
> I agree with this.
> 
> --Srini
> 
> "Brian F. G. Bidulock" wrote:
> 
> > ksrinivas,
> >
> > ksrinivas wrote:                                  
>                               (Wed, 12 Feb 2003
> 16:10:48)
> > > Brian,
> > >
> > > "Brian F. G. Bidulock" wrote:
> > >
> > > > Srini,
> > > >
> > > > ksrinivas wrote:                              
>                                   (Wed, 12 Feb 2003
> 13:42:16)
> > > > > Brian,
> > > > >
> > > > >      Can you tell me when an SGP should send
> Notify(AS-PENDING) and
> > > > > Notify(Insufficient ASP Resources Active in
> AS)?. I felt RFC and IG of
> > > > > M3UA is not clear on these two.
> > > > >
> > > > > --Srini
> > > >
> > > > It think that it is pretty straight-forward.
> > > >
> > > > NTFY(Insufficient ASPs) is sent for an AS by
> an SGP when the number of
> > > > ASPs in the ASP-ACTIVE state for the AS falls
> below "n", but not to
> > > > zero, (as a result of ASP failures or
> deactivations); where "n" is the
> > > > minimum number of active ASPs defined for the
> AS.  So this notification
> > > > is only sent if "n" is greater than 1.
> > > >
> > > > NTFY(AS-PENDING) is sent for an AS by an SGP
> when the number of ASPs in
> > > > the ASP-ACTIVE state for the AS falls to zero
> (as a result of ASP
> > > > failures or deactivations).
> > > >
> > >
> > > I understood the same way initially. But the
> section 3.15.2 of M3UA IG-02
> > > has confused me. The following is the text from
> this section.
> > >
> > > "AC2PN: one ASP in ASP-ACTIVE state moves to
> ASP-INACTIVE or ASP-DOWN
> > > states, causing the number of ASPs in ASP-ACTIVE
> drop below n."
> > >
> > > So the AS moves to AS-PENDING state when ASPs in
> ASP-ACTIVE drop below 'n'
> > > (but not necessarily zero). Do we not have to
> send Notify(AS-PENDING) when
> > > AS has moved to AS-PENDING state?.
> >
> > I see.
> >
> > I largely disagree with the IG section 3.15.2. 
> Consider that the new
> > description of the AS-PENDING state is:
> >
> >     AS-PENDING: An active ASP has transitioned
> from ASP-ACTIVE to ASP-
> >     INACTIVE or ASP-DOWN and it was the number n
> active ASP in the AS.  A
> >     recovery timer T(r) SHOULD be started and all
> incoming signalling messages
> >     SHOULD be queued by the SGP. If an ASP becomes
> ASP-ACTIVE before T(r)
> >     expires, the AS is moved to the AS-ACTIVE
> state and all the queued
> >     messages will be sent to the ASP.
> >
> > I think that it would be undesireable for an SGP
> to queue "incoming"
> > signalling messages when the number of active ASPs
> is less than "n" for a
> > number of reasons stated on the "n+k redundancy"
> thread.
> >
> > For example, at quiet periods, when there is not
> much traffic, is there any
> > reason to stop service because one ASP happens to
> be inactive?  (Remember that
> > this only counts for load-sharing and broadcast
> modes.)
> >
> > I believe that the number of active ASPs less than
> "n" corresponds to the
> > AS-ACTIVE state (as in the RFC).
> >
> > I disagree with the modified state machine as
> arranged between Tolga and
> > Javier with little input from others.
> >
> > As far as I'm concerned the AC2PN transition is
> caused by the last active ASP
> > moving to the ASP-INACTIVE or ASP-DOWN state or
> only if the SGP follows the
> > "MAY" (and withholds traffic for m < n).
> >
> > --brian
> >
> > --
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran


__________________________________________________
Do you Yahoo!?
Yahoo! Shopping - Send Flowers for Valentine's Day
http://shopping.yahoo.com
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 08:19:56 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA01933
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 08:19:55 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CDTJM14774
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 08:29:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CDT4p14748;
	Wed, 12 Feb 2003 08:29:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CDS1p14712
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 08:28:01 -0500
Received: from stewart.chicago.il.us (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA01917;
	Wed, 12 Feb 2003 08:18:05 -0500 (EST)
Received: from stewart.chicago.il.us (stewart.chicago.il.us [127.0.0.1])
	by stewart.chicago.il.us (8.12.6/8.12.3) with ESMTP id h1CDLlu6003679;
	Wed, 12 Feb 2003 07:21:48 -0600 (CST)
	(envelope-from randall@stewart.chicago.il.us)
Message-ID: <3E4A4A6B.4010405@stewart.chicago.il.us>
Date: Wed, 12 Feb 2003 07:21:47 -0600
From: Randall Stewart <randall@stewart.chicago.il.us>
User-Agent: Mozilla/5.0 (X11; U; FreeBSD i386; en-US; rv:1.1) Gecko/20021005
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: bidulock@openss7.org
CC: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>,
        Caitlin Bestler <cait@asomi.com>, tsvwg@ietf.org,
        Tuexen Michael <michael.tuexen@siemens.com>, sigtran@ietf.org
References: <A2AA39E1-3DFB-11D7-90B4-0030654C1AB6@lurchi.franken.de> <3E49899A.6010809@stewart.chicago.il.us> <20030211191543.A14481@openss7.org> <3E4A36FF.8050205@stewart.chicago.il.us> <20030212052223.A29464@openss7.org> <3E4A3E62.4050207@stewart.chicago.il.us> <20030212053843.C29464@openss7.org>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] Re: [Tsvwg] A question of Un-Ordered SCTP
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Brian:

I am really not interested in adding a bit in
the header for RDDP. I think it is better
to do something minimalist that meets there
needs instead of burning a bit (of only 5 left)
for one particular application.

Having a second sequence number space that is
allowed.. but not required.. and Michaels suggested
TLV in INIT/INIT-ACK would be more than adequate for
RDDP or for that matter any other application that
wants data right away but yet would like to know
presentation order to the sending SCTP.

R

Brian F. G. Bidulock wrote:
> Randall,
> 
> So why not introduce another bit into the data chunk header?
> 
> Say, the O-bit which indicates that data is ordered at the
> sender but not at the receiver?
> 
> An SCTP which does not know about the O-bit will assume that
> ordered delivery is required.  Receivers which know about
> the O-bit will know that unordered deliver is allowed.
> 
> That way you can start counting at 0 and leave unordered
> data chunks alone.
> 
> While your at it, why not place the offset of the data chunk
> in the final message in the DATA chunk header so that the
> payload is identical for RDDP and non-RDDP SCTPs.  That way,
> the non-RDDP receiver will be able to ignore the O-bit and
> offset, and deliver the data ordered.  The RDDP receiver
> will be able to observe the O-bit and offset and direct place
> the data chunk.
> 
> --brian
> 
> 
> Randall Stewart wrote:                                                             (Wed, 12 Feb 2003 06:30:26)
> 
>>Brian F. G. Bidulock wrote:
>>
>>>Randall,
>>>
>>>For the M2PA application, I think that reserving
>>>'0' for unordered messages for which the order of
>>>sending is unimportant is still a good idea.
>>>So the number space from 1 to 0xffff indicates
>>>importance of order.  '0' indicates no importance
>>>of order.  As in M2PA's case, a sender could mix
>>>both on the same stream.
>>>
>>>There then becomes three different types of messages:
>>>
>>>1) Unordered messages for which the order of sending is
>>>   unimportant (U-bit SSN=0),
>>>
>>>2) unordered messages for which the order of sending is
>>>   important (U-bit SSN!=0), and
>>>
>>>3) ordered messages for which the order of sending is
>>>   important (~U-bit).
>>>
>>>And this also handles compatibility because previous
>>>unordered messages fall into category (1) by nature
>>>of the fact that SSN==0.
>>>
>>>But starting counting at 1, you get rid of the ability
>>>to indicate that the order of sending is unimportant and
>>>force the receiving ULP to handle ordering of all received
>>>unordered data.
>>>
>>>So, in short, start counting at 1, regardless of whether
>>>there is an indicator in the INIT/INIT-ACK.
>>>
>>
>>That might be ok, but in the above does not that mean
>>when you reach 0xffff you end up sending a type
>>1) message... unless of course you want to have
>>a counting scheme where you always skip 0... which
>>is a bit more code on every sequence increase...
>>
>>And I do still think you will need to "know" ahead of
>>time if both sides support 2) otherwise you would not
>>know to embed your own SSN (in cases where you must
>>emulate 2)
>>
>>R
>>
>>
>>>--brian
>>>
>>>Randall Stewart wrote:                                                             (Wed, 12 Feb 2003 05:58:55)
>>>
>>>
>>>>Brian/Michael:
>>>>
>>>>I do like the idea of keeping things the same (aka same
>>>>methodology)... this is good.
>>>>
>>>>It is also neat to be able to tell the difference between
>>>>an implementation that supports this feature and one that
>>>>does not.  But there are positives and negatives, I think,
>>>>to each idea...
>>>>
>>>>A) For using a TLV, (as Michael suggests)
>>>>   we really need to put this into a seperate
>>>>   draft.. when I do think one could easily fit
>>>>   this into the IG as Brian as suggested.
>>>>
>>>>B) montonically increasing numbers would eventually
>>>>   wrap when the SSN hit FFFF back to 0... thus
>>>>   the application would need to have some sort
>>>>   of state to track if the peer supported the
>>>>   feature.
>>>>
>>>>It seems to me it is only the first message that
>>>>arrives that benefits from the start at 0. If the
>>>>application assumes a "not supported", then by
>>>>starting the SSN at 1 the app can flip the flag
>>>
>>>>from "not supported" to "supported" on the arrival
>>>
>>>>of the first message.
>>>>
>>>>If the application instead set its flag to "supported"
>>>>the first message would not tell this information and
>>>>instead it would only be upon the second arriving message
>>>>that the app could make the flag determination... causing
>>>>either a delay in delivery of the first message (yuck) or
>>>>just toss it up and wait.
>>>>
>>>>More importantly, I am wondering, if the implementation does
>>>>NOT support this, what will the app do? Change its
>>>>message structure? If so it will need to have some clues
>>>>(as the initiator) that its implementation does not
>>>>support the feature .. possibly a HAVE_SCTP_UNORDERED_SSN
>>>>in sctp.h... but the problem is you would not know about
>>>>the other guy until its first unordered message arrives.
>>>>So if you have to send the first message what do you do?
>>>>
>>>>if HAVE_SCTP_UNORDERED_SSN is 0.. no problem.. you
>>>>send a message that has the imbedded sequence number.. but
>>>>what do you send when HAVE_SCTP_UNRODERED_SSN is set to 1?
>>>>
>>>>If you send the imbedded sequence in the message anyway whats
>>>>the point of this little trick? The second message (which you
>>>>may have to send before knowing receiving a unordered
>>>>msg from the peer) would still have this, and maybe the third
>>>>etc...
>>>>
>>>>Hmm.. the more I think on this the more it becomes clear that
>>>>you must know ahead of time i.e. when the association comes
>>>>up to "do the right thing".... which goes back to Michaels
>>>>idea of a TLV.
>>>>
>>>>I have been wondering for quite some time now if we don't
>>>>just need a TLV in the INIT/INIT-ACK that indicate features
>>>>in general... something like a bit mask or byte array indicating
>>>>each feature...
>>>>
>>>>addip
>>>>delip
>>>>prsctp
>>>>setprimary
>>>>unordered-assigns-sequences
>>>>
>>>>etc..
>>>>
>>>>Maybe this is a better way to go.. and we could write a short
>>>>draft dictating this bit/byte TLV... code as a 0x80 class.. that
>>>>way old implementations will throw it away... (we really don't
>>>>care if they don't understand it)...
>>>>
>>>>I have been avoiding this.. but it might be the best approach..
>>>>
>>>>Then again, I guess it could also be put in addip or prsctp since
>>>>it would effect them as well :> Then one could put the
>>>>other change in the IG and thus have a very easy change that
>>>>is in the IG and a minor update to either addip or prsctp...
>>>>
>>>>
>>>>What do you all think?
>>>>
>>>>R
>>>>
>>>>Brian F. G. Bidulock wrote:
>>>>
>>>>
>>>>>Randall,
>>>>>
>>>>>I agree to having SCTP set the sequence number and, as you said in
>>>>>your first note, SCTP passing the sequence number to the ULP on
>>>>>the receiving side is already done for ordered traffic.  Some even
>>>>>pass the zero now.
>>>>>
>>>>>I suppose that SCTP would set this sequence number when it is given
>>>>>the data for delivery at the sending ULP and the sequence would represent
>>>>>the order that the data was presented to SCTP by the ULP at the sending
>>>>>side.
>>>>>
>>>>>I think it would be good to see the requirement that the sequence
>>>>>number begin at 1 and be monotonically increasing in the saem fashion
>>>>>as ordered streams.
>>>>>
>>>>>That way RDDP and other applications could perform their own reordering
>>>>>at the receiver.  I think it would be a good thing to have SCTP set
>>>>>this value.
>>>>>
>>>>>In fact, I can see another application for this feature.  M2PA
>>>>>(SIGTRAN protocol) needs in some circumstances to send data unordered
>>>>>with the hope of expedited delivery (does not want control information
>>>>>stuck behind large amounts of data in an ordered stream), however, the
>>>>>order of processing of the expedited control information is important.
>>>>>
>>>>>This facility would permit M2PA to send such information unordered
>>>>>to expedite it and the receiver could perform the reordering itself.
>>>>>Up until now, the only solution was to place a sequence number in with
>>>>>the control information.
>>>>>
>>>>>Could we perhaps add a section to the SCTP IG.  "Problem statement:
>>>>>the sender wishes to send unordered data but the receive needs to know
>>>>>the original order of sending..." or some such.
>>>>>
>>>>>It should be quite simple to add the feature to existing SCTP
>>>>>implementations.
>>>>>
>>>>>--brian
>>>>>
>>>>>
>>>>>On Tue, 11 Feb 2003, Randall Stewart wrote:
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>>Michael:
>>>>>>
>>>>>>This seems like a valid method.. it gets around Brian's objection
>>>>>>(which was a good point) and yet still provides the application with
>>>>>>a hint about what the assigned sequence order was on the unordered
>>>>>>messages...
>>>>>>
>>>>>>R
>>>>>>
>>>>>>Michael Tuexen wrote:
>>>>>>
>>>>>>
>>>>>>
>>>>>>>Caitlin,
>>>>>>>
>>>>>>>thank you very much for the clarification.
>>>>>>>
>>>>>>>I do not like the idea to pass the SSN transparently from the ULP to the 
>>>>>>>ULP.
>>>>>>>But having two sequence spaces is OK for me. This means that the SSN at the
>>>>>>>sending side is controlled by the SCTP and given to the ULP at the 
>>>>>>>receiving side.
>>>>>>>
>>>>>>>This would be a clean solution. So I voting for have two separate number 
>>>>>>>spaces
>>>>>>>for user data. Since the one for ordered delivery exists already in RFC 
>>>>>>>2960, this
>>>>>>>only means that a new one is introduced for the unordered delivery user 
>>>>>>>data.
>>>>>>>
>>>>>>>Best regards
>>>>>>>Michael
>>>>>>>
>>>>>>>On Tuesday, Feb 11, 2003, at 17:37 Europe/Berlin, Caitlin Bestler wrote:
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>>On 2/11/03, Tuexen Michael wrote:
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>>Dear all,
>>>>>>>>>
>>>>>>>>>does this mean that you want two sequence spaces per
>>>>>>>>>stream: one for the 'ordered' user data and one for the
>>>>>>>>>unordered user data?
>>>>>>>>>
>>>>>>>>
>>>>>>>>The RDDP/SCTP adaptation only uses unordered delivery, so
>>>>>>>>that question did not really arise.
>>>>>>>>
>>>>>>>>But if you were to generalize the problem, I believe the
>>>>>>>>answer would be yes.
>>>>>>>>
>>>>>>>>The SSN for ordered delivery would be unchanged. The
>>>>>>>>addition would be providing "advisory ordering" for
>>>>>>>>unordered delivery. The general characteristic is that
>>>>>>>>the receiving ULP wants the data in whatever order it
>>>>>>>>arrives, but wants to know what the original order was.
>>>>>>>>
>>>>>>>>In addition to RDDP, RTP over UDP is another example of
>>>>>>>>an application that wants to be aware of the original
>>>>>>>>ordering, even while accepting the data out-of-order.
>>>>>>>>
>>>>>>>>Caitlin Bestler
>>>>>>>>http://asomi.com/CaitlinBestler/
>>>>>>>>_______________________________________________
>>>>>>>>tsvwg mailing list
>>>>>>>>tsvwg@ietf.org
>>>>>>>>https://www1.ietf.org/mailman/listinfo/tsvwg
>>>>>>>>
>>>>>>>
>>>>>>>_______________________________________________
>>>>>>>tsvwg mailing list
>>>>>>>tsvwg@ietf.org
>>>>>>>https://www1.ietf.org/mailman/listinfo/tsvwg
>>>>>>>
>>>>>>>
>>>>>>
>>>>>>
>>>>>>-- 
>>>>>>Randall R. Stewart
>>>>>>randall@stewart.chicago.il.us 815-342-5222 (cell phone)
>>>>>>
>>>>>>_______________________________________________
>>>>>>tsvwg mailing list
>>>>>>tsvwg@ietf.org
>>>>>>https://www1.ietf.org/mailman/listinfo/tsvwg
>>>>>
>>>>>
>>>>-- 
>>>>Randall R. Stewart
>>>>randall@stewart.chicago.il.us 815-342-5222 (cell phone)
>>>
>>>
>>
>>-- 
>>Randall R. Stewart
>>randall@stewart.chicago.il.us 815-342-5222 (cell phone)
> 
> 


-- 
Randall R. Stewart
randall@stewart.chicago.il.us 815-342-5222 (cell phone)

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 08:21:54 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA01992
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 08:21:54 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CDVIC14867
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 08:31:18 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CDV5p14856;
	Wed, 12 Feb 2003 08:31:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CDU1p14817
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 08:30:01 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA01947
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 08:20:05 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1CDNjqe023398;
	Wed, 12 Feb 2003 07:23:47 -0600 (CST)
Received: from axes-mach01.usa.alcatel.com by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id RAA14328; Wed, 12 Feb 2003 17:52:53 -0500 (GMT)
Message-ID: <3E4A4AFD.BC64022D@axes-mach01.usa.alcatel.com>
Date: Wed, 12 Feb 2003 18:54:13 +0530
From: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Tolga Asveren <tolga_asveren@yahoo.com>
CC: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
References: <20030212131558.41891.qmail@web12801.mail.yahoo.com>
Content-Type: text/plain; charset=x-user-defined
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Tolga,

   Can you please tell me when to send Notify(AS-PENDING) and
Notify(Insufficient ASP Resources Active in AS)?.

--Srini

Tolga Asveren wrote:

>  If the traffic mode is 1+k I agree with that
> statement but if it is n+k, I disagree.
>
>    Tolga
> --- ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
> wrote:
> > Brian,
> >
> > >>I believe that the number of active ASPs less than
> > "n" corresponds to the AS-ACTIVE state (as in >>the
> > RFC).
> >
> > I agree with this.
> >
> > --Srini
> >
> > "Brian F. G. Bidulock" wrote:
> >
> > > ksrinivas,
> > >
> > > ksrinivas wrote:
> >                               (Wed, 12 Feb 2003
> > 16:10:48)
> > > > Brian,
> > > >
> > > > "Brian F. G. Bidulock" wrote:
> > > >
> > > > > Srini,
> > > > >
> > > > > ksrinivas wrote:
> >                                   (Wed, 12 Feb 2003
> > 13:42:16)
> > > > > > Brian,
> > > > > >
> > > > > >      Can you tell me when an SGP should send
> > Notify(AS-PENDING) and
> > > > > > Notify(Insufficient ASP Resources Active in
> > AS)?. I felt RFC and IG of
> > > > > > M3UA is not clear on these two.
> > > > > >
> > > > > > --Srini
> > > > >
> > > > > It think that it is pretty straight-forward.
> > > > >
> > > > > NTFY(Insufficient ASPs) is sent for an AS by
> > an SGP when the number of
> > > > > ASPs in the ASP-ACTIVE state for the AS falls
> > below "n", but not to
> > > > > zero, (as a result of ASP failures or
> > deactivations); where "n" is the
> > > > > minimum number of active ASPs defined for the
> > AS.  So this notification
> > > > > is only sent if "n" is greater than 1.
> > > > >
> > > > > NTFY(AS-PENDING) is sent for an AS by an SGP
> > when the number of ASPs in
> > > > > the ASP-ACTIVE state for the AS falls to zero
> > (as a result of ASP
> > > > > failures or deactivations).
> > > > >
> > > >
> > > > I understood the same way initially. But the
> > section 3.15.2 of M3UA IG-02
> > > > has confused me. The following is the text from
> > this section.
> > > >
> > > > "AC2PN: one ASP in ASP-ACTIVE state moves to
> > ASP-INACTIVE or ASP-DOWN
> > > > states, causing the number of ASPs in ASP-ACTIVE
> > drop below n."
> > > >
> > > > So the AS moves to AS-PENDING state when ASPs in
> > ASP-ACTIVE drop below 'n'
> > > > (but not necessarily zero). Do we not have to
> > send Notify(AS-PENDING) when
> > > > AS has moved to AS-PENDING state?.
> > >
> > > I see.
> > >
> > > I largely disagree with the IG section 3.15.2.
> > Consider that the new
> > > description of the AS-PENDING state is:
> > >
> > >     AS-PENDING: An active ASP has transitioned
> > from ASP-ACTIVE to ASP-
> > >     INACTIVE or ASP-DOWN and it was the number n
> > active ASP in the AS.  A
> > >     recovery timer T(r) SHOULD be started and all
> > incoming signalling messages
> > >     SHOULD be queued by the SGP. If an ASP becomes
> > ASP-ACTIVE before T(r)
> > >     expires, the AS is moved to the AS-ACTIVE
> > state and all the queued
> > >     messages will be sent to the ASP.
> > >
> > > I think that it would be undesireable for an SGP
> > to queue "incoming"
> > > signalling messages when the number of active ASPs
> > is less than "n" for a
> > > number of reasons stated on the "n+k redundancy"
> > thread.
> > >
> > > For example, at quiet periods, when there is not
> > much traffic, is there any
> > > reason to stop service because one ASP happens to
> > be inactive?  (Remember that
> > > this only counts for load-sharing and broadcast
> > modes.)
> > >
> > > I believe that the number of active ASPs less than
> > "n" corresponds to the
> > > AS-ACTIVE state (as in the RFC).
> > >
> > > I disagree with the modified state machine as
> > arranged between Tolga and
> > > Javier with little input from others.
> > >
> > > As far as I'm concerned the AC2PN transition is
> > caused by the last active ASP
> > > moving to the ASP-INACTIVE or ASP-DOWN state or
> > only if the SGP follows the
> > > "MAY" (and withholds traffic for m < n).
> > >
> > > --brian
> > >
> > > --
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
>
> __________________________________________________
> Do you Yahoo!?
> Yahoo! Shopping - Send Flowers for Valentine's Day
> http://shopping.yahoo.com
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 08:30:24 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA02130
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 08:30:24 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CDdmN15865
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 08:39:48 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CDdVp15852;
	Wed, 12 Feb 2003 08:39:32 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CDcsp15812
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 08:38:54 -0500
Received: from web12808.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id IAA02122
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 08:28:58 -0500 (EST)
Message-ID: <20030212133241.34194.qmail@web12808.mail.yahoo.com>
Received: from [62.29.64.191] by web12808.mail.yahoo.com via HTTP; Wed, 12 Feb 2003 05:32:41 PST
Date: Wed, 12 Feb 2003 05:32:41 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
In-Reply-To: <3E4A4AFD.BC64022D@axes-mach01.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Srini,

IMO Notify(AS Pending) is sent when AS chnages to
AS-Pending state. Notify(Insuff. ASP resources) is
sent, when there is no AS state change but SGP wants
to inform INACTIVE ASPs that it wants them to switch
to ACTIVE state. I think, the difference is required
v.s. preferred. For example if I am sure that I will
need 2 ACTIVE ASPs then I would use n=2. But if I
think that mostly 1 is enough but according to traffic
load I might need a second one I would use n=1 and use
NTFY(Insuff. ASP resources Active) when necessary. In
fact for that scenario, the number of preferred ASPs
is not predefined, i.e. it changes with traffic load,
so I wouldn't think it can be prearrenged between SGP
and ASP. But if one is sure about the required number
of ASPs then this value can be configured both on ASP
and SGP and an AS state chnage based on that value
seems appropriate to me.

    Tolga
--- ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
wrote:
> Tolga,
> 
>    Can you please tell me when to send
> Notify(AS-PENDING) and
> Notify(Insufficient ASP Resources Active in AS)?.
> 
> --Srini
> 
> Tolga Asveren wrote:
> 
> >  If the traffic mode is 1+k I agree with that
> > statement but if it is n+k, I disagree.
> >
> >    Tolga
> > --- ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
> > wrote:
> > > Brian,
> > >
> > > >>I believe that the number of active ASPs less
> than
> > > "n" corresponds to the AS-ACTIVE state (as in
> >>the
> > > RFC).
> > >
> > > I agree with this.
> > >
> > > --Srini
> > >
> > > "Brian F. G. Bidulock" wrote:
> > >
> > > > ksrinivas,
> > > >
> > > > ksrinivas wrote:
> > >                               (Wed, 12 Feb 2003
> > > 16:10:48)
> > > > > Brian,
> > > > >
> > > > > "Brian F. G. Bidulock" wrote:
> > > > >
> > > > > > Srini,
> > > > > >
> > > > > > ksrinivas wrote:
> > >                                   (Wed, 12 Feb
> 2003
> > > 13:42:16)
> > > > > > > Brian,
> > > > > > >
> > > > > > >      Can you tell me when an SGP should
> send
> > > Notify(AS-PENDING) and
> > > > > > > Notify(Insufficient ASP Resources Active
> in
> > > AS)?. I felt RFC and IG of
> > > > > > > M3UA is not clear on these two.
> > > > > > >
> > > > > > > --Srini
> > > > > >
> > > > > > It think that it is pretty
> straight-forward.
> > > > > >
> > > > > > NTFY(Insufficient ASPs) is sent for an AS
> by
> > > an SGP when the number of
> > > > > > ASPs in the ASP-ACTIVE state for the AS
> falls
> > > below "n", but not to
> > > > > > zero, (as a result of ASP failures or
> > > deactivations); where "n" is the
> > > > > > minimum number of active ASPs defined for
> the
> > > AS.  So this notification
> > > > > > is only sent if "n" is greater than 1.
> > > > > >
> > > > > > NTFY(AS-PENDING) is sent for an AS by an
> SGP
> > > when the number of ASPs in
> > > > > > the ASP-ACTIVE state for the AS falls to
> zero
> > > (as a result of ASP
> > > > > > failures or deactivations).
> > > > > >
> > > > >
> > > > > I understood the same way initially. But the
> > > section 3.15.2 of M3UA IG-02
> > > > > has confused me. The following is the text
> from
> > > this section.
> > > > >
> > > > > "AC2PN: one ASP in ASP-ACTIVE state moves to
> > > ASP-INACTIVE or ASP-DOWN
> > > > > states, causing the number of ASPs in
> ASP-ACTIVE
> > > drop below n."
> > > > >
> > > > > So the AS moves to AS-PENDING state when
> ASPs in
> > > ASP-ACTIVE drop below 'n'
> > > > > (but not necessarily zero). Do we not have
> to
> > > send Notify(AS-PENDING) when
> > > > > AS has moved to AS-PENDING state?.
> > > >
> > > > I see.
> > > >
> > > > I largely disagree with the IG section 3.15.2.
> > > Consider that the new
> > > > description of the AS-PENDING state is:
> > > >
> > > >     AS-PENDING: An active ASP has transitioned
> > > from ASP-ACTIVE to ASP-
> > > >     INACTIVE or ASP-DOWN and it was the number
> n
> > > active ASP in the AS.  A
> > > >     recovery timer T(r) SHOULD be started and
> all
> > > incoming signalling messages
> > > >     SHOULD be queued by the SGP. If an ASP
> becomes
> > > ASP-ACTIVE before T(r)
> > > >     expires, the AS is moved to the AS-ACTIVE
> > > state and all the queued
> > > >     messages will be sent to the ASP.
> > > >
> > > > I think that it would be undesireable for an
> SGP
> > > to queue "incoming"
> > > > signalling messages when the number of active
> ASPs
> > > is less than "n" for a
> > > > number of reasons stated on the "n+k
> redundancy"
> > > thread.
> > > >
> > > > For example, at quiet periods, when there is
> not
> > > much traffic, is there any
> > > > reason to stop service because one ASP happens
> to
> > > be inactive?  (Remember that
> > > > this only counts for load-sharing and
> broadcast
> > > modes.)
> > > >
> > > > I believe that the number of active ASPs less
> than
> > > "n" corresponds to the
> > > > AS-ACTIVE state (as in the RFC).
> > > >
> > > > I disagree with the modified state machine as
> > > arranged between Tolga and
> > > > Javier with little input from others.
> > > >
> > > > As far as I'm concerned the AC2PN transition
> is
> > > caused by the last active ASP
> > > > moving to the ASP-INACTIVE or ASP-DOWN state
> or
> > > only if the SGP follows the
> > > > "MAY" (and withholds traffic for m < n).
> > > >
> > > > --brian
> > > >
> > > > --
> > > > Brian F. G. Bidulock
> > > > bidulock@openss7.org
> > > > http://www.openss7.org/
> > > >
> _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> > __________________________________________________
> > Do you Yahoo!?
> > Yahoo! Shopping - Send Flowers for Valentine's Day
> > http://shopping.yahoo.com
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 


__________________________________________________
Do you Yahoo!?
Yahoo! Shopping - Send Flowers for Valentine's Day
http://shopping.yahoo.com
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 08:35:23 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA02246
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 08:35:23 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CDilq16083
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 08:44:47 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CDi5p16061;
	Wed, 12 Feb 2003 08:44:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CDhqp16043
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 08:43:52 -0500
Received: from domino2.nc.tekelec.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA02204
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 08:33:56 -0500 (EST)
From: jeffrey.craig@tekelec.com
Received: from domino4.nc.tekelec.com ([198.89.40.9])
          by domino2.nc.tekelec.com (Lotus Domino Release 5.0.10)
          with ESMTP id 2003021208311278:32280 ;
          Wed, 12 Feb 2003 08:31:12 -0500 
Subject: Re: [Sigtran] M2PA draft comments and questions.
To: bidulock@openss7.org
Cc: John Li <jli@iname.com>, sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.7  March 21, 2001
Message-ID: <OF2DF310C4.A6055B45-ON85256CCB.00467B35@nc.tekelec.com>
Date: Wed, 12 Feb 2003 08:33:17 -0500
MIME-Version: 1.0
X-MIMETrack: Serialize by Router on Domino4/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/12/2003 08:33:17 AM,
	Itemize by SMTP Server on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/12/2003 08:31:12 AM,
	Serialize by Router on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/12/2003 08:31:23 AM,
	Serialize complete at 02/12/2003 08:31:23 AM
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hello Guys,

Another round... comments are embedded.

Regards,

Jeff


On 02/11/2003 10:33:11 PM sigtran-admin wrote:

>John,
>
>John Li
>wrote:

>(Tue, 11 Feb 2003 21:41:07)
>>
>> Hi, Jeff,
>> Thanks for the responses. Further comments are inline.
>> Thanks,
>> John
>>
>> -----Original Message-----
>> From: jeffrey.craig@tekelec.com [mailto:jeffrey.craig@tekelec.com]
>> Sent: Tuesday, February 11, 2003 1:24 PM
>> To: John Li
>> Cc: sigtran@ietf.org
>> Subject: Re: [Sigtran] M2PA draft comments and questions.
>>
>>
>>
>> Hello John,
>>
>> Since no one else is responding, I'll answer.
>>
>> My responses are embedded below.
>>
>> Regards,
>>
>> Jeff
>>
>> On 02/11/2003 10:09:47 AM sigtran-admin wrote:
>>
>> >All,
>> >I sent the following comments and questions to the group a while ago,
and
>> have
>> >not got responses yet. here goes again.
>> >thanks.
>> >John
>> >
>> >
>> >1.   In section 1.5: the following statement is somewhat confusing.
>> >"The SG is an IPSP equipped with both traditional SS7 and IP
>> >network connections. In effect, the Signaling Gateway acts as a
>> >Signal Transfer Point (STP)"
>> >
>> >Perhaps we can say "The SG interconnects SS7 and IP networks by
>> >converting signaling traffic" or just refer to the definition in
>> >section 1.2
>>
>> I am not sure what defect this removes, and so I have no comment on
this.
>>
>> >
>> >
>> >2.   Section 2.1.5 Message length
>> >Should we specify the valid value range of the length?
>> >
>>
>> No. Min/Max message length is dependent on message type,
>> and in the case of user data is governed by MTP3.
>>
>> [John] the max payload MTP2 can carry is 272 octets. So by leaving that
>> unspecified, it would be fine if M2PA used in a pure IP environment, but
>> would have issues when used in a SG, who will be doing the segmentation
and
>> reassembling when receiving oversize(>272) messages from peer M2PA
before it
>> can be passed to the ss7 network?
>
>MTP3b does not limit to 272 octets.  Neither does Q.703 Annex A.
>
>This is a network restriction understood by MTP3.  M2PA imposes no
>restriction on message size.
>
>>
>> >3.   Section 2.3.1 User Data
>> >
>> >statement "M2PA does not add padding to the MTP3 message.".
>> >I think we should make it a requirement: "M2PA should not add padding
>> >to the MTP3 message."
>> >
>>
>> Sounds good to me.
>
>It is necessary to say MUST rather than SHOULD.
>
>>
>> >
>> >4.   In section 4.1.3 Link alignment
>> >
>> >perhaps we can specify the recommended practice for choosing
>> >emergency proving vs. normal proving during alignment procedure.
>> >
>>
>> This is governed by MTP3 procedures and is outside the scope
>> of the M2PA draft.
>>
>> [John] I thought about the scope issue as well. But since the draft
covers
>> MTP3 changeover in 2.4.6, by the same token, it would be very useful to
add
>> a paragraph to explain the two different alignment procedures in section
>> 2.4: Procedures to Support the MTP3/MTP2 Interface.
>
>
>Changeover is discussed because the chnageover message contains
>the MTP2 sequence number.  XCO/XCA need to be used due to M2PA's
>larger sequence number.
>
>Emergency vs. Normal proving is adequately addressed by MTP3 which
>is cited as a reference.
>
>>
>> >
>> >5.   Section 4.1.8:
>> >"If a Link Status Alignment message with an unsupported
>> >received by the newer version, the receiving end's M2PA
>> >SHALL NOT complete the alignment procedure."
>> >
>> >can we clearly define the behavior of not completing the alignment
>> >procedure?
>>
>> I believe that you are suggesting that the text be reworded to state
>> the following:
>>
>> "If a Link Status Alignment message with an unsupported version
>> is received by the newer version, the receiving end's M2PA
>> SHOULD fail the link."
>>
>> I think this was already understood, but the change improves clarity.
>>
>> [John] Further, I am thinking in this case, what is the procedure to
prevent
>> further alignment when the M2PA knows the incompatible versions: for
>> example, should M2PA notify its upper user about the incompatible so
that
>> the upper layer will not initiate alignment again. what is the status to
the
>> M2PA user to indicate it should not try to re-align if so, etc.
>
>Management procedures are implementation dependent.

(Jeff)
I agree that requesting that the upper layer stop trying to align the link
is
outside the scope of the M2PA draft. There is no MTP3 primitive for this.
There is nothing preventing the implementor from maintaining and
reporting 'M2PA version error' measurements.

The initial question was: "What is the definition of 'SHALL NOT complete
the alignment procedure'".

The draft currently implies that the unsupported LSA is simply ignored.
This means that alignment will ultimately fail due to T2 expiry by the
receiver of the unsupported LSA. This also means that the sender is
likely to advance to PROVING. I see no problem with this.

So the text at 4.1.8 should be clarified:

OLD TEXT
In case a newer version of M2PA is incompatible with an older version,
the newer version SHOULD recognize this and prevent the alignment of
the link. If a Link Status Alignment message with an unsupported
version is received by the newer version, the receiving end's M2PA
SHALL NOT complete the alignment procedure.

NEW TEXT
In case a newer version of M2PA is incompatible with an older version,
the newer version SHOULD recognize this and prevent the alignment of
the link. If a Link Status Alignment message with an unsupported
version is received by the newer version, the receiving end's M2PA
MUST discard the message.

>
>--brian
>
>--
>Brian F. G. Bidulock
>bidulock@openss7.org
>http://www.openss7.org/
>_______________________________________________
>Sigtran mailing list
>Sigtran@ietf.org
>https://www1.ietf.org/mailman/listinfo/sigtran


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 09:10:08 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA02861
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 09:10:08 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CEJXp18657
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 09:19:33 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CEJJp18650;
	Wed, 12 Feb 2003 09:19:19 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CEIkp18622
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 09:18:46 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA02850
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 09:08:49 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1CECUqe011404;
	Wed, 12 Feb 2003 08:12:31 -0600 (CST)
Received: from axes-mach01.usa.alcatel.com by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id SAA14962; Wed, 12 Feb 2003 18:41:36 -0500 (GMT)
Message-ID: <3E4A5668.4E899EE7@axes-mach01.usa.alcatel.com>
Date: Wed, 12 Feb 2003 19:42:57 +0530
From: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Tolga Asveren <tolga_asveren@yahoo.com>
CC: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
References: <20030212133241.34194.qmail@web12808.mail.yahoo.com>
Content-Type: text/plain; charset=x-user-defined
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Tolga,

>>Notify(AS Pending) is sent when AS changes to AS-Pending state.

So as per section 3.15.2 of M3UA IG, AC2PN causes AS state change to
AS-PENDING and forces to send Notify(AS Pending). i.e., when ASPs in the
ASP-ACTIVE state for the AS falls below "n".

>>For example if I am sure that I will need 2 ACTIVE ASPs then I would
use n=2. But if I think that mostly 1 is enough but according to traffic
load I might need a second one I would use n=1 and use NTFY(Insuff. ASP
resources Active) when necessary.

Do you mean to say that NTFY(Insuff. ASP resources Active) would be used
to make more than 'n' ASPs ACTIVE when necessary.

--Srini

Tolga Asveren wrote:

> Srini,
>
> IMO Notify(AS Pending) is sent when AS chnages to
> AS-Pending state. Notify(Insuff. ASP resources) is
> sent, when there is no AS state change but SGP wants
> to inform INACTIVE ASPs that it wants them to switch
> to ACTIVE state. I think, the difference is required
> v.s. preferred. For example if I am sure that I will
> need 2 ACTIVE ASPs then I would use n=2. But if I
> think that mostly 1 is enough but according to traffic
> load I might need a second one I would use n=1 and use
> NTFY(Insuff. ASP resources Active) when necessary. In
> fact for that scenario, the number of preferred ASPs
> is not predefined, i.e. it changes with traffic load,
> so I wouldn't think it can be prearrenged between SGP
> and ASP. But if one is sure about the required number
> of ASPs then this value can be configured both on ASP
> and SGP and an AS state chnage based on that value
> seems appropriate to me.
>
>     Tolga
> --- ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
> wrote:
> > Tolga,
> >
> >    Can you please tell me when to send
> > Notify(AS-PENDING) and
> > Notify(Insufficient ASP Resources Active in AS)?.
> >
> > --Srini
> >
> > Tolga Asveren wrote:
> >
> > >  If the traffic mode is 1+k I agree with that
> > > statement but if it is n+k, I disagree.
> > >
> > >    Tolga
> > > --- ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
> > > wrote:
> > > > Brian,
> > > >
> > > > >>I believe that the number of active ASPs less
> > than
> > > > "n" corresponds to the AS-ACTIVE state (as in
> > >>the
> > > > RFC).
> > > >
> > > > I agree with this.
> > > >
> > > > --Srini
> > > >
> > > > "Brian F. G. Bidulock" wrote:
> > > >
> > > > > ksrinivas,
> > > > >
> > > > > ksrinivas wrote:
> > > >                               (Wed, 12 Feb 2003
> > > > 16:10:48)
> > > > > > Brian,
> > > > > >
> > > > > > "Brian F. G. Bidulock" wrote:
> > > > > >
> > > > > > > Srini,
> > > > > > >
> > > > > > > ksrinivas wrote:
> > > >                                   (Wed, 12 Feb
> > 2003
> > > > 13:42:16)
> > > > > > > > Brian,
> > > > > > > >
> > > > > > > >      Can you tell me when an SGP should
> > send
> > > > Notify(AS-PENDING) and
> > > > > > > > Notify(Insufficient ASP Resources Active
> > in
> > > > AS)?. I felt RFC and IG of
> > > > > > > > M3UA is not clear on these two.
> > > > > > > >
> > > > > > > > --Srini
> > > > > > >
> > > > > > > It think that it is pretty
> > straight-forward.
> > > > > > >
> > > > > > > NTFY(Insufficient ASPs) is sent for an AS
> > by
> > > > an SGP when the number of
> > > > > > > ASPs in the ASP-ACTIVE state for the AS
> > falls
> > > > below "n", but not to
> > > > > > > zero, (as a result of ASP failures or
> > > > deactivations); where "n" is the
> > > > > > > minimum number of active ASPs defined for
> > the
> > > > AS.  So this notification
> > > > > > > is only sent if "n" is greater than 1.
> > > > > > >
> > > > > > > NTFY(AS-PENDING) is sent for an AS by an
> > SGP
> > > > when the number of ASPs in
> > > > > > > the ASP-ACTIVE state for the AS falls to
> > zero
> > > > (as a result of ASP
> > > > > > > failures or deactivations).
> > > > > > >
> > > > > >
> > > > > > I understood the same way initially. But the
> > > > section 3.15.2 of M3UA IG-02
> > > > > > has confused me. The following is the text
> > from
> > > > this section.
> > > > > >
> > > > > > "AC2PN: one ASP in ASP-ACTIVE state moves to
> > > > ASP-INACTIVE or ASP-DOWN
> > > > > > states, causing the number of ASPs in
> > ASP-ACTIVE
> > > > drop below n."
> > > > > >
> > > > > > So the AS moves to AS-PENDING state when
> > ASPs in
> > > > ASP-ACTIVE drop below 'n'
> > > > > > (but not necessarily zero). Do we not have
> > to
> > > > send Notify(AS-PENDING) when
> > > > > > AS has moved to AS-PENDING state?.
> > > > >
> > > > > I see.
> > > > >
> > > > > I largely disagree with the IG section 3.15.2.
> > > > Consider that the new
> > > > > description of the AS-PENDING state is:
> > > > >
> > > > >     AS-PENDING: An active ASP has transitioned
> > > > from ASP-ACTIVE to ASP-
> > > > >     INACTIVE or ASP-DOWN and it was the number
> > n
> > > > active ASP in the AS.  A
> > > > >     recovery timer T(r) SHOULD be started and
> > all
> > > > incoming signalling messages
> > > > >     SHOULD be queued by the SGP. If an ASP
> > becomes
> > > > ASP-ACTIVE before T(r)
> > > > >     expires, the AS is moved to the AS-ACTIVE
> > > > state and all the queued
> > > > >     messages will be sent to the ASP.
> > > > >
> > > > > I think that it would be undesireable for an
> > SGP
> > > > to queue "incoming"
> > > > > signalling messages when the number of active
> > ASPs
> > > > is less than "n" for a
> > > > > number of reasons stated on the "n+k
> > redundancy"
> > > > thread.
> > > > >
> > > > > For example, at quiet periods, when there is
> > not
> > > > much traffic, is there any
> > > > > reason to stop service because one ASP happens
> > to
> > > > be inactive?  (Remember that
> > > > > this only counts for load-sharing and
> > broadcast
> > > > modes.)
> > > > >
> > > > > I believe that the number of active ASPs less
> > than
> > > > "n" corresponds to the
> > > > > AS-ACTIVE state (as in the RFC).
> > > > >
> > > > > I disagree with the modified state machine as
> > > > arranged between Tolga and
> > > > > Javier with little input from others.
> > > > >
> > > > > As far as I'm concerned the AC2PN transition
> > is
> > > > caused by the last active ASP
> > > > > moving to the ASP-INACTIVE or ASP-DOWN state
> > or
> > > > only if the SGP follows the
> > > > > "MAY" (and withholds traffic for m < n).
> > > > >
> > > > > --brian
> > > > >
> > > > > --
> > > > > Brian F. G. Bidulock
> > > > > bidulock@openss7.org
> > > > > http://www.openss7.org/
> > > > >
> > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > >
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >
> > > __________________________________________________
> > > Do you Yahoo!?
> > > Yahoo! Shopping - Send Flowers for Valentine's Day
> > > http://shopping.yahoo.com
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> >
>
> __________________________________________________
> Do you Yahoo!?
> Yahoo! Shopping - Send Flowers for Valentine's Day
> http://shopping.yahoo.com
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 09:32:28 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA03696
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 09:32:28 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CEfs320338
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 09:41:54 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CEfYp20293;
	Wed, 12 Feb 2003 09:41:36 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1BAGXp02288
	for <sigtran@optimus.ietf.org>; Tue, 11 Feb 2003 05:16:33 -0500
Received: from beamer.mchh.siemens.de (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA25557
	for <sigtran@ietf.org>; Tue, 11 Feb 2003 05:07:11 -0500 (EST)
Received: from blues.mchh.siemens.de ([139.21.204.206])
	by beamer.mchh.siemens.de (8.9.3/8.9.3) with ESMTP id LAA08627;
	Tue, 11 Feb 2003 11:10:51 +0100 (MET)
Received: from mchh274e.demchh201e.icn.siemens.de ([139.21.200.84])
	by blues.mchh.siemens.de (8.9.1/8.9.1) with ESMTP id LAA16712;
	Tue, 11 Feb 2003 11:10:25 +0100 (MET)
Received: by MCHH274E with Internet Mail Service (5.5.2656.59)
	id <D13JRD2J>; Tue, 11 Feb 2003 11:10:00 +0100
Message-ID: <3D72C18DF140D311A54F0008C7DBAC8D02917438@MCHH252E>
From: Gaiser Franz <franz.gaiser@siemens.com>
To: sigtran@ietf.org
Cc: "'Tom George'" <Tom.George@alcatel.com>
Date: Tue, 11 Feb 2003 11:09:58 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2656.59)
Content-Type: text/plain;
	charset="ISO-8859-1"
Subject: [Sigtran] M2PA draft 07 comment: LPO
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Hi all,

I have an additional comment/remark to the description of handling Processor Outage in draft 07of M2PA. The handling of the following situation is not clear to me: - Proving Period completed (T4 expired) and Local Processor Outage is occured. In the case without LPO the m2pa would start sending LS Ready to the peer and moves to state ALIGNED READY. 
For the above situation (with LPO) the handling is not clearly stated. Is LS Ready and/or LS PO to be send ?According to Q.703  the m2pa should send LS PO message instead of LS Ready and moves to state ALIGNED NOT READY.



regards, franz
I
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 09:32:30 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA03711
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 09:32:30 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CEfuV20351
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 09:41:56 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CEfip20310;
	Wed, 12 Feb 2003 09:41:44 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1C8Wmp28295
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 03:32:48 -0500
Received: from gorilla.mchh.siemens.de (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA25932
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 03:22:59 -0500 (EST)
Received: from moody.mchh.siemens.de ([139.21.205.85])
	by gorilla.mchh.siemens.de (8.9.3/8.9.3) with ESMTP id JAA23700;
	Wed, 12 Feb 2003 09:26:42 +0100 (MET)
Received: from mchh246e.demchh201e.icn.siemens.de (mchh246e.mchh.siemens.de [139.21.200.56])
	by moody.mchh.siemens.de (8.9.1/8.9.1) with ESMTP id JAA21181;
	Wed, 12 Feb 2003 09:26:42 +0100 (MET)
Received: by mchh246e.mchh.siemens.de with Internet Mail Service (5.5.2656.59)
	id <1Z1M2LBM>; Wed, 12 Feb 2003 09:25:53 +0100
Message-ID: <F92978223B01D311ACFF0008C71EE19D015DFA9C@MCHH225E>
From: Tuexen Michael <michael.tuexen@siemens.com>
To: "'Tom George'" <Tom.George@alcatel.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: RE: [Sigtran] M2PA Security section (was: m2pa comment: author li
	st)
Date: Wed, 12 Feb 2003 09:25:51 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2656.59)
Content-Type: text/plain;
	charset="ISO-8859-1"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Hi,


why not simply referring to the SIGTRAN security doc instead
of reproducing part of it?

Best regards
Michael

Michael Tuexen   Tel.:   +49 89 722 47210
Siemens AG       Fax:    +49 89 722 48212
ICN CP SE C 7    e-mail: Michael.Tuexen@siemens.com

> -----Original Message-----
> From: Tom George [mailto:Tom.George@alcatel.com]
> Sent: Tuesday, February 11, 2003 10:58 PM
> Cc: 'sigtran@ietf.org'
> Subject: [Sigtran] M2PA Security section (was: m2pa comment: author
> list)
> 
> 
> On Jan 26, 2003 "Ong, Lyndon" wrote:
>   [...]
> > The security section probably needs to be updated based on the
> > security framework document (which should come out this week).
>   [...]
> 
> Here is my first stab at revising section 6 of the M2PA 
> draft. Comments
> are welcome, especially since I am not familiar with the security
> protocols.
> 
> tomG.
> 
> =============================
> 6. Security
> 
> M2PA is designed to carry signaling messages for telephony
> services. As such, M2PA MUST involve the security needs of several
> parties: 
> 
>    - the end users of the services
> 
>    - the network providers
> 
>    - the applications involved
> 
> Additional requirements MAY come from local regulation.  
> 
> While these parties may have some overlapping security needs, their
> needs may not be identical. Any security solution SHOULD fulfill all
> of the different parties' needs.
> 
> 6.1 Security Requirements
> 
> As a transport protocol, M2PA has the following security objectives:
> 
>    - Availability of reliable and timely user data transport.
> 
>    - Integrity of user data transport.
> 
>    - Confidentiality of user data.
> 
> M2PA runs on top of SCTP [RFC2960].  SCTP provides certain transport
> related security features, as discussed in [Security]. However, these
> security features MAY NOT be adequate for all of the involved parties'
> requirements.
> 
> When M2PA is running in professionally managed corporate or service
> provider network, it is reasonable to expect that this network
> includes an appropriate security policy framework. The "Site Security
> Handbook" [RFC2196] SHOULD be consulted for guidance.
> 
> When the network in which M2PA runs involves more than one party
> (e.g., a non-private network), it MAY NOT be reasonable to expect that
> all parties have implemented security in a sufficient manner.  In such
> a case, it is recommended that IPSec or TLS be used to ensure
> confidentiality of the user payload. Consult [Security] for more
> guidance on the use of security protocols. 
> 
> 6.2 Protecting Confidentiality
> 
> Particularly for mobile users, the requirement for confidentiality MAY
> include the masking of IP addresses and ports. In this case
> application-level encryption is not sufficient. IPSec ESP SHOULD be
> used instead [RFC2401].  Regardless of which level performs the
> encryption, the IPSec ISAKMP service SHOULD be used for key
> management.
> 
> =============================
> 
> {Add an informative reference:}
> 
>   [Security]
>        J. Loughney, M. Tuexen, and J. Pastor-Balbas, "Security
>        Considerations for SIGTRAN Protocols,"
>        draft-ietf-sigtran-security-01.txt (January 2003).
> 
> =============================
> 
> -- 
> * Tom George       PB7-24001 (near B-21)      
> Tom.George@alcatel.com  *
> * 972-519-3168   Intranet:  
http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 09:32:43 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA03728
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 09:32:43 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CEg9V20423
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 09:42:09 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CEfpp20325;
	Wed, 12 Feb 2003 09:41:51 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1C8kvp29497
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 03:46:57 -0500
Received: from gorilla.mchh.siemens.de (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA26156;
	Wed, 12 Feb 2003 03:37:06 -0500 (EST)
Received: from blues.mchh.siemens.de ([139.21.204.206])
	by gorilla.mchh.siemens.de (8.9.3/8.9.3) with ESMTP id JAA25330;
	Wed, 12 Feb 2003 09:40:45 +0100 (MET)
Received: from mchh253e.mchh.siemens.de (mchh253e.mchh.siemens.de [139.21.200.63])
	by blues.mchh.siemens.de (8.9.1/8.9.1) with ESMTP id JAA23879;
	Wed, 12 Feb 2003 09:40:19 +0100 (MET)
Received: by MCHH253E with Internet Mail Service (5.5.2653.19)
	id <D131R42R>; Wed, 12 Feb 2003 09:39:55 +0100
Message-ID: <F92978223B01D311ACFF0008C71EE19D015DFA9E@MCHH225E>
From: Tuexen Michael <michael.tuexen@siemens.com>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>,
        Randall Stewart
	 <randall@stewart.chicago.il.us>
Cc: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>,
        Caitlin Bestler
	 <cait@asomi.com>, tsvwg@ietf.org,
        Tuexen Michael
	 <michael.tuexen@siemens.com>, sigtran@ietf.org
Date: Wed, 12 Feb 2003 09:39:41 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Sigtran] RE: [Tsvwg] A question of Un-Ordered SCTP
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Brian,
see my comments below.
Best regards
Michael

Michael Tuexen   Tel.:   +49 89 722 47210
Siemens AG       Fax:    +49 89 722 48212
ICN CP SE C 7    e-mail: Michael.Tuexen@siemens.com

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Wednesday, February 12, 2003 3:16 AM
> To: Randall Stewart
> Cc: Michael Tuexen; Caitlin Bestler; tsvwg@ietf.org; Tuexen Michael;
> sigtran@ietf.org
> Subject: Re: [Tsvwg] A question of Un-Ordered SCTP
> 
> 
> Randall,
> 
> I agree to having SCTP set the sequence number and, as you said in
> your first note, SCTP passing the sequence number to the ULP on
> the receiving side is already done for ordered traffic.  Some even
> pass the zero now.
> 
> I suppose that SCTP would set this sequence number when it is given
> the data for delivery at the sending ULP and the sequence 
> would represent
> the order that the data was presented to SCTP by the ULP at 
> the sending
> side.
> 
> I think it would be good to see the requirement that the sequence
> number begin at 1 and be monotonically increasing in the saem fashion
> as ordered streams.
> 
I would prefer that the two sequence spaces are handled the same
way. Thus both start with SSN=0.
> That way RDDP and other applications could perform their own 
> reordering
> at the receiver.  I think it would be a good thing to have SCTP set
> this value.
> 
> In fact, I can see another application for this feature.  M2PA
> (SIGTRAN protocol) needs in some circumstances to send data unordered
> with the hope of expedited delivery (does not want control information
> stuck behind large amounts of data in an ordered stream), however, the
> order of processing of the expedited control information is important.
> 
> This facility would permit M2PA to send such information unordered
> to expedite it and the receiver could perform the reordering itself.
> Up until now, the only solution was to place a sequence number in with
> the control information.
> 
> Could we perhaps add a section to the SCTP IG.  "Problem statement:
> the sender wishes to send unordered data but the receive needs to know
> the original order of sending..." or some such.
> 
> It should be quite simple to add the feature to existing SCTP
> implementations.
> 
> --brian
> 
> 
> On Tue, 11 Feb 2003, Randall Stewart wrote:
> 
> > Michael:
> > 
> > This seems like a valid method.. it gets around Brian's objection
> > (which was a good point) and yet still provides the application with
> > a hint about what the assigned sequence order was on the unordered
> > messages...
> > 
> > R
> > 
> > Michael Tuexen wrote:
> > > Caitlin,
> > > 
> > > thank you very much for the clarification.
> > > 
> > > I do not like the idea to pass the SSN transparently from 
> the ULP to the 
> > > ULP.
> > > But having two sequence spaces is OK for me. This means 
> that the SSN at the
> > > sending side is controlled by the SCTP and given to the 
> ULP at the 
> > > receiving side.
> > > 
> > > This would be a clean solution. So I voting for have two 
> separate number 
> > > spaces
> > > for user data. Since the one for ordered delivery exists 
> already in RFC 
> > > 2960, this
> > > only means that a new one is introduced for the unordered 
> delivery user 
> > > data.
> > > 
> > > Best regards
> > > Michael
> > > 
> > > On Tuesday, Feb 11, 2003, at 17:37 Europe/Berlin, Caitlin 
> Bestler wrote:
> > > 
> > >> On 2/11/03, Tuexen Michael wrote:
> > >>
> > >>> Dear all,
> > >>>
> > >>> does this mean that you want two sequence spaces per
> > >>> stream: one for the 'ordered' user data and one for the
> > >>> unordered user data?
> > >>>
> > >> The RDDP/SCTP adaptation only uses unordered delivery, so
> > >> that question did not really arise.
> > >>
> > >> But if you were to generalize the problem, I believe the
> > >> answer would be yes.
> > >>
> > >> The SSN for ordered delivery would be unchanged. The
> > >> addition would be providing "advisory ordering" for
> > >> unordered delivery. The general characteristic is that
> > >> the receiving ULP wants the data in whatever order it
> > >> arrives, but wants to know what the original order was.
> > >>
> > >> In addition to RDDP, RTP over UDP is another example of
> > >> an application that wants to be aware of the original
> > >> ordering, even while accepting the data out-of-order.
> > >>
> > >> Caitlin Bestler
> > >> http://asomi.com/CaitlinBestler/
> > >> _______________________________________________
> > >> tsvwg mailing list
> > >> tsvwg@ietf.org
> > >> https://www1.ietf.org/mailman/listinfo/tsvwg
> > >>
> > > 
> > > _______________________________________________
> > > tsvwg mailing list
> > > tsvwg@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/tsvwg
> > > 
> > > 
> > 
> > 
> > -- 
> > Randall R. Stewart
> > randall@stewart.chicago.il.us 815-342-5222 (cell phone)
> > 
> > _______________________________________________
> > tsvwg mailing list
> > tsvwg@ietf.org
> > https://www1.ietf.org/mailman/listinfo/tsvwg
> 
> -- 
> Brian F. G. Bidulock    ¦ The reasonable man adapts himself to the ¦
> bidulock@openss7.org    ¦ world; the unreasonable one persists in  ¦
> http://www.openss7.org/ ¦ trying  to adapt the  world  to himself. ¦
>                         ¦ Therefore  all  progress  depends on the ¦
>                         ¦ unreasonable man. -- George Bernard Shaw ¦
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 09:32:45 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA03741
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 09:32:45 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CEgBb20440
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 09:42:11 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CEfwp20370;
	Wed, 12 Feb 2003 09:41:58 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CCIMp09668
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 07:18:22 -0500
Received: from beamer.mchh.siemens.de (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA00082;
	Wed, 12 Feb 2003 07:08:28 -0500 (EST)
Received: from blues.mchh.siemens.de ([139.21.204.206])
	by beamer.mchh.siemens.de (8.9.3/8.9.3) with ESMTP id NAA22625;
	Wed, 12 Feb 2003 13:12:11 +0100 (MET)
Received: from mchh273e.demchh201e.icn.siemens.de (mchh273e.mchh.siemens.de [139.21.200.83])
	by blues.mchh.siemens.de (8.9.1/8.9.1) with ESMTP id NAA27398;
	Wed, 12 Feb 2003 13:11:43 +0100 (MET)
Received: by mchh273e.mchh.siemens.de with Internet Mail Service (5.5.2656.59)
	id <1JB5XSNB>; Wed, 12 Feb 2003 13:11:19 +0100
Message-ID: <F92978223B01D311ACFF0008C71EE19D015DFAA8@MCHH225E>
From: Tuexen Michael <michael.tuexen@siemens.com>
To: "'Randall Stewart'" <randall@stewart.chicago.il.us>, bidulock@openss7.org,
        Michael Tuexen <Michael.Tuexen@lurchi.franken.de>
Cc: Caitlin Bestler <cait@asomi.com>, tsvwg@ietf.org,
        Tuexen Michael
	 <michael.tuexen@siemens.com>, sigtran@ietf.org
Date: Wed, 12 Feb 2003 13:11:19 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2656.59)
Content-Type: text/plain;
	charset="ISO-8859-1"
Subject: [Sigtran] RE: [Tsvwg] A question of Un-Ordered SCTP
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Randy,

I think having a FEATURES parameter would be a good idea. I do not
like the idea of bitmask. Instead of this I would propose something
like a FEATURES parameter which looks like:

       0                   1                   2                   3
       0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
      |   FEATURES_PARAMETER_ID       |       Parameter Length        |
      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
      |                         FEATURE_ID 1                          |
      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
      |                         FEATURE_ID 2                          |
      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
      |                         FEATURE_ID 3                          |
      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
      |                         FEATURE_ID 4                          |
      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

Each feature will be an IANA assiged 32 bit uint. This is extendable.

Best regards
Michael

Michael Tuexen   Tel.:   +49 89 722 47210
Siemens AG       Fax:    +49 89 722 48212
ICN CP SE C 7    e-mail: Michael.Tuexen@siemens.com

> -----Original Message-----
> From: Randall Stewart [mailto:randall@stewart.chicago.il.us]
> Sent: Wednesday, February 12, 2003 12:59 PM
> To: bidulock@openss7.org; Michael Tuexen
> Cc: Caitlin Bestler; tsvwg@ietf.org; Tuexen Michael; sigtran@ietf.org
> Subject: Re: [Tsvwg] A question of Un-Ordered SCTP
> 
> 
> Brian/Michael:
> 
> I do like the idea of keeping things the same (aka same
> methodology)... this is good.
> 
> It is also neat to be able to tell the difference between
> an implementation that supports this feature and one that
> does not.  But there are positives and negatives, I think,
> to each idea...
> 
> A) For using a TLV, (as Michael suggests)
>     we really need to put this into a seperate
>     draft.. when I do think one could easily fit
>     this into the IG as Brian as suggested.
> 
> B) montonically increasing numbers would eventually
>     wrap when the SSN hit FFFF back to 0... thus
>     the application would need to have some sort
>     of state to track if the peer supported the
>     feature.
> 
> It seems to me it is only the first message that
> arrives that benefits from the start at 0. If the
> application assumes a "not supported", then by
> starting the SSN at 1 the app can flip the flag
> from "not supported" to "supported" on the arrival
> of the first message.
> 
> If the application instead set its flag to "supported"
> the first message would not tell this information and
> instead it would only be upon the second arriving message
> that the app could make the flag determination... causing
> either a delay in delivery of the first message (yuck) or
> just toss it up and wait.
> 
> More importantly, I am wondering, if the implementation does
> NOT support this, what will the app do? Change its
> message structure? If so it will need to have some clues
> (as the initiator) that its implementation does not
> support the feature .. possibly a HAVE_SCTP_UNORDERED_SSN
> in sctp.h... but the problem is you would not know about
> the other guy until its first unordered message arrives.
> So if you have to send the first message what do you do?
> 
> if HAVE_SCTP_UNORDERED_SSN is 0.. no problem.. you
> send a message that has the imbedded sequence number.. but
> what do you send when HAVE_SCTP_UNRODERED_SSN is set to 1?
> 
> If you send the imbedded sequence in the message anyway whats
> the point of this little trick? The second message (which you
> may have to send before knowing receiving a unordered
> msg from the peer) would still have this, and maybe the third
> etc...
> 
> Hmm.. the more I think on this the more it becomes clear that
> you must know ahead of time i.e. when the association comes
> up to "do the right thing".... which goes back to Michaels
> idea of a TLV.
> 
> I have been wondering for quite some time now if we don't
> just need a TLV in the INIT/INIT-ACK that indicate features
> in general... something like a bit mask or byte array indicating
> each feature...
> 
> addip
> delip
> prsctp
> setprimary
> unordered-assigns-sequences
> 
> etc..
> 
> Maybe this is a better way to go.. and we could write a short
> draft dictating this bit/byte TLV... code as a 0x80 class.. that
> way old implementations will throw it away... (we really don't
> care if they don't understand it)...
> 
> I have been avoiding this.. but it might be the best approach..
> 
> Then again, I guess it could also be put in addip or prsctp since
> it would effect them as well :> Then one could put the
> other change in the IG and thus have a very easy change that
> is in the IG and a minor update to either addip or prsctp...
> 
> 
> What do you all think?
> 
> R
> 
> Brian F. G. Bidulock wrote:
> > Randall,
> > 
> > I agree to having SCTP set the sequence number and, as you said in
> > your first note, SCTP passing the sequence number to the ULP on
> > the receiving side is already done for ordered traffic.  Some even
> > pass the zero now.
> > 
> > I suppose that SCTP would set this sequence number when it is given
> > the data for delivery at the sending ULP and the sequence 
> would represent
> > the order that the data was presented to SCTP by the ULP at 
> the sending
> > side.
> > 
> > I think it would be good to see the requirement that the sequence
> > number begin at 1 and be monotonically increasing in the 
> saem fashion
> > as ordered streams.
> > 
> > That way RDDP and other applications could perform their 
> own reordering
> > at the receiver.  I think it would be a good thing to have SCTP set
> > this value.
> > 
> > In fact, I can see another application for this feature.  M2PA
> > (SIGTRAN protocol) needs in some circumstances to send data 
> unordered
> > with the hope of expedited delivery (does not want control 
> information
> > stuck behind large amounts of data in an ordered stream), 
> however, the
> > order of processing of the expedited control information is 
> important.
> > 
> > This facility would permit M2PA to send such information unordered
> > to expedite it and the receiver could perform the reordering itself.
> > Up until now, the only solution was to place a sequence 
> number in with
> > the control information.
> > 
> > Could we perhaps add a section to the SCTP IG.  "Problem statement:
> > the sender wishes to send unordered data but the receive 
> needs to know
> > the original order of sending..." or some such.
> > 
> > It should be quite simple to add the feature to existing SCTP
> > implementations.
> > 
> > --brian
> > 
> > 
> > On Tue, 11 Feb 2003, Randall Stewart wrote:
> > 
> > 
> >>Michael:
> >>
> >>This seems like a valid method.. it gets around Brian's objection
> >>(which was a good point) and yet still provides the application with
> >>a hint about what the assigned sequence order was on the unordered
> >>messages...
> >>
> >>R
> >>
> >>Michael Tuexen wrote:
> >>
> >>>Caitlin,
> >>>
> >>>thank you very much for the clarification.
> >>>
> >>>I do not like the idea to pass the SSN transparently from 
> the ULP to the 
> >>>ULP.
> >>>But having two sequence spaces is OK for me. This means 
> that the SSN at the
> >>>sending side is controlled by the SCTP and given to the ULP at the 
> >>>receiving side.
> >>>
> >>>This would be a clean solution. So I voting for have two 
> separate number 
> >>>spaces
> >>>for user data. Since the one for ordered delivery exists 
> already in RFC 
> >>>2960, this
> >>>only means that a new one is introduced for the unordered 
> delivery user 
> >>>data.
> >>>
> >>>Best regards
> >>>Michael
> >>>
> >>>On Tuesday, Feb 11, 2003, at 17:37 Europe/Berlin, Caitlin 
> Bestler wrote:
> >>>
> >>>
> >>>>On 2/11/03, Tuexen Michael wrote:
> >>>>
> >>>>
> >>>>>Dear all,
> >>>>>
> >>>>>does this mean that you want two sequence spaces per
> >>>>>stream: one for the 'ordered' user data and one for the
> >>>>>unordered user data?
> >>>>>
> >>>>
> >>>>The RDDP/SCTP adaptation only uses unordered delivery, so
> >>>>that question did not really arise.
> >>>>
> >>>>But if you were to generalize the problem, I believe the
> >>>>answer would be yes.
> >>>>
> >>>>The SSN for ordered delivery would be unchanged. The
> >>>>addition would be providing "advisory ordering" for
> >>>>unordered delivery. The general characteristic is that
> >>>>the receiving ULP wants the data in whatever order it
> >>>>arrives, but wants to know what the original order was.
> >>>>
> >>>>In addition to RDDP, RTP over UDP is another example of
> >>>>an application that wants to be aware of the original
> >>>>ordering, even while accepting the data out-of-order.
> >>>>
> >>>>Caitlin Bestler
> >>>>http://asomi.com/CaitlinBestler/
> >>>>_______________________________________________
> >>>>tsvwg mailing list
> >>>>tsvwg@ietf.org
> >>>>https://www1.ietf.org/mailman/listinfo/tsvwg
> >>>>
> >>>
> >>>_______________________________________________
> >>>tsvwg mailing list
> >>>tsvwg@ietf.org
> >>>https://www1.ietf.org/mailman/listinfo/tsvwg
> >>>
> >>>
> >>
> >>
> >>-- 
> >>Randall R. Stewart
> >>randall@stewart.chicago.il.us 815-342-5222 (cell phone)
> >>
> >>_______________________________________________
> >>tsvwg mailing list
> >>tsvwg@ietf.org
> >>https://www1.ietf.org/mailman/listinfo/tsvwg
> > 
> > 
> 
> 
> -- 
> Randall R. Stewart
> randall@stewart.chicago.il.us 815-342-5222 (cell phone)
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 09:32:58 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA03755
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 09:32:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CEgNO20472
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 09:42:23 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CEg5p20401;
	Wed, 12 Feb 2003 09:42:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CCiIp11766
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 07:44:18 -0500
Received: from beamer.mchh.siemens.de (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA00907;
	Wed, 12 Feb 2003 07:34:23 -0500 (EST)
Received: from moody.mchh.siemens.de ([139.21.205.85])
	by beamer.mchh.siemens.de (8.9.3/8.9.3) with ESMTP id NAA24579;
	Wed, 12 Feb 2003 13:38:06 +0100 (MET)
Received: from mchh247e.demchh201e.icn.siemens.de (mchh247e.mchh.siemens.de [139.21.200.57])
	by moody.mchh.siemens.de (8.9.1/8.9.1) with ESMTP id NAA14552;
	Wed, 12 Feb 2003 13:38:05 +0100 (MET)
Received: by mchh247e.mchh.siemens.de with Internet Mail Service (5.5.2656.59)
	id <146820DR>; Wed, 12 Feb 2003 13:37:16 +0100
Message-ID: <F92978223B01D311ACFF0008C71EE19D015DFAAA@MCHH225E>
From: Tuexen Michael <michael.tuexen@siemens.com>
To: "'Randall Stewart'" <randall@stewart.chicago.il.us>,
        Tuexen Michael
	 <michael.tuexen@siemens.com>
Cc: bidulock@openss7.org, Michael Tuexen
	 <Michael.Tuexen@lurchi.franken.de>,
        Caitlin Bestler <cait@asomi.com>, tsvwg@ietf.org, sigtran@ietf.org
Date: Wed, 12 Feb 2003 13:37:14 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2656.59)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Sigtran] RE: [Tsvwg] A question of Un-Ordered SCTP
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

32 bits were easier to draw... No problem in making
them 16 bit unsigned. 

Best regards
Michael

Michael Tuexen   Tel.:   +49 89 722 47210
Siemens AG       Fax:    +49 89 722 48212
ICN CP SE C 7    e-mail: Michael.Tuexen@siemens.com

> -----Original Message-----
> From: Randall Stewart [mailto:randall@stewart.chicago.il.us]
> Sent: Wednesday, February 12, 2003 1:25 PM
> To: Tuexen Michael
> Cc: bidulock@openss7.org; Michael Tuexen; Caitlin Bestler;
> tsvwg@ietf.org; sigtran@ietf.org
> Subject: Re: [Tsvwg] A question of Un-Ordered SCTP
> 
> 
> This looks good.. but would it be
> better to make the FEATURE_ID 16 bits..
> 
> .. my reasoning is that cookies are getting
> bigger and it might be best to control there
> length to some degree... I could go either way
> but I think 65,535 features is more than enough
> to choose from :>
> 
> R
> 
> Tuexen Michael wrote:
> > Randy,
> > 
> > I think having a FEATURES parameter would be a good idea. I do not
> > like the idea of bitmask. Instead of this I would propose something
> > like a FEATURES parameter which looks like:
> > 
> >        0                   1                   2                   3
> >        0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 
> 6 7 8 9 0 1
> >       
> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
> >       |   FEATURES_PARAMETER_ID       |       Parameter 
> Length        |
> >       
> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
> >       |                         FEATURE_ID 1                
>           |
> >       
> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
> >       |                         FEATURE_ID 2                
>           |
> >       
> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
> >       |                         FEATURE_ID 3                
>           |
> >       
> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
> >       |                         FEATURE_ID 4                
>           |
> >       
> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
> > 
> > Each feature will be an IANA assiged 32 bit uint. This is 
> extendable.
> > 
> > Best regards
> > Michael
> > 
> > Michael Tuexen   Tel.:   +49 89 722 47210
> > Siemens AG       Fax:    +49 89 722 48212
> > ICN CP SE C 7    e-mail: Michael.Tuexen@siemens.com
> > 
> > 
> >>-----Original Message-----
> >>From: Randall Stewart [mailto:randall@stewart.chicago.il.us]
> >>Sent: Wednesday, February 12, 2003 12:59 PM
> >>To: bidulock@openss7.org; Michael Tuexen
> >>Cc: Caitlin Bestler; tsvwg@ietf.org; Tuexen Michael; 
> sigtran@ietf.org
> >>Subject: Re: [Tsvwg] A question of Un-Ordered SCTP
> >>
> >>
> >>Brian/Michael:
> >>
> >>I do like the idea of keeping things the same (aka same
> >>methodology)... this is good.
> >>
> >>It is also neat to be able to tell the difference between
> >>an implementation that supports this feature and one that
> >>does not.  But there are positives and negatives, I think,
> >>to each idea...
> >>
> >>A) For using a TLV, (as Michael suggests)
> >>    we really need to put this into a seperate
> >>    draft.. when I do think one could easily fit
> >>    this into the IG as Brian as suggested.
> >>
> >>B) montonically increasing numbers would eventually
> >>    wrap when the SSN hit FFFF back to 0... thus
> >>    the application would need to have some sort
> >>    of state to track if the peer supported the
> >>    feature.
> >>
> >>It seems to me it is only the first message that
> >>arrives that benefits from the start at 0. If the
> >>application assumes a "not supported", then by
> >>starting the SSN at 1 the app can flip the flag
> >>from "not supported" to "supported" on the arrival
> >>of the first message.
> >>
> >>If the application instead set its flag to "supported"
> >>the first message would not tell this information and
> >>instead it would only be upon the second arriving message
> >>that the app could make the flag determination... causing
> >>either a delay in delivery of the first message (yuck) or
> >>just toss it up and wait.
> >>
> >>More importantly, I am wondering, if the implementation does
> >>NOT support this, what will the app do? Change its
> >>message structure? If so it will need to have some clues
> >>(as the initiator) that its implementation does not
> >>support the feature .. possibly a HAVE_SCTP_UNORDERED_SSN
> >>in sctp.h... but the problem is you would not know about
> >>the other guy until its first unordered message arrives.
> >>So if you have to send the first message what do you do?
> >>
> >>if HAVE_SCTP_UNORDERED_SSN is 0.. no problem.. you
> >>send a message that has the imbedded sequence number.. but
> >>what do you send when HAVE_SCTP_UNRODERED_SSN is set to 1?
> >>
> >>If you send the imbedded sequence in the message anyway whats
> >>the point of this little trick? The second message (which you
> >>may have to send before knowing receiving a unordered
> >>msg from the peer) would still have this, and maybe the third
> >>etc...
> >>
> >>Hmm.. the more I think on this the more it becomes clear that
> >>you must know ahead of time i.e. when the association comes
> >>up to "do the right thing".... which goes back to Michaels
> >>idea of a TLV.
> >>
> >>I have been wondering for quite some time now if we don't
> >>just need a TLV in the INIT/INIT-ACK that indicate features
> >>in general... something like a bit mask or byte array indicating
> >>each feature...
> >>
> >>addip
> >>delip
> >>prsctp
> >>setprimary
> >>unordered-assigns-sequences
> >>
> >>etc..
> >>
> >>Maybe this is a better way to go.. and we could write a short
> >>draft dictating this bit/byte TLV... code as a 0x80 class.. that
> >>way old implementations will throw it away... (we really don't
> >>care if they don't understand it)...
> >>
> >>I have been avoiding this.. but it might be the best approach..
> >>
> >>Then again, I guess it could also be put in addip or prsctp since
> >>it would effect them as well :> Then one could put the
> >>other change in the IG and thus have a very easy change that
> >>is in the IG and a minor update to either addip or prsctp...
> >>
> >>
> >>What do you all think?
> >>
> >>R
> >>
> >>Brian F. G. Bidulock wrote:
> >>
> >>>Randall,
> >>>
> >>>I agree to having SCTP set the sequence number and, as you said in
> >>>your first note, SCTP passing the sequence number to the ULP on
> >>>the receiving side is already done for ordered traffic.  Some even
> >>>pass the zero now.
> >>>
> >>>I suppose that SCTP would set this sequence number when it is given
> >>>the data for delivery at the sending ULP and the sequence 
> >>
> >>would represent
> >>
> >>>the order that the data was presented to SCTP by the ULP at 
> >>
> >>the sending
> >>
> >>>side.
> >>>
> >>>I think it would be good to see the requirement that the sequence
> >>>number begin at 1 and be monotonically increasing in the 
> >>
> >>saem fashion
> >>
> >>>as ordered streams.
> >>>
> >>>That way RDDP and other applications could perform their 
> >>
> >>own reordering
> >>
> >>>at the receiver.  I think it would be a good thing to have SCTP set
> >>>this value.
> >>>
> >>>In fact, I can see another application for this feature.  M2PA
> >>>(SIGTRAN protocol) needs in some circumstances to send data 
> >>
> >>unordered
> >>
> >>>with the hope of expedited delivery (does not want control 
> >>
> >>information
> >>
> >>>stuck behind large amounts of data in an ordered stream), 
> >>
> >>however, the
> >>
> >>>order of processing of the expedited control information is 
> >>
> >>important.
> >>
> >>>This facility would permit M2PA to send such information unordered
> >>>to expedite it and the receiver could perform the 
> reordering itself.
> >>>Up until now, the only solution was to place a sequence 
> >>
> >>number in with
> >>
> >>>the control information.
> >>>
> >>>Could we perhaps add a section to the SCTP IG.  "Problem statement:
> >>>the sender wishes to send unordered data but the receive 
> >>
> >>needs to know
> >>
> >>>the original order of sending..." or some such.
> >>>
> >>>It should be quite simple to add the feature to existing SCTP
> >>>implementations.
> >>>
> >>>--brian
> >>>
> >>>
> >>>On Tue, 11 Feb 2003, Randall Stewart wrote:
> >>>
> >>>
> >>>
> >>>>Michael:
> >>>>
> >>>>This seems like a valid method.. it gets around Brian's objection
> >>>>(which was a good point) and yet still provides the 
> application with
> >>>>a hint about what the assigned sequence order was on the unordered
> >>>>messages...
> >>>>
> >>>>R
> >>>>
> >>>>Michael Tuexen wrote:
> >>>>
> >>>>
> >>>>>Caitlin,
> >>>>>
> >>>>>thank you very much for the clarification.
> >>>>>
> >>>>>I do not like the idea to pass the SSN transparently from 
> >>>>
> >>the ULP to the 
> >>
> >>>>>ULP.
> >>>>>But having two sequence spaces is OK for me. This means 
> >>>>
> >>that the SSN at the
> >>
> >>>>>sending side is controlled by the SCTP and given to the 
> ULP at the 
> >>>>>receiving side.
> >>>>>
> >>>>>This would be a clean solution. So I voting for have two 
> >>>>
> >>separate number 
> >>
> >>>>>spaces
> >>>>>for user data. Since the one for ordered delivery exists 
> >>>>
> >>already in RFC 
> >>
> >>>>>2960, this
> >>>>>only means that a new one is introduced for the unordered 
> >>>>
> >>delivery user 
> >>
> >>>>>data.
> >>>>>
> >>>>>Best regards
> >>>>>Michael
> >>>>>
> >>>>>On Tuesday, Feb 11, 2003, at 17:37 Europe/Berlin, Caitlin 
> >>>>
> >>Bestler wrote:
> >>
> >>>>>
> >>>>>>On 2/11/03, Tuexen Michael wrote:
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>>>Dear all,
> >>>>>>>
> >>>>>>>does this mean that you want two sequence spaces per
> >>>>>>>stream: one for the 'ordered' user data and one for the
> >>>>>>>unordered user data?
> >>>>>>>
> >>>>>>
> >>>>>>The RDDP/SCTP adaptation only uses unordered delivery, so
> >>>>>>that question did not really arise.
> >>>>>>
> >>>>>>But if you were to generalize the problem, I believe the
> >>>>>>answer would be yes.
> >>>>>>
> >>>>>>The SSN for ordered delivery would be unchanged. The
> >>>>>>addition would be providing "advisory ordering" for
> >>>>>>unordered delivery. The general characteristic is that
> >>>>>>the receiving ULP wants the data in whatever order it
> >>>>>>arrives, but wants to know what the original order was.
> >>>>>>
> >>>>>>In addition to RDDP, RTP over UDP is another example of
> >>>>>>an application that wants to be aware of the original
> >>>>>>ordering, even while accepting the data out-of-order.
> >>>>>>
> >>>>>>Caitlin Bestler
> >>>>>>http://asomi.com/CaitlinBestler/
> >>>>>>_______________________________________________
> >>>>>>tsvwg mailing list
> >>>>>>tsvwg@ietf.org
> >>>>>>https://www1.ietf.org/mailman/listinfo/tsvwg
> >>>>>>
> >>>>>
> >>>>>_______________________________________________
> >>>>>tsvwg mailing list
> >>>>>tsvwg@ietf.org
> >>>>>https://www1.ietf.org/mailman/listinfo/tsvwg
> >>>>>
> >>>>>
> >>>>
> >>>>
> >>>>-- 
> >>>>Randall R. Stewart
> >>>>randall@stewart.chicago.il.us 815-342-5222 (cell phone)
> >>>>
> >>>>_______________________________________________
> >>>>tsvwg mailing list
> >>>>tsvwg@ietf.org
> >>>>https://www1.ietf.org/mailman/listinfo/tsvwg
> >>>
> >>>
> >>
> >>-- 
> >>Randall R. Stewart
> >>randall@stewart.chicago.il.us 815-342-5222 (cell phone)
> >>
> > 
> > 
> > 
> 
> 
> -- 
> Randall R. Stewart
> randall@stewart.chicago.il.us 815-342-5222 (cell phone)
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 09:33:00 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA03768
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 09:33:00 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CEgPv20485
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 09:42:25 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CEgCp20455;
	Wed, 12 Feb 2003 09:42:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CCrvp12337
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 07:53:57 -0500
Received: from gorilla.mchh.siemens.de (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA01166;
	Wed, 12 Feb 2003 07:44:01 -0500 (EST)
Received: from blues.mchh.siemens.de ([139.21.204.206])
	by gorilla.mchh.siemens.de (8.9.3/8.9.3) with ESMTP id NAA18331;
	Wed, 12 Feb 2003 13:47:42 +0100 (MET)
Received: from mchh253e.mchh.siemens.de (mchh253e.mchh.siemens.de [139.21.200.63])
	by blues.mchh.siemens.de (8.9.1/8.9.1) with ESMTP id NAA02502;
	Wed, 12 Feb 2003 13:47:18 +0100 (MET)
Received: by MCHH253E with Internet Mail Service (5.5.2653.19)
	id <D131R71H>; Wed, 12 Feb 2003 13:46:53 +0100
Message-ID: <F92978223B01D311ACFF0008C71EE19D015DFAAB@MCHH225E>
From: Tuexen Michael <michael.tuexen@siemens.com>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>,
        Randall Stewart
	 <randall@stewart.chicago.il.us>
Cc: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>,
        Caitlin Bestler
	 <cait@asomi.com>, tsvwg@ietf.org,
        Tuexen Michael
	 <michael.tuexen@siemens.com>, sigtran@ietf.org
Date: Wed, 12 Feb 2003 13:46:48 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Sigtran] RE: [Tsvwg] A question of Un-Ordered SCTP
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Brian,

I disagree. The upper layer can say that he is 
interested in the order or not by using the
ordered delivery service or not. The peer gets
in both cases an indication of the sending sequence.

You are proposing, that the ULP say I'm not interested
in ordered delivery, but I would like to use unordered
delivery. And now you have some Upper layer information
that some messages belong to two classes (which you
want to code SSN = 0 or != 0). From an SCTP service
point of view there is NO difference. All messages
marked for unordered delivery are given to the ULP
as soon as possible.

I think you want to encode some upper layer information
by using SSN=0 or SSN=!0. For me this looks like layer violation.

Best regards
Michael

Michael Tuexen   Tel.:   +49 89 722 47210
Siemens AG       Fax:    +49 89 722 48212
ICN CP SE C 7    e-mail: Michael.Tuexen@siemens.com

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Wednesday, February 12, 2003 1:22 PM
> To: Randall Stewart
> Cc: Michael Tuexen; Caitlin Bestler; tsvwg@ietf.org; Tuexen Michael;
> sigtran@ietf.org
> Subject: Re: [Tsvwg] A question of Un-Ordered SCTP
> 
> 
> Randall,
> 
> For the M2PA application, I think that reserving
> '0' for unordered messages for which the order of
> sending is unimportant is still a good idea.
> So the number space from 1 to 0xffff indicates
> importance of order.  '0' indicates no importance
> of order.  As in M2PA's case, a sender could mix
> both on the same stream.
> 
> There then becomes three different types of messages:
> 
> 1) Unordered messages for which the order of sending is
>    unimportant (U-bit SSN=0),
> 
> 2) unordered messages for which the order of sending is
>    important (U-bit SSN!=0), and
> 
> 3) ordered messages for which the order of sending is
>    important (~U-bit).
> 
> And this also handles compatibility because previous
> unordered messages fall into category (1) by nature
> of the fact that SSN==0.
> 
> But starting counting at 1, you get rid of the ability
> to indicate that the order of sending is unimportant and
> force the receiving ULP to handle ordering of all received
> unordered data.
> 
> So, in short, start counting at 1, regardless of whether
> there is an indicator in the INIT/INIT-ACK.
> 
> --brian
> 
> Randall Stewart wrote:                                        
>                      (Wed, 12 Feb 2003 05:58:55)
> > Brian/Michael:
> > 
> > I do like the idea of keeping things the same (aka same
> > methodology)... this is good.
> > 
> > It is also neat to be able to tell the difference between
> > an implementation that supports this feature and one that
> > does not.  But there are positives and negatives, I think,
> > to each idea...
> > 
> > A) For using a TLV, (as Michael suggests)
> >     we really need to put this into a seperate
> >     draft.. when I do think one could easily fit
> >     this into the IG as Brian as suggested.
> > 
> > B) montonically increasing numbers would eventually
> >     wrap when the SSN hit FFFF back to 0... thus
> >     the application would need to have some sort
> >     of state to track if the peer supported the
> >     feature.
> > 
> > It seems to me it is only the first message that
> > arrives that benefits from the start at 0. If the
> > application assumes a "not supported", then by
> > starting the SSN at 1 the app can flip the flag
> > from "not supported" to "supported" on the arrival
> > of the first message.
> > 
> > If the application instead set its flag to "supported"
> > the first message would not tell this information and
> > instead it would only be upon the second arriving message
> > that the app could make the flag determination... causing
> > either a delay in delivery of the first message (yuck) or
> > just toss it up and wait.
> > 
> > More importantly, I am wondering, if the implementation does
> > NOT support this, what will the app do? Change its
> > message structure? If so it will need to have some clues
> > (as the initiator) that its implementation does not
> > support the feature .. possibly a HAVE_SCTP_UNORDERED_SSN
> > in sctp.h... but the problem is you would not know about
> > the other guy until its first unordered message arrives.
> > So if you have to send the first message what do you do?
> > 
> > if HAVE_SCTP_UNORDERED_SSN is 0.. no problem.. you
> > send a message that has the imbedded sequence number.. but
> > what do you send when HAVE_SCTP_UNRODERED_SSN is set to 1?
> > 
> > If you send the imbedded sequence in the message anyway whats
> > the point of this little trick? The second message (which you
> > may have to send before knowing receiving a unordered
> > msg from the peer) would still have this, and maybe the third
> > etc...
> > 
> > Hmm.. the more I think on this the more it becomes clear that
> > you must know ahead of time i.e. when the association comes
> > up to "do the right thing".... which goes back to Michaels
> > idea of a TLV.
> > 
> > I have been wondering for quite some time now if we don't
> > just need a TLV in the INIT/INIT-ACK that indicate features
> > in general... something like a bit mask or byte array indicating
> > each feature...
> > 
> > addip
> > delip
> > prsctp
> > setprimary
> > unordered-assigns-sequences
> > 
> > etc..
> > 
> > Maybe this is a better way to go.. and we could write a short
> > draft dictating this bit/byte TLV... code as a 0x80 class.. that
> > way old implementations will throw it away... (we really don't
> > care if they don't understand it)...
> > 
> > I have been avoiding this.. but it might be the best approach..
> > 
> > Then again, I guess it could also be put in addip or prsctp since
> > it would effect them as well :> Then one could put the
> > other change in the IG and thus have a very easy change that
> > is in the IG and a minor update to either addip or prsctp...
> > 
> > 
> > What do you all think?
> > 
> > R
> > 
> > Brian F. G. Bidulock wrote:
> > > Randall,
> > > 
> > > I agree to having SCTP set the sequence number and, as you said in
> > > your first note, SCTP passing the sequence number to the ULP on
> > > the receiving side is already done for ordered traffic.  Some even
> > > pass the zero now.
> > > 
> > > I suppose that SCTP would set this sequence number when 
> it is given
> > > the data for delivery at the sending ULP and the sequence 
> would represent
> > > the order that the data was presented to SCTP by the ULP 
> at the sending
> > > side.
> > > 
> > > I think it would be good to see the requirement that the sequence
> > > number begin at 1 and be monotonically increasing in the 
> saem fashion
> > > as ordered streams.
> > > 
> > > That way RDDP and other applications could perform their 
> own reordering
> > > at the receiver.  I think it would be a good thing to 
> have SCTP set
> > > this value.
> > > 
> > > In fact, I can see another application for this feature.  M2PA
> > > (SIGTRAN protocol) needs in some circumstances to send 
> data unordered
> > > with the hope of expedited delivery (does not want 
> control information
> > > stuck behind large amounts of data in an ordered stream), 
> however, the
> > > order of processing of the expedited control information 
> is important.
> > > 
> > > This facility would permit M2PA to send such information unordered
> > > to expedite it and the receiver could perform the 
> reordering itself.
> > > Up until now, the only solution was to place a sequence 
> number in with
> > > the control information.
> > > 
> > > Could we perhaps add a section to the SCTP IG.  "Problem 
> statement:
> > > the sender wishes to send unordered data but the receive 
> needs to know
> > > the original order of sending..." or some such.
> > > 
> > > It should be quite simple to add the feature to existing SCTP
> > > implementations.
> > > 
> > > --brian
> > > 
> > > 
> > > On Tue, 11 Feb 2003, Randall Stewart wrote:
> > > 
> > > 
> > >>Michael:
> > >>
> > >>This seems like a valid method.. it gets around Brian's objection
> > >>(which was a good point) and yet still provides the 
> application with
> > >>a hint about what the assigned sequence order was on the unordered
> > >>messages...
> > >>
> > >>R
> > >>
> > >>Michael Tuexen wrote:
> > >>
> > >>>Caitlin,
> > >>>
> > >>>thank you very much for the clarification.
> > >>>
> > >>>I do not like the idea to pass the SSN transparently 
> from the ULP to the 
> > >>>ULP.
> > >>>But having two sequence spaces is OK for me. This means 
> that the SSN at the
> > >>>sending side is controlled by the SCTP and given to the 
> ULP at the 
> > >>>receiving side.
> > >>>
> > >>>This would be a clean solution. So I voting for have two 
> separate number 
> > >>>spaces
> > >>>for user data. Since the one for ordered delivery exists 
> already in RFC 
> > >>>2960, this
> > >>>only means that a new one is introduced for the 
> unordered delivery user 
> > >>>data.
> > >>>
> > >>>Best regards
> > >>>Michael
> > >>>
> > >>>On Tuesday, Feb 11, 2003, at 17:37 Europe/Berlin, 
> Caitlin Bestler wrote:
> > >>>
> > >>>
> > >>>>On 2/11/03, Tuexen Michael wrote:
> > >>>>
> > >>>>
> > >>>>>Dear all,
> > >>>>>
> > >>>>>does this mean that you want two sequence spaces per
> > >>>>>stream: one for the 'ordered' user data and one for the
> > >>>>>unordered user data?
> > >>>>>
> > >>>>
> > >>>>The RDDP/SCTP adaptation only uses unordered delivery, so
> > >>>>that question did not really arise.
> > >>>>
> > >>>>But if you were to generalize the problem, I believe the
> > >>>>answer would be yes.
> > >>>>
> > >>>>The SSN for ordered delivery would be unchanged. The
> > >>>>addition would be providing "advisory ordering" for
> > >>>>unordered delivery. The general characteristic is that
> > >>>>the receiving ULP wants the data in whatever order it
> > >>>>arrives, but wants to know what the original order was.
> > >>>>
> > >>>>In addition to RDDP, RTP over UDP is another example of
> > >>>>an application that wants to be aware of the original
> > >>>>ordering, even while accepting the data out-of-order.
> > >>>>
> > >>>>Caitlin Bestler
> > >>>>http://asomi.com/CaitlinBestler/
> > >>>>_______________________________________________
> > >>>>tsvwg mailing list
> > >>>>tsvwg@ietf.org
> > >>>>https://www1.ietf.org/mailman/listinfo/tsvwg
> > >>>>
> > >>>
> > >>>_______________________________________________
> > >>>tsvwg mailing list
> > >>>tsvwg@ietf.org
> > >>>https://www1.ietf.org/mailman/listinfo/tsvwg
> > >>>
> > >>>
> > >>
> > >>
> > >>-- 
> > >>Randall R. Stewart
> > >>randall@stewart.chicago.il.us 815-342-5222 (cell phone)
> > >>
> > >>_______________________________________________
> > >>tsvwg mailing list
> > >>tsvwg@ietf.org
> > >>https://www1.ietf.org/mailman/listinfo/tsvwg
> > > 
> > > 
> > 
> > 
> > -- 
> > Randall R. Stewart
> > randall@stewart.chicago.il.us 815-342-5222 (cell phone)
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 09:47:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA04585
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 09:47:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CEvNq21405
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 09:57:23 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CEv7p21377;
	Wed, 12 Feb 2003 09:57:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CEtDp21280
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 09:55:13 -0500
Received: from rtp-core-2.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA04463
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 09:45:16 -0500 (EST)
Received: from cia.cisco.com (IDENT:mirapoint@cia.cisco.com [161.44.85.200])
	by rtp-core-2.cisco.com (8.12.6/8.12.6) with ESMTP id h1CEmvNh024501;
	Wed, 12 Feb 2003 09:48:57 -0500 (EST)
Received: from kmorneau-w2k.cisco.com (rtp-vpn1-533.cisco.com [10.82.226.21])
	by cia.cisco.com (Mirapoint)
	with ESMTP id ABP06248;
	Wed, 12 Feb 2003 09:38:57 -0500 (EST)
Message-Id: <4.3.2.7.2.20030212094830.04b6cf50@cia.cisco.com>
X-Sender: kmorneau@cia.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Wed, 12 Feb 2003 09:48:55 -0500
To: Tuexen Michael <michael.tuexen@siemens.com>,
        "'Tom George'" <Tom.George@alcatel.com>
From: Ken Morneault <kmorneau@cisco.com>
Subject: RE: [Sigtran] M2PA Security section (was: m2pa comment: author
  li st)
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
In-Reply-To: <F92978223B01D311ACFF0008C71EE19D015DFA9C@MCHH225E>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>



Tom,

Also, it should be a "normative" reference.

Ken

At 09:25 AM 2/12/2003 +0100, Tuexen Michael wrote:
>Hi,
>
>
>why not simply referring to the SIGTRAN security doc instead
>of reproducing part of it?
>
>Best regards
>Michael
>
>Michael Tuexen   Tel.:   +49 89 722 47210
>Siemens AG       Fax:    +49 89 722 48212
>ICN CP SE C 7    e-mail: Michael.Tuexen@siemens.com
>
>> -----Original Message-----
>> From: Tom George [mailto:Tom.George@alcatel.com]
>> Sent: Tuesday, February 11, 2003 10:58 PM
>> Cc: 'sigtran@ietf.org'
>> Subject: [Sigtran] M2PA Security section (was: m2pa comment: author
>> list)
>> 
>> 
>> On Jan 26, 2003 "Ong, Lyndon" wrote:
>>   [...]
>> > The security section probably needs to be updated based on the
>> > security framework document (which should come out this week).
>>   [...]
>> 
>> Here is my first stab at revising section 6 of the M2PA 
>> draft. Comments
>> are welcome, especially since I am not familiar with the security
>> protocols.
>> 
>> tomG.
>> 
>> =============================
>> 6. Security
>> 
>> M2PA is designed to carry signaling messages for telephony
>> services. As such, M2PA MUST involve the security needs of several
>> parties: 
>> 
>>    - the end users of the services
>> 
>>    - the network providers
>> 
>>    - the applications involved
>> 
>> Additional requirements MAY come from local regulation.  
>> 
>> While these parties may have some overlapping security needs, their
>> needs may not be identical. Any security solution SHOULD fulfill all
>> of the different parties' needs.
>> 
>> 6.1 Security Requirements
>> 
>> As a transport protocol, M2PA has the following security objectives:
>> 
>>    - Availability of reliable and timely user data transport.
>> 
>>    - Integrity of user data transport.
>> 
>>    - Confidentiality of user data.
>> 
>> M2PA runs on top of SCTP [RFC2960].  SCTP provides certain transport
>> related security features, as discussed in [Security]. However, these
>> security features MAY NOT be adequate for all of the involved parties'
>> requirements.
>> 
>> When M2PA is running in professionally managed corporate or service
>> provider network, it is reasonable to expect that this network
>> includes an appropriate security policy framework. The "Site Security
>> Handbook" [RFC2196] SHOULD be consulted for guidance.
>> 
>> When the network in which M2PA runs involves more than one party
>> (e.g., a non-private network), it MAY NOT be reasonable to expect that
>> all parties have implemented security in a sufficient manner.  In such
>> a case, it is recommended that IPSec or TLS be used to ensure
>> confidentiality of the user payload. Consult [Security] for more
>> guidance on the use of security protocols. 
>> 
>> 6.2 Protecting Confidentiality
>> 
>> Particularly for mobile users, the requirement for confidentiality MAY
>> include the masking of IP addresses and ports. In this case
>> application-level encryption is not sufficient. IPSec ESP SHOULD be
>> used instead [RFC2401].  Regardless of which level performs the
>> encryption, the IPSec ISAKMP service SHOULD be used for key
>> management.
>> 
>> =============================
>> 
>> {Add an informative reference:}
>> 
>>   [Security]
>>        J. Loughney, M. Tuexen, and J. Pastor-Balbas, "Security
>>        Considerations for SIGTRAN Protocols,"
>>        draft-ietf-sigtran-security-01.txt (January 2003).
>> 
>> =============================
>> 
>> -- 
>> * Tom George       PB7-24001 (near B-21)      
>> Tom.George@alcatel.com  *
>> * 972-519-3168   Intranet:  
>http://www.ssd.usa.alcatel.com/~tgeorge/  *
>* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
>* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
>* to advertise or represent its position,  operations, or activities. *
>_______________________________________________
>Sigtran mailing list
>Sigtran@ietf.org
>https://www1.ietf.org/mailman/listinfo/sigtran
>_______________________________________________
>Sigtran mailing list
>Sigtran@ietf.org
>https://www1.ietf.org/mailman/listinfo/sigtran 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 10:03:00 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA05366
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 10:03:00 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CFCQN22728
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 10:12:26 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CFCBp22708;
	Wed, 12 Feb 2003 10:12:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CFBtp22685
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 10:11:55 -0500
Received: from domino2.nc.tekelec.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA05344
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 10:01:57 -0500 (EST)
From: jeffrey.craig@tekelec.com
Received: from domino4.nc.tekelec.com ([198.89.40.9])
          by domino2.nc.tekelec.com (Lotus Domino Release 5.0.10)
          with ESMTP id 2003021209591974:32494 ;
          Wed, 12 Feb 2003 09:59:19 -0500 
Subject: Re: [Sigtran] M2PA draft 7 question
To: wayne n taylor <waynet@cisco.com>
Cc: Tom George <Tom.George@alcatel.com>, bidulock@openss7.org,
        sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.7  March 21, 2001
Message-ID: <OFE387804C.FEA1E263-ON85256CCB.0050BD83@nc.tekelec.com>
Date: Wed, 12 Feb 2003 10:01:23 -0500
MIME-Version: 1.0
X-MIMETrack: Serialize by Router on Domino4/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/12/2003 10:01:24 AM,
	Itemize by SMTP Server on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/12/2003 09:59:19 AM,
	Serialize by Router on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/12/2003 09:59:25 AM,
	Serialize complete at 02/12/2003 09:59:25 AM
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hello Guys,

I think that we have consensus that the M2PA finite state machine
should match the Q.703 SDLs, where possible. It has been several
M2PA revisions now that the state machine has been updated, but
fallen short of matching Q.703.

I agree with Brian's suggested M2PA changes (below) with the exception of
the
following:

>> The first thing a link should send is LS OOS.
>>
>> In the out of service state, it would be desireable for the out of
>> service side to always send LS OOS in response to other messages.
>>
>> Also, the out of service side should be permitted to send LS OOS
>> periodically.

What does 'first thing' mean? (upon START? upon association becoming
established, as the current draft states?)

What problem is being solved by these three changes? Has anyone
observed alignment loops at plugtests or in testing?

Where in Q.703 does it specify these behaviors?

Do we have consensus that the changes will not result in a "potential loop"
during alignment, as was the initial concern? I am not convinced that
these three behaviors are necessary or desirable.

Regards,

Jeff

On 02/11/2003 04:06:34 PM sigtran-admin wrote:

>Tom,
>as i described earlier, my concern is the potential loop that could
>occur during the initial LS Alignment exchange based on the requirement
>that both peers must send and receive LS Alignment messages. Brian has
>shown in his response (and i agree) that the scenario i described will
>not happen if the procedures defined in the SDLs of the Q.703 spec is
>adopted for m2pa in regards to the initial "alignment" exchange.
>
>wayne
>
>"Brian F. G. Bidulock" wrote:
>>
>> Tom,
>>
>> The first thing a link should send is LS OOS.
>>
>> In the out of service state, it would be desireable for the out of
>> service side to always send LS OOS in response to other messages.
>>
>> Also, the out of service side should be permitted to send LS OOS
>> periodically.
>>
>> With regards to Q.703 state machines, LS OOS in response to alignment
>> will not abort alignment and T2 should still expire.  See test case
>> "3.1.2 Timer T2" in draft-bidulock-sigtran-m2pa-test-02.txt.
>>
>> T2 is defined as the duration of time that the M2PA peer wil wait to
>> receive a status "Alignment" message after sending a status "Alignment"
>> message.
>>
>> As status Alignment will abort an otherwise in-service link, other
>> link status received while running T2 should be ignored.
>>
>> Any different behaviour could interfere with MTP Level 3's link
>> restoration timers and procedures.
>>
>> More comments below...
>>
>> Tom George
>wrote:

>(Mon, 10 Feb 2003 16:07:21)
>> > Wayne,
>> >
>> > Yes, I want both sides to send and receive a LS Alignment message to
>> > complete alignment.
>> >
>> > To fix the problem you describe, how about this" If M2PA has not
>> > received MTP3 Start and receives LS Alignment, M2PA replies with LS
OOS
>> > (instead of not replying at all and counting on the other end's T2
>> > expiring).
>> >
>> > tomG.
>> >
>> > wayne n taylor wrote:
>> > >
>> > > Hi All,
>> > >
>> > > i have a couple of comments/questions about the initial exchange of
the
>> > > Link Status Alignment messages. there seems to be some inconsistency
in
>> > > the text and finite state machine.
>> > >
>> > > section 4.1.3 paragraph 4:
>> > >
>> > > "Once the association is established and M2PA has received a Start
>> > > Request from MTP3, M2PA sends the Link Status Alignment message to
its
>> > > peer. If M2PA has not already received the Link Status Alignment
>> > > message from its peer, then M2PA starts timer T2."
>> > >
>> > > what state would the local M2PA be in to receive a LS Alignment
message
>> > > from its peer prior to getting a MTP3 Start ?. i assume the OOS
state,
>> > > but the finite state machine does not show LS Alignment as an input.
>> > >
>> > > section 4.1.3 paragraph 5:
>> > >
>> > > "(Note that if the remote M2PA has not received a Start Request from
>> > > its MTP3, it will not send the Link Status Alignment message to the
>> > > local M2PA. Eventually timer T2 in the local M2PA will expire. If
the
>> > > remote M2PA receives a Start Request from its MTP3 and sends Link
>> > > Status Alignment before the local M2PA's timer T2 expires, the
>> > > alignment procedure can continue.)"
>> > >
>> > > assume the local M2PA receives a Start Request from MTP3, sends LS
>> > > Alignment and start T2, but the remote peer has not received a Start
>> > > Request and is in OOS. the current finite state machine does not
show
>> > > what to do with the LS Alignment from the local peer so lets assume
the
>> > > LS Alignment is dropped. the remote peer receives a Start Request
from
>> > > MTP3, sends LS Alignment and start T2 (since the remote peer didnt
get
>> > > the local peers LS Alignment). assume further that timer T2 is still
>> > > running on the local peer. the text from section 4.1.3 above states
the
>> > > "alignment procedure can continue". in this case the local peer sent
a
>> > > LS Alignment and received a LS Alignment. the remote peer only sent
a LS
>> > > Alignment and did not receive an LS Alignment (since LS Alignment
>> > > messages are not retried). eventually T2 on the remote peer will
expire,
>> > > T3 on the local peer will expire and the alignment procedure starts
over
>> > > again. it seems possible for the peers to get caught in a loop
trying to
>> > > exchange LS Alignment messages.
>>
>> See draft-bidulock-sigtran-m2pa-test-02.txt Section "3.1.3 Timer T3".
You
>> will see that what you describe does not happen if you follow Q.703
SDLs.
>> In Q.703 once SIO is sent SIN is sufficient to complete alignment.  So,
>> one side sends ALIGNMENT and starts T2, the other side sends ALIGNMENT
and
>> starts T2.  The first side, receives the alignment, stops T2 and sends
>> PROVING-NORMAL or PROVING-EMERGENCY and starts T3.  When the second side
>> receives PROVING-NORMAL or PROVING-EMERGENCY it stops T2 and sends
>> PROVING-NORMAL or PROVING-EMERGENCY and starts T4.  When the first side
>> receives PROVING-NORMAL it stops T3 and starts T4.  When either T4
expires,
>> the side sends IN-SERVICE and starts T1.  When the side receives
IN-SERVICE
>> it stops T1 and indicates in_service to MTP3.
>>
>> This behavior can be seen on Figure 9/Q.703 (sheet 2 of 6) - Initial
alignment
>> control.  When the IAC starts it sends SIO and starts T2.  If it
receives
>> any of SIO, SIN or SIE it stops T2 and send SIN or SIE depending on
emergency
>> status and starts T3.  Receipt of a subsequent SIN or SIE (sheet 3 of 6)
>> stops T3 and starts T4.  Note that if it receives SIO again, it ignores
it
>> at that point and continues to send SIE or SIN.
>>
>> These is no reason for M2PA not to follow the Q.703 SDLs in this regard.
>>
>> A statement that might clarify is that the M2PA sends LS ALIGNMENT and
if
>> it receives ALIGNMENT or PROVING it stops T2 and PROVING and starts T3.
>> Note that it takes two PROVING messages in this case to move the peer
into
>> the proving state.  One PROVING message acts like an ALIGNMENT (cancels
>> T2 and starts T3), the next acts like the first PROVING message (cancels
>> T3 and starts T4).
>>
>> --brian
>>
>> > >
>> > > section 4.1.3 paragraph 9:
>> > >
>> > > "When M2PA has both sent and received the Link Status Alignment
>> > > message, it has completed alignment.  M2PA starts the aligned timer
T3
>> > > and moves to the proving state."
>> > >
>> > > is it a requirement for both the local and remote peers to
send/receive
>> > > LS Alignment for alignment to be complete ?. the only way to
transition
>> > > from Initial Alignment to Proving is to receive LS Alignment. in the
>> > > scenario above, both peers do not send and receive LS Alignment
>> > > messages.
>> > >
>> > > regards,
>> > >
>> > > wayne
>> >
>> > --
>> > * Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com
*
>> > * 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/
*
>> > * Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075
*
>> > * I actively disclaim speaking for Alcatel USA, Inc. I do not purport
*
>> > * to advertise or represent its position,  operations, or activities.
*
>> > _______________________________________________
>> > Sigtran mailing list
>> > Sigtran@ietf.org
>> > https://www1.ietf.org/mailman/listinfo/sigtran
>>
>> --
>> Brian F. G. Bidulock
>> bidulock@openss7.org
>> http://www.openss7.org/
>> _______________________________________________
>> Sigtran mailing list
>> Sigtran@ietf.org
>> https://www1.ietf.org/mailman/listinfo/sigtran
>_______________________________________________
>Sigtran mailing list
>Sigtran@ietf.org
>https://www1.ietf.org/mailman/listinfo/sigtran


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 10:13:07 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA05535
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 10:13:07 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CFMVQ23146
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 10:22:31 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CFMCp23137;
	Wed, 12 Feb 2003 10:22:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CFLpp23097
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 10:21:51 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA05486
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 10:11:53 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:kOla3fYKi8UiRZAmfBRWYJEwDYnE2+fw@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1CFFZD19887;
	Wed, 12 Feb 2003 08:15:36 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1CFFZw23324;
	Wed, 12 Feb 2003 08:15:35 -0700
Date: Wed, 12 Feb 2003 08:15:35 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: jeffrey.craig@tekelec.com
Cc: wayne n taylor <waynet@cisco.com>, Tom George <Tom.George@alcatel.com>,
        sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft 7 question
Message-ID: <20030212081535.A21356@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: jeffrey.craig@tekelec.com,
	wayne n taylor <waynet@cisco.com>,
	Tom George <Tom.George@alcatel.com>, sigtran@ietf.org
References: <OFE387804C.FEA1E263-ON85256CCB.0050BD83@nc.tekelec.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OFE387804C.FEA1E263-ON85256CCB.0050BD83@nc.tekelec.com>; from jeffrey.craig@tekelec.com on Wed, Feb 12, 2003 at 10:01:23AM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Jeff,

jeffrey.craig@tekelec.com wrote:                                             (Wed, 12 Feb 2003 10:01:23)
> 
> Hello Guys,
> 
> I think that we have consensus that the M2PA finite state machine
> should match the Q.703 SDLs, where possible. It has been several
> M2PA revisions now that the state machine has been updated, but
> fallen short of matching Q.703.
> 
> I agree with Brian's suggested M2PA changes (below) with the exception of
> the
> following:
> 
> >> The first thing a link should send is LS OOS.
> >>
> >> In the out of service state, it would be desireable for the out of
> >> service side to always send LS OOS in response to other messages.
> >>
> >> Also, the out of service side should be permitted to send LS OOS
> >> periodically.
> 
> What does 'first thing' mean? (upon START? upon association becoming
> established, as the current draft states?)

First thing after association established.

> 
> What problem is being solved by these three changes? Has anyone
> observed alignment loops at plugtests or in testing?

AFAIK the draft allows LS to be sent periodically.

Also, sending LS OOS in response to other messages while in the out of service
state handles the SCTP Restart scenario without having to directoy address
SCTP restart.

> 
> Where in Q.703 does it specify these behaviors?

Q.703 repeats LSSU, particularly SIOS, when the link is initially powered
on, regardless of any other messages while in the out of service state, and
continuously while in the out of service state.

> 
> Do we have consensus that the changes will not result in a "potential loop"
> during alignment, as was the initial concern? I am not convinced that
> these three behaviors are necessary or desirable.

What loop?  There is no loop in the Q.703 SDLs for sending SIOS while
in the out of service state.

--brian

> 
> Regards,
> 
> Jeff
> 
> On 02/11/2003 04:06:34 PM sigtran-admin wrote:
> 
> >Tom,
> >as i described earlier, my concern is the potential loop that could
> >occur during the initial LS Alignment exchange based on the requirement
> >that both peers must send and receive LS Alignment messages. Brian has
> >shown in his response (and i agree) that the scenario i described will
> >not happen if the procedures defined in the SDLs of the Q.703 spec is
> >adopted for m2pa in regards to the initial "alignment" exchange.
> >
> >wayne
> >
> >"Brian F. G. Bidulock" wrote:
> >>
> >> Tom,
> >>
> >> The first thing a link should send is LS OOS.
> >>
> >> In the out of service state, it would be desireable for the out of
> >> service side to always send LS OOS in response to other messages.
> >>
> >> Also, the out of service side should be permitted to send LS OOS
> >> periodically.
> >>
> >> With regards to Q.703 state machines, LS OOS in response to alignment
> >> will not abort alignment and T2 should still expire.  See test case
> >> "3.1.2 Timer T2" in draft-bidulock-sigtran-m2pa-test-02.txt.
> >>
> >> T2 is defined as the duration of time that the M2PA peer wil wait to
> >> receive a status "Alignment" message after sending a status "Alignment"
> >> message.
> >>
> >> As status Alignment will abort an otherwise in-service link, other
> >> link status received while running T2 should be ignored.
> >>
> >> Any different behaviour could interfere with MTP Level 3's link
> >> restoration timers and procedures.
> >>
> >> More comments below...
> >>
> >> Tom George
> >wrote:
> 
> >(Mon, 10 Feb 2003 16:07:21)
> >> > Wayne,
> >> >
> >> > Yes, I want both sides to send and receive a LS Alignment message to
> >> > complete alignment.
> >> >
> >> > To fix the problem you describe, how about this" If M2PA has not
> >> > received MTP3 Start and receives LS Alignment, M2PA replies with LS
> OOS
> >> > (instead of not replying at all and counting on the other end's T2
> >> > expiring).
> >> >
> >> > tomG.
> >> >
> >> > wayne n taylor wrote:
> >> > >
> >> > > Hi All,
> >> > >
> >> > > i have a couple of comments/questions about the initial exchange of
> the
> >> > > Link Status Alignment messages. there seems to be some inconsistency
> in
> >> > > the text and finite state machine.
> >> > >
> >> > > section 4.1.3 paragraph 4:
> >> > >
> >> > > "Once the association is established and M2PA has received a Start
> >> > > Request from MTP3, M2PA sends the Link Status Alignment message to
> its
> >> > > peer. If M2PA has not already received the Link Status Alignment
> >> > > message from its peer, then M2PA starts timer T2."
> >> > >
> >> > > what state would the local M2PA be in to receive a LS Alignment
> message
> >> > > from its peer prior to getting a MTP3 Start ?. i assume the OOS
> state,
> >> > > but the finite state machine does not show LS Alignment as an input.
> >> > >
> >> > > section 4.1.3 paragraph 5:
> >> > >
> >> > > "(Note that if the remote M2PA has not received a Start Request from
> >> > > its MTP3, it will not send the Link Status Alignment message to the
> >> > > local M2PA. Eventually timer T2 in the local M2PA will expire. If
> the
> >> > > remote M2PA receives a Start Request from its MTP3 and sends Link
> >> > > Status Alignment before the local M2PA's timer T2 expires, the
> >> > > alignment procedure can continue.)"
> >> > >
> >> > > assume the local M2PA receives a Start Request from MTP3, sends LS
> >> > > Alignment and start T2, but the remote peer has not received a Start
> >> > > Request and is in OOS. the current finite state machine does not
> show
> >> > > what to do with the LS Alignment from the local peer so lets assume
> the
> >> > > LS Alignment is dropped. the remote peer receives a Start Request
> from
> >> > > MTP3, sends LS Alignment and start T2 (since the remote peer didnt
> get
> >> > > the local peers LS Alignment). assume further that timer T2 is still
> >> > > running on the local peer. the text from section 4.1.3 above states
> the
> >> > > "alignment procedure can continue". in this case the local peer sent
> a
> >> > > LS Alignment and received a LS Alignment. the remote peer only sent
> a LS
> >> > > Alignment and did not receive an LS Alignment (since LS Alignment
> >> > > messages are not retried). eventually T2 on the remote peer will
> expire,
> >> > > T3 on the local peer will expire and the alignment procedure starts
> over
> >> > > again. it seems possible for the peers to get caught in a loop
> trying to
> >> > > exchange LS Alignment messages.
> >>
> >> See draft-bidulock-sigtran-m2pa-test-02.txt Section "3.1.3 Timer T3".
> You
> >> will see that what you describe does not happen if you follow Q.703
> SDLs.
> >> In Q.703 once SIO is sent SIN is sufficient to complete alignment.  So,
> >> one side sends ALIGNMENT and starts T2, the other side sends ALIGNMENT
> and
> >> starts T2.  The first side, receives the alignment, stops T2 and sends
> >> PROVING-NORMAL or PROVING-EMERGENCY and starts T3.  When the second side
> >> receives PROVING-NORMAL or PROVING-EMERGENCY it stops T2 and sends
> >> PROVING-NORMAL or PROVING-EMERGENCY and starts T4.  When the first side
> >> receives PROVING-NORMAL it stops T3 and starts T4.  When either T4
> expires,
> >> the side sends IN-SERVICE and starts T1.  When the side receives
> IN-SERVICE
> >> it stops T1 and indicates in_service to MTP3.
> >>
> >> This behavior can be seen on Figure 9/Q.703 (sheet 2 of 6) - Initial
> alignment
> >> control.  When the IAC starts it sends SIO and starts T2.  If it
> receives
> >> any of SIO, SIN or SIE it stops T2 and send SIN or SIE depending on
> emergency
> >> status and starts T3.  Receipt of a subsequent SIN or SIE (sheet 3 of 6)
> >> stops T3 and starts T4.  Note that if it receives SIO again, it ignores
> it
> >> at that point and continues to send SIE or SIN.
> >>
> >> These is no reason for M2PA not to follow the Q.703 SDLs in this regard.
> >>
> >> A statement that might clarify is that the M2PA sends LS ALIGNMENT and
> if
> >> it receives ALIGNMENT or PROVING it stops T2 and PROVING and starts T3.
> >> Note that it takes two PROVING messages in this case to move the peer
> into
> >> the proving state.  One PROVING message acts like an ALIGNMENT (cancels
> >> T2 and starts T3), the next acts like the first PROVING message (cancels
> >> T3 and starts T4).
> >>
> >> --brian
> >>
> >> > >
> >> > > section 4.1.3 paragraph 9:
> >> > >
> >> > > "When M2PA has both sent and received the Link Status Alignment
> >> > > message, it has completed alignment.  M2PA starts the aligned timer
> T3
> >> > > and moves to the proving state."
> >> > >
> >> > > is it a requirement for both the local and remote peers to
> send/receive
> >> > > LS Alignment for alignment to be complete ?. the only way to
> transition
> >> > > from Initial Alignment to Proving is to receive LS Alignment. in the
> >> > > scenario above, both peers do not send and receive LS Alignment
> >> > > messages.
> >> > >
> >> > > regards,
> >> > >
> >> > > wayne
> >> >
> >> > --
> >> > * Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com
> *
> >> > * 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/
> *
> >> > * Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075
> *
> >> > * I actively disclaim speaking for Alcatel USA, Inc. I do not purport
> *
> >> > * to advertise or represent its position,  operations, or activities.
> *
> >> > _______________________________________________
> >> > Sigtran mailing list
> >> > Sigtran@ietf.org
> >> > https://www1.ietf.org/mailman/listinfo/sigtran
> >>
> >> --
> >> Brian F. G. Bidulock
> >> bidulock@openss7.org
> >> http://www.openss7.org/
> >> _______________________________________________
> >> Sigtran mailing list
> >> Sigtran@ietf.org
> >> https://www1.ietf.org/mailman/listinfo/sigtran
> >_______________________________________________
> >Sigtran mailing list
> >Sigtran@ietf.org
> >https://www1.ietf.org/mailman/listinfo/sigtran
> 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 10:19:06 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA05866
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 10:19:06 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CFSVV23539
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 10:28:31 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CFSHp23481;
	Wed, 12 Feb 2003 10:28:17 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CFRIp23427
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 10:27:18 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA05749
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 10:17:22 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:83ZjCDach9sC2lb/0gRfOB87omfuP4sL@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1CFL4D19968;
	Wed, 12 Feb 2003 08:21:04 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1CFL3v24120;
	Wed, 12 Feb 2003 08:21:03 -0700
Date: Wed, 12 Feb 2003 08:21:03 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: jeffrey.craig@tekelec.com
Cc: John Li <jli@iname.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft comments and questions.
Message-ID: <20030212082103.B21356@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: jeffrey.craig@tekelec.com, John Li <jli@iname.com>,
	sigtran@ietf.org
References: <OF2DF310C4.A6055B45-ON85256CCB.00467B35@nc.tekelec.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF2DF310C4.A6055B45-ON85256CCB.00467B35@nc.tekelec.com>; from jeffrey.craig@tekelec.com on Wed, Feb 12, 2003 at 08:33:17AM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Jeff,

jeffrey.craig@tekelec.com wrote:                                             (Wed, 12 Feb 2003 08:33:17)
> >
> >Management procedures are implementation dependent.
> 
> (Jeff)
> I agree that requesting that the upper layer stop trying to align the link
> is
> outside the scope of the M2PA draft. There is no MTP3 primitive for this.
> There is nothing preventing the implementor from maintaining and
> reporting 'M2PA version error' measurements.
> 
> The initial question was: "What is the definition of 'SHALL NOT complete
> the alignment procedure'".
> 
> The draft currently implies that the unsupported LSA is simply ignored.
> This means that alignment will ultimately fail due to T2 expiry by the
> receiver of the unsupported LSA. This also means that the sender is
> likely to advance to PROVING. I see no problem with this.

The sender should not necessarily advance to proving, unless it likes
the verion of a received ALIGNMENT or PROVING message.

> 
> So the text at 4.1.8 should be clarified:
> 
> OLD TEXT
> In case a newer version of M2PA is incompatible with an older version,
> the newer version SHOULD recognize this and prevent the alignment of
> the link. If a Link Status Alignment message with an unsupported
> version is received by the newer version, the receiving end's M2PA
> SHALL NOT complete the alignment procedure.
> 
> NEW TEXT
> In case a newer version of M2PA is incompatible with an older version,
> the newer version SHOULD recognize this and prevent the alignment of
> the link. If a Link Status Alignment message with an unsupported
> version is received by the newer version, the receiving end's M2PA
> MUST discard the message.

I would say that if it receives any message with an unsupported version
it SHOULD discard the message and not act upon it.

(SHOULD not MUST, because "unsupported" is not clearly defined.)

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 10:23:33 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA05976
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 10:23:33 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CFWwE23793
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 10:32:58 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CFWbp23786;
	Wed, 12 Feb 2003 10:32:37 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CFVwp23734
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 10:31:58 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA05956;
	Wed, 12 Feb 2003 10:21:57 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:jgx+alQfkp9Pc8elrGS5BJhxdzCWB8w9@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1CFPdD20020;
	Wed, 12 Feb 2003 08:25:39 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1CFPdY24714;
	Wed, 12 Feb 2003 08:25:39 -0700
Date: Wed, 12 Feb 2003 08:25:39 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Randall Stewart <randall@stewart.chicago.il.us>
Cc: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>,
        Caitlin Bestler <cait@asomi.com>, tsvwg@ietf.org,
        Tuexen Michael <michael.tuexen@siemens.com>, sigtran@ietf.org
Message-ID: <20030212082539.C21356@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Randall Stewart <randall@stewart.chicago.il.us>,
	Michael Tuexen <Michael.Tuexen@lurchi.franken.de>,
	Caitlin Bestler <cait@asomi.com>, tsvwg@ietf.org,
	Tuexen Michael <michael.tuexen@siemens.com>, sigtran@ietf.org
References: <A2AA39E1-3DFB-11D7-90B4-0030654C1AB6@lurchi.franken.de> <3E49899A.6010809@stewart.chicago.il.us> <20030211191543.A14481@openss7.org> <3E4A36FF.8050205@stewart.chicago.il.us> <20030212052223.A29464@openss7.org> <3E4A3E62.4050207@stewart.chicago.il.us> <20030212053843.C29464@openss7.org> <3E4A4A6B.4010405@stewart.chicago.il.us>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E4A4A6B.4010405@stewart.chicago.il.us>; from randall@stewart.chicago.il.us on Wed, Feb 12, 2003 at 07:21:47AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Subject: [Sigtran] Re: [Tsvwg] A question of Un-Ordered SCTP
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Randall,

Then send them ordered and let the RDDP/SCTP receiver perform
direct data placement without waiting for reordering.  Ordered
data chunks already have usable stream sequence numbers.  It
appears that you want to use the U-bit to tell the receiver that
it doesn't need to order the messages when all you need is
unordered delivery at the receiver.  Ordered delivery across 
the SCTP ULP API is not a requirement as it is in TCP.

--brian

Randall Stewart wrote:                                                             (Wed, 12 Feb 2003 07:21:47)
> Brian:
> 
> I am really not interested in adding a bit in
> the header for RDDP. I think it is better
> to do something minimalist that meets there
> needs instead of burning a bit (of only 5 left)
> for one particular application.
> 
> Having a second sequence number space that is
> allowed.. but not required.. and Michaels suggested
> TLV in INIT/INIT-ACK would be more than adequate for
> RDDP or for that matter any other application that
> wants data right away but yet would like to know
> presentation order to the sending SCTP.
> 
> R
> 
> Brian F. G. Bidulock wrote:
> > Randall,
> > 
> > So why not introduce another bit into the data chunk header?
> > 
> > Say, the O-bit which indicates that data is ordered at the
> > sender but not at the receiver?
> > 
> > An SCTP which does not know about the O-bit will assume that
> > ordered delivery is required.  Receivers which know about
> > the O-bit will know that unordered deliver is allowed.
> > 
> > That way you can start counting at 0 and leave unordered
> > data chunks alone.
> > 
> > While your at it, why not place the offset of the data chunk
> > in the final message in the DATA chunk header so that the
> > payload is identical for RDDP and non-RDDP SCTPs.  That way,
> > the non-RDDP receiver will be able to ignore the O-bit and
> > offset, and deliver the data ordered.  The RDDP receiver
> > will be able to observe the O-bit and offset and direct place
> > the data chunk.
> > 
> > --brian
> > 
> > 
> > Randall Stewart wrote:                                                             (Wed, 12 Feb 2003 06:30:26)
> > 
> >>Brian F. G. Bidulock wrote:
> >>
> >>>Randall,
> >>>
> >>>For the M2PA application, I think that reserving
> >>>'0' for unordered messages for which the order of
> >>>sending is unimportant is still a good idea.
> >>>So the number space from 1 to 0xffff indicates
> >>>importance of order.  '0' indicates no importance
> >>>of order.  As in M2PA's case, a sender could mix
> >>>both on the same stream.
> >>>
> >>>There then becomes three different types of messages:
> >>>
> >>>1) Unordered messages for which the order of sending is
> >>>   unimportant (U-bit SSN=0),
> >>>
> >>>2) unordered messages for which the order of sending is
> >>>   important (U-bit SSN!=0), and
> >>>
> >>>3) ordered messages for which the order of sending is
> >>>   important (~U-bit).
> >>>
> >>>And this also handles compatibility because previous
> >>>unordered messages fall into category (1) by nature
> >>>of the fact that SSN==0.
> >>>
> >>>But starting counting at 1, you get rid of the ability
> >>>to indicate that the order of sending is unimportant and
> >>>force the receiving ULP to handle ordering of all received
> >>>unordered data.
> >>>
> >>>So, in short, start counting at 1, regardless of whether
> >>>there is an indicator in the INIT/INIT-ACK.
> >>>
> >>
> >>That might be ok, but in the above does not that mean
> >>when you reach 0xffff you end up sending a type
> >>1) message... unless of course you want to have
> >>a counting scheme where you always skip 0... which
> >>is a bit more code on every sequence increase...
> >>
> >>And I do still think you will need to "know" ahead of
> >>time if both sides support 2) otherwise you would not
> >>know to embed your own SSN (in cases where you must
> >>emulate 2)
> >>
> >>R
> >>
> >>
> >>>--brian
> >>>
> >>>Randall Stewart wrote:                                                             (Wed, 12 Feb 2003 05:58:55)
> >>>
> >>>
> >>>>Brian/Michael:
> >>>>
> >>>>I do like the idea of keeping things the same (aka same
> >>>>methodology)... this is good.
> >>>>
> >>>>It is also neat to be able to tell the difference between
> >>>>an implementation that supports this feature and one that
> >>>>does not.  But there are positives and negatives, I think,
> >>>>to each idea...
> >>>>
> >>>>A) For using a TLV, (as Michael suggests)
> >>>>   we really need to put this into a seperate
> >>>>   draft.. when I do think one could easily fit
> >>>>   this into the IG as Brian as suggested.
> >>>>
> >>>>B) montonically increasing numbers would eventually
> >>>>   wrap when the SSN hit FFFF back to 0... thus
> >>>>   the application would need to have some sort
> >>>>   of state to track if the peer supported the
> >>>>   feature.
> >>>>
> >>>>It seems to me it is only the first message that
> >>>>arrives that benefits from the start at 0. If the
> >>>>application assumes a "not supported", then by
> >>>>starting the SSN at 1 the app can flip the flag
> >>>
> >>>>from "not supported" to "supported" on the arrival
> >>>
> >>>>of the first message.
> >>>>
> >>>>If the application instead set its flag to "supported"
> >>>>the first message would not tell this information and
> >>>>instead it would only be upon the second arriving message
> >>>>that the app could make the flag determination... causing
> >>>>either a delay in delivery of the first message (yuck) or
> >>>>just toss it up and wait.
> >>>>
> >>>>More importantly, I am wondering, if the implementation does
> >>>>NOT support this, what will the app do? Change its
> >>>>message structure? If so it will need to have some clues
> >>>>(as the initiator) that its implementation does not
> >>>>support the feature .. possibly a HAVE_SCTP_UNORDERED_SSN
> >>>>in sctp.h... but the problem is you would not know about
> >>>>the other guy until its first unordered message arrives.
> >>>>So if you have to send the first message what do you do?
> >>>>
> >>>>if HAVE_SCTP_UNORDERED_SSN is 0.. no problem.. you
> >>>>send a message that has the imbedded sequence number.. but
> >>>>what do you send when HAVE_SCTP_UNRODERED_SSN is set to 1?
> >>>>
> >>>>If you send the imbedded sequence in the message anyway whats
> >>>>the point of this little trick? The second message (which you
> >>>>may have to send before knowing receiving a unordered
> >>>>msg from the peer) would still have this, and maybe the third
> >>>>etc...
> >>>>
> >>>>Hmm.. the more I think on this the more it becomes clear that
> >>>>you must know ahead of time i.e. when the association comes
> >>>>up to "do the right thing".... which goes back to Michaels
> >>>>idea of a TLV.
> >>>>
> >>>>I have been wondering for quite some time now if we don't
> >>>>just need a TLV in the INIT/INIT-ACK that indicate features
> >>>>in general... something like a bit mask or byte array indicating
> >>>>each feature...
> >>>>
> >>>>addip
> >>>>delip
> >>>>prsctp
> >>>>setprimary
> >>>>unordered-assigns-sequences
> >>>>
> >>>>etc..
> >>>>
> >>>>Maybe this is a better way to go.. and we could write a short
> >>>>draft dictating this bit/byte TLV... code as a 0x80 class.. that
> >>>>way old implementations will throw it away... (we really don't
> >>>>care if they don't understand it)...
> >>>>
> >>>>I have been avoiding this.. but it might be the best approach..
> >>>>
> >>>>Then again, I guess it could also be put in addip or prsctp since
> >>>>it would effect them as well :> Then one could put the
> >>>>other change in the IG and thus have a very easy change that
> >>>>is in the IG and a minor update to either addip or prsctp...
> >>>>
> >>>>
> >>>>What do you all think?
> >>>>
> >>>>R
> >>>>
> >>>>Brian F. G. Bidulock wrote:
> >>>>
> >>>>
> >>>>>Randall,
> >>>>>
> >>>>>I agree to having SCTP set the sequence number and, as you said in
> >>>>>your first note, SCTP passing the sequence number to the ULP on
> >>>>>the receiving side is already done for ordered traffic.  Some even
> >>>>>pass the zero now.
> >>>>>
> >>>>>I suppose that SCTP would set this sequence number when it is given
> >>>>>the data for delivery at the sending ULP and the sequence would represent
> >>>>>the order that the data was presented to SCTP by the ULP at the sending
> >>>>>side.
> >>>>>
> >>>>>I think it would be good to see the requirement that the sequence
> >>>>>number begin at 1 and be monotonically increasing in the saem fashion
> >>>>>as ordered streams.
> >>>>>
> >>>>>That way RDDP and other applications could perform their own reordering
> >>>>>at the receiver.  I think it would be a good thing to have SCTP set
> >>>>>this value.
> >>>>>
> >>>>>In fact, I can see another application for this feature.  M2PA
> >>>>>(SIGTRAN protocol) needs in some circumstances to send data unordered
> >>>>>with the hope of expedited delivery (does not want control information
> >>>>>stuck behind large amounts of data in an ordered stream), however, the
> >>>>>order of processing of the expedited control information is important.
> >>>>>
> >>>>>This facility would permit M2PA to send such information unordered
> >>>>>to expedite it and the receiver could perform the reordering itself.
> >>>>>Up until now, the only solution was to place a sequence number in with
> >>>>>the control information.
> >>>>>
> >>>>>Could we perhaps add a section to the SCTP IG.  "Problem statement:
> >>>>>the sender wishes to send unordered data but the receive needs to know
> >>>>>the original order of sending..." or some such.
> >>>>>
> >>>>>It should be quite simple to add the feature to existing SCTP
> >>>>>implementations.
> >>>>>
> >>>>>--brian
> >>>>>
> >>>>>
> >>>>>On Tue, 11 Feb 2003, Randall Stewart wrote:
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>>Michael:
> >>>>>>
> >>>>>>This seems like a valid method.. it gets around Brian's objection
> >>>>>>(which was a good point) and yet still provides the application with
> >>>>>>a hint about what the assigned sequence order was on the unordered
> >>>>>>messages...
> >>>>>>
> >>>>>>R
> >>>>>>
> >>>>>>Michael Tuexen wrote:
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>>>Caitlin,
> >>>>>>>
> >>>>>>>thank you very much for the clarification.
> >>>>>>>
> >>>>>>>I do not like the idea to pass the SSN transparently from the ULP to the 
> >>>>>>>ULP.
> >>>>>>>But having two sequence spaces is OK for me. This means that the SSN at the
> >>>>>>>sending side is controlled by the SCTP and given to the ULP at the 
> >>>>>>>receiving side.
> >>>>>>>
> >>>>>>>This would be a clean solution. So I voting for have two separate number 
> >>>>>>>spaces
> >>>>>>>for user data. Since the one for ordered delivery exists already in RFC 
> >>>>>>>2960, this
> >>>>>>>only means that a new one is introduced for the unordered delivery user 
> >>>>>>>data.
> >>>>>>>
> >>>>>>>Best regards
> >>>>>>>Michael
> >>>>>>>
> >>>>>>>On Tuesday, Feb 11, 2003, at 17:37 Europe/Berlin, Caitlin Bestler wrote:
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>>>On 2/11/03, Tuexen Michael wrote:
> >>>>>>>>
> >>>>>>>>
> >>>>>>>>
> >>>>>>>>
> >>>>>>>>>Dear all,
> >>>>>>>>>
> >>>>>>>>>does this mean that you want two sequence spaces per
> >>>>>>>>>stream: one for the 'ordered' user data and one for the
> >>>>>>>>>unordered user data?
> >>>>>>>>>
> >>>>>>>>
> >>>>>>>>The RDDP/SCTP adaptation only uses unordered delivery, so
> >>>>>>>>that question did not really arise.
> >>>>>>>>
> >>>>>>>>But if you were to generalize the problem, I believe the
> >>>>>>>>answer would be yes.
> >>>>>>>>
> >>>>>>>>The SSN for ordered delivery would be unchanged. The
> >>>>>>>>addition would be providing "advisory ordering" for
> >>>>>>>>unordered delivery. The general characteristic is that
> >>>>>>>>the receiving ULP wants the data in whatever order it
> >>>>>>>>arrives, but wants to know what the original order was.
> >>>>>>>>
> >>>>>>>>In addition to RDDP, RTP over UDP is another example of
> >>>>>>>>an application that wants to be aware of the original
> >>>>>>>>ordering, even while accepting the data out-of-order.
> >>>>>>>>
> >>>>>>>>Caitlin Bestler
> >>>>>>>>http://asomi.com/CaitlinBestler/
> >>>>>>>>_______________________________________________
> >>>>>>>>tsvwg mailing list
> >>>>>>>>tsvwg@ietf.org
> >>>>>>>>https://www1.ietf.org/mailman/listinfo/tsvwg
> >>>>>>>>
> >>>>>>>
> >>>>>>>_______________________________________________
> >>>>>>>tsvwg mailing list
> >>>>>>>tsvwg@ietf.org
> >>>>>>>https://www1.ietf.org/mailman/listinfo/tsvwg
> >>>>>>>
> >>>>>>>
> >>>>>>
> >>>>>>
> >>>>>>-- 
> >>>>>>Randall R. Stewart
> >>>>>>randall@stewart.chicago.il.us 815-342-5222 (cell phone)
> >>>>>>
> >>>>>>_______________________________________________
> >>>>>>tsvwg mailing list
> >>>>>>tsvwg@ietf.org
> >>>>>>https://www1.ietf.org/mailman/listinfo/tsvwg
> >>>>>
> >>>>>
> >>>>-- 
> >>>>Randall R. Stewart
> >>>>randall@stewart.chicago.il.us 815-342-5222 (cell phone)
> >>>
> >>>
> >>
> >>-- 
> >>Randall R. Stewart
> >>randall@stewart.chicago.il.us 815-342-5222 (cell phone)
> > 
> > 
> 
> 
> -- 
> Randall R. Stewart
> randall@stewart.chicago.il.us 815-342-5222 (cell phone)

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 10:33:32 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA06292
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 10:33:32 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CFgwi24925
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 10:42:58 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CFgZp24910;
	Wed, 12 Feb 2003 10:42:36 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CFfpp24881
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 10:41:51 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA06274
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 10:31:54 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1CFZZqe028671;
	Wed, 12 Feb 2003 09:35:35 -0600 (CST)
Message-ID: <3E4A69C7.46BD6983@alcatel.com>
Date: Wed, 12 Feb 2003 09:35:35 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Security section (was: m2pa comment: author list)
References: <2135200C183FD5119588009027DE572302836C0A@webdev-owa.oni.com> <3E4971EF.BDD6AAE3@alcatel.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

It was brought to my attention that in section 6.2, "wireless" would be
more accurate than "mobile". 

tomG.

Tom George wrote:
> 
> On Jan 26, 2003 "Ong, Lyndon" wrote:
>   [...]
> > The security section probably needs to be updated based on the
> > security framework document (which should come out this week).
>   [...]
> 
> Here is my first stab at revising section 6 of the M2PA draft. Comments
> are welcome, especially since I am not familiar with the security
> protocols.
> 
> tomG.
> 
> =============================
> 6. Security
> 
> M2PA is designed to carry signaling messages for telephony
> services. As such, M2PA MUST involve the security needs of several
> parties:
> 
>    - the end users of the services
> 
>    - the network providers
> 
>    - the applications involved
> 
> Additional requirements MAY come from local regulation.
> 
> While these parties may have some overlapping security needs, their
> needs may not be identical. Any security solution SHOULD fulfill all
> of the different parties' needs.
> 
> 6.1 Security Requirements
> 
> As a transport protocol, M2PA has the following security objectives:
> 
>    - Availability of reliable and timely user data transport.
> 
>    - Integrity of user data transport.
> 
>    - Confidentiality of user data.
> 
> M2PA runs on top of SCTP [RFC2960].  SCTP provides certain transport
> related security features, as discussed in [Security]. However, these
> security features MAY NOT be adequate for all of the involved parties'
> requirements.
> 
> When M2PA is running in professionally managed corporate or service
> provider network, it is reasonable to expect that this network
> includes an appropriate security policy framework. The "Site Security
> Handbook" [RFC2196] SHOULD be consulted for guidance.
> 
> When the network in which M2PA runs involves more than one party
> (e.g., a non-private network), it MAY NOT be reasonable to expect that
> all parties have implemented security in a sufficient manner.  In such
> a case, it is recommended that IPSec or TLS be used to ensure
> confidentiality of the user payload. Consult [Security] for more
> guidance on the use of security protocols.
> 
> 6.2 Protecting Confidentiality
> 
> Particularly for mobile users, the requirement for confidentiality MAY
> include the masking of IP addresses and ports. In this case
> application-level encryption is not sufficient. IPSec ESP SHOULD be
> used instead [RFC2401].  Regardless of which level performs the
> encryption, the IPSec ISAKMP service SHOULD be used for key
> management.
> 
> =============================
> 
> {Add an informative reference:}
> 
>   [Security]
>        J. Loughney, M. Tuexen, and J. Pastor-Balbas, "Security
>        Considerations for SIGTRAN Protocols,"
>        draft-ietf-sigtran-security-01.txt (January 2003).
> 
> =============================
> 
> --
> * Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
> * 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
> * Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
> * I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
> * to advertise or represent its position,  operations, or activities. *
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 10:35:59 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA06420
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 10:35:59 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CFjOh25074
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 10:45:24 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CFj5p25050;
	Wed, 12 Feb 2003 10:45:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CFiGp25014
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 10:44:16 -0500
Received: from rtp-core-2.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA06332
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 10:34:19 -0500 (EST)
Received: from dingdong.cisco.com (IDENT:mirapoint@dingdong.cisco.com [64.102.17.16])
	by rtp-core-2.cisco.com (8.12.6/8.12.6) with ESMTP id h1CFbwNh027598;
	Wed, 12 Feb 2003 10:37:58 -0500 (EST)
Received: from cisco.com (dhcp-64-102-86-148.cisco.com [64.102.86.148])
	by dingdong.cisco.com (Mirapoint Messaging Server MOS 3.2.1-GA)
	with ESMTP id ABK70932;
	Wed, 12 Feb 2003 10:37:56 -0500 (EST)
Message-ID: <3E4A6A54.46FFD0A6@cisco.com>
Date: Wed, 12 Feb 2003 10:37:56 -0500
From: wayne n taylor <waynet@cisco.com>
X-Mailer: Mozilla 4.79 [en]C-CCK-MCD   (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: jeffrey.craig@tekelec.com
CC: Tom George <Tom.George@alcatel.com>, bidulock@openss7.org,
        sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft 7 question
References: <OFE387804C.FEA1E263-ON85256CCB.0050BD83@nc.tekelec.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Hi Guys,

see comments inline ...

jeffrey.craig@tekelec.com wrote:
> 
> Hello Guys,
> 
> I think that we have consensus that the M2PA finite state machine
> should match the Q.703 SDLs, where possible. It has been several
> M2PA revisions now that the state machine has been updated, but
> fallen short of matching Q.703.
> 
> I agree with Brian's suggested M2PA changes (below) with the exception of
> the
> following:
> 
> >> The first thing a link should send is LS OOS.
> >>
> >> In the out of service state, it would be desireable for the out of
> >> service side to always send LS OOS in response to other messages.
> >>
> >> Also, the out of service side should be permitted to send LS OOS
> >> periodically.
> 
> What does 'first thing' mean? (upon START? upon association becoming
> established, as the current draft states?)
> 
> What problem is being solved by these three changes? Has anyone
> observed alignment loops at plugtests or in testing?
>
[waynet] i am not aware of anyone reporting alignment loops at plugtests
or testing. with the requirement that each peer must both send and
receive a LS Alignment message, it is pretty easy to create a loop. all
that is needed to form a loop is to have the peers to keep missing each
others LS Alignment message as described in my earlier comments. Brian's
earlier comments describes how the Q.703 SDLs eliminates this behavior.

> Where in Q.703 does it specify these behaviors?
> 
> Do we have consensus that the changes will not result in a "potential loop"
> during alignment, as was the initial concern? I am not convinced that
> these three behaviors are necessary or desirable.

[waynet] i dont think Brian's suggestions concerning the sending of LS
OOS is intended as a solution to the loop concerns i mentioned. is
comments about following the Q.703 SDLs to eliminate a potential loop is
valid.
> 
> Regards,
> 
> Jeff
> 
> On 02/11/2003 04:06:34 PM sigtran-admin wrote:
> 
> >Tom,
> >as i described earlier, my concern is the potential loop that could
> >occur during the initial LS Alignment exchange based on the requirement
> >that both peers must send and receive LS Alignment messages. Brian has
> >shown in his response (and i agree) that the scenario i described will
> >not happen if the procedures defined in the SDLs of the Q.703 spec is
> >adopted for m2pa in regards to the initial "alignment" exchange.
> >
> >wayne
> >
> >"Brian F. G. Bidulock" wrote:
> >>
> >> Tom,
> >>
> >> The first thing a link should send is LS OOS.
> >>
> >> In the out of service state, it would be desireable for the out of
> >> service side to always send LS OOS in response to other messages.
> >>
> >> Also, the out of service side should be permitted to send LS OOS
> >> periodically.
> >>
> >> With regards to Q.703 state machines, LS OOS in response to alignment
> >> will not abort alignment and T2 should still expire.  See test case
> >> "3.1.2 Timer T2" in draft-bidulock-sigtran-m2pa-test-02.txt.
> >>
> >> T2 is defined as the duration of time that the M2PA peer wil wait to
> >> receive a status "Alignment" message after sending a status "Alignment"
> >> message.
> >>
> >> As status Alignment will abort an otherwise in-service link, other
> >> link status received while running T2 should be ignored.
> >>
> >> Any different behaviour could interfere with MTP Level 3's link
> >> restoration timers and procedures.
> >>
> >> More comments below...
> >>
> >> Tom George
> >wrote:
> 
> >(Mon, 10 Feb 2003 16:07:21)
> >> > Wayne,
> >> >
> >> > Yes, I want both sides to send and receive a LS Alignment message to
> >> > complete alignment.
> >> >
> >> > To fix the problem you describe, how about this" If M2PA has not
> >> > received MTP3 Start and receives LS Alignment, M2PA replies with LS
> OOS
> >> > (instead of not replying at all and counting on the other end's T2
> >> > expiring).
> >> >
> >> > tomG.
> >> >
> >> > wayne n taylor wrote:
> >> > >
> >> > > Hi All,
> >> > >
> >> > > i have a couple of comments/questions about the initial exchange of
> the
> >> > > Link Status Alignment messages. there seems to be some inconsistency
> in
> >> > > the text and finite state machine.
> >> > >
> >> > > section 4.1.3 paragraph 4:
> >> > >
> >> > > "Once the association is established and M2PA has received a Start
> >> > > Request from MTP3, M2PA sends the Link Status Alignment message to
> its
> >> > > peer. If M2PA has not already received the Link Status Alignment
> >> > > message from its peer, then M2PA starts timer T2."
> >> > >
> >> > > what state would the local M2PA be in to receive a LS Alignment
> message
> >> > > from its peer prior to getting a MTP3 Start ?. i assume the OOS
> state,
> >> > > but the finite state machine does not show LS Alignment as an input.
> >> > >
> >> > > section 4.1.3 paragraph 5:
> >> > >
> >> > > "(Note that if the remote M2PA has not received a Start Request from
> >> > > its MTP3, it will not send the Link Status Alignment message to the
> >> > > local M2PA. Eventually timer T2 in the local M2PA will expire. If
> the
> >> > > remote M2PA receives a Start Request from its MTP3 and sends Link
> >> > > Status Alignment before the local M2PA's timer T2 expires, the
> >> > > alignment procedure can continue.)"
> >> > >
> >> > > assume the local M2PA receives a Start Request from MTP3, sends LS
> >> > > Alignment and start T2, but the remote peer has not received a Start
> >> > > Request and is in OOS. the current finite state machine does not
> show
> >> > > what to do with the LS Alignment from the local peer so lets assume
> the
> >> > > LS Alignment is dropped. the remote peer receives a Start Request
> from
> >> > > MTP3, sends LS Alignment and start T2 (since the remote peer didnt
> get
> >> > > the local peers LS Alignment). assume further that timer T2 is still
> >> > > running on the local peer. the text from section 4.1.3 above states
> the
> >> > > "alignment procedure can continue". in this case the local peer sent
> a
> >> > > LS Alignment and received a LS Alignment. the remote peer only sent
> a LS
> >> > > Alignment and did not receive an LS Alignment (since LS Alignment
> >> > > messages are not retried). eventually T2 on the remote peer will
> expire,
> >> > > T3 on the local peer will expire and the alignment procedure starts
> over
> >> > > again. it seems possible for the peers to get caught in a loop
> trying to
> >> > > exchange LS Alignment messages.
> >>
> >> See draft-bidulock-sigtran-m2pa-test-02.txt Section "3.1.3 Timer T3".
> You
> >> will see that what you describe does not happen if you follow Q.703
> SDLs.
> >> In Q.703 once SIO is sent SIN is sufficient to complete alignment.  So,
> >> one side sends ALIGNMENT and starts T2, the other side sends ALIGNMENT
> and
> >> starts T2.  The first side, receives the alignment, stops T2 and sends
> >> PROVING-NORMAL or PROVING-EMERGENCY and starts T3.  When the second side
> >> receives PROVING-NORMAL or PROVING-EMERGENCY it stops T2 and sends
> >> PROVING-NORMAL or PROVING-EMERGENCY and starts T4.  When the first side
> >> receives PROVING-NORMAL it stops T3 and starts T4.  When either T4
> expires,
> >> the side sends IN-SERVICE and starts T1.  When the side receives
> IN-SERVICE
> >> it stops T1 and indicates in_service to MTP3.
> >>
> >> This behavior can be seen on Figure 9/Q.703 (sheet 2 of 6) - Initial
> alignment
> >> control.  When the IAC starts it sends SIO and starts T2.  If it
> receives
> >> any of SIO, SIN or SIE it stops T2 and send SIN or SIE depending on
> emergency
> >> status and starts T3.  Receipt of a subsequent SIN or SIE (sheet 3 of 6)
> >> stops T3 and starts T4.  Note that if it receives SIO again, it ignores
> it
> >> at that point and continues to send SIE or SIN.
> >>
> >> These is no reason for M2PA not to follow the Q.703 SDLs in this regard.
> >>
> >> A statement that might clarify is that the M2PA sends LS ALIGNMENT and
> if
> >> it receives ALIGNMENT or PROVING it stops T2 and PROVING and starts T3.
> >> Note that it takes two PROVING messages in this case to move the peer
> into
> >> the proving state.  One PROVING message acts like an ALIGNMENT (cancels
> >> T2 and starts T3), the next acts like the first PROVING message (cancels
> >> T3 and starts T4).
> >>
> >> --brian
> >>
> >> > >
> >> > > section 4.1.3 paragraph 9:
> >> > >
> >> > > "When M2PA has both sent and received the Link Status Alignment
> >> > > message, it has completed alignment.  M2PA starts the aligned timer
> T3
> >> > > and moves to the proving state."
> >> > >
> >> > > is it a requirement for both the local and remote peers to
> send/receive
> >> > > LS Alignment for alignment to be complete ?. the only way to
> transition
> >> > > from Initial Alignment to Proving is to receive LS Alignment. in the
> >> > > scenario above, both peers do not send and receive LS Alignment
> >> > > messages.
> >> > >
> >> > > regards,
> >> > >
> >> > > wayne
> >> >
> >> > --
> >> > * Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com
> *
> >> > * 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/
> *
> >> > * Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075
> *
> >> > * I actively disclaim speaking for Alcatel USA, Inc. I do not purport
> *
> >> > * to advertise or represent its position,  operations, or activities.
> *
> >> > _______________________________________________
> >> > Sigtran mailing list
> >> > Sigtran@ietf.org
> >> > https://www1.ietf.org/mailman/listinfo/sigtran
> >>
> >> --
> >> Brian F. G. Bidulock
> >> bidulock@openss7.org
> >> http://www.openss7.org/
> >> _______________________________________________
> >> Sigtran mailing list
> >> Sigtran@ietf.org
> >> https://www1.ietf.org/mailman/listinfo/sigtran
> >_______________________________________________
> >Sigtran mailing list
> >Sigtran@ietf.org
> >https://www1.ietf.org/mailman/listinfo/sigtran
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 10:36:54 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA06499
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 10:36:54 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CFkJU25170
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 10:46:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CFk2p25121;
	Wed, 12 Feb 2003 10:46:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CFjKp25064
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 10:45:20 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA06376
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 10:35:24 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1CFd6qe029440;
	Wed, 12 Feb 2003 09:39:06 -0600 (CST)
Message-ID: <3E4A6A9A.8747D276@alcatel.com>
Date: Wed, 12 Feb 2003 09:39:06 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Security section (was: m2pa comment: authorli st)
References: <4.3.2.7.2.20030212094830.04b6cf50@cia.cisco.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

OK with me. I will put it in the normative section. 

Ken Morneault wrote:
> 
> Tom,
> 
> Also, it should be a "normative" reference.
> 
> Ken
  [...]

> >> =============================
> >>
> >> {Add an informative reference:}
> >>
> >>   [Security]
> >>        J. Loughney, M. Tuexen, and J. Pastor-Balbas, "Security
> >>        Considerations for SIGTRAN Protocols,"
> >>        draft-ietf-sigtran-security-01.txt (January 2003).
> >>
> >> =============================

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 10:37:50 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA06546
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 10:37:50 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CFlFL25235
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 10:47:15 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CFl3p25218;
	Wed, 12 Feb 2003 10:47:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CFkSp25191
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 10:46:28 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA06486
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 10:36:32 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:xjIvJpV6YgG+a4iz9nWbX9AG5Uux/Nzy@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1CFeFD20452;
	Wed, 12 Feb 2003 08:40:15 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1CFeEM26834;
	Wed, 12 Feb 2003 08:40:14 -0700
Date: Wed, 12 Feb 2003 08:40:14 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tolga Asveren <tolga_asveren@yahoo.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030212084014.D21356@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <3E4A452A.85F1083@axes-mach01.usa.alcatel.com> <20030212131558.41891.qmail@web12801.mail.yahoo.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20030212131558.41891.qmail@web12801.mail.yahoo.com>; from tolga_asveren@yahoo.com on Wed, Feb 12, 2003 at 05:15:58AM -0800
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Tolga,

I disagree strongly.  When we build n+1 redundant systems for carrier
grade operation we do not stop service when a processor is still able
to provide service.

If I have 3 processors in a 2+1 arrangement (where 2 are needed to handle
the maximum engineered load and 1 is spare), I do not stop service just
because I have only 1 processor alive.  It just means that the system
will not be able to handle the maximum engineered load with 1 processor.

Overload is a different issue.  If the 1 remaining processor is overloaded,
it should use overload controls to keep itself alive.  In M3UA's case this
could be overload controls at ISUP level or nodal TFC at MTP level or even
SCTP flow control (failing to service read buffer).

Having the SG move to the AS-PENDING state when the number of active ASPs
fall below "n" (2 in the case above) is completely undesirable.

I thought I made this clear on the "n+k redudancy" thread.

I strongly disagree with the changes of section 3.15 of the IG.

--brian


Tolga Asveren wrote:                                                                 (Wed, 12 Feb 2003 05:15:58)
>  If the traffic mode is 1+k I agree with that
> statement but if it is n+k, I disagree.
> 
>    Tolga
> --- ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
> wrote:

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 10:42:10 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA06689
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 10:42:10 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CFpaq25406
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 10:51:36 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CFp7p25386;
	Wed, 12 Feb 2003 10:51:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CFosp25370
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 10:50:54 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA06621
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 10:40:57 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:AQpZmx8ZyA+Lgywrc+fReIQVKfcF0crS@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1CFiaD20513;
	Wed, 12 Feb 2003 08:44:36 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1CFial27496;
	Wed, 12 Feb 2003 08:44:36 -0700
Date: Wed, 12 Feb 2003 08:44:36 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Gaiser Franz <franz.gaiser@siemens.com>
Cc: sigtran@ietf.org, "'Tom George'" <Tom.George@alcatel.com>
Subject: Re: [Sigtran] M2PA draft 07 comment: LPO
Message-ID: <20030212084436.E21356@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Gaiser Franz <franz.gaiser@siemens.com>, sigtran@ietf.org,
	'Tom George' <Tom.George@alcatel.com>
References: <3D72C18DF140D311A54F0008C7DBAC8D02917438@MCHH252E>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3D72C18DF140D311A54F0008C7DBAC8D02917438@MCHH252E>; from franz.gaiser@siemens.com on Tue, Feb 11, 2003 at 11:09:58AM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Gaiser,

Gaiser Franz wrote:                                                                 (Tue, 11 Feb 2003 11:09:58)
> Hi all,
> 
> I have an additional comment/remark to the description of handling Processor
> Outage in draft 07of M2PA. The handling of the following situation is not
> clear to me: - Proving Period completed (T4 expired) and Local Processor
> Outage is occured. In the case without LPO the m2pa would start sending LS
> Ready to the peer and moves to state ALIGNED READY. 

> For the above situation (with LPO) the handling is not clearly stated. Is LS
> Ready and/or LS PO to be send ?According to Q.703  the m2pa should send LS
> PO message instead of LS Ready and moves to state ALIGNED NOT READY.

Then that's what M2PA should do.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 11:02:11 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA07331
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 11:02:11 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CGBbC26894
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 11:11:37 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CGBIp26873;
	Wed, 12 Feb 2003 11:11:18 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CGAdp26834
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 11:10:39 -0500
Received: from domino2.nc.tekelec.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA07307
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 11:00:40 -0500 (EST)
From: jeffrey.craig@tekelec.com
Received: from domino4.nc.tekelec.com ([198.89.40.9])
          by domino2.nc.tekelec.com (Lotus Domino Release 5.0.10)
          with ESMTP id 2003021210580281:32663 ;
          Wed, 12 Feb 2003 10:58:02 -0500 
Subject: Re: [Sigtran] M2PA draft 7 question
To: bidulock@openss7.org
Cc: wayne n taylor <waynet@cisco.com>, Tom George <Tom.George@alcatel.com>,
        sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.7  March 21, 2001
Message-ID: <OFD3354CD0.0495FC34-ON85256CCB.00555985@nc.tekelec.com>
Date: Wed, 12 Feb 2003 11:00:06 -0500
MIME-Version: 1.0
X-MIMETrack: Serialize by Router on Domino4/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/12/2003 11:00:07 AM,
	Itemize by SMTP Server on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/12/2003 10:58:02 AM,
	Serialize by Router on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/12/2003 10:58:07 AM,
	Serialize complete at 02/12/2003 10:58:07 AM
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hello Brian,

More below....

Regards,

Jeff


On 02/12/2003 10:15:35 AM "Brian F. G. Bidulock" wrote:

>Jeff,
>
>jeffrey.craig@tekelec.com wrote:

>(Wed, 12 Feb 2003 10:01:23)
>>
>> Hello Guys,
>>
>> I think that we have consensus that the M2PA finite state machine
>> should match the Q.703 SDLs, where possible. It has been several
>> M2PA revisions now that the state machine has been updated, but
>> fallen short of matching Q.703.
>>
>> I agree with Brian's suggested M2PA changes (below) with the exception
of
>> the
>> following:
>>
>> >> The first thing a link should send is LS OOS.
>> >>
>> >> In the out of service state, it would be desireable for the out of
>> >> service side to always send LS OOS in response to other messages.
>> >>
>> >> Also, the out of service side should be permitted to send LS OOS
>> >> periodically.
>>
>> What does 'first thing' mean? (upon START? upon association becoming
>> established, as the current draft states?)
>
>First thing after association established.

Okay, good.

>
>>
>> What problem is being solved by these three changes? Has anyone
>> observed alignment loops at plugtests or in testing?
>
>AFAIK the draft allows LS to be sent periodically.

I don't know about this, since

"After the association is established, M2PA SHALL send a Link Status
Out of Service message to its peer."

means 'send one LSO' to me.

Besides, you didn't answer my questions.

>
>Also, sending LS OOS in response to other messages while in the out of
service
>state handles the SCTP Restart scenario without having to directoy address
>SCTP restart.
>
>>
>> Where in Q.703 does it specify these behaviors?
>
>Q.703 repeats LSSU, particularly SIOS, when the link is initially powered
>on, regardless of any other messages while in the out of service state,
and
>continuously while in the out of service state.

Ah. I see it now -- not in the SDLs, but -- the last sentence in Q.703 7.2
implies
repeated sending of SIOS prior to Intial Alignment:

    "Status indication OS is sent on completion of 'power on' (see Figures
2
    and 8) until initial alignment is started."

Note the 'until'. Okay. So these proposed changes do match Q.703.

Let me attempt to clarify the proposed changes with regards to out of
service state:

OLD TEXT 4.1.3
After the association is established, M2PA SHALL send a Link Status
Out of Service message to its peer.

NEW TEXT 4.1.3
After the association is established, M2PA SHALL send a Link Status
Out of Service message to its peer and, prior to receiving a Start
Request from MTP3, MAY periodically send Link Status Out of
Service messages. During the Out of Service State M2PA SHOULD
respond to received messages other than Link Status Out of Service
by sending Link Status Out of Service.

>
>>
>> Do we have consensus that the changes will not result in a "potential
loop"
>> during alignment, as was the initial concern? I am not convinced that
>> these three behaviors are necessary or desirable.
>
>What loop?  There is no loop in the Q.703 SDLs for sending SIOS while
>in the out of service state.

You ask me? I'm just trying to understand what 'problem' is being
addressed by proposed changes listed in this thread.

>
>--brian
>
>>
>> Regards,
>>
>> Jeff
>>
>> On 02/11/2003 04:06:34 PM sigtran-admin wrote:
>>
>> >Tom,
>> >as i described earlier, my concern is the potential loop that could
>> >occur during the initial LS Alignment exchange based on the requirement
>> >that both peers must send and receive LS Alignment messages. Brian has
>> >shown in his response (and i agree) that the scenario i described will
>> >not happen if the procedures defined in the SDLs of the Q.703 spec is
>> >adopted for m2pa in regards to the initial "alignment" exchange.
>> >
>> >wayne
>> >
>> >"Brian F. G. Bidulock" wrote:
>> >>
>> >> Tom,
>> >>
>> >> The first thing a link should send is LS OOS.
>> >>
>> >> In the out of service state, it would be desireable for the out of
>> >> service side to always send LS OOS in response to other messages.
>> >>
>> >> Also, the out of service side should be permitted to send LS OOS
>> >> periodically.
>> >>
>> >> With regards to Q.703 state machines, LS OOS in response to alignment
>> >> will not abort alignment and T2 should still expire.  See test case
>> >> "3.1.2 Timer T2" in draft-bidulock-sigtran-m2pa-test-02.txt.
>> >>
>> >> T2 is defined as the duration of time that the M2PA peer wil wait to
>> >> receive a status "Alignment" message after sending a status
"Alignment"
>> >> message.
>> >>
>> >> As status Alignment will abort an otherwise in-service link, other
>> >> link status received while running T2 should be ignored.
>> >>
>> >> Any different behaviour could interfere with MTP Level 3's link
>> >> restoration timers and procedures.
>> >>
>> >> More comments below...
>> >>
>> >> Tom George
>> >wrote:
>>
>> >(Mon, 10 Feb 2003 16:07:21)
>> >> > Wayne,
>> >> >
>> >> > Yes, I want both sides to send and receive a LS Alignment message
to
>> >> > complete alignment.
>> >> >
>> >> > To fix the problem you describe, how about this" If M2PA has not
>> >> > received MTP3 Start and receives LS Alignment, M2PA replies with LS
>> OOS
>> >> > (instead of not replying at all and counting on the other end's T2
>> >> > expiring).
>> >> >
>> >> > tomG.
>> >> >
>> >> > wayne n taylor wrote:
>> >> > >
>> >> > > Hi All,
>> >> > >
>> >> > > i have a couple of comments/questions about the initial exchange
of
>> the
>> >> > > Link Status Alignment messages. there seems to be some
inconsistency
>> in
>> >> > > the text and finite state machine.
>> >> > >
>> >> > > section 4.1.3 paragraph 4:
>> >> > >
>> >> > > "Once the association is established and M2PA has received a
Start
>> >> > > Request from MTP3, M2PA sends the Link Status Alignment message
to
>> its
>> >> > > peer. If M2PA has not already received the Link Status Alignment
>> >> > > message from its peer, then M2PA starts timer T2."
>> >> > >
>> >> > > what state would the local M2PA be in to receive a LS Alignment
>> message
>> >> > > from its peer prior to getting a MTP3 Start ?. i assume the OOS
>> state,
>> >> > > but the finite state machine does not show LS Alignment as an
input.
>> >> > >
>> >> > > section 4.1.3 paragraph 5:
>> >> > >
>> >> > > "(Note that if the remote M2PA has not received a Start Request
from
>> >> > > its MTP3, it will not send the Link Status Alignment message to
the
>> >> > > local M2PA. Eventually timer T2 in the local M2PA will expire. If
>> the
>> >> > > remote M2PA receives a Start Request from its MTP3 and sends Link
>> >> > > Status Alignment before the local M2PA's timer T2 expires, the
>> >> > > alignment procedure can continue.)"
>> >> > >
>> >> > > assume the local M2PA receives a Start Request from MTP3, sends
LS
>> >> > > Alignment and start T2, but the remote peer has not received a
Start
>> >> > > Request and is in OOS. the current finite state machine does not
>> show
>> >> > > what to do with the LS Alignment from the local peer so lets
assume
>> the
>> >> > > LS Alignment is dropped. the remote peer receives a Start Request
>> from
>> >> > > MTP3, sends LS Alignment and start T2 (since the remote peer
didnt
>> get
>> >> > > the local peers LS Alignment). assume further that timer T2 is
still
>> >> > > running on the local peer. the text from section 4.1.3 above
states
>> the
>> >> > > "alignment procedure can continue". in this case the local peer
sent
>> a
>> >> > > LS Alignment and received a LS Alignment. the remote peer only
sent
>> a LS
>> >> > > Alignment and did not receive an LS Alignment (since LS Alignment
>> >> > > messages are not retried). eventually T2 on the remote peer will
>> expire,
>> >> > > T3 on the local peer will expire and the alignment procedure
starts
>> over
>> >> > > again. it seems possible for the peers to get caught in a loop
>> trying to
>> >> > > exchange LS Alignment messages.
>> >>
>> >> See draft-bidulock-sigtran-m2pa-test-02.txt Section "3.1.3 Timer T3".
>> You
>> >> will see that what you describe does not happen if you follow Q.703
>> SDLs.
>> >> In Q.703 once SIO is sent SIN is sufficient to complete alignment.
So,
>> >> one side sends ALIGNMENT and starts T2, the other side sends
ALIGNMENT
>> and
>> >> starts T2.  The first side, receives the alignment, stops T2 and
sends
>> >> PROVING-NORMAL or PROVING-EMERGENCY and starts T3.  When the second
side
>> >> receives PROVING-NORMAL or PROVING-EMERGENCY it stops T2 and sends
>> >> PROVING-NORMAL or PROVING-EMERGENCY and starts T4.  When the first
side
>> >> receives PROVING-NORMAL it stops T3 and starts T4.  When either T4
>> expires,
>> >> the side sends IN-SERVICE and starts T1.  When the side receives
>> IN-SERVICE
>> >> it stops T1 and indicates in_service to MTP3.
>> >>
>> >> This behavior can be seen on Figure 9/Q.703 (sheet 2 of 6) - Initial
>> alignment
>> >> control.  When the IAC starts it sends SIO and starts T2.  If it
>> receives
>> >> any of SIO, SIN or SIE it stops T2 and send SIN or SIE depending on
>> emergency
>> >> status and starts T3.  Receipt of a subsequent SIN or SIE (sheet 3 of
6)
>> >> stops T3 and starts T4.  Note that if it receives SIO again, it
ignores
>> it
>> >> at that point and continues to send SIE or SIN.
>> >>
>> >> These is no reason for M2PA not to follow the Q.703 SDLs in this
regard.
>> >>
>> >> A statement that might clarify is that the M2PA sends LS ALIGNMENT
and
>> if
>> >> it receives ALIGNMENT or PROVING it stops T2 and PROVING and starts
T3.
>> >> Note that it takes two PROVING messages in this case to move the peer
>> into
>> >> the proving state.  One PROVING message acts like an ALIGNMENT
(cancels
>> >> T2 and starts T3), the next acts like the first PROVING message
(cancels
>> >> T3 and starts T4).
>> >>
>> >> --brian
>> >>
>> >> > >
>> >> > > section 4.1.3 paragraph 9:
>> >> > >
>> >> > > "When M2PA has both sent and received the Link Status Alignment
>> >> > > message, it has completed alignment.  M2PA starts the aligned
timer
>> T3
>> >> > > and moves to the proving state."
>> >> > >
>> >> > > is it a requirement for both the local and remote peers to
>> send/receive
>> >> > > LS Alignment for alignment to be complete ?. the only way to
>> transition
>> >> > > from Initial Alignment to Proving is to receive LS Alignment. in
the
>> >> > > scenario above, both peers do not send and receive LS Alignment
>> >> > > messages.
>> >> > >
>> >> > > regards,
>> >> > >
>> >> > > wayne
>> >> >
>> >> > --
>> >> > * Tom George       PB7-24001 (near B-21)
Tom.George@alcatel.com
>> *
>> >> > * 972-519-3168   Intranet:
http://www.ssd.usa.alcatel.com/~tgeorge/
>> *
>> >> > * Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas
75075
>> *
>> >> > * I actively disclaim speaking for Alcatel USA, Inc. I do not
purport
>> *
>> >> > * to advertise or represent its position,  operations, or
activities.
>> *
>> >> > _______________________________________________
>> >> > Sigtran mailing list
>> >> > Sigtran@ietf.org
>> >> > https://www1.ietf.org/mailman/listinfo/sigtran
>> >>
>> >> --
>> >> Brian F. G. Bidulock
>> >> bidulock@openss7.org
>> >> http://www.openss7.org/
>> >> _______________________________________________
>> >> Sigtran mailing list
>> >> Sigtran@ietf.org
>> >> https://www1.ietf.org/mailman/listinfo/sigtran
>> >_______________________________________________
>> >Sigtran mailing list
>> >Sigtran@ietf.org
>> >https://www1.ietf.org/mailman/listinfo/sigtran
>>
>
>--
>Brian F. G. Bidulock
>bidulock@openss7.org
>http://www.openss7.org/


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 11:56:20 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA08557
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 11:56:20 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CH5li05209
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 12:05:47 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CH5Vp05176;
	Wed, 12 Feb 2003 12:05:32 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CH0ip04724
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 12:00:44 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA08289
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 11:50:44 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:3of8GHS3tRaFm5hs4yMqlqFj6fWxkzRG@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1CGsRD21770;
	Wed, 12 Feb 2003 09:54:27 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1CGsRh05959;
	Wed, 12 Feb 2003 09:54:27 -0700
Date: Wed, 12 Feb 2003 09:54:26 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: jeffrey.craig@tekelec.com
Cc: wayne n taylor <waynet@cisco.com>, Tom George <Tom.George@alcatel.com>,
        sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft 7 question
Message-ID: <20030212095426.A5499@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: jeffrey.craig@tekelec.com,
	wayne n taylor <waynet@cisco.com>,
	Tom George <Tom.George@alcatel.com>, sigtran@ietf.org
References: <OFD3354CD0.0495FC34-ON85256CCB.00555985@nc.tekelec.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OFD3354CD0.0495FC34-ON85256CCB.00555985@nc.tekelec.com>; from jeffrey.craig@tekelec.com on Wed, Feb 12, 2003 at 11:00:06AM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

jeffrey.craig,

jeffrey.craig@tekelec.com wrote:                                             (Wed, 12 Feb 2003 11:00:06)
> 
> Hello Brian,
> 
> More below....
> 
> Regards,
> 
> Jeff
> 
> 
> On 02/12/2003 10:15:35 AM "Brian F. G. Bidulock" wrote:
> 
> >Jeff,
> >
> >jeffrey.craig@tekelec.com wrote:
> 
> >(Wed, 12 Feb 2003 10:01:23)
> >>
> >> Hello Guys,
> >>
> >> I think that we have consensus that the M2PA finite state machine
> >> should match the Q.703 SDLs, where possible. It has been several
> >> M2PA revisions now that the state machine has been updated, but
> >> fallen short of matching Q.703.
> >>
> >> I agree with Brian's suggested M2PA changes (below) with the exception
> of
> >> the
> >> following:
> >>
> >> >> The first thing a link should send is LS OOS.
> >> >>
> >> >> In the out of service state, it would be desireable for the out of
> >> >> service side to always send LS OOS in response to other messages.
> >> >>
> >> >> Also, the out of service side should be permitted to send LS OOS
> >> >> periodically.
> >>
> >> What does 'first thing' mean? (upon START? upon association becoming
> >> established, as the current draft states?)
> >
> >First thing after association established.
> 
> Okay, good.

There's another "first thing".

Suppose that an implementation separates the SCTP association
from the M2PA.  In this case, the M2PA could go away and come
back without the SCTP association being dropped.  In that case
the M2PA should first send LS OOS when it comes back in the
Out of Service state.

--brian

> 
> >
> >>
> >> What problem is being solved by these three changes? Has anyone
> >> observed alignment loops at plugtests or in testing?
> >
> >AFAIK the draft allows LS to be sent periodically.
> 
> I don't know about this, since
> 
> "After the association is established, M2PA SHALL send a Link Status
> Out of Service message to its peer."
> 
> means 'send one LSO' to me.
> 
> Besides, you didn't answer my questions.
> 
> >
> >Also, sending LS OOS in response to other messages while in the out of
> service
> >state handles the SCTP Restart scenario without having to directoy address
> >SCTP restart.
> >
> >>
> >> Where in Q.703 does it specify these behaviors?
> >
> >Q.703 repeats LSSU, particularly SIOS, when the link is initially powered
> >on, regardless of any other messages while in the out of service state,
> and
> >continuously while in the out of service state.
> 
> Ah. I see it now -- not in the SDLs, but -- the last sentence in Q.703 7.2
> implies
> repeated sending of SIOS prior to Intial Alignment:
> 
>     "Status indication OS is sent on completion of 'power on' (see Figures
> 2
>     and 8) until initial alignment is started."
> 
> Note the 'until'. Okay. So these proposed changes do match Q.703.
> 
> Let me attempt to clarify the proposed changes with regards to out of
> service state:
> 
> OLD TEXT 4.1.3
> After the association is established, M2PA SHALL send a Link Status
> Out of Service message to its peer.
> 
> NEW TEXT 4.1.3
> After the association is established, M2PA SHALL send a Link Status
> Out of Service message to its peer and, prior to receiving a Start
> Request from MTP3, MAY periodically send Link Status Out of
> Service messages. During the Out of Service State M2PA SHOULD
> respond to received messages other than Link Status Out of Service
> by sending Link Status Out of Service.
> 
> >
> >>
> >> Do we have consensus that the changes will not result in a "potential
> loop"
> >> during alignment, as was the initial concern? I am not convinced that
> >> these three behaviors are necessary or desirable.
> >
> >What loop?  There is no loop in the Q.703 SDLs for sending SIOS while
> >in the out of service state.
> 
> You ask me? I'm just trying to understand what 'problem' is being
> addressed by proposed changes listed in this thread.
> 
> >
> >--brian
> >
> >>
> >> Regards,
> >>
> >> Jeff
> >>
> >> On 02/11/2003 04:06:34 PM sigtran-admin wrote:
> >>
> >> >Tom,
> >> >as i described earlier, my concern is the potential loop that could
> >> >occur during the initial LS Alignment exchange based on the requirement
> >> >that both peers must send and receive LS Alignment messages. Brian has
> >> >shown in his response (and i agree) that the scenario i described will
> >> >not happen if the procedures defined in the SDLs of the Q.703 spec is
> >> >adopted for m2pa in regards to the initial "alignment" exchange.
> >> >
> >> >wayne
> >> >
> >> >"Brian F. G. Bidulock" wrote:
> >> >>
> >> >> Tom,
> >> >>
> >> >> The first thing a link should send is LS OOS.
> >> >>
> >> >> In the out of service state, it would be desireable for the out of
> >> >> service side to always send LS OOS in response to other messages.
> >> >>
> >> >> Also, the out of service side should be permitted to send LS OOS
> >> >> periodically.
> >> >>
> >> >> With regards to Q.703 state machines, LS OOS in response to alignment
> >> >> will not abort alignment and T2 should still expire.  See test case
> >> >> "3.1.2 Timer T2" in draft-bidulock-sigtran-m2pa-test-02.txt.
> >> >>
> >> >> T2 is defined as the duration of time that the M2PA peer wil wait to
> >> >> receive a status "Alignment" message after sending a status
> "Alignment"
> >> >> message.
> >> >>
> >> >> As status Alignment will abort an otherwise in-service link, other
> >> >> link status received while running T2 should be ignored.
> >> >>
> >> >> Any different behaviour could interfere with MTP Level 3's link
> >> >> restoration timers and procedures.
> >> >>
> >> >> More comments below...
> >> >>
> >> >> Tom George
> >> >wrote:
> >>
> >> >(Mon, 10 Feb 2003 16:07:21)
> >> >> > Wayne,
> >> >> >
> >> >> > Yes, I want both sides to send and receive a LS Alignment message
> to
> >> >> > complete alignment.
> >> >> >
> >> >> > To fix the problem you describe, how about this" If M2PA has not
> >> >> > received MTP3 Start and receives LS Alignment, M2PA replies with LS
> >> OOS
> >> >> > (instead of not replying at all and counting on the other end's T2
> >> >> > expiring).
> >> >> >
> >> >> > tomG.
> >> >> >
> >> >> > wayne n taylor wrote:
> >> >> > >
> >> >> > > Hi All,
> >> >> > >
> >> >> > > i have a couple of comments/questions about the initial exchange
> of
> >> the
> >> >> > > Link Status Alignment messages. there seems to be some
> inconsistency
> >> in
> >> >> > > the text and finite state machine.
> >> >> > >
> >> >> > > section 4.1.3 paragraph 4:
> >> >> > >
> >> >> > > "Once the association is established and M2PA has received a
> Start
> >> >> > > Request from MTP3, M2PA sends the Link Status Alignment message
> to
> >> its
> >> >> > > peer. If M2PA has not already received the Link Status Alignment
> >> >> > > message from its peer, then M2PA starts timer T2."
> >> >> > >
> >> >> > > what state would the local M2PA be in to receive a LS Alignment
> >> message
> >> >> > > from its peer prior to getting a MTP3 Start ?. i assume the OOS
> >> state,
> >> >> > > but the finite state machine does not show LS Alignment as an
> input.
> >> >> > >
> >> >> > > section 4.1.3 paragraph 5:
> >> >> > >
> >> >> > > "(Note that if the remote M2PA has not received a Start Request
> from
> >> >> > > its MTP3, it will not send the Link Status Alignment message to
> the
> >> >> > > local M2PA. Eventually timer T2 in the local M2PA will expire. If
> >> the
> >> >> > > remote M2PA receives a Start Request from its MTP3 and sends Link
> >> >> > > Status Alignment before the local M2PA's timer T2 expires, the
> >> >> > > alignment procedure can continue.)"
> >> >> > >
> >> >> > > assume the local M2PA receives a Start Request from MTP3, sends
> LS
> >> >> > > Alignment and start T2, but the remote peer has not received a
> Start
> >> >> > > Request and is in OOS. the current finite state machine does not
> >> show
> >> >> > > what to do with the LS Alignment from the local peer so lets
> assume
> >> the
> >> >> > > LS Alignment is dropped. the remote peer receives a Start Request
> >> from
> >> >> > > MTP3, sends LS Alignment and start T2 (since the remote peer
> didnt
> >> get
> >> >> > > the local peers LS Alignment). assume further that timer T2 is
> still
> >> >> > > running on the local peer. the text from section 4.1.3 above
> states
> >> the
> >> >> > > "alignment procedure can continue". in this case the local peer
> sent
> >> a
> >> >> > > LS Alignment and received a LS Alignment. the remote peer only
> sent
> >> a LS
> >> >> > > Alignment and did not receive an LS Alignment (since LS Alignment
> >> >> > > messages are not retried). eventually T2 on the remote peer will
> >> expire,
> >> >> > > T3 on the local peer will expire and the alignment procedure
> starts
> >> over
> >> >> > > again. it seems possible for the peers to get caught in a loop
> >> trying to
> >> >> > > exchange LS Alignment messages.
> >> >>
> >> >> See draft-bidulock-sigtran-m2pa-test-02.txt Section "3.1.3 Timer T3".
> >> You
> >> >> will see that what you describe does not happen if you follow Q.703
> >> SDLs.
> >> >> In Q.703 once SIO is sent SIN is sufficient to complete alignment.
> So,
> >> >> one side sends ALIGNMENT and starts T2, the other side sends
> ALIGNMENT
> >> and
> >> >> starts T2.  The first side, receives the alignment, stops T2 and
> sends
> >> >> PROVING-NORMAL or PROVING-EMERGENCY and starts T3.  When the second
> side
> >> >> receives PROVING-NORMAL or PROVING-EMERGENCY it stops T2 and sends
> >> >> PROVING-NORMAL or PROVING-EMERGENCY and starts T4.  When the first
> side
> >> >> receives PROVING-NORMAL it stops T3 and starts T4.  When either T4
> >> expires,
> >> >> the side sends IN-SERVICE and starts T1.  When the side receives
> >> IN-SERVICE
> >> >> it stops T1 and indicates in_service to MTP3.
> >> >>
> >> >> This behavior can be seen on Figure 9/Q.703 (sheet 2 of 6) - Initial
> >> alignment
> >> >> control.  When the IAC starts it sends SIO and starts T2.  If it
> >> receives
> >> >> any of SIO, SIN or SIE it stops T2 and send SIN or SIE depending on
> >> emergency
> >> >> status and starts T3.  Receipt of a subsequent SIN or SIE (sheet 3 of
> 6)
> >> >> stops T3 and starts T4.  Note that if it receives SIO again, it
> ignores
> >> it
> >> >> at that point and continues to send SIE or SIN.
> >> >>
> >> >> These is no reason for M2PA not to follow the Q.703 SDLs in this
> regard.
> >> >>
> >> >> A statement that might clarify is that the M2PA sends LS ALIGNMENT
> and
> >> if
> >> >> it receives ALIGNMENT or PROVING it stops T2 and PROVING and starts
> T3.
> >> >> Note that it takes two PROVING messages in this case to move the peer
> >> into
> >> >> the proving state.  One PROVING message acts like an ALIGNMENT
> (cancels
> >> >> T2 and starts T3), the next acts like the first PROVING message
> (cancels
> >> >> T3 and starts T4).
> >> >>
> >> >> --brian
> >> >>
> >> >> > >
> >> >> > > section 4.1.3 paragraph 9:
> >> >> > >
> >> >> > > "When M2PA has both sent and received the Link Status Alignment
> >> >> > > message, it has completed alignment.  M2PA starts the aligned
> timer
> >> T3
> >> >> > > and moves to the proving state."
> >> >> > >
> >> >> > > is it a requirement for both the local and remote peers to
> >> send/receive
> >> >> > > LS Alignment for alignment to be complete ?. the only way to
> >> transition
> >> >> > > from Initial Alignment to Proving is to receive LS Alignment. in
> the
> >> >> > > scenario above, both peers do not send and receive LS Alignment
> >> >> > > messages.
> >> >> > >
> >> >> > > regards,
> >> >> > >
> >> >> > > wayne
> >> >> >
> >> >> > --
> >> >> > * Tom George       PB7-24001 (near B-21)
> Tom.George@alcatel.com
> >> *
> >> >> > * 972-519-3168   Intranet:
> http://www.ssd.usa.alcatel.com/~tgeorge/
> >> *
> >> >> > * Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas
> 75075
> >> *
> >> >> > * I actively disclaim speaking for Alcatel USA, Inc. I do not
> purport
> >> *
> >> >> > * to advertise or represent its position,  operations, or
> activities.
> >> *
> >> >> > _______________________________________________
> >> >> > Sigtran mailing list
> >> >> > Sigtran@ietf.org
> >> >> > https://www1.ietf.org/mailman/listinfo/sigtran
> >> >>
> >> >> --
> >> >> Brian F. G. Bidulock
> >> >> bidulock@openss7.org
> >> >> http://www.openss7.org/
> >> >> _______________________________________________
> >> >> Sigtran mailing list
> >> >> Sigtran@ietf.org
> >> >> https://www1.ietf.org/mailman/listinfo/sigtran
> >> >_______________________________________________
> >> >Sigtran mailing list
> >> >Sigtran@ietf.org
> >> >https://www1.ietf.org/mailman/listinfo/sigtran
> >>
> >
> >--
> >Brian F. G. Bidulock
> >bidulock@openss7.org
> >http://www.openss7.org/
> 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 14:53:09 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA13632
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 14:53:08 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CK2eG16716
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 15:02:40 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CK2Np16700;
	Wed, 12 Feb 2003 15:02:25 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CK1np16653
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 15:01:49 -0500
Received: from web12808.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id OAA13550
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 14:51:46 -0500 (EST)
Message-ID: <20030212195530.20271.qmail@web12808.mail.yahoo.com>
Received: from [62.29.77.73] by web12808.mail.yahoo.com via HTTP; Wed, 12 Feb 2003 11:55:30 PST
Date: Wed, 12 Feb 2003 11:55:30 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
In-Reply-To: <20030212084014.D21356@openss7.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Brian,
--- "Brian F. G. Bidulock" <bidulock@openss7.org>
wrote:
> Tolga,
> 
> I disagree strongly.  When we build n+1 redundant
> systems for carrier
> grade operation we do not stop service when a
> processor is still able
> to provide service.
> 
> If I have 3 processors in a 2+1 arrangement (where 2
> are needed to handle
> the maximum engineered load and 1 is spare), I do
> not stop service just
> because I have only 1 processor alive.  It just
> means that the system
> will not be able to handle the maximum engineered
> load with 1 processor.
[TOLGA]If in that configuration it is not mandatory to
stop traffic when 1 processor is alive, then IMO the
appropaite traffic mode should be 1+n.
> 
> Overload is a different issue.  If the 1 remaining
> processor is overloaded,
> it should use overload controls to keep itself
> alive.  In M3UA's case this
> could be overload controls at ISUP level or nodal
> TFC at MTP level or even
> SCTP flow control (failing to service read buffer).
> 
> Having the SG move to the AS-PENDING state when the
> number of active ASPs
> fall below "n" (2 in the case above) is completely
> undesirable.
> 
> I thought I made this clear on the "n+k redudancy"
> thread.
> 
> I strongly disagree with the changes of section 3.15
> of the IG.
[TOLGA] When n=1 for the satet machine in IG, it will
be the same like the state machine in IG. So, will it
still cause a serious problem? (but let's discuss and
finalize this issue)


__________________________________________________
Do you Yahoo!?
Yahoo! Shopping - Send Flowers for Valentine's Day
http://shopping.yahoo.com
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 14:54:46 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA13717
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 14:54:46 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CK4Ih16802
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 15:04:18 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CK44p16783;
	Wed, 12 Feb 2003 15:04:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CK3op16754
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 15:03:50 -0500
Received: from web12806.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id OAA13666
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 14:53:46 -0500 (EST)
Message-ID: <20030212195730.6289.qmail@web12806.mail.yahoo.com>
Received: from [62.29.77.73] by web12806.mail.yahoo.com via HTTP; Wed, 12 Feb 2003 11:57:30 PST
Date: Wed, 12 Feb 2003 11:57:30 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
In-Reply-To: <3E4A5668.4E899EE7@axes-mach01.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Srini,
--- ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
wrote:
> Tolga,
> 
> >>Notify(AS Pending) is sent when AS changes to
> AS-Pending state.
> 
> So as per section 3.15.2 of M3UA IG, AC2PN causes AS
> state change to
> AS-PENDING and forces to send Notify(AS Pending).
> i.e., when ASPs in the
> ASP-ACTIVE state for the AS falls below "n".
[TOLGA] That is my understanding. And of course n can
be 1.
> 
> >>For example if I am sure that I will need 2 ACTIVE
> ASPs then I would
> use n=2. But if I think that mostly 1 is enough but
> according to traffic
> load I might need a second one I would use n=1 and
> use NTFY(Insuff. ASP
> resources Active) when necessary.
> 
> Do you mean to say that NTFY(Insuff. ASP resources
> Active) would be used
> to make more than 'n' ASPs ACTIVE when necessary.
[TOLGA]That is my understanding.
> 
> --Srini
> 
> Tolga Asveren wrote:
> 
> > Srini,
> >
> > IMO Notify(AS Pending) is sent when AS chnages to
> > AS-Pending state. Notify(Insuff. ASP resources) is
> > sent, when there is no AS state change but SGP
> wants
> > to inform INACTIVE ASPs that it wants them to
> switch
> > to ACTIVE state. I think, the difference is
> required
> > v.s. preferred. For example if I am sure that I
> will
> > need 2 ACTIVE ASPs then I would use n=2. But if I
> > think that mostly 1 is enough but according to
> traffic
> > load I might need a second one I would use n=1 and
> use
> > NTFY(Insuff. ASP resources Active) when necessary.
> In
> > fact for that scenario, the number of preferred
> ASPs
> > is not predefined, i.e. it changes with traffic
> load,
> > so I wouldn't think it can be prearrenged between
> SGP
> > and ASP. But if one is sure about the required
> number
> > of ASPs then this value can be configured both on
> ASP
> > and SGP and an AS state chnage based on that value
> > seems appropriate to me.
> >
> >     Tolga
> > --- ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
> > wrote:
> > > Tolga,
> > >
> > >    Can you please tell me when to send
> > > Notify(AS-PENDING) and
> > > Notify(Insufficient ASP Resources Active in
> AS)?.
> > >
> > > --Srini
> > >
> > > Tolga Asveren wrote:
> > >
> > > >  If the traffic mode is 1+k I agree with that
> > > > statement but if it is n+k, I disagree.
> > > >
> > > >    Tolga
> > > > --- ksrinivas
> <ksrini@axes-mach01.usa.alcatel.com>
> > > > wrote:
> > > > > Brian,
> > > > >
> > > > > >>I believe that the number of active ASPs
> less
> > > than
> > > > > "n" corresponds to the AS-ACTIVE state (as
> in
> > > >>the
> > > > > RFC).
> > > > >
> > > > > I agree with this.
> > > > >
> > > > > --Srini
> > > > >
> > > > > "Brian F. G. Bidulock" wrote:
> > > > >
> > > > > > ksrinivas,
> > > > > >
> > > > > > ksrinivas wrote:
> > > > >                               (Wed, 12 Feb
> 2003
> > > > > 16:10:48)
> > > > > > > Brian,
> > > > > > >
> > > > > > > "Brian F. G. Bidulock" wrote:
> > > > > > >
> > > > > > > > Srini,
> > > > > > > >
> > > > > > > > ksrinivas wrote:
> > > > >                                   (Wed, 12
> Feb
> > > 2003
> > > > > 13:42:16)
> > > > > > > > > Brian,
> > > > > > > > >
> > > > > > > > >      Can you tell me when an SGP
> should
> > > send
> > > > > Notify(AS-PENDING) and
> > > > > > > > > Notify(Insufficient ASP Resources
> Active
> > > in
> > > > > AS)?. I felt RFC and IG of
> > > > > > > > > M3UA is not clear on these two.
> > > > > > > > >
> > > > > > > > > --Srini
> > > > > > > >
> > > > > > > > It think that it is pretty
> > > straight-forward.
> > > > > > > >
> > > > > > > > NTFY(Insufficient ASPs) is sent for an
> AS
> > > by
> > > > > an SGP when the number of
> > > > > > > > ASPs in the ASP-ACTIVE state for the
> AS
> > > falls
> > > > > below "n", but not to
> > > > > > > > zero, (as a result of ASP failures or
> > > > > deactivations); where "n" is the
> > > > > > > > minimum number of active ASPs defined
> for
> > > the
> > > > > AS.  So this notification
> > > > > > > > is only sent if "n" is greater than 1.
> > > > > > > >
> > > > > > > > NTFY(AS-PENDING) is sent for an AS by
> an
> > > SGP
> > > > > when the number of ASPs in
> > > > > > > > the ASP-ACTIVE state for the AS falls
> to
> > > zero
> > > > > (as a result of ASP
> > > > > > > > failures or deactivations).
> > > > > > > >
> > > > > > >
> > > > > > > I understood the same way initially. But
> the
> > > > > section 3.15.2 of M3UA IG-02
> > > > > > > has confused me. The following is the
> text
> > > from
> > > > > this section.
> > > > > > >
> > > > > > > "AC2PN: one ASP in ASP-ACTIVE state
> moves to
> > > > > ASP-INACTIVE or ASP-DOWN
> > > > > > > states, causing the number of ASPs in
> > > ASP-ACTIVE
> > > > > drop below n."
> > > > > > >
> > > > > > > So the AS moves to AS-PENDING state when
> > > ASPs in
> > > > > ASP-ACTIVE drop below 'n'
> > > > > > > (but not necessarily zero). Do we not
> have
> > > to
> > > > > send Notify(AS-PENDING) when
> > > > > > > AS has moved to AS-PENDING state?.
> > > > > >
> > > > > > I see.
> > > > > >
> > > > > > I largely disagree with the IG section
> 3.15.2.
> > > > > Consider that the new
> > > > > > description of the AS-PENDING state is:
> > > > > >
> > > > > >     AS-PENDING: An active ASP has
> transitioned
> > > > > from ASP-ACTIVE to ASP-
> > > > > >     INACTIVE or ASP-DOWN and it was the
> number
> > > n
> > > > > active ASP in the AS.  A
> > > > > >     recovery timer T(r) SHOULD be started
> and
> > > all
> > > > > incoming signalling messages
> > > > > >     SHOULD be queued by the SGP. If an ASP
> > > becomes
> 
=== message truncated ===


__________________________________________________
Do you Yahoo!?
Yahoo! Shopping - Send Flowers for Valentine's Day
http://shopping.yahoo.com
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 15:07:51 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA14354
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 15:07:51 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CKHNS17927
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 15:17:23 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CKHAp17917;
	Wed, 12 Feb 2003 15:17:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CKGrp17902
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 15:16:53 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA14340
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 15:06:50 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:JjnS3++XOYG/FSgnjwCk6tNGPPT5IeX8@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1CKAYD25779;
	Wed, 12 Feb 2003 13:10:34 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1CKAYD02770;
	Wed, 12 Feb 2003 13:10:34 -0700
Date: Wed, 12 Feb 2003 13:10:33 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tolga Asveren <tolga_asveren@yahoo.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030212131033.A1284@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030212084014.D21356@openss7.org> <20030212195530.20271.qmail@web12808.mail.yahoo.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20030212195530.20271.qmail@web12808.mail.yahoo.com>; from tolga_asveren@yahoo.com on Wed, Feb 12, 2003 at 11:55:30AM -0800
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Tolga,

Tolga Asveren wrote:                                                                 (Wed, 12 Feb 2003 11:55:30)
> Brian,
> --- "Brian F. G. Bidulock" <bidulock@openss7.org>
> wrote:
> > Tolga,
> > 
> > I disagree strongly.  When we build n+1 redundant
> > systems for carrier
> > grade operation we do not stop service when a
> > processor is still able
> > to provide service.
> > 
> > If I have 3 processors in a 2+1 arrangement (where 2
> > are needed to handle
> > the maximum engineered load and 1 is spare), I do
> > not stop service just
> > because I have only 1 processor alive.  It just
> > means that the system
> > will not be able to handle the maximum engineered
> > load with 1 processor.
> [TOLGA]If in that configuration it is not mandatory to
> stop traffic when 1 processor is alive, then IMO the
> appropaite traffic mode should be 1+n.

It is NEVER the case that it is mandatory to stop traffic
when the number of active processors falls below anyting
but 1.

You are interpreting the n in n+k to mean "the number of
preocessors beneath which the system will stop working."

This is not what n in n+k is used for in telephony.  The
n is "the number of precessors beneath which the system
will not be able to process the engineered maximum load."

You simply have the wrong concept of n.

> > 
> > Overload is a different issue.  If the 1 remaining
> > processor is overloaded,
> > it should use overload controls to keep itself
> > alive.  In M3UA's case this
> > could be overload controls at ISUP level or nodal
> > TFC at MTP level or even
> > SCTP flow control (failing to service read buffer).
> > 
> > Having the SG move to the AS-PENDING state when the
> > number of active ASPs
> > fall below "n" (2 in the case above) is completely
> > undesirable.
> > 
> > I thought I made this clear on the "n+k redudancy"
> > thread.
> > 
> > I strongly disagree with the changes of section 3.15
> > of the IG.
> [TOLGA] When n=1 for the satet machine in IG, it will
> be the same like the state machine in IG. So, will it
> still cause a serious problem? (but let's discuss and
> finalize this issue)

You have the wrong n.

AS-PENDING state should only be entered when the all the
ASPs are inactive.

Insuffient ASPs active should be sent when the number of
active processors falls below n (n is the number of
processors required to process the engineered maximum load).

If the ASPs want to follow your non-telephony concept of n,
they are welcome to deactivate when they receive the
Insuffient ASPs notification.

Strike section 3.15. from the IG and leave things the way
they are in the RFC.  That will resolve the issue.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 15:28:12 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA14832
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 15:28:12 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CKbj619301
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 15:37:45 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CKbEp18875;
	Wed, 12 Feb 2003 15:37:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CKaAp18583
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 15:36:10 -0500
Received: from domino2.nc.tekelec.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA14767
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 15:26:06 -0500 (EST)
From: jeffrey.craig@tekelec.com
Received: from domino4.nc.tekelec.com ([198.89.40.9])
          by domino2.nc.tekelec.com (Lotus Domino Release 5.0.10)
          with ESMTP id 2003021215233213:33397 ;
          Wed, 12 Feb 2003 15:23:32 -0500 
Subject: Re: [Sigtran] M2PA draft 7 question -- alignment conclusion
To: bidulock@openss7.org
Cc: wayne n taylor <waynet@cisco.com>, Tom George <Tom.George@alcatel.com>,
        sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.7  March 21, 2001
Message-ID: <OFCE38CC73.0B5491B2-ON85256CCB.006AFB69@nc.tekelec.com>
Date: Wed, 12 Feb 2003 15:25:35 -0500
MIME-Version: 1.0
X-MIMETrack: Serialize by Router on Domino4/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/12/2003 03:25:37 PM,
	Itemize by SMTP Server on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/12/2003 03:23:32 PM,
	Serialize by Router on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/12/2003 03:23:34 PM,
	Serialize complete at 02/12/2003 03:23:34 PM
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hello Brian,

This thread has gotten too long. Let's try and conclude it with
a summary of the recommended changes to the M2PA draft.
The intent is to match Q.703.

Regards,

Jeff

----------
Figure 7 is updated to show a transition from INITIAL_ALIGNMENT
to PROVING upon reception of either a Link Status Alignment or
Link Status Proving Normal or Link Status Proving Emergency
message. (Q.703 figure 4, figure 9 sheet 2)


----------
OLD TEXT 4.1.3
After the association is established, M2PA SHALL send a Link Status
Out of Service message to its peer.

NEW TEXT 4.1.3
After the association is established, M2PA SHALL send a Link Status
Out of Service message to its peer and, prior to receiving a Start
Request from MTP3, MAY periodically send Link Status Out of
Service messages. During the Out of Service State M2PA SHOULD
respond to received messages other than Link Status Out of Service
by sending Link Status Out of Service.


----------
OLD TEXT 4.1.3
M2PA stops timer T2 when it has received the Link Status Alignment
message from its peer.

NEW TEXT 4.1.3
M2PA stops timer T2 when it has received a Link Status Alignment
message or Link Status Proving Normal or Link Status Emergency
message from its peer.


----------
OLD TEXT 4.1.3
Note: Between the time M2PA sends Link Status Alignment to its peer
and receives Link Status Alignment from its peer, M2PA may receive
Link Status Out of Service from its peer. This message is
ignored. After receiving Link Status Alignment from the peer, receipt
of a Link Status Out of Service message causes M2PA to send Out of
Service to MTP3 and return to the Out of Service state.

NEW TEXT 4.1.3
Note: Between the time M2PA sends Link Status Alignment to its peer
and receives either Link Status Alignment or Link Status Proving
Normal or Link Status Proving Emergency from its peer, M2PA may
receive Link Status Out of Service from its peer. This message is
ignored. After receiving Link Status Alignment or Link Status
Proving Normal or Link Status Proving Emergency from its peer,
receipt of a Link Status Out of Service message causes M2PA to
send Out of Service to MTP3 and return to the Out of Service state.

----------
OLD TEXT 4.1.3

When M2PA has both sent and received the Link Status Alignment
message, it has completed alignment.  M2PA starts the aligned timer
T3 and moves to the proving state.

M2PA stops timer T3 when it receives a Proving Normal or Proving
Emergency message and starts proving period timer T4.

NEW TEXT 4.1.3

When M2PA has sent Link Status Alignment and received either a Link
Status Alignment or a Link Status Proving Normal or a Link Status
Proving Emergency message, it has completed alignment. M2PA starts
the aligned timer T3 and moves to the proving state. Note that
timer T3 is started even if reception of a proving message caused
the transition to the proving state.

M2PA stops timer T3 when it receives a Link Status Proving Normal
or Link Status Proving Emergency message and starts proving period
timer T4.


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 15:35:31 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA15024
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 15:35:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CKj4L19659
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 15:45:04 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CKiap19597;
	Wed, 12 Feb 2003 15:44:36 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CKhjp19542
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 15:43:45 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA14945
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 15:33:40 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1CKbOeW010216;
	Wed, 12 Feb 2003 14:37:24 -0600 (CST)
Message-ID: <3E4AB084.4987B82F@alcatel.com>
Date: Wed, 12 Feb 2003 14:37:24 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Security section (was: m2pa comment: author list)
References: <F92978223B01D311ACFF0008C71EE19D015DFA9C@MCHH225E>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Michael,

This is a good idea. I could replace the text in 6.1 with a reference to
the SIGTRAN security draft.

Why was the "Site Security Handbook" RFC 2196 not mentioned in the
SIGTRAN security draft? I think I picked up my reference to this
document from the M3UA RFC. 

There is no mention of the consideration about wireless users as in 6.2.
Should this be in the SIGTRAN security draft, M2PA, both, or neither?

tomG.


Tuexen Michael wrote:
> 
> Hi,
> 
> why not simply referring to the SIGTRAN security doc instead
> of reproducing part of it?
> 
> Best regards
> Michael
> 
> Michael Tuexen   Tel.:   +49 89 722 47210
> Siemens AG       Fax:    +49 89 722 48212
> ICN CP SE C 7    e-mail: Michael.Tuexen@siemens.com
> 
> > -----Original Message-----
> > From: Tom George [mailto:Tom.George@alcatel.com]
> > Sent: Tuesday, February 11, 2003 10:58 PM
> > Cc: 'sigtran@ietf.org'
> > Subject: [Sigtran] M2PA Security section (was: m2pa comment: author
> > list)
> >
> >
> > On Jan 26, 2003 "Ong, Lyndon" wrote:
> >   [...]
> > > The security section probably needs to be updated based on the
> > > security framework document (which should come out this week).
> >   [...]
> >
> > Here is my first stab at revising section 6 of the M2PA
> > draft. Comments
> > are welcome, especially since I am not familiar with the security
> > protocols.
> >
> > tomG.
> >
> > =============================
> > 6. Security
> >
> > M2PA is designed to carry signaling messages for telephony
> > services. As such, M2PA MUST involve the security needs of several
> > parties:
> >
> >    - the end users of the services
> >
> >    - the network providers
> >
> >    - the applications involved
> >
> > Additional requirements MAY come from local regulation.
> >
> > While these parties may have some overlapping security needs, their
> > needs may not be identical. Any security solution SHOULD fulfill all
> > of the different parties' needs.
> >
> > 6.1 Security Requirements
> >
> > As a transport protocol, M2PA has the following security objectives:
> >
> >    - Availability of reliable and timely user data transport.
> >
> >    - Integrity of user data transport.
> >
> >    - Confidentiality of user data.
> >
> > M2PA runs on top of SCTP [RFC2960].  SCTP provides certain transport
> > related security features, as discussed in [Security]. However, these
> > security features MAY NOT be adequate for all of the involved parties'
> > requirements.
> >
> > When M2PA is running in professionally managed corporate or service
> > provider network, it is reasonable to expect that this network
> > includes an appropriate security policy framework. The "Site Security
> > Handbook" [RFC2196] SHOULD be consulted for guidance.
> >
> > When the network in which M2PA runs involves more than one party
> > (e.g., a non-private network), it MAY NOT be reasonable to expect that
> > all parties have implemented security in a sufficient manner.  In such
> > a case, it is recommended that IPSec or TLS be used to ensure
> > confidentiality of the user payload. Consult [Security] for more
> > guidance on the use of security protocols.
> >
> > 6.2 Protecting Confidentiality
> >
> > Particularly for mobile users, the requirement for confidentiality MAY
> > include the masking of IP addresses and ports. In this case
> > application-level encryption is not sufficient. IPSec ESP SHOULD be
> > used instead [RFC2401].  Regardless of which level performs the
> > encryption, the IPSec ISAKMP service SHOULD be used for key
> > management.
> >
> > =============================
> >
> > {Add an informative reference:}
> >
> >   [Security]
> >        J. Loughney, M. Tuexen, and J. Pastor-Balbas, "Security
> >        Considerations for SIGTRAN Protocols,"
> >        draft-ietf-sigtran-security-01.txt (January 2003).
> >
> > =============================
> >
> > --
> > * Tom George       PB7-24001 (near B-21)
> > Tom.George@alcatel.com  *
> > * 972-519-3168   Intranet:
> http://www.ssd.usa.alcatel.com/~tgeorge/  *
> * Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
> * I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
> * to advertise or represent its position,  operations, or activities. *
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 15:37:37 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA15087
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 15:37:37 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CKlAG19734
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 15:47:10 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CKk3p19708;
	Wed, 12 Feb 2003 15:46:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CKj7p19672
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 15:45:07 -0500
Received: from web12806.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id PAA15002
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 15:35:02 -0500 (EST)
Message-ID: <20030212203847.15909.qmail@web12806.mail.yahoo.com>
Received: from [62.29.77.73] by web12806.mail.yahoo.com via HTTP; Wed, 12 Feb 2003 12:38:47 PST
Date: Wed, 12 Feb 2003 12:38:47 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
In-Reply-To: <20030212131033.A1284@openss7.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Brian,
--- "Brian F. G. Bidulock" <bidulock@openss7.org>
wrote:
> Tolga,
> 
> Tolga Asveren wrote:                                
>                                 (Wed, 12 Feb 2003
> 11:55:30)
> > Brian,
> > --- "Brian F. G. Bidulock" <bidulock@openss7.org>
> > wrote:
> > > Tolga,
> > > 
> > > I disagree strongly.  When we build n+1
> redundant
> > > systems for carrier
> > > grade operation we do not stop service when a
> > > processor is still able
> > > to provide service.
> > > 
> > > If I have 3 processors in a 2+1 arrangement
> (where 2
> > > are needed to handle
> > > the maximum engineered load and 1 is spare), I
> do
> > > not stop service just
> > > because I have only 1 processor alive.  It just
> > > means that the system
> > > will not be able to handle the maximum
> engineered
> > > load with 1 processor.
> > [TOLGA]If in that configuration it is not
> mandatory to
> > stop traffic when 1 processor is alive, then IMO
> the
> > appropaite traffic mode should be 1+n.
> 
> It is NEVER the case that it is mandatory to stop
> traffic
> when the number of active processors falls below
> anyting
> but 1.
> 
> You are interpreting the n in n+k to mean "the
> number of
> preocessors beneath which the system will stop
> working."
> 
> This is not what n in n+k is used for in telephony. 
> The
> n is "the number of precessors beneath which the
> system
> will not be able to process the engineered maximum
> load."
> 
> You simply have the wrong concept of n.
[TOLGA]If the consensus is that there is no need for
the concept of n+k traffic mode, where n is mandatory,
I would consider this also as a solution (in fact this
was one of the proposals made)But this also requires
some changes to be done to RFC (please see below)
> 
> > > 
> > > Overload is a different issue.  If the 1
> remaining
> > > processor is overloaded,
> > > it should use overload controls to keep itself
> > > alive.  In M3UA's case this
> > > could be overload controls at ISUP level or
> nodal
> > > TFC at MTP level or even
> > > SCTP flow control (failing to service read
> buffer).
> > > 
> > > Having the SG move to the AS-PENDING state when
> the
> > > number of active ASPs
> > > fall below "n" (2 in the case above) is
> completely
> > > undesirable.
> > > 
> > > I thought I made this clear on the "n+k
> redudancy"
> > > thread.
> > > 
> > > I strongly disagree with the changes of section
> 3.15
> > > of the IG.
> > [TOLGA] When n=1 for the satet machine in IG, it
> will
> > be the same like the state machine in IG. So, will
> it
> > still cause a serious problem? (but let's discuss
> and
> > finalize this issue)
> 
> You have the wrong n.
> 
> AS-PENDING state should only be entered when the all
> the
> ASPs are inactive.
> 
> Insuffient ASPs active should be sent when the
> number of
> active processors falls below n (n is the number of
> processors required to process the engineered
> maximum load).
> 
> If the ASPs want to follow your non-telephony
> concept of n,
> they are welcome to deactivate when they receive the
> Insuffient ASPs notification.
> 
> Strike section 3.15. from the IG and leave things
> the way
> they are in the RFC.  That will resolve the issue.
[TOLGA]I disagree with your statement that leaving RFC
as it is provides a solution.
4.3.4.3 ASP Active Procedures*6th paragraph before the
end)
"For the n+k redundancy case, ASPs which are in that
AS should coordinate among themselves the number of
active ASPs in the AS, and should start sending
traffic only after n ASPs are active."

Again in 4.3.4.3 just one paragraphs before
"An SGP or IPSP, upon recepition of an ASP Active
message for the first ASP in a Loadshare AS, MAY
choose not to direct traffic to a newly active ASP
until it determines that there are sufficient
resources to handle the expected load(e.g., until
there are "n" ASPs in state ASP-ACTIVE in the AS). In
this case, the SGP or IPSP SHOULD withold the
Notify(AS-ACTIVE) until there are sufficient
resources."

Now IMO the first paragraph assumes that depending on
n AS state changes whereas the second one doesn't so
there is a constradiction between them. If we use n
with the meaning you suggest, then I wouldn't think it
should have a role to decide when to start/stop
traffic. In that case, does it have any impact on the
procedures between ASP/SGP? Like that, the only
necessary thing to be indicated in the RFC/IG IMO is
to say, when SGP wants more ACTIVE ASPs for whatever
reason, it should send NTFY(Insuff. ASP resources
Active) and there is no need to mention anything about
n.(and I would think the first paragraph above should
be deleted and the second one modified to make this
work properly -all this assuming there is no ralistic
scenario where n>1 is mandatory-)



__________________________________________________
Do you Yahoo!?
Yahoo! Shopping - Send Flowers for Valentine's Day
http://shopping.yahoo.com
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 15:43:45 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA15261
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 15:43:45 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CKrIk20069
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 15:53:18 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CKr5p20040;
	Wed, 12 Feb 2003 15:53:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CKqxp20020
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 15:52:59 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA15235
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 15:42:55 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:PEst+rQOqcoVRwNOOotGbIjYfWkz84I5@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1CKkdD26388;
	Wed, 12 Feb 2003 13:46:39 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1CKkd007819;
	Wed, 12 Feb 2003 13:46:39 -0700
Date: Wed, 12 Feb 2003 13:46:39 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: jeffrey.craig@tekelec.com
Cc: wayne n taylor <waynet@cisco.com>, Tom George <Tom.George@alcatel.com>,
        sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft 7 question -- alignment conclusion
Message-ID: <20030212134639.C1284@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: jeffrey.craig@tekelec.com,
	wayne n taylor <waynet@cisco.com>,
	Tom George <Tom.George@alcatel.com>, sigtran@ietf.org
References: <OFCE38CC73.0B5491B2-ON85256CCB.006AFB69@nc.tekelec.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OFCE38CC73.0B5491B2-ON85256CCB.006AFB69@nc.tekelec.com>; from jeffrey.craig@tekelec.com on Wed, Feb 12, 2003 at 03:25:35PM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

jeffrey.craig,

There is the requirement that M2PA send repeated PROVING-NORMAL or
PROVING-EMERGENCY messages while running T3 for this to work.

--brian

jeffrey.craig@tekelec.com wrote:                                             (Wed, 12 Feb 2003 15:25:35)
> 
> Hello Brian,
> 
> This thread has gotten too long. Let's try and conclude it with
> a summary of the recommended changes to the M2PA draft.
> The intent is to match Q.703.
> 
> Regards,
> 
> Jeff
> 
> ----------
> Figure 7 is updated to show a transition from INITIAL_ALIGNMENT
> to PROVING upon reception of either a Link Status Alignment or
> Link Status Proving Normal or Link Status Proving Emergency
> message. (Q.703 figure 4, figure 9 sheet 2)
> 
> 
> ----------
> OLD TEXT 4.1.3
> After the association is established, M2PA SHALL send a Link Status
> Out of Service message to its peer.
> 
> NEW TEXT 4.1.3
> After the association is established, M2PA SHALL send a Link Status
> Out of Service message to its peer and, prior to receiving a Start
> Request from MTP3, MAY periodically send Link Status Out of
> Service messages. During the Out of Service State M2PA SHOULD
> respond to received messages other than Link Status Out of Service
> by sending Link Status Out of Service.
> 
> 
> ----------
> OLD TEXT 4.1.3
> M2PA stops timer T2 when it has received the Link Status Alignment
> message from its peer.
> 
> NEW TEXT 4.1.3
> M2PA stops timer T2 when it has received a Link Status Alignment
> message or Link Status Proving Normal or Link Status Emergency
> message from its peer.
> 
> 
> ----------
> OLD TEXT 4.1.3
> Note: Between the time M2PA sends Link Status Alignment to its peer
> and receives Link Status Alignment from its peer, M2PA may receive
> Link Status Out of Service from its peer. This message is
> ignored. After receiving Link Status Alignment from the peer, receipt
> of a Link Status Out of Service message causes M2PA to send Out of
> Service to MTP3 and return to the Out of Service state.
> 
> NEW TEXT 4.1.3
> Note: Between the time M2PA sends Link Status Alignment to its peer
> and receives either Link Status Alignment or Link Status Proving
> Normal or Link Status Proving Emergency from its peer, M2PA may
> receive Link Status Out of Service from its peer. This message is
> ignored. After receiving Link Status Alignment or Link Status
> Proving Normal or Link Status Proving Emergency from its peer,
> receipt of a Link Status Out of Service message causes M2PA to
> send Out of Service to MTP3 and return to the Out of Service state.
> 
> ----------
> OLD TEXT 4.1.3
> 
> When M2PA has both sent and received the Link Status Alignment
> message, it has completed alignment.  M2PA starts the aligned timer
> T3 and moves to the proving state.
> 
> M2PA stops timer T3 when it receives a Proving Normal or Proving
> Emergency message and starts proving period timer T4.
> 
> NEW TEXT 4.1.3
> 
> When M2PA has sent Link Status Alignment and received either a Link
> Status Alignment or a Link Status Proving Normal or a Link Status
> Proving Emergency message, it has completed alignment. M2PA starts
> the aligned timer T3 and moves to the proving state. Note that
> timer T3 is started even if reception of a proving message caused
> the transition to the proving state.
> 
> M2PA stops timer T3 when it receives a Link Status Proving Normal
> or Link Status Proving Emergency message and starts proving period
> timer T4.
> 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 16:15:20 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA15831
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 16:15:20 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CLOs521708
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 16:24:54 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CLOap21689;
	Wed, 12 Feb 2003 16:24:37 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CLNZp21637
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 16:23:35 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA15800
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 16:13:30 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:5+YV04Q1YBGLp9D/G3fkmlfvGVj475vs@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1CLHED26955;
	Wed, 12 Feb 2003 14:17:14 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1CLHEt12491;
	Wed, 12 Feb 2003 14:17:14 -0700
Date: Wed, 12 Feb 2003 14:17:14 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tolga Asveren <tolga_asveren@yahoo.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030212141714.D1284@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030212131033.A1284@openss7.org> <20030212203847.15909.qmail@web12806.mail.yahoo.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20030212203847.15909.qmail@web12806.mail.yahoo.com>; from tolga_asveren@yahoo.com on Wed, Feb 12, 2003 at 12:38:47PM -0800
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Tolga,

Tolga Asveren wrote:                                                                 (Wed, 12 Feb 2003 12:38:47)
> > 
> > Strike section 3.15. from the IG and leave things
> > the way
> > they are in the RFC.  That will resolve the issue.
> [TOLGA]I disagree with your statement that leaving RFC
> as it is provides a solution.
> 4.3.4.3 ASP Active Procedures*6th paragraph before the
> end)
> "For the n+k redundancy case, ASPs which are in that
> AS should coordinate among themselves the number of
> active ASPs in the AS, and should start sending
> traffic only after n ASPs are active."
> 
> Again in 4.3.4.3 just one paragraphs before
> "An SGP or IPSP, upon recepition of an ASP Active
> message for the first ASP in a Loadshare AS, MAY
> choose not to direct traffic to a newly active ASP
> until it determines that there are sufficient
> resources to handle the expected load(e.g., until
> there are "n" ASPs in state ASP-ACTIVE in the AS). In
> this case, the SGP or IPSP SHOULD withold the
> Notify(AS-ACTIVE) until there are sufficient
> resources."

There is no problem with the above statement.  Because
the newly active AS was previously inactive, the
inaccessibiliy of the MTP-SAP has already been indicated
to the SS7 network.  There is no traffic to offer until
the SG says otherwise.  It is the SG's responsibility to
determine MTP-SAP (SPMC) state as appropriate, and one
of the things it may consider is the number of ASPs
active within a time window for a newly active AS.

For example, one could set a timer when the first ASP
comes active and if n is surpassed before the timer
expires, indicate availability of the MTP-SAP to the
SS7 network, or if the timer expires, indicate availability
of the MTP-SAP to the SS7 network anyway.

The section above was meant to address the situation where
ASPs activate in a staggered fashion due to reboot and under
traffic backpressure it would behoove the SG to wait for
a short period to let as many come active as are going to
come active (or as many as can handle the engineered max
load, in this case "n") before indicating accessibility of
the AS to the SS7 network.

The first section goes hand in hand with the second.  If
the SG is going to hesitate before indicating that the
MTP-SAP is active, then the ASPs should as well (otherwise
the SG will be forced to send messages on behalf of the
MTP-SAP and can no longer resonably report it inaccessible).

The other statement in the RFC (that the ASPs wait for
NTFY(AS Active) and the SGP witholds NTFY(AS Active) until
"n" ASPs are active) goes hand in hand with these two.

If the SG cares to delay indicating that the MTP-SAP is
accessible to the SS7 network, it should also delay the
sending of NTFY(AS Active).  The ASPs can then know to
suppress sending of messages until receiving NTFY(AS
Active) from the SG and will know when the SG feels
that the MTP-SAP should be declared acessible to the SS7
network.


> 
> Now IMO the first paragraph assumes that depending on
> n AS state changes whereas the second one doesn't so
> there is a constradiction between them. If we use n
> with the meaning you suggest, then I wouldn't think it
> should have a role to decide when to start/stop
> traffic.

The decision to start traffic and the decision to stop
traffic are two separate decisions.

In deciding when to start traffic after a failure, it
would be smart to consider the ability of the system to
handle the upcoming load.  This is because ocillations
due to overload has a more disasterous effect on
telephone calls than outage (resource in otherwise
unaffected nodes are consumed to no purpose).  Also,
after an outage occurs there is invariably steep
increase in the traffic that will be initially offered
due to reattempts, particularly where the AS represents
an entire point code.  I have seen outages where the
reattempt pressure was five to ten times the traffic
levels offered just before the failure.

In deciding when to stop traffic the only reasonable
criteria for a telephony system is to only stop traffic
when it is impossible to process any traffic (i.e,
when the number of active ASPs for the AS falls to zero).
Anything else would fail system reliability engineering
considerations and would be excessively costly to the
operator in terms of cash flow.

It appears that it is not a contradiction at all.  The
SG can just not be required to apply the start traffic
criteria to the stop traffic decision, or visa versa.

> In that case, does it have any impact on the
> procedures between ASP/SGP? Like that, the only
> necessary thing to be indicated in the RFC/IG IMO is
> to say, when SGP wants more ACTIVE ASPs for whatever
> reason, it should send NTFY(Insuff. ASP resources
> Active) and there is no need to mention anything about
> n.(and I would think the first paragraph above should
> be deleted and the second one modified to make this
> work properly -all this assuming there is no ralistic
> scenario where n>1 is mandatory-)

The paragraphs are fine and the meaning of "n" is consistent
across both the start decision and the stop decision.  "n"
represents the number of active ASPs necessary to handle
the maximum engineered load.  In the start decision case,
it is best to wait for "n" ASPs to become available before
flooding them with the reattempt backpressure.  In the
stop decision case, it is best to wait for the AS to go
inactive before pending on the AS.

Again, IMO the RFC is ok as it stands.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 17:06:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA17392
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 17:06:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CMGUr25341
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 17:16:30 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CMGEp25330;
	Wed, 12 Feb 2003 17:16:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CMFmp25303
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 17:15:48 -0500
Received: from web12805.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id RAA17355
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 17:05:43 -0500 (EST)
Message-ID: <20030212220925.31809.qmail@web12805.mail.yahoo.com>
Received: from [62.29.85.164] by web12805.mail.yahoo.com via HTTP; Wed, 12 Feb 2003 14:09:25 PST
Date: Wed, 12 Feb 2003 14:09:25 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
In-Reply-To: <20030212141714.D1284@openss7.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Brian,
--- "Brian F. G. Bidulock" <bidulock@openss7.org>
wrote:
> Tolga,
> 
> Tolga Asveren wrote:                                
>                                 (Wed, 12 Feb 2003
> 12:38:47)
> > > 
> > > Strike section 3.15. from the IG and leave
> things
> > > the way
> > > they are in the RFC.  That will resolve the
> issue.
> > [TOLGA]I disagree with your statement that leaving
> RFC
> > as it is provides a solution.
> > 4.3.4.3 ASP Active Procedures*6th paragraph before
> the
> > end)
> > "For the n+k redundancy case, ASPs which are in
> that
> > AS should coordinate among themselves the number
> of
> > active ASPs in the AS, and should start sending
> > traffic only after n ASPs are active."
> > 
> > Again in 4.3.4.3 just one paragraphs before
> > "An SGP or IPSP, upon recepition of an ASP Active
> > message for the first ASP in a Loadshare AS, MAY
> > choose not to direct traffic to a newly active ASP
> > until it determines that there are sufficient
> > resources to handle the expected load(e.g., until
> > there are "n" ASPs in state ASP-ACTIVE in the AS).
> In
> > this case, the SGP or IPSP SHOULD withold the
> > Notify(AS-ACTIVE) until there are sufficient
> > resources."
> 
> There is no problem with the above statement. 
> Because
> the newly active AS was previously inactive, the
> inaccessibiliy of the MTP-SAP has already been
> indicated
> to the SS7 network.  There is no traffic to offer
> until
> the SG says otherwise.  It is the SG's
> responsibility to
> determine MTP-SAP (SPMC) state as appropriate, and
> one
> of the things it may consider is the number of ASPs
> active within a time window for a newly active AS.
> 
> For example, one could set a timer when the first
> ASP
> comes active and if n is surpassed before the timer
> expires, indicate availability of the MTP-SAP to the
> SS7 network, or if the timer expires, indicate
> availability
> of the MTP-SAP to the SS7 network anyway.
> 
> The section above was meant to address the situation
> where
> ASPs activate in a staggered fashion due to reboot
> and under
> traffic backpressure it would behoove the SG to wait
> for
> a short period to let as many come active as are
> going to
> come active (or as many as can handle the engineered
> max
> load, in this case "n") before indicating
> accessibility of
> the AS to the SS7 network.
> 
> The first section goes hand in hand with the second.
>  If
> the SG is going to hesitate before indicating that
> the
> MTP-SAP is active, then the ASPs should as well
> (otherwise
> the SG will be forced to send messages on behalf of
> the
> MTP-SAP and can no longer resonably report it
> inaccessible).
> 
> The other statement in the RFC (that the ASPs wait
> for
> NTFY(AS Active) and the SGP witholds NTFY(AS Active)
> until
> "n" ASPs are active) goes hand in hand with these
> two.
[TOLGA]Is there a place in RFC/IG, which indicates
that ASPs SHOULD wait for NTFY(AS Active) to start
send/receive traffic? To my knowledge there isn't and
this depends just on receiving ASPAC-ACK and on the
number of ACTIVE ASPs -based on the above paragraph-
IMO the not nice thing with those two statements is
that ASPs SHOULD wait for n ACTIVE ASPs to start
sending traffic, whereas SGP MAY withhold traffic.
There should be a consistency between them.
> 
> If the SG cares to delay indicating that the MTP-SAP
> is
> accessible to the SS7 network, it should also delay
> the
> sending of NTFY(AS Active).  The ASPs can then know
> to
> suppress sending of messages until receiving NTFY(AS
> Active) from the SG and will know when the SG feels
> that the MTP-SAP should be declared acessible to the
> SS7
> network.
> 
> 
> > 
> > Now IMO the first paragraph assumes that depending
> on
> > n AS state changes whereas the second one doesn't
> so
> > there is a constradiction between them. If we use
> n
> > with the meaning you suggest, then I wouldn't
> think it
> > should have a role to decide when to start/stop
> > traffic.
> 
> The decision to start traffic and the decision to
> stop
> traffic are two separate decisions.
> 
> In deciding when to start traffic after a failure,
> it
> would be smart to consider the ability of the system
> to
> handle the upcoming load.  This is because
> ocillations
> due to overload has a more disasterous effect on
> telephone calls than outage (resource in otherwise
> unaffected nodes are consumed to no purpose).  Also,
> after an outage occurs there is invariably steep
> increase in the traffic that will be initially
> offered
> due to reattempts, particularly where the AS
> represents
> an entire point code.  I have seen outages where the
> reattempt pressure was five to ten times the traffic
> levels offered just before the failure.
> 
> In deciding when to stop traffic the only reasonable
> criteria for a telephony system is to only stop
> traffic
> when it is impossible to process any traffic (i.e,
> when the number of active ASPs for the AS falls to
> zero).
> Anything else would fail system reliability
> engineering
> considerations and would be excessively costly to
> the
> operator in terms of cash flow.
> 
> It appears that it is not a contradiction at all. 
> The
> SG can just not be required to apply the start
> traffic
> criteria to the stop traffic decision, or visa
> versa.
> 
> > In that case, does it have any impact on the
> > procedures between ASP/SGP? Like that, the only
> > necessary thing to be indicated in the RFC/IG IMO
> is
> > to say, when SGP wants more ACTIVE ASPs for
> whatever
> > reason, it should send NTFY(Insuff. ASP resources
> > Active) and there is no need to mention anything
> about
> > n.(and I would think the first paragraph above
> should
> > be deleted and the second one modified to make
> this
> > work properly -all this assuming there is no
> ralistic
> > scenario where n>1 is mandatory-)
> 
> The paragraphs are fine and the meaning of "n" is
> consistent
> across both the start decision and the stop
> decision.  "n"
> represents the number of active ASPs necessary to
> handle
> the maximum engineered load.  In the start decision
> case,
> it is best to wait for "n" ASPs to become available
> before
> flooding them with the reattempt backpressure.  In
> the
> stop decision case, it is best to wait for the AS to
> go
> inactive before pending on the AS.
> 
> Again, IMO the RFC is ok as it stands.
[TOLGA]If you think that for start case it is best to
wait for n, then SG SHOULD withhold traffic instead of
MAY withhold traffic(like ASP SHOULD wait for n ACTIVE
ASPs) In fact as far as I recall the request to change
this MAY to SHOULD was the reason, why the debate
about n+k started. IMO how ASP and SGP behave
regarding n should be the same.
If this MAY is changed to SHOULD, I would think still
an update is necessary to AS state machine for
AS-INACTIVE to AS-ACTIVE transition (nth ASP goes
ACTIVE instead of 1st ASP goes ACTIVE and we don't
touch transition from ACTIVE to PENDING)
If nobody thinks that there is a realistic scenario,
where n>1 is mandatory, i.e. less than n ACTIVE ASP
will change AS state to PENDING-, then I think the
approach you explained makes sense.
> 
> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/


__________________________________________________
Do you Yahoo!?
Yahoo! Shopping - Send Flowers for Valentine's Day
http://shopping.yahoo.com
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 18:33:42 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA18792
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 18:33:41 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CNhH529941
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 18:43:17 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CNh0p29928;
	Wed, 12 Feb 2003 18:43:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CNfTp29877
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 18:41:29 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA18771
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 18:31:23 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1CNZ5qe003500;
	Wed, 12 Feb 2003 17:35:06 -0600 (CST)
Message-ID: <3E4ADA29.42A1DA9D@alcatel.com>
Date: Wed, 12 Feb 2003 17:35:05 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft 7 question -- alignment conclusion
References: <OFCE38CC73.0B5491B2-ON85256CCB.006AFB69@nc.tekelec.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Jeff,

If M2PA sends LS Alignment and receives a LS Proving, I don't see the
point in starting T3. Since T3 is supposed to make sure that the other
side sends a LS Proving, why not just stop T2, skip T3, and begin
Proving?

tomG.

jeffrey.craig@tekelec.com wrote:
  [...]
> ----------
> OLD TEXT 4.1.3
> 
> When M2PA has both sent and received the Link Status Alignment
> message, it has completed alignment.  M2PA starts the aligned timer
> T3 and moves to the proving state.
> 
> M2PA stops timer T3 when it receives a Proving Normal or Proving
> Emergency message and starts proving period timer T4.
> 
> NEW TEXT 4.1.3
> 
> When M2PA has sent Link Status Alignment and received either a Link
> Status Alignment or a Link Status Proving Normal or a Link Status
> Proving Emergency message, it has completed alignment. M2PA starts
> the aligned timer T3 and moves to the proving state. Note that
> timer T3 is started even if reception of a proving message caused
> the transition to the proving state.
> 
> M2PA stops timer T3 when it receives a Link Status Proving Normal
> or Link Status Proving Emergency message and starts proving period
> timer T4.
  [...]

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 18:49:48 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA19113
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 18:49:48 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1CNxO430368
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 18:59:24 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CNxAp30347;
	Wed, 12 Feb 2003 18:59:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1CNwip30325
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 18:58:44 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA19099
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 18:48:37 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1CNqKqe005507;
	Wed, 12 Feb 2003 17:52:20 -0600 (CST)
Message-ID: <3E4ADE34.4F4EC572@alcatel.com>
Date: Wed, 12 Feb 2003 17:52:20 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft 07 comment: LPO
References: <3D72C18DF140D311A54F0008C7DBAC8D02917438@MCHH252E> <20030212084436.E21356@openss7.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Franz,

I updated the diagram because of another discussion. I put an excerpt
from it further down in this message. I think it shows the transition
you asked about.

But there is nothing in the text that says this. I have this paragraph: 

======
If there is a local processor outage condition during the alignment
procedure, M2PA sends Link Status Processor Outage to its peer. When
the local processor outage condition ends, then M2PA SHALL send Link
Status Processor Outage Ended to its peer. M2PA SHALL attempt to
complete the alignment procedure during the local processor outage
condition.
======

Perhaps I should add something like this: "If the local processor outage
condition exists when M2PA completes Proving, M2PA shall enter the
ALIGNED NOT READY state."


======
  |                       |   +-----+                  |                
  |                       |   |     | LPO/LPR          |                
  |                       V   V     |                  |                
  |                 +-----------+   |                  |
  |                 |  PROVING  |---+                  |
  |<----------------|           |--------------------->+
  |      MTP3 Stop  +-----------+     SCTP Comm Error  |
  |   OR T3 Expiry        |  |     OR SCTP Comm Lost   |
  |   OR Receive LS OOS   |  |                         |
  |                       |  +-------------------------|--------+
  |                       |     T4 Expiry              |        |      
  |                       |     Send LS PO             |        |      
  |                       |                            |        |      
  |                       | T4 Expiry                  |        |      
  |                       | Send LS Ready              |        |      
  |                       |                            |        |      
  |                       |                            |        |      
  |                       V           SCTP Comm Error  |        V      
  |                 +-----------+  OR SCTP Comm Lost   |  +-----------+
  |                 |  ALIGNED  |--------------------->+  |  ALIGNED  |
  +<----------------|   READY   |<----------------------->| NOT READY |
  |      MTP3 Stop  +-----------+     LPO/LPR             +-----------+
  |   OR T1 Expiry        |        OR Receive LS PO/PR       |       |
  |   OR Receive LS OOS   |                                  |   T1  | 
  |   OR Receive LS Align | Receive LS Ready      Receive    | Expiry|
  |                       | OR Receive User Data    LS Ready | OR Rcv|
  |                       |                                  |    LS |
  |                       |                                  |  Align|
======


tomG.

"Brian F. G. Bidulock" wrote:
> 
> Gaiser,
> 
> Gaiser Franz wrote:     (Tue, 11 Feb 2003 11:09:58)
> > Hi all,
> >
> > I have an additional comment/remark to the description of handling Processor
> > Outage in draft 07of M2PA. The handling of the following situation is not
> > clear to me: - Proving Period completed (T4 expired) and Local Processor
> > Outage is occured. In the case without LPO the m2pa would start sending LS
> > Ready to the peer and moves to state ALIGNED READY.
> 
> > For the above situation (with LPO) the handling is not clearly stated. Is LS
> > Ready and/or LS PO to be send ?According to Q.703  the m2pa should send LS
> > PO message instead of LS Ready and moves to state ALIGNED NOT READY.
> 
> Then that's what M2PA should do.
> 
> --brian
> 
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 21:54:36 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA22152
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 21:54:36 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1D34Ht07672
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 22:04:17 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1D33qp07659;
	Wed, 12 Feb 2003 22:03:53 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1D32tp07623
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 22:02:55 -0500
Received: from mail.ritt.com.cn (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA22136
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 21:52:42 -0500 (EST)
Received: from [10.8.11.33] by mail_server
  (ArGoSoft Mail Server Pro for WinNT/2000/XP, Version 1.8 (1.8.1.5)); Thu, 13 Feb 2003 10:50:52 
Message-ID: <001201c2d30b$b37b52e0$1dfea8c0@cqrail.com>
From: "Lu Jun" <lujun@mail.ritt.com.cn>
To: "sigtran" <sigtran@ietf.org>
Subject: [Sigtran] M2UA message acknowlede
Date: Thu, 13 Feb 2003 10:57:51 +0800
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----=_NextPart_000_000F_01C2D34E.C0C5E620"
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 5.50.4133.2400
X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

This is a multi-part message in MIME format.

------=_NextPart_000_000F_01C2D34E.C0C5E620
Content-Type: text/plain;
	charset="gb2312"
Content-Transfer-Encoding: quoted-printable

HI

I have a question about the correlation ID for M2UA
Does all the message must be ack by DataACK message with correlation ID =
.
If the receive the DATA without correlation ID , how the peer M2UA to =
ack this message,
does it use the SCTP stream number to ack the received DATA?

LuJun

------=_NextPart_000_000F_01C2D34E.C0C5E620
Content-Type: text/html;
	charset="gb2312"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; charset=3Dgb2312">
<META content=3D"MSHTML 5.50.4134.100" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3D"Times New Roman" size=3D4>HI</FONT></DIV>
<DIV><FONT face=3D"Times New Roman" size=3D4></FONT>&nbsp;</DIV>
<DIV><FONT size=3D2><FONT face=3D"Times New Roman" =
size=3D4>I&nbsp;have</FONT>=20
</FONT><FONT face=3D"Times New Roman" size=3D4>a question about the =
correlation ID=20
for M2UA</FONT></DIV>
<DIV><FONT face=3D"Times New Roman" size=3D4>Does all the message must =
be ack by=20
DataACK message with correlation ID .</FONT></DIV>
<DIV><FONT face=3D"Times New Roman" size=3D4>If the receive the DATA =
without=20
correlation ID , how the peer M2UA to ack this message,</FONT></DIV>
<DIV><FONT face=3D"Times New Roman" size=3D4>does it use the SCTP stream =
number to=20
ack the received DATA?</FONT></DIV>
<DIV><FONT face=3D"Times New Roman" size=3D4></FONT>&nbsp;</DIV>
<DIV><FONT face=3D"Times New Roman" =
size=3D4>LuJun</FONT></DIV></BODY></HTML>

------=_NextPart_000_000F_01C2D34E.C0C5E620--


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 21:57:35 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA22221
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 21:57:35 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1D37GA08360
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 22:07:16 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1D373p07930;
	Wed, 12 Feb 2003 22:07:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1D36Wp07774
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 22:06:32 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA22202
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 21:56:19 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:abfg3O24UAn4psZaSQQIjcuMVO+anzTt@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1D304D32524;
	Wed, 12 Feb 2003 20:00:04 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1D303I25776;
	Wed, 12 Feb 2003 20:00:03 -0700
Date: Wed, 12 Feb 2003 20:00:02 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tolga Asveren <tolga_asveren@yahoo.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030212200002.C17731@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030212141714.D1284@openss7.org> <20030212220925.31809.qmail@web12805.mail.yahoo.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20030212220925.31809.qmail@web12805.mail.yahoo.com>; from tolga_asveren@yahoo.com on Wed, Feb 12, 2003 at 02:09:25PM -0800
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Tolga,

Tolga Asveren wrote:                                                                 (Wed, 12 Feb 2003 14:09:25)

> [TOLGA]Is there a place in RFC/IG, which indicates
> that ASPs SHOULD wait for NTFY(AS Active) to start
> send/receive traffic? To my knowledge there isn't and
> this depends just on receiving ASPAC-ACK and on the
> number of ACTIVE ASPs -based on the above paragraph-
> IMO the not nice thing with those two statements is
> that ASPs SHOULD wait for n ACTIVE ASPs to start
> sending traffic, whereas SGP MAY withhold traffic.
> There should be a consistency between them.

Also, section 4.3.4.3 middle of page 82, top of page 83 (just after
the sections you quoted):

Loadshare:

   An SGP or IPSP, upon reception of an ASP Active message for the first
   ASP in a Loadshare AS, MAY choose not to direct traffic to a newly
   active ASP until it determines that there are sufficient resources to
   handle the expected load (e.g., until there are "n" ASPs in state
   ASP-ACTIVE in the AS).  In this case, the SGP or IPSP SHOULD withhold
   the Notify (AS-ACTIVE) until there are sufficient resources.

Broadcast:

   An SGP or IPSP, upon reception of an ASP Active message for the first
   ASP in a Broadcast AS, MAY choose not to direct traffic to a newly
   active ASP until it determines that there are sufficient resources to
   handle the expected load (e.g., until there are "n" ASPs in state
   ASP-ACTIVE in the AS).  In this case, the SGP or IPSP SHOULD withhold
   the Notify (AS-ACTIVE) until there are sufficient resources.

As the SGP SHOULD withhold the Notify (AS-ACTIVE), the ASP can rely on
its being withheld as an indication that the SGP is not directing traffic
to to the active ASPs for the AS because insufficient ASPs have been
activated.  Because the ASP SHOULD withhold traffic in the case of
insufficient ASPs, the ASP SHOULD NOT start traffic until it receives
Notify (AS-ACTIVE) from the SGP.  That seems pretty straightforward to me.

Also, see example 5.1.4.

> [TOLGA]If you think that for start case it is best to
> wait for n, then SG SHOULD withhold traffic instead of
> MAY withhold traffic(like ASP SHOULD wait for n ACTIVE
> ASPs) In fact as far as I recall the request to change
> this MAY to SHOULD was the reason, why the debate
> about n+k started. IMO how ASP and SGP behave
> regarding n should be the same.

I disagree with SHOULD.  Directing traffic on activation requires
and exchange between the SG and the SS7 network.  These exchanges
are outside the scope of the RFC and the RFC cannot place requirements
on the SS7 network.  Also, in the case of SI granularity AS, the
SG cannot "direct traffic" to the AS.  It will take the expiry of UFC
timers at the remote signalling points before any traffic will be
restarted or directed.  By saying MAY, we are saying that the SG is
graciously permitted to do what it needs to do with respect to the
SS7 network.

However, we do then say that if the SG does not direct traffic that
it SHOULD withold the Notify (AS-ACTIVE) to let the ASP know what it
is doing.  This is SG/ASP interaction and is within the scope of the
RFC.

Also, what the ASPs do with regard to the SG is directly within the
scope of the RFC and the RFC can place the SHOULD requirement on the
ASPs.  The ASPs should not restart traffic until it knows that n ASPs
are up.  And how does it tell this?  Well the RFC alludes to some
inter-ASP coordination which is also outside thes scope of the RFC
(APS to ASP protocol).  The only requirement in the RFC that allows
the ASP to determine that n ASPs are active is that the SG has sent
Notify (AS-ACTIVE) for the AS.

If you want to clarify "SHOULD wait until n ASPs are active" to
"SHOULD wait for Notify (AS-ACTIVE)", feel free, as it seems very
apparent to me anyway (and was the result of the thread just after
last call on the subject, too).

> If this MAY is changed to SHOULD,

MAY cannot change to SHOULD because it is an action towards the
SS7 network upon which M3UA cannot place requirements.

> I would think still
> an update is necessary to AS state machine for
> AS-INACTIVE to AS-ACTIVE transition (nth ASP goes
> ACTIVE instead of 1st ASP goes ACTIVE and we don't
> touch transition from ACTIVE to PENDING)

No, the state transition is still valid, it is just that the SGP
SHOULD withhold the Notify (AS-ACTIVE) until the nth ASP is active
iff it withholds traffic.

The current state machine in the RFC is fine.

> If nobody thinks that there is a realistic scenario,
> where n>1 is mandatory, i.e. less than n ACTIVE ASP
> will change AS state to PENDING-, then I think the
> approach you explained makes sense.

Putting perfectly good processors out of service by blocking all
traffic to them and creating a complete service outage where one is
not necessary could not possibly be construed as realistic by any
operator.  I'd hate to be in front of Congress explaning that one!

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 21:59:33 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA22275
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 21:59:33 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1D39Ew08598
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 22:09:14 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1D393p08586;
	Wed, 12 Feb 2003 22:09:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1D38ep08544
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 22:08:40 -0500
Received: from mail.ritt.com.cn (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA22236
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 21:58:26 -0500 (EST)
Received: from [10.8.11.33] by mail_server
  (ArGoSoft Mail Server Pro for WinNT/2000/XP, Version 1.8 (1.8.1.5)); Thu, 13 Feb 2003 10:56:37 
Message-ID: <001201c2d30c$81475840$1dfea8c0@cqrail.com>
From: "Lu Jun" <lujun@mail.ritt.com.cn>
To: "sigtran" <sigtran@ietf.org>
Subject: [Sigtran] M2UA message acknowledge
Date: Thu, 13 Feb 2003 11:03:36 +0800
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----=_NextPart_000_000F_01C2D34F.8EA12DC0"
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 5.50.4133.2400
X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

This is a multi-part message in MIME format.

------=_NextPart_000_000F_01C2D34F.8EA12DC0
Content-Type: text/plain;
	charset="gb2312"
Content-Transfer-Encoding: quoted-printable

Hi
I have another question for M2UA implement on SG
does the NO.7 link message ack independent on the ack of M2UA?
it means that the message can be acked by BSN number ,but this message =
may not
acked by remote M2UA ASP, and stored in SG M2UA transmit queue .

Lujun

------=_NextPart_000_000F_01C2D34F.8EA12DC0
Content-Type: text/html;
	charset="gb2312"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; charset=3Dgb2312">
<META content=3D"MSHTML 5.50.4134.100" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT size=3D2>Hi</FONT></DIV>
<DIV><FONT size=3D2>I have another question for M2UA implement on =
SG</FONT></DIV>
<DIV><FONT size=3D2>does the NO.7 link message ack independent on the =
ack of=20
M2UA?</FONT></DIV>
<DIV><FONT size=3D2>it means that the message can be acked by BSN number =
,but this=20
message may not</FONT></DIV>
<DIV><FONT size=3D2>acked by remote M2UA ASP, and stored in SG M2UA =
transmit=20
queue&nbsp;.</FONT></DIV>
<DIV><FONT size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT size=3D2>Lujun</FONT></DIV></BODY></HTML>

------=_NextPart_000_000F_01C2D34F.8EA12DC0--


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 22:01:36 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA22355
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 22:01:36 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1D3BHi08732
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 22:11:17 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1D3B5p08715;
	Wed, 12 Feb 2003 22:11:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1D3AMp08674
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 22:10:22 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA22318
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 22:00:10 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:Ib19rnmalgf5B9oODhvyZtdQ6kqCfBFe@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1D33tD32665;
	Wed, 12 Feb 2003 20:03:55 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1D33tv26386;
	Wed, 12 Feb 2003 20:03:55 -0700
Date: Wed, 12 Feb 2003 20:03:54 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tom George <Tom.George@alcatel.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft 7 question -- alignment conclusion
Message-ID: <20030212200354.D17731@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tom George <Tom.George@alcatel.com>, sigtran@ietf.org
References: <OFCE38CC73.0B5491B2-ON85256CCB.006AFB69@nc.tekelec.com> <3E4ADA29.42A1DA9D@alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E4ADA29.42A1DA9D@alcatel.com>; from Tom.George@alcatel.com on Wed, Feb 12, 2003 at 05:35:05PM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Tom,

Tom George wrote:                                                                      (Wed, 12 Feb 2003 17:35:05)
> Jeff,
> 
> If M2PA sends LS Alignment and receives a LS Proving, I don't see the
> point in starting T3. Since T3 is supposed to make sure that the other
> side sends a LS Proving, why not just stop T2, skip T3, and begin
> Proving?

'cause that's what Q.703 SDLs do.  They treat the first SIN/SIE as
though it was an SIO, starts T3 and still require the next SIN/SIE
before stopping T3.

--brian


> 
> tomG.
> 
> jeffrey.craig@tekelec.com wrote:
>   [...]
> > ----------
> > OLD TEXT 4.1.3
> > 
> > When M2PA has both sent and received the Link Status Alignment
> > message, it has completed alignment.  M2PA starts the aligned timer
> > T3 and moves to the proving state.
> > 
> > M2PA stops timer T3 when it receives a Proving Normal or Proving
> > Emergency message and starts proving period timer T4.
> > 
> > NEW TEXT 4.1.3
> > 
> > When M2PA has sent Link Status Alignment and received either a Link
> > Status Alignment or a Link Status Proving Normal or a Link Status
> > Proving Emergency message, it has completed alignment. M2PA starts
> > the aligned timer T3 and moves to the proving state. Note that
> > timer T3 is started even if reception of a proving message caused
> > the transition to the proving state.
> > 
> > M2PA stops timer T3 when it receives a Link Status Proving Normal
> > or Link Status Proving Emergency message and starts proving period
> > timer T4.
>   [...]
> 
> -- 
> * Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
> * 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
> * Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
> * I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
> * to advertise or represent its position,  operations, or activities. *
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 22:08:36 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA22563
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 22:08:36 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1D3IHu09026
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 22:18:17 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1D3I6p09012;
	Wed, 12 Feb 2003 22:18:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1D3Hap08991
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 22:17:36 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA22540
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 22:07:24 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:i8+rALkgEJotFAskwRs4g/WeDoxm3Ene@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1D3B9D32761;
	Wed, 12 Feb 2003 20:11:09 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1D3B9i27370;
	Wed, 12 Feb 2003 20:11:09 -0700
Date: Wed, 12 Feb 2003 20:11:09 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Lu Jun <lujun@mail.ritt.com.cn>
Cc: sigtran <sigtran@ietf.org>
Subject: Re: [Sigtran] M2UA message acknowledge
Message-ID: <20030212201109.E17731@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Lu Jun <lujun@mail.ritt.com.cn>,
	sigtran <sigtran@ietf.org>
References: <001201c2d30c$81475840$1dfea8c0@cqrail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <001201c2d30c$81475840$1dfea8c0@cqrail.com>; from lujun@mail.ritt.com.cn on Thu, Feb 13, 2003 at 11:03:36AM +0800
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Lu,

The purpose of M2UA's ack is to indicate buffer depth rather than to
acknowledge messages.  No retransmission and no negative ack is possible.
M2UA at the SG should not wait for the M2UA data ack before acking MSUs.

The SG can gauge the depth of the "Receive Buffer" by the depth of its
internal receive buffer and the number of (or byte count of) messages
sent to the ASP awaiting acknowledgement.  It can make determinations of
receive congestion based on this information.

The ASP can gauge the depth of the "Transmit Buffer" by the depth of is
internal transmit buffer and the number of (or byte count of) messages
sent to the SG awaiting acknowledgement.  It can make determinations of
transmit congestion (and congestion level) based on this information.

It is not required that these function be peformed in the fashion I
explain above.  It is only required that if the sender has included the
Correlation Id parameter in the Data message that the receive must
ack the message so that it can perform the functions it desires.  In
this way there is interoperability between implementations that provide
procedure such as those above and implementations that do not.

--brian

Lu Jun wrote:                                                                           (Thu, 13 Feb 2003 11:03:36)
> 
>    Hi
> 
>    I have another question for M2UA implement on SG
> 
>    does the NO.7 link message ack independent on the ack of M2UA?
> 
>    it means that the message can be acked by BSN number ,but this message
>    may not
> 
>    acked by remote M2UA ASP, and stored in SG M2UA transmit queue .
> 
> 
> 
>    Lujun

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 22:09:33 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA22598
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 22:09:33 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1D3JER09098
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 22:19:14 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1D3J4p09091;
	Wed, 12 Feb 2003 22:19:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1D3IWp09053
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 22:18:32 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA22553
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 22:08:20 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:bd3yX3imMXywPvqUfPVBm6R8R1EZgFCv@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1D3C5D00306;
	Wed, 12 Feb 2003 20:12:05 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1D3C5i27502;
	Wed, 12 Feb 2003 20:12:05 -0700
Date: Wed, 12 Feb 2003 20:12:04 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Lu Jun <lujun@mail.ritt.com.cn>
Cc: sigtran <sigtran@ietf.org>
Subject: Re: [Sigtran] M2UA message acknowlede
Message-ID: <20030212201204.F17731@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Lu Jun <lujun@mail.ritt.com.cn>,
	sigtran <sigtran@ietf.org>
References: <001201c2d30b$b37b52e0$1dfea8c0@cqrail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <001201c2d30b$b37b52e0$1dfea8c0@cqrail.com>; from lujun@mail.ritt.com.cn on Thu, Feb 13, 2003 at 10:57:51AM +0800
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Lu,

M2UA does not ack messages sent without a Correlation Id parameter.

--brian

Lu Jun wrote:                                                                           (Thu, 13 Feb 2003 10:57:51)
> 
>    HI
> 
> 
> 
>    I have a question about the correlation ID for M2UA
> 
>    Does  all  the message must be ack by DataACK message with correlation
>    ID .
> 
>    If  the receive the DATA without correlation ID , how the peer M2UA to
>    ack this message,
> 
>    does it use the SCTP stream number to ack the received DATA?
> 
> 
> 
>    LuJun

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 22:42:36 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA23375
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 22:42:36 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1D3qGi10899
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 22:52:16 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1D3q5p10890;
	Wed, 12 Feb 2003 22:52:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1D3pop10837
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 22:51:50 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA23363
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 22:41:37 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1D3jIeW004864;
	Wed, 12 Feb 2003 21:45:20 -0600 (CST)
Received: from axes-mach01.usa.alcatel.com by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id IAA22329; Thu, 13 Feb 2003 08:14:14 -0500 (GMT)
Message-ID: <3E4B14E5.535F7963@axes-mach01.usa.alcatel.com>
Date: Thu, 13 Feb 2003 09:15:41 +0530
From: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: bidulock@openss7.org
CC: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
References: <3E4A452A.85F1083@axes-mach01.usa.alcatel.com> <20030212131558.41891.qmail@web12801.mail.yahoo.com> <20030212084014.D21356@openss7.org>
Content-Type: text/plain; charset=x-user-defined
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Tolga,

    Moreover the SG has to buffer the traffic in such a situation(when load is more).

--Srini

"Brian F. G. Bidulock" wrote:

> Tolga,
>
> I disagree strongly.  When we build n+1 redundant systems for carrier
> grade operation we do not stop service when a processor is still able
> to provide service.
>
> If I have 3 processors in a 2+1 arrangement (where 2 are needed to handle
> the maximum engineered load and 1 is spare), I do not stop service just
> because I have only 1 processor alive.  It just means that the system
> will not be able to handle the maximum engineered load with 1 processor.
>
> Overload is a different issue.  If the 1 remaining processor is overloaded,
> it should use overload controls to keep itself alive.  In M3UA's case this
> could be overload controls at ISUP level or nodal TFC at MTP level or even
> SCTP flow control (failing to service read buffer).
>
> Having the SG move to the AS-PENDING state when the number of active ASPs
> fall below "n" (2 in the case above) is completely undesirable.
>
> I thought I made this clear on the "n+k redudancy" thread.
>
> I strongly disagree with the changes of section 3.15 of the IG.
>
> --brian
>
> Tolga Asveren wrote:                                                                 (Wed, 12 Feb 2003 05:15:58)
> >  If the traffic mode is 1+k I agree with that
> > statement but if it is n+k, I disagree.
> >
> >    Tolga
> > --- ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
> > wrote:
>
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 12 23:24:29 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA23964
	for <sigtran-archive@odin.ietf.org>; Wed, 12 Feb 2003 23:24:29 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1D4YAn12717
	for sigtran-archive@odin.ietf.org; Wed, 12 Feb 2003 23:34:10 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1D4Xpp12670;
	Wed, 12 Feb 2003 23:33:52 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1D4WZp12630
	for <sigtran@optimus.ietf.org>; Wed, 12 Feb 2003 23:32:35 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA23947
	for <sigtran@ietf.org>; Wed, 12 Feb 2003 23:22:22 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1D4Q1qe000333;
	Wed, 12 Feb 2003 22:26:03 -0600 (CST)
Received: from axes-mach01.usa.alcatel.com by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id IAA23493; Thu, 13 Feb 2003 08:55:06 -0500 (GMT)
Message-ID: <3E4B1E7A.5CD88963@axes-mach01.usa.alcatel.com>
Date: Thu, 13 Feb 2003 09:56:34 +0530
From: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Tolga Asveren <tolga_asveren@yahoo.com>
CC: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
References: <20030212195730.6289.qmail@web12806.mail.yahoo.com>
Content-Type: text/plain; charset=x-user-defined
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Tolga,

IMO, in n+k redundancy model, 'n' are needed to handle the maximum
engineered load and the remaining 'k' are to be used as spare. When any
of the Active ASP of 'n' is down/inactive, ASPs of the remaining 'k'
(spare) should be notified for insufficient ASP resources inorder to
move them to ASP Active state to again make 'n' ASP Active to handle the
maximum engineered load.

--Srini

Tolga Asveren wrote:

> Srini,
> --- ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
> wrote:
> > Tolga,
> >
> > >>Notify(AS Pending) is sent when AS changes to
> > AS-Pending state.
> >
> > So as per section 3.15.2 of M3UA IG, AC2PN causes AS
> > state change to
> > AS-PENDING and forces to send Notify(AS Pending).
> > i.e., when ASPs in the
> > ASP-ACTIVE state for the AS falls below "n".
> [TOLGA] That is my understanding. And of course n can
> be 1.
> >
> > >>For example if I am sure that I will need 2 ACTIVE
> > ASPs then I would
> > use n=2. But if I think that mostly 1 is enough but
> > according to traffic
> > load I might need a second one I would use n=1 and
> > use NTFY(Insuff. ASP
> > resources Active) when necessary.
> >
> > Do you mean to say that NTFY(Insuff. ASP resources
> > Active) would be used
> > to make more than 'n' ASPs ACTIVE when necessary.
> [TOLGA]That is my understanding.
> >
> > --Srini
> >
> > Tolga Asveren wrote:
> >
> > > Srini,
> > >
> > > IMO Notify(AS Pending) is sent when AS chnages to
> > > AS-Pending state. Notify(Insuff. ASP resources) is
> > > sent, when there is no AS state change but SGP
> > wants
> > > to inform INACTIVE ASPs that it wants them to
> > switch
> > > to ACTIVE state. I think, the difference is
> > required
> > > v.s. preferred. For example if I am sure that I
> > will
> > > need 2 ACTIVE ASPs then I would use n=2. But if I
> > > think that mostly 1 is enough but according to
> > traffic
> > > load I might need a second one I would use n=1 and
> > use
> > > NTFY(Insuff. ASP resources Active) when necessary.
> > In
> > > fact for that scenario, the number of preferred
> > ASPs
> > > is not predefined, i.e. it changes with traffic
> > load,
> > > so I wouldn't think it can be prearrenged between
> > SGP
> > > and ASP. But if one is sure about the required
> > number
> > > of ASPs then this value can be configured both on
> > ASP
> > > and SGP and an AS state chnage based on that value
> > > seems appropriate to me.
> > >
> > >     Tolga
> > > --- ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
> > > wrote:
> > > > Tolga,
> > > >
> > > >    Can you please tell me when to send
> > > > Notify(AS-PENDING) and
> > > > Notify(Insufficient ASP Resources Active in
> > AS)?.
> > > >
> > > > --Srini
> > > >
> > > > Tolga Asveren wrote:
> > > >
> > > > >  If the traffic mode is 1+k I agree with that
> > > > > statement but if it is n+k, I disagree.
> > > > >
> > > > >    Tolga
> > > > > --- ksrinivas
> > <ksrini@axes-mach01.usa.alcatel.com>
> > > > > wrote:
> > > > > > Brian,
> > > > > >
> > > > > > >>I believe that the number of active ASPs
> > less
> > > > than
> > > > > > "n" corresponds to the AS-ACTIVE state (as
> > in
> > > > >>the
> > > > > > RFC).
> > > > > >
> > > > > > I agree with this.
> > > > > >
> > > > > > --Srini
> > > > > >
> > > > > > "Brian F. G. Bidulock" wrote:
> > > > > >
> > > > > > > ksrinivas,
> > > > > > >
> > > > > > > ksrinivas wrote:
> > > > > >                               (Wed, 12 Feb
> > 2003
> > > > > > 16:10:48)
> > > > > > > > Brian,
> > > > > > > >
> > > > > > > > "Brian F. G. Bidulock" wrote:
> > > > > > > >
> > > > > > > > > Srini,
> > > > > > > > >
> > > > > > > > > ksrinivas wrote:
> > > > > >                                   (Wed, 12
> > Feb
> > > > 2003
> > > > > > 13:42:16)
> > > > > > > > > > Brian,
> > > > > > > > > >
> > > > > > > > > >      Can you tell me when an SGP
> > should
> > > > send
> > > > > > Notify(AS-PENDING) and
> > > > > > > > > > Notify(Insufficient ASP Resources
> > Active
> > > > in
> > > > > > AS)?. I felt RFC and IG of
> > > > > > > > > > M3UA is not clear on these two.
> > > > > > > > > >
> > > > > > > > > > --Srini
> > > > > > > > >
> > > > > > > > > It think that it is pretty
> > > > straight-forward.
> > > > > > > > >
> > > > > > > > > NTFY(Insufficient ASPs) is sent for an
> > AS
> > > > by
> > > > > > an SGP when the number of
> > > > > > > > > ASPs in the ASP-ACTIVE state for the
> > AS
> > > > falls
> > > > > > below "n", but not to
> > > > > > > > > zero, (as a result of ASP failures or
> > > > > > deactivations); where "n" is the
> > > > > > > > > minimum number of active ASPs defined
> > for
> > > > the
> > > > > > AS.  So this notification
> > > > > > > > > is only sent if "n" is greater than 1.
> > > > > > > > >
> > > > > > > > > NTFY(AS-PENDING) is sent for an AS by
> > an
> > > > SGP
> > > > > > when the number of ASPs in
> > > > > > > > > the ASP-ACTIVE state for the AS falls
> > to
> > > > zero
> > > > > > (as a result of ASP
> > > > > > > > > failures or deactivations).
> > > > > > > > >
> > > > > > > >
> > > > > > > > I understood the same way initially. But
> > the
> > > > > > section 3.15.2 of M3UA IG-02
> > > > > > > > has confused me. The following is the
> > text
> > > > from
> > > > > > this section.
> > > > > > > >
> > > > > > > > "AC2PN: one ASP in ASP-ACTIVE state
> > moves to
> > > > > > ASP-INACTIVE or ASP-DOWN
> > > > > > > > states, causing the number of ASPs in
> > > > ASP-ACTIVE
> > > > > > drop below n."
> > > > > > > >
> > > > > > > > So the AS moves to AS-PENDING state when
> > > > ASPs in
> > > > > > ASP-ACTIVE drop below 'n'
> > > > > > > > (but not necessarily zero). Do we not
> > have
> > > > to
> > > > > > send Notify(AS-PENDING) when
> > > > > > > > AS has moved to AS-PENDING state?.
> > > > > > >
> > > > > > > I see.
> > > > > > >
> > > > > > > I largely disagree with the IG section
> > 3.15.2.
> > > > > > Consider that the new
> > > > > > > description of the AS-PENDING state is:
> > > > > > >
> > > > > > >     AS-PENDING: An active ASP has
> > transitioned
> > > > > > from ASP-ACTIVE to ASP-
> > > > > > >     INACTIVE or ASP-DOWN and it was the
> > number
> > > > n
> > > > > > active ASP in the AS.  A
> > > > > > >     recovery timer T(r) SHOULD be started
> > and
> > > > all
> > > > > > incoming signalling messages
> > > > > > >     SHOULD be queued by the SGP. If an ASP
> > > > becomes
> >
> === message truncated ===
>
> __________________________________________________
> Do you Yahoo!?
> Yahoo! Shopping - Send Flowers for Valentine's Day
> http://shopping.yahoo.com
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 02:29:02 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA06239
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 02:29:02 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1D7cmm00353
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 02:38:48 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1D7cUp00336;
	Thu, 13 Feb 2003 02:38:32 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1D7blp32762
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 02:37:47 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA06226
	for <sigtran@ietf.org>; Thu, 13 Feb 2003 02:27:30 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1D7VAJT013703;
	Thu, 13 Feb 2003 01:31:11 -0600 (CST)
Received: from axes-mach01.usa.alcatel.com by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id MAA28087; Thu, 13 Feb 2003 12:00:08 -0500 (GMT)
Message-ID: <3E4B49D9.2D602CE1@axes-mach01.usa.alcatel.com>
Date: Thu, 13 Feb 2003 13:01:37 +0530
From: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: bidulock@openss7.org
CC: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
References: <20030212141714.D1284@openss7.org> <20030212220925.31809.qmail@web12805.mail.yahoo.com> <20030212200002.C17731@openss7.org>
Content-Type: text/plain; charset=x-user-defined
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Brian,

    I agree with your comments. An AS is moved from AS-PENDING to AS-ACTIVE on the first ASP becoming Active. Am I
right?. In that case, Should we send all the bufferred or queued traffic at SG to the first ASP that becomes Active
immediately or wait until n number of ASPs become Active?.

--Srini

"Brian F. G. Bidulock" wrote:

> Tolga,
>
> Tolga Asveren wrote:                                                                 (Wed, 12 Feb 2003 14:09:25)
>
> > [TOLGA]Is there a place in RFC/IG, which indicates
> > that ASPs SHOULD wait for NTFY(AS Active) to start
> > send/receive traffic? To my knowledge there isn't and
> > this depends just on receiving ASPAC-ACK and on the
> > number of ACTIVE ASPs -based on the above paragraph-
> > IMO the not nice thing with those two statements is
> > that ASPs SHOULD wait for n ACTIVE ASPs to start
> > sending traffic, whereas SGP MAY withhold traffic.
> > There should be a consistency between them.
>
> Also, section 4.3.4.3 middle of page 82, top of page 83 (just after
> the sections you quoted):
>
> Loadshare:
>
>    An SGP or IPSP, upon reception of an ASP Active message for the first
>    ASP in a Loadshare AS, MAY choose not to direct traffic to a newly
>    active ASP until it determines that there are sufficient resources to
>    handle the expected load (e.g., until there are "n" ASPs in state
>    ASP-ACTIVE in the AS).  In this case, the SGP or IPSP SHOULD withhold
>    the Notify (AS-ACTIVE) until there are sufficient resources.
>
> Broadcast:
>
>    An SGP or IPSP, upon reception of an ASP Active message for the first
>    ASP in a Broadcast AS, MAY choose not to direct traffic to a newly
>    active ASP until it determines that there are sufficient resources to
>    handle the expected load (e.g., until there are "n" ASPs in state
>    ASP-ACTIVE in the AS).  In this case, the SGP or IPSP SHOULD withhold
>    the Notify (AS-ACTIVE) until there are sufficient resources.
>
> As the SGP SHOULD withhold the Notify (AS-ACTIVE), the ASP can rely on
> its being withheld as an indication that the SGP is not directing traffic
> to to the active ASPs for the AS because insufficient ASPs have been
> activated.  Because the ASP SHOULD withhold traffic in the case of
> insufficient ASPs, the ASP SHOULD NOT start traffic until it receives
> Notify (AS-ACTIVE) from the SGP.  That seems pretty straightforward to me.
>
> Also, see example 5.1.4.
>
> > [TOLGA]If you think that for start case it is best to
> > wait for n, then SG SHOULD withhold traffic instead of
> > MAY withhold traffic(like ASP SHOULD wait for n ACTIVE
> > ASPs) In fact as far as I recall the request to change
> > this MAY to SHOULD was the reason, why the debate
> > about n+k started. IMO how ASP and SGP behave
> > regarding n should be the same.
>
> I disagree with SHOULD.  Directing traffic on activation requires
> and exchange between the SG and the SS7 network.  These exchanges
> are outside the scope of the RFC and the RFC cannot place requirements
> on the SS7 network.  Also, in the case of SI granularity AS, the
> SG cannot "direct traffic" to the AS.  It will take the expiry of UFC
> timers at the remote signalling points before any traffic will be
> restarted or directed.  By saying MAY, we are saying that the SG is
> graciously permitted to do what it needs to do with respect to the
> SS7 network.
>
> However, we do then say that if the SG does not direct traffic that
> it SHOULD withold the Notify (AS-ACTIVE) to let the ASP know what it
> is doing.  This is SG/ASP interaction and is within the scope of the
> RFC.
>
> Also, what the ASPs do with regard to the SG is directly within the
> scope of the RFC and the RFC can place the SHOULD requirement on the
> ASPs.  The ASPs should not restart traffic until it knows that n ASPs
> are up.  And how does it tell this?  Well the RFC alludes to some
> inter-ASP coordination which is also outside thes scope of the RFC
> (APS to ASP protocol).  The only requirement in the RFC that allows
> the ASP to determine that n ASPs are active is that the SG has sent
> Notify (AS-ACTIVE) for the AS.
>
> If you want to clarify "SHOULD wait until n ASPs are active" to
> "SHOULD wait for Notify (AS-ACTIVE)", feel free, as it seems very
> apparent to me anyway (and was the result of the thread just after
> last call on the subject, too).
>
> > If this MAY is changed to SHOULD,
>
> MAY cannot change to SHOULD because it is an action towards the
> SS7 network upon which M3UA cannot place requirements.
>
> > I would think still
> > an update is necessary to AS state machine for
> > AS-INACTIVE to AS-ACTIVE transition (nth ASP goes
> > ACTIVE instead of 1st ASP goes ACTIVE and we don't
> > touch transition from ACTIVE to PENDING)
>
> No, the state transition is still valid, it is just that the SGP
> SHOULD withhold the Notify (AS-ACTIVE) until the nth ASP is active
> iff it withholds traffic.
>
> The current state machine in the RFC is fine.
>
> > If nobody thinks that there is a realistic scenario,
> > where n>1 is mandatory, i.e. less than n ACTIVE ASP
> > will change AS state to PENDING-, then I think the
> > approach you explained makes sense.
>
> Putting perfectly good processors out of service by blocking all
> traffic to them and creating a complete service outage where one is
> not necessary could not possibly be construed as realistic by any
> operator.  I'd hate to be in front of Congress explaning that one!
>
> --brian
>
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 07:25:33 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA10682
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 07:25:33 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DCZOJ16148
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 07:35:24 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DCYwp16119;
	Thu, 13 Feb 2003 07:35:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DCX3p16079
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 07:33:03 -0500
Received: from web12808.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id HAA10646
	for <sigtran@ietf.org>; Thu, 13 Feb 2003 07:22:40 -0500 (EST)
Message-ID: <20030213122623.77496.qmail@web12808.mail.yahoo.com>
Received: from [62.29.64.91] by web12808.mail.yahoo.com via HTTP; Thu, 13 Feb 2003 04:26:23 PST
Date: Thu, 13 Feb 2003 04:26:23 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
In-Reply-To: <20030212200002.C17731@openss7.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Brian,
--- "Brian F. G. Bidulock" <bidulock@openss7.org>
wrote:
> Tolga,
> 
> Tolga Asveren wrote:                                
>                                 (Wed, 12 Feb 2003
> 14:09:25)
> 
> > [TOLGA]Is there a place in RFC/IG, which indicates
> > that ASPs SHOULD wait for NTFY(AS Active) to start
> > send/receive traffic? To my knowledge there isn't
> and
> > this depends just on receiving ASPAC-ACK and on
> the
> > number of ACTIVE ASPs -based on the above
> paragraph-
> > IMO the not nice thing with those two statements
> is
> > that ASPs SHOULD wait for n ACTIVE ASPs to start
> > sending traffic, whereas SGP MAY withhold traffic.
> > There should be a consistency between them.
> 
[..snip..]
> 
> As the SGP SHOULD withhold the Notify (AS-ACTIVE),
> the ASP can rely on
> its being withheld as an indication that the SGP is
> not directing traffic
> to to the active ASPs for the AS because
> insufficient ASPs have been
> activated.  Because the ASP SHOULD withhold traffic
> in the case of
> insufficient ASPs, the ASP SHOULD NOT start traffic
> until it receives
> Notify (AS-ACTIVE) from the SGP.  That seems pretty
> straightforward to me.
[TOLGA]This is not straightforward at all to me. NTFY
is used only as an advisory message. Probably you also
recall that I proposed to use NTFY(AS-ACTIVE) like you
want to use above (just before the text was frozen)
but we end up with the sentence saying that ASP SHOULD
coordinate among themselves number of ACTIVE ASPs and
SHOULD start sending traffic based on that.
> 
> Also, see example 5.1.4.
> 
> > [TOLGA]If you think that for start case it is best
> to
> > wait for n, then SG SHOULD withhold traffic
> instead of
> > MAY withhold traffic(like ASP SHOULD wait for n
> ACTIVE
> > ASPs) In fact as far as I recall the request to
> change
> > this MAY to SHOULD was the reason, why the debate
> > about n+k started. IMO how ASP and SGP behave
> > regarding n should be the same.
> 
> I disagree with SHOULD.  Directing traffic on
> activation requires
> and exchange between the SG and the SS7 network. 
> These exchanges
> are outside the scope of the RFC and the RFC cannot
> place requirements
> on the SS7 network.  Also, in the case of SI
> granularity AS, the
> SG cannot "direct traffic" to the AS.  It will take
> the expiry of UFC
> timers at the remote signalling points before any
> traffic will be
> restarted or directed.  By saying MAY, we are saying
> that the SG is
> graciously permitted to do what it needs to do with
> respect to the
> SS7 network.
[TOLGA]IMO, Unless necessary changes are done to make
ASP start traffic based on NTFY, if MAY is not changed
to SHOULD, there is the possibility of inconsistent
behavior between ASP and SGP. Let's say n is 3 and SG
decides to send traffic with the first ACTIVE ASP,
ASPs won't send traffic based on the current M3UA RFC.
If we decide to use NTFY to make ASPs aware when to
start traffic, then still I would think concept of n+k
is not necessary on procedures between ASp/SGP. SGP
decides when traffic will start and will inform ASPs
about that.
> 
> However, we do then say that if the SG does not
> direct traffic that
> it SHOULD withold the Notify (AS-ACTIVE) to let the
> ASP know what it
> is doing.  This is SG/ASP interaction and is within
> the scope of the
> RFC.
> 
> Also, what the ASPs do with regard to the SG is
> directly within the
> scope of the RFC and the RFC can place the SHOULD
> requirement on the
> ASPs.  The ASPs should not restart traffic until it
> knows that n ASPs
> are up.  And how does it tell this?  Well the RFC
> alludes to some
> inter-ASP coordination which is also outside thes
> scope of the RFC
> (APS to ASP protocol).  The only requirement in the
> RFC that allows
> the ASP to determine that n ASPs are active is that
> the SG has sent
> Notify (AS-ACTIVE) for the AS.
[TOLGA]Nowhere it is stated that NTFY is to be
interpreted so stronglt by ASP and till now we always
interpreted NTFY only as an advisory message.
> 
> If you want to clarify "SHOULD wait until n ASPs are
> active" to
> "SHOULD wait for Notify (AS-ACTIVE)", feel free, as
> it seems very
> apparent to me anyway (and was the result of the
> thread just after
> last call on the subject, too).
[TOLGA]This is quite a change logically (but not
necessarily I disagree, what do the others think?). 
> 
> > If this MAY is changed to SHOULD,
> 
> MAY cannot change to SHOULD because it is an action
> towards the
> SS7 network upon which M3UA cannot place
> requirements.
> 
> > I would think still
> > an update is necessary to AS state machine for
> > AS-INACTIVE to AS-ACTIVE transition (nth ASP goes
> > ACTIVE instead of 1st ASP goes ACTIVE and we don't
> > touch transition from ACTIVE to PENDING)
> 
> No, the state transition is still valid, it is just
> that the SGP
> SHOULD withhold the Notify (AS-ACTIVE) until the nth
> ASP is active
> iff it withholds traffic.
[TOLGA] Something not clear to me, let's assume SG
wants to withhold traffic till 3 ASPs are ACTIVE. Will
it advertise PC of AS available to SS7 network with
the first ACTIVE ASP? 
It is strange to me thinking that AS state changes to
ACTIVE and NTFY(AS ACTIVE) is withhold. I would think,
AS chnages to ACTIVE and NTFY is sent. But when AS
changes to ACTIVE state can be decided by SG itself.
> 
> The current state machine in the RFC is fine.
> 
> > If nobody thinks that there is a realistic
> scenario,
> > where n>1 is mandatory, i.e. less than n ACTIVE
> ASP
> > will change AS state to PENDING-, then I think the
> > approach you explained makes sense.
> 
> Putting perfectly good processors out of service by
> blocking all
> traffic to them and creating a complete service
> outage where one is
> not necessary could not possibly be construed as
> realistic by any
> operator.  I'd hate to be in front of Congress
> explaning that one!
[TOLGA]I was thinking on the cases, where one is sure
that with less than n ACTIVE ASP, congestion will
happen soon, so just to act a bit in advance.


__________________________________________________
Do you Yahoo!?
Yahoo! Shopping - Send Flowers for Valentine's Day
http://shopping.yahoo.com
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 07:26:28 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA10728
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 07:26:28 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DCaKj16212
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 07:36:20 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DCa3p16195;
	Thu, 13 Feb 2003 07:36:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DCZnp16172
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 07:35:49 -0500
Received: from web12801.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id HAA10680
	for <sigtran@ietf.org>; Thu, 13 Feb 2003 07:25:26 -0500 (EST)
Message-ID: <20030213122910.88257.qmail@web12801.mail.yahoo.com>
Received: from [62.29.64.91] by web12801.mail.yahoo.com via HTTP; Thu, 13 Feb 2003 04:29:10 PST
Date: Thu, 13 Feb 2003 04:29:10 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
In-Reply-To: <3E4B14E5.535F7963@axes-mach01.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Srini,
--- ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
wrote:
> Tolga,
> 
>     Moreover the SG has to buffer the traffic in
> such a situation(when load is more).
[TOLGA] I don' think so, because if n is interpreted
as mandatory, AS won't be advetised as available to
SS7 netwrok till n ASPs are ACTIVE, so there won't be
any load at all.
> 
> --Srini
> 
> "Brian F. G. Bidulock" wrote:
> 
> > Tolga,
> >
> > I disagree strongly.  When we build n+1 redundant
> systems for carrier
> > grade operation we do not stop service when a
> processor is still able
> > to provide service.
> >
> > If I have 3 processors in a 2+1 arrangement (where
> 2 are needed to handle
> > the maximum engineered load and 1 is spare), I do
> not stop service just
> > because I have only 1 processor alive.  It just
> means that the system
> > will not be able to handle the maximum engineered
> load with 1 processor.
> >
> > Overload is a different issue.  If the 1 remaining
> processor is overloaded,
> > it should use overload controls to keep itself
> alive.  In M3UA's case this
> > could be overload controls at ISUP level or nodal
> TFC at MTP level or even
> > SCTP flow control (failing to service read
> buffer).
> >
> > Having the SG move to the AS-PENDING state when
> the number of active ASPs
> > fall below "n" (2 in the case above) is completely
> undesirable.
> >
> > I thought I made this clear on the "n+k redudancy"
> thread.
> >
> > I strongly disagree with the changes of section
> 3.15 of the IG.
> >
> > --brian
> >
> > Tolga Asveren wrote:                              
>                                   (Wed, 12 Feb 2003
> 05:15:58)
> > >  If the traffic mode is 1+k I agree with that
> > > statement but if it is n+k, I disagree.
> > >
> > >    Tolga
> > > --- ksrinivas
> <ksrini@axes-mach01.usa.alcatel.com>
> > > wrote:
> >
> > --
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran


__________________________________________________
Do you Yahoo!?
Yahoo! Shopping - Send Flowers for Valentine's Day
http://shopping.yahoo.com
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 07:47:34 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA11054
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 07:47:34 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DCvQQ17657
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 07:57:26 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DCvBp17624;
	Thu, 13 Feb 2003 07:57:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DCuEp17575
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 07:56:14 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA10993
	for <sigtran@ietf.org>; Thu, 13 Feb 2003 07:45:51 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1DCnWeW018125;
	Thu, 13 Feb 2003 06:49:34 -0600 (CST)
Received: from axes-mach01.usa.alcatel.com by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id RAA06168; Thu, 13 Feb 2003 17:18:40 -0500 (GMT)
Message-ID: <3E4B9484.91CC9DF3@axes-mach01.usa.alcatel.com>
Date: Thu, 13 Feb 2003 18:20:13 +0530
From: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Tolga Asveren <tolga_asveren@yahoo.com>
CC: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
References: <20030213122910.88257.qmail@web12801.mail.yahoo.com>
Content-Type: text/plain; charset=x-user-defined
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Tolga,

   The situation which I meant is "When an ASP in the AS goes Inactive
by making number of Active ASPs less than 'n' and also at the same
movement traffic is more". In this situation if the traffic is stopped
sending to the AS and tried buffering, I think we will run
out-of-buffers in no time.

--Srini

Tolga Asveren wrote:

> Srini,
> --- ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
> wrote:
> > Tolga,
> >
> >     Moreover the SG has to buffer the traffic in
> > such a situation(when load is more).
> [TOLGA] I don' think so, because if n is interpreted
> as mandatory, AS won't be advetised as available to
> SS7 netwrok till n ASPs are ACTIVE, so there won't be
> any load at all.
> >
> > --Srini
> >
> > "Brian F. G. Bidulock" wrote:
> >
> > > Tolga,
> > >
> > > I disagree strongly.  When we build n+1 redundant
> > systems for carrier
> > > grade operation we do not stop service when a
> > processor is still able
> > > to provide service.
> > >
> > > If I have 3 processors in a 2+1 arrangement (where
> > 2 are needed to handle
> > > the maximum engineered load and 1 is spare), I do
> > not stop service just
> > > because I have only 1 processor alive.  It just
> > means that the system
> > > will not be able to handle the maximum engineered
> > load with 1 processor.
> > >
> > > Overload is a different issue.  If the 1 remaining
> > processor is overloaded,
> > > it should use overload controls to keep itself
> > alive.  In M3UA's case this
> > > could be overload controls at ISUP level or nodal
> > TFC at MTP level or even
> > > SCTP flow control (failing to service read
> > buffer).
> > >
> > > Having the SG move to the AS-PENDING state when
> > the number of active ASPs
> > > fall below "n" (2 in the case above) is completely
> > undesirable.
> > >
> > > I thought I made this clear on the "n+k redudancy"
> > thread.
> > >
> > > I strongly disagree with the changes of section
> > 3.15 of the IG.
> > >
> > > --brian
> > >
> > > Tolga Asveren wrote:
> >                                   (Wed, 12 Feb 2003
> > 05:15:58)
> > > >  If the traffic mode is 1+k I agree with that
> > > > statement but if it is n+k, I disagree.
> > > >
> > > >    Tolga
> > > > --- ksrinivas
> > <ksrini@axes-mach01.usa.alcatel.com>
> > > > wrote:
> > >
> > > --
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
>
> __________________________________________________
> Do you Yahoo!?
> Yahoo! Shopping - Send Flowers for Valentine's Day
> http://shopping.yahoo.com
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 07:59:32 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA11149
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 07:59:32 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DD9Od18687
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 08:09:24 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DD96p18667;
	Thu, 13 Feb 2003 08:09:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DD8Rp18650
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 08:08:27 -0500
Received: from domino2.nc.tekelec.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA11141
	for <sigtran@ietf.org>; Thu, 13 Feb 2003 07:58:03 -0500 (EST)
From: jeffrey.craig@tekelec.com
Received: from domino4.nc.tekelec.com ([198.89.40.9])
          by domino2.nc.tekelec.com (Lotus Domino Release 5.0.10)
          with ESMTP id 2003021307552871:34090 ;
          Thu, 13 Feb 2003 07:55:28 -0500 
Subject: Re: [Sigtran] M2PA draft 07 comment: LPO
To: Tom George <Tom.George@alcatel.com>
Cc: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.7  March 21, 2001
Message-ID: <OF42B79CB5.F6706E3B-ON85256CCC.0046C21D@nc.tekelec.com>
Date: Thu, 13 Feb 2003 07:57:32 -0500
MIME-Version: 1.0
X-MIMETrack: Serialize by Router on Domino4/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/13/2003 07:57:33 AM,
	Itemize by SMTP Server on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/13/2003 07:55:28 AM,
	Serialize by Router on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/13/2003 07:55:30 AM,
	Serialize complete at 02/13/2003 07:55:30 AM
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hello Tom,

The updated M2PA state machine figure still does not
match the Q.703 SDLs.

Here follows my attempt to get a closer match....

Regards,

Jeff

                   +-----------+
                   |   POWER   |
                   |    OFF    |
                   +-----------+
                         |
                         | Power On
                         |
       Flush Buffers     |  +-----------------------------------------+
       OR Retrieval      |  |                                         |
              +-------+  |  |                                         |
              |       |  V  V                                         |
              |    +-----------+                                      |
              +--->|  OUT OF   |                                      |
  +--------------->|  SERVICE  |<--+                                  |
  |                +-----------+   | Link Configured                  |
  |                      |   |     | (Associate)                      |
  |                      |   +-----+                                  |
  |     Receive LS OOS   |                                            |
  |           +-------+  | MTP3 Start                                 |
  |           |       |  V                                            |
  |           |    +-----------+                                      |
  |           +--->|    NOT    |                                      |
  +<---------------|  ALIGNED  |------------------------------------->+
  |     MTP3 Stop  +-----------+     SCTP Comm Error                  |
  |  OR T2 Expiry        |        OR SCTP Comm Lost                   |
  |                      |                                            |
  |                      |    Receive LS Align                        |
  |                      | OR Receive LS Prov Norm                    |
  |     Receive LS Align | OR Receive LS Prov Emer                    |
  |           +-------+  |                                            |
  |           |       |  V                                            |
  |           |    +-----------+                                      |
  |           +--->|  ALIGNED  |                                      |
  +<---------------|           |------------------------------------->+
  |     MTP3 Stop  +-----------+     SCTP Comm Error                  |
  |  OR T3 Expiry      ^    |     OR SCTP Comm Lost                   |
  |  OR Receive LS OOS |    |                                         |
  |                    |    |    Receive LS Prov Norm                 |
  |                    |    | OR Receive LS Prov Emer                 |
  |                    |    |                                         |
  |   Receive LS Align |    |                                         |
  |                    |    |                                         |
  |                    |    |                                         |
  | Receive LS Prov X  |    |                                         |
  |           +------+ |    |                                         |
  |           |      | |    V                                         |
  |           |    +-----------+                                      |
  |           +--->|  PROVING  |                                      |
  |<---------------|           |------------------------------------->+
  |     MTP3 Stop  +-----------+     SCTP Comm Error                  |
  |  OR Receive LS OOS   |        OR SCTP Comm Lost                   |
  |                      |                                            |
  |                      | T4 Expiry                                  |
  |                      |                                            |
  |                      |             MTP3 Stop            +-------->+
  |                      |             OR Receive LS OOS    |         |
  |                      |             OR SCTP Comm Error   |         |
  |                      |             OR SCTP Comm Lost    |         |
  |                      |             OR T1 Expiry         |         |
  |                      |             OR M2PA Link Failure |         |
  |                      |             OR Receive LS Align  |         |
  |                      V                                  |         |
  |                +-----------+                   +-----------+      |
  |                |  ALIGNED  |                   |  ALIGNED  |      |
  +<---------------|   READY   |<----------------->| NOT READY |      |
  |     MTP3 Stop  +-----------+     LPO/LPR       +-----------+      |
  |  OR T1 Expiry        |   |                         | Rcv LS Ready |
  |  OR Receive LS OOS   |   |                         | OR Rcv Data  |
  |  OR Receive LS Align |   +------------------------>+ OR Rcv LS PO |
  |  OR SCTP Comm Error  |   Receive LS PO             |              |
  |  OR SCTP Comm Lost   |                             |              |
  |                      |    Receive LS Ready         |              |
  |                      | OR Receive User Data        |              |
  |                      V                             V              |
  |                +-----------+ LPO or Rcv LS PO  +-----------+      |
  |                |    IN     |------------------>| PROCESSOR |      |
  +<---------------|  SERVICE  |<------------------|  OUTAGE   |      |
                   +-----------+    MTP3 Flush     +-----------+      |
       MTP3 Stop                 OR MTP3 Continue           |         |
    OR Receive LS OOS                                       |         |
    OR SCTP Comm Error                    MTP3 Stop         |         |
|
    OR SCTP Comm Lost                  OR Receive LS OOS    |         |
    OR T6 Expiry                       OR SCTP Comm Error   |         |
    OR T7 Expiry                       OR SCTP Comm Lost    |         |
    OR M2PA Link Failure               OR M2PA Link Failure |         |
    OR Receive LS Align                OR Receive LS Align  +-------->+

























                                                                                                                                  
                      Tom George                                                                                                  
                      <Tom.George@alca         To:                                                                                
                      tel.com>                 cc:      sigtran@ietf.org                                                          
                      Sent by:                 Subject: Re: [Sigtran] M2PA draft 07 comment: LPO                                  
                      sigtran-admin@ie                                                                                            
                      tf.org                                                                                                      
                                                                                                                                  
                                                                                                                                  
                      02/12/03 06:52                                                                                              
                      PM                                                                                                          
                                                                                                                                  
                                                                                                                                  




Franz,

I updated the diagram because of another discussion. I put an excerpt
from it further down in this message. I think it shows the transition
you asked about.

But there is nothing in the text that says this. I have this paragraph:

======
If there is a local processor outage condition during the alignment
procedure, M2PA sends Link Status Processor Outage to its peer. When
the local processor outage condition ends, then M2PA SHALL send Link
Status Processor Outage Ended to its peer. M2PA SHALL attempt to
complete the alignment procedure during the local processor outage
condition.
======

Perhaps I should add something like this: "If the local processor outage
condition exists when M2PA completes Proving, M2PA shall enter the
ALIGNED NOT READY state."


======
  |                       |   +-----+                  |
  |                       |   |     | LPO/LPR          |
  |                       V   V     |                  |
  |                 +-----------+   |                  |
  |                 |  PROVING  |---+                  |
  |<----------------|           |--------------------->+
  |      MTP3 Stop  +-----------+     SCTP Comm Error  |
  |   OR T3 Expiry        |  |     OR SCTP Comm Lost   |
  |   OR Receive LS OOS   |  |                         |
  |                       |  +-------------------------|--------+
  |                       |     T4 Expiry              |        |
  |                       |     Send LS PO             |        |
  |                       |                            |        |
  |                       | T4 Expiry                  |        |
  |                       | Send LS Ready              |        |
  |                       |                            |        |
  |                       |                            |        |
  |                       V           SCTP Comm Error  |        V
  |                 +-----------+  OR SCTP Comm Lost   |  +-----------+
  |                 |  ALIGNED  |--------------------->+  |  ALIGNED  |
  +<----------------|   READY   |<----------------------->| NOT READY |
  |      MTP3 Stop  +-----------+     LPO/LPR             +-----------+
  |   OR T1 Expiry        |        OR Receive LS PO/PR       |       |
  |   OR Receive LS OOS   |                                  |   T1  |
  |   OR Receive LS Align | Receive LS Ready      Receive    | Expiry|
  |                       | OR Receive User Data    LS Ready | OR Rcv|
  |                       |                                  |    LS |
  |                       |                                  |  Align|
======


tomG.

"Brian F. G. Bidulock" wrote:
>
> Gaiser,
>
> Gaiser Franz wrote:     (Tue, 11 Feb 2003 11:09:58)
> > Hi all,
> >
> > I have an additional comment/remark to the description of handling
Processor
> > Outage in draft 07of M2PA. The handling of the following situation is
not
> > clear to me: - Proving Period completed (T4 expired) and Local
Processor
> > Outage is occured. In the case without LPO the m2pa would start sending
LS
> > Ready to the peer and moves to state ALIGNED READY.
>
> > For the above situation (with LPO) the handling is not clearly stated.
Is LS
> > Ready and/or LS PO to be send ?According to Q.703  the m2pa should send
LS
> > PO message instead of LS Ready and moves to state ALIGNED NOT READY.
>
> Then that's what M2PA should do.
>
> --brian
>
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 08:08:37 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA11401
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 08:08:36 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DDITc19174
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 08:18:29 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DDI9p19143;
	Thu, 13 Feb 2003 08:18:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DDHqp19113
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 08:17:52 -0500
Received: from web12802.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id IAA11354
	for <sigtran@ietf.org>; Thu, 13 Feb 2003 08:07:28 -0500 (EST)
Message-ID: <20030213131112.12844.qmail@web12802.mail.yahoo.com>
Received: from [62.29.64.91] by web12802.mail.yahoo.com via HTTP; Thu, 13 Feb 2003 05:11:12 PST
Date: Thu, 13 Feb 2003 05:11:12 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
In-Reply-To: <3E4B9484.91CC9DF3@axes-mach01.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Srini,
--- ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
wrote:
> Tolga,
> 
>    The situation which I meant is "When an ASP in
> the AS goes Inactive
> by making number of Active ASPs less than 'n' and
> also at the same
> movement traffic is more". In this situation if the
> traffic is stopped
> sending to the AS and tried buffering, I think we
> will run
> out-of-buffers in no time.
[TOLGA]I understand your concern but is the situation
different if you have one big ASP, which can handle a
lot of traffic? When it goes INACTIVE switching AS to
pending state, still one might run out of buffers, if
no new ASP is ACTIVE soon. For n mandatory (not
because I think we should go in that direction) case,
IMO it is similar, there are n-1 ACTIVE ASPs, and if
no new ASP becomes ACTIVE soon, one migth run out of
buffers.
> 
> --Srini
> 
> Tolga Asveren wrote:
> 
> > Srini,
> > --- ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
> > wrote:
> > > Tolga,
> > >
> > >     Moreover the SG has to buffer the traffic in
> > > such a situation(when load is more).
> > [TOLGA] I don' think so, because if n is
> interpreted
> > as mandatory, AS won't be advetised as available
> to
> > SS7 netwrok till n ASPs are ACTIVE, so there won't
> be
> > any load at all.
> > >
> > > --Srini
> > >
> > > "Brian F. G. Bidulock" wrote:
> > >
> > > > Tolga,
> > > >
> > > > I disagree strongly.  When we build n+1
> redundant
> > > systems for carrier
> > > > grade operation we do not stop service when a
> > > processor is still able
> > > > to provide service.
> > > >
> > > > If I have 3 processors in a 2+1 arrangement
> (where
> > > 2 are needed to handle
> > > > the maximum engineered load and 1 is spare), I
> do
> > > not stop service just
> > > > because I have only 1 processor alive.  It
> just
> > > means that the system
> > > > will not be able to handle the maximum
> engineered
> > > load with 1 processor.
> > > >
> > > > Overload is a different issue.  If the 1
> remaining
> > > processor is overloaded,
> > > > it should use overload controls to keep itself
> > > alive.  In M3UA's case this
> > > > could be overload controls at ISUP level or
> nodal
> > > TFC at MTP level or even
> > > > SCTP flow control (failing to service read
> > > buffer).
> > > >
> > > > Having the SG move to the AS-PENDING state
> when
> > > the number of active ASPs
> > > > fall below "n" (2 in the case above) is
> completely
> > > undesirable.
> > > >
> > > > I thought I made this clear on the "n+k
> redudancy"
> > > thread.
> > > >
> > > > I strongly disagree with the changes of
> section
> > > 3.15 of the IG.
> > > >
> > > > --brian
> > > >
> > > > Tolga Asveren wrote:
> > >                                   (Wed, 12 Feb
> 2003
> > > 05:15:58)
> > > > >  If the traffic mode is 1+k I agree with
> that
> > > > > statement but if it is n+k, I disagree.
> > > > >
> > > > >    Tolga
> > > > > --- ksrinivas
> > > <ksrini@axes-mach01.usa.alcatel.com>
> > > > > wrote:
> > > >
> > > > --
> > > > Brian F. G. Bidulock
> > > > bidulock@openss7.org
> > > > http://www.openss7.org/
> > > >
> _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> > __________________________________________________
> > Do you Yahoo!?
> > Yahoo! Shopping - Send Flowers for Valentine's Day
> > http://shopping.yahoo.com
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 


__________________________________________________
Do you Yahoo!?
Yahoo! Shopping - Send Flowers for Valentine's Day
http://shopping.yahoo.com
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 08:55:03 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA12230
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 08:55:02 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DE4uO21587
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 09:04:56 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DE4Zp21557;
	Thu, 13 Feb 2003 09:04:36 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DE36p21499
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 09:03:06 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA12184
	for <sigtran@ietf.org>; Thu, 13 Feb 2003 08:52:41 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1DDuMeW025652;
	Thu, 13 Feb 2003 07:56:24 -0600 (CST)
Received: from axes-mach01.usa.alcatel.com by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id SAA07237; Thu, 13 Feb 2003 18:25:29 -0500 (GMT)
Message-ID: <3E4BA42E.9CC47855@axes-mach01.usa.alcatel.com>
Date: Thu, 13 Feb 2003 19:27:02 +0530
From: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Tolga Asveren <tolga_asveren@yahoo.com>
CC: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
References: <20030213131112.12844.qmail@web12802.mail.yahoo.com>
Content-Type: text/plain; charset=x-user-defined
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Tolga,

Tolga Asveren wrote:

> Srini,
> --- ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
> wrote:
> > Tolga,
> >
> >    The situation which I meant is "When an ASP in
> > the AS goes Inactive
> > by making number of Active ASPs less than 'n' and
> > also at the same
> > movement traffic is more". In this situation if the
> > traffic is stopped
> > sending to the AS and tried buffering, I think we
> > will run
> > out-of-buffers in no time.
> [TOLGA]I understand your concern but is the situation
> different if you have one big ASP, which can handle a
> lot of traffic? When it goes INACTIVE switching AS to
> pending state, still one might run out of buffers, if
> no new ASP is ACTIVE soon. For n mandatory (not
> because I think we should go in that direction) case,
> IMO it is similar, there are n-1 ACTIVE ASPs, and if
> no new ASP becomes ACTIVE soon, one migth run out of
> buffers.

Atleast for n > 1 case, as Brian suggested, can we not decrease incoming
traffic by initiating the congestion procedures and distribute the
existing traffic over the available Active ASPs?.

--Srini

>
> >
> > --Srini
> >
> > Tolga Asveren wrote:
> >
> > > Srini,
> > > --- ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
> > > wrote:
> > > > Tolga,
> > > >
> > > >     Moreover the SG has to buffer the traffic in
> > > > such a situation(when load is more).
> > > [TOLGA] I don' think so, because if n is
> > interpreted
> > > as mandatory, AS won't be advetised as available
> > to
> > > SS7 netwrok till n ASPs are ACTIVE, so there won't
> > be
> > > any load at all.
> > > >
> > > > --Srini
> > > >
> > > > "Brian F. G. Bidulock" wrote:
> > > >
> > > > > Tolga,
> > > > >
> > > > > I disagree strongly.  When we build n+1
> > redundant
> > > > systems for carrier
> > > > > grade operation we do not stop service when a
> > > > processor is still able
> > > > > to provide service.
> > > > >
> > > > > If I have 3 processors in a 2+1 arrangement
> > (where
> > > > 2 are needed to handle
> > > > > the maximum engineered load and 1 is spare), I
> > do
> > > > not stop service just
> > > > > because I have only 1 processor alive.  It
> > just
> > > > means that the system
> > > > > will not be able to handle the maximum
> > engineered
> > > > load with 1 processor.
> > > > >
> > > > > Overload is a different issue.  If the 1
> > remaining
> > > > processor is overloaded,
> > > > > it should use overload controls to keep itself
> > > > alive.  In M3UA's case this
> > > > > could be overload controls at ISUP level or
> > nodal
> > > > TFC at MTP level or even
> > > > > SCTP flow control (failing to service read
> > > > buffer).
> > > > >
> > > > > Having the SG move to the AS-PENDING state
> > when
> > > > the number of active ASPs
> > > > > fall below "n" (2 in the case above) is
> > completely
> > > > undesirable.
> > > > >
> > > > > I thought I made this clear on the "n+k
> > redudancy"
> > > > thread.
> > > > >
> > > > > I strongly disagree with the changes of
> > section
> > > > 3.15 of the IG.
> > > > >
> > > > > --brian
> > > > >
> > > > > Tolga Asveren wrote:
> > > >                                   (Wed, 12 Feb
> > 2003
> > > > 05:15:58)
> > > > > >  If the traffic mode is 1+k I agree with
> > that
> > > > > > statement but if it is n+k, I disagree.
> > > > > >
> > > > > >    Tolga
> > > > > > --- ksrinivas
> > > > <ksrini@axes-mach01.usa.alcatel.com>
> > > > > > wrote:
> > > > >
> > > > > --
> > > > > Brian F. G. Bidulock
> > > > > bidulock@openss7.org
> > > > > http://www.openss7.org/
> > > > >
> > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > >
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >
> > > __________________________________________________
> > > Do you Yahoo!?
> > > Yahoo! Shopping - Send Flowers for Valentine's Day
> > > http://shopping.yahoo.com
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> >
>
> __________________________________________________
> Do you Yahoo!?
> Yahoo! Shopping - Send Flowers for Valentine's Day
> http://shopping.yahoo.com
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 08:55:07 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA12245
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 08:55:07 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DE51g21610
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 09:05:01 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DE4mp21580;
	Thu, 13 Feb 2003 09:04:48 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DE3Up21518
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 09:03:30 -0500
Received: from domino2.nc.tekelec.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA12187
	for <sigtran@ietf.org>; Thu, 13 Feb 2003 08:53:05 -0500 (EST)
From: jeffrey.craig@tekelec.com
Received: from domino4.nc.tekelec.com ([198.89.40.9])
          by domino2.nc.tekelec.com (Lotus Domino Release 5.0.10)
          with ESMTP id 2003021308503124:34181 ;
          Thu, 13 Feb 2003 08:50:31 -0500 
Subject: Re: AW: [Sigtran] M2PA draft 07 comment: LPO
To: Gaiser Franz <franz.gaiser@siemens.com>
Cc: sigtran@ietf.org, Tom George <Tom.George@alcatel.com>
X-Mailer: Lotus Notes Release 5.0.7  March 21, 2001
Message-ID: <OFB928FB8F.3D1D3FC6-ON85256CCC.004BF844@nc.tekelec.com>
Date: Thu, 13 Feb 2003 08:52:36 -0500
MIME-Version: 1.0
X-MIMETrack: Serialize by Router on Domino4/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/13/2003 08:52:36 AM,
	Itemize by SMTP Server on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/13/2003 08:50:31 AM,
	Serialize by Router on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/13/2003 08:50:33 AM,
	Serialize complete at 02/13/2003 08:50:33 AM
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hello Franz,

>an remark to your M2PA state event machine. In the case
>'Proving' (T4 expired) is finished and local Processor
>Outage is occurred, I missing an arrow from state
>'PROVING' to state 'ALIGNED_NOT_READY'.

You are correct, it's an error. Here is another try....
(It would be a lot easier if M2PA simply referred to Q.703 SDLs.)

Regards,

Jeff

                   +-----------+
                   |   POWER   |
                   |    OFF    |
                   +-----------+
                         |
                         | Power On
                         |
       Flush Buffers     |  +-----------------------------------------+
       OR Retrieval      |  |                                         |
              +-------+  |  |                                         |
              |       |  V  V                                         |
              |    +-----------+                                      |
              +--->|  OUT OF   |                                      |
  +--------------->|  SERVICE  |<--+                                  |
  |                +-----------+   | Link Configured                  |
  |                      |   |     | (Associate)                      |
  |                      |   +-----+                                  |
  |     Receive LS OOS   |                                            |
  |           +-------+  | MTP3 Start                                 |
  |           |       |  V                                            |
  |           |    +-----------+                                      |
  |           +--->|    NOT    |                                      |
  +<---------------|  ALIGNED  |------------------------------------->+
  |     MTP3 Stop  +-----------+     SCTP Comm Error                  |
  |  OR T2 Expiry        |        OR SCTP Comm Lost                   |
  |                      |                                            |
  |                      |    Receive LS Align                        |
  |                      | OR Receive LS Prov Norm                    |
  |     Receive LS Align | OR Receive LS Prov Emer                    |
  |           +-------+  |                                            |
  |           |       |  V                                            |
  |           |    +-----------+                                      |
  |           +--->|  ALIGNED  |                                      |
  +<---------------|           |------------------------------------->+
  |     MTP3 Stop  +-----------+     SCTP Comm Error                  |
  |  OR T3 Expiry      ^    |     OR SCTP Comm Lost                   |
  |  OR Receive LS OOS |    |                                         |
  |                    |    |    Receive LS Prov Norm                 |
  |                    |    | OR Receive LS Prov Emer                 |
  |                    |    |                                         |
  |   Receive LS Align |    |                                         |
  |                    |    |                                         |
  |                    |    |                                         |
  | Receive LS Prov X  |    |                                         |
  |           +------+ |    |                                         |
  |           |      | |    V                                         |
  |           |    +-----------+                                      |
  |           +--->|  PROVING  |------------------------------------->+
  |<---------------|           |------+          SCTP Comm Error      |
  |     MTP3 Stop  +-----------+      |         OR SCTP Comm Lost     |
  |  OR Receive LS OOS   |            |                               |
  |                      |  T4 Expiry |                               |
  |                      |   (LPO)    |                               |
  |                      |            |                               |
  |                      |            | MTP3 Stop            +------->+
  |                      |            | OR Receive LS OOS    |        |
  |                      |            | OR SCTP Comm Error   |        |
  |                      |            | OR SCTP Comm Lost    |        |
  |                      |            | OR T1 Expiry         |        |
  |           T4 Expiry  |            | OR M2PA Link Failure |        |
  |            (No LPO)  |            | OR Receive LS Align  |        |
  |                      V            |                      |        |
  |                +-----------+ LPO  V            +-----------+      |
  |                |  ALIGNED  |------+----------->|  ALIGNED  |      |
  +<---------------|   READY   |<------------------| NOT READY |      |
  |     MTP3 Stop  +-----------+       LPR         +-----------+      |
  |  OR T1 Expiry        |   |                         | Rcv LS Ready |
  |  OR Receive LS OOS   |   |                         | OR Rcv Data  |
  |  OR Receive LS Align |   +------------------------>+ OR Rcv LS PO |
  |  OR SCTP Comm Error  |   Receive LS PO             |              |
  |  OR SCTP Comm Lost   |                             |              |
  |                      |    Receive LS Ready         |              |
  |                      | OR Receive User Data        |              |
  |                      V                             V              |
  |                +-----------+ LPO or Rcv LS PO  +-----------+      |
  |                |    IN     |------------------>| PROCESSOR |      |
  +<---------------|  SERVICE  |<------------------|  OUTAGE   |      |
                   +-----------+    MTP3 Flush     +-----------+      |
       MTP3 Stop                 OR MTP3 Continue            |        |
    OR Receive LS OOS                                        |        |
    OR SCTP Comm Error                     MTP3 Stop         |        |
    OR SCTP Comm Lost                   OR Receive LS OOS    |        |
    OR T6 Expiry                        OR SCTP Comm Error   |        |
    OR T7 Expiry                        OR SCTP Comm Lost    |        |
    OR M2PA Link Failure                OR M2PA Link Failure |        |
    OR Receive LS Align                 OR Receive LS Align  +------->+


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 08:55:30 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA12260
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 08:55:30 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DE5Nn21656
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 09:05:23 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DE58p21641;
	Thu, 13 Feb 2003 09:05:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DE4Mp21540
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 09:04:22 -0500
Received: from stewart.chicago.il.us (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA12205;
	Thu, 13 Feb 2003 08:53:57 -0500 (EST)
Received: from stewart.chicago.il.us (stewart.chicago.il.us [127.0.0.1])
	by stewart.chicago.il.us (8.12.6/8.12.3) with ESMTP id h1DDvJQE000570;
	Thu, 13 Feb 2003 07:57:19 -0600 (CST)
	(envelope-from randall@stewart.chicago.il.us)
Message-ID: <3E4BA43E.6010807@stewart.chicago.il.us>
Date: Thu, 13 Feb 2003 07:57:18 -0600
From: Randall Stewart <randall@stewart.chicago.il.us>
User-Agent: Mozilla/5.0 (X11; U; FreeBSD i386; en-US; rv:1.1) Gecko/20021005
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Tuexen Michael <michael.tuexen@siemens.com>,
        Caitlin Bestler <cait@asomi.com>
CC: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>, tsvwg@ietf.org,
        sigtran@ietf.org
References: <F92978223B01D311ACFF0008C71EE19D015DFAAB@MCHH225E>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] Re: [Tsvwg] A question of Un-Ordered SCTP
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Michael/Caitlin:

Ok, I think that there is enough mixed sentiment on this issue
that here is how we should proceed:

a) A CAPABILITIES TLV should probably be created, possibly in
    the add-ip draft? That could then indicate different supported
    cababilities like, add/del/set-primary/pr-sctp and our new
    idea unreliable with sender sequencing.

b) A new small draft should be created outlining:
     1) a seperate sequence space,
     2) how it would be used with un-ordered
     3) a recommendation to the receivers to pass
        the unordered SSN to the ULP as is, I know
        in my impelementation I did this anyway.. it
        takes extra code to NOT do this :> so I think
        thats pretty non-controversial.
     4) specify a capability value for this.

c) Update the pr-sctp draft to use the capabilites as
    well and get rid of its specialized TLV to annoucne
    pr-sctp.

Get all of this submitted and see how the general wg feels about
it..

If you would like I will send you a rought draft in a
week or two.. sound ok?

R


-- 
Randall R. Stewart
randall@stewart.chicago.il.us 815-342-5222 (cell phone)

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 09:15:59 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA12598
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 09:15:59 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DEPrg23289
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 09:25:53 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DEPcp23270;
	Thu, 13 Feb 2003 09:25:38 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DEOjp23217
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 09:24:45 -0500
Received: from gorilla.mchh.siemens.de (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA12540;
	Thu, 13 Feb 2003 09:14:19 -0500 (EST)
Received: from blues.mchh.siemens.de ([139.21.204.206])
	by gorilla.mchh.siemens.de (8.9.3/8.9.3) with ESMTP id PAA23532;
	Thu, 13 Feb 2003 15:18:01 +0100 (MET)
Received: from mchh273e.demchh201e.icn.siemens.de (mchh273e.mchh.siemens.de [139.21.200.83])
	by blues.mchh.siemens.de (8.9.1/8.9.1) with ESMTP id PAA15070;
	Thu, 13 Feb 2003 15:17:36 +0100 (MET)
Received: by mchh273e.mchh.siemens.de with Internet Mail Service (5.5.2656.59)
	id <1JB5YMD4>; Thu, 13 Feb 2003 15:17:12 +0100
Message-ID: <F92978223B01D311ACFF0008C71EE19D015DFAB4@MCHH225E>
From: Tuexen Michael <michael.tuexen@siemens.com>
To: "'Randall Stewart'" <randall@stewart.chicago.il.us>,
        Tuexen Michael
	 <michael.tuexen@siemens.com>,
        Caitlin Bestler <cait@asomi.com>
Cc: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>, tsvwg@ietf.org,
        sigtran@ietf.org
Date: Thu, 13 Feb 2003 15:17:11 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2656.59)
Content-Type: text/plain;
	charset="ISO-8859-1"
Subject: [Sigtran] RE: [Tsvwg] A question of Un-Ordered SCTP
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Sounds good to me.

Best regards
Michael

Michael Tuexen   Tel.:   +49 89 722 47210
Siemens AG       Fax:    +49 89 722 48212
ICN CP SE C 7    e-mail: Michael.Tuexen@siemens.com

> -----Original Message-----
> From: Randall Stewart [mailto:randall@stewart.chicago.il.us]
> Sent: Thursday, February 13, 2003 2:57 PM
> To: Tuexen Michael; Caitlin Bestler
> Cc: Michael Tuexen; tsvwg@ietf.org; sigtran@ietf.org
> Subject: Re: [Tsvwg] A question of Un-Ordered SCTP
> 
> 
> Michael/Caitlin:
> 
> Ok, I think that there is enough mixed sentiment on this issue
> that here is how we should proceed:
> 
> a) A CAPABILITIES TLV should probably be created, possibly in
>     the add-ip draft? That could then indicate different supported
>     cababilities like, add/del/set-primary/pr-sctp and our new
>     idea unreliable with sender sequencing.
> 
> b) A new small draft should be created outlining:
>      1) a seperate sequence space,
>      2) how it would be used with un-ordered
>      3) a recommendation to the receivers to pass
>         the unordered SSN to the ULP as is, I know
>         in my impelementation I did this anyway.. it
>         takes extra code to NOT do this :> so I think
>         thats pretty non-controversial.
>      4) specify a capability value for this.
> 
> c) Update the pr-sctp draft to use the capabilites as
>     well and get rid of its specialized TLV to annoucne
>     pr-sctp.
> 
> Get all of this submitted and see how the general wg feels about
> it..
> 
> If you would like I will send you a rought draft in a
> week or two.. sound ok?
> 
> R
> 
> 
> -- 
> Randall R. Stewart
> randall@stewart.chicago.il.us 815-342-5222 (cell phone)
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 09:28:31 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA12850
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 09:28:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DEcQK24455
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 09:38:26 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DEc9p24442;
	Thu, 13 Feb 2003 09:38:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DEbcp24393
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 09:37:38 -0500
Received: from domino2.nc.tekelec.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA12832;
	Thu, 13 Feb 2003 09:27:12 -0500 (EST)
From: jeffrey.craig@tekelec.com
Received: from domino4.nc.tekelec.com ([198.89.40.9])
          by domino2.nc.tekelec.com (Lotus Domino Release 5.0.10)
          with ESMTP id 2003021309243673:34265 ;
          Thu, 13 Feb 2003 09:24:36 -0500 
Subject: Re: [Sigtran] M2PA draft 7 question -- alignment conclusion
To: bidulock@openss7.org
Cc: sigtran@ietf.org, sigtran-admin@ietf.org,
        Tom George <Tom.George@alcatel.com>, wayne n taylor <waynet@cisco.com>
X-Mailer: Lotus Notes Release 5.0.7  March 21, 2001
Message-ID: <OF2BD7F60F.80D04655-ON85256CCC.004D9BA1@nc.tekelec.com>
Date: Thu, 13 Feb 2003 09:26:41 -0500
MIME-Version: 1.0
X-MIMETrack: Serialize by Router on Domino4/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/13/2003 09:26:42 AM,
	Itemize by SMTP Server on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/13/2003 09:24:36 AM,
	Serialize by Router on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/13/2003 09:24:39 AM,
	Serialize complete at 02/13/2003 09:24:39 AM
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hello Brian,

I should know better than put 'conclusion' in the title of
a SIGTRAN post.

>There is the requirement that M2PA send repeated PROVING-NORMAL or
>PROVING-EMERGENCY messages while running T3 for this to work.

So what do you recommend? Leave M2PA alignment as it is?
Make it match Q.703 with some noted exceptions? I don't
see mention in Q.703 of repeated SIN/SIE during T3. Perhaps
you could explain.

I now think that it is a bad idea to change M2PA such that
during OOS it responds to received LSA with LSO.
I think this introduces a failed alignment loop, which was the
original topic of this thread. M2PA should continue to receive
LSA during OOS and record the reception so that it can be acted
upon when MTP3 start is subsequently received.

Case where M2PA during OOS responds to received
LSA by sending LSO. Results in continuous failure
to align.


MTP3     |    M2PA     |    M2PA     |   MTP3
         |             |             |
         |(clear LSA)  |(clear LSA)  |
         | OOS/Estab   | OOS/Estab   |
         |             |             |
start -->+             |             |
         | INIT ALIGN  |             |
         |      LSA -->+             |
         | T2 Start    |             |
         |             +<-- LSO      |
         |             |             |
         +<-- OSI      |             |
         | T2 Stop     |             |
         |(clear LSA)  |             |
         | OOS/Estab   |             |
         |             |             |
         |             |             +<-- start
         |             | INIT ALIGN  |
         |             +<-- LSA      |
         |             | T2 Start    |
         |      LSO -->+             |
         |             |             |
         |             |     OSI --->+
         |             | T2 Stop     |
         |             |(clear LSA)  |
         |             | OOS/Estab   |
         |             |             |
start -->+             |             |
         /         <repeats>         /


----------
OLD TEXT 4.1.3
After the association is established, M2PA SHALL send a Link Status
Out of Service message to its peer.

NEW TEXT 4.1.3 (revision #2)
After the association is established, M2PA SHALL send a Link Status
Out of Service message to its peer and, prior to receiving a Start
Request from MTP3, MAY periodically send Link Status Out of
Service messages. During the Out of Service State M2PA SHOULD
discard all received messages except Link Status Out of Service and
Link Status Alignment and respond to each discarded message by sending
a Link Status Out of Service message.

Regards,

Jeff



                                                                                                                                  
                      "Brian F. G.                                                                                                
                      Bidulock"                To:      jeffrey.craig@tekelec.com                                                 
                      <bidulock@openss         cc:      wayne n taylor <waynet@cisco.com>, Tom George <Tom.George@alcatel.com>,   
                      7.org>                   sigtran@ietf.org                                                                   
                      Sent by:                 Subject: Re: [Sigtran] M2PA draft 7 question -- alignment conclusion               
                      sigtran-admin@ie                                                                                            
                      tf.org                                                                                                      
                                                                                                                                  
                                                                                                                                  
                      02/12/03 03:46                                                                                              
                      PM                                                                                                          
                      Please respond                                                                                              
                      to bidulock                                                                                                 
                                                                                                                                  
                                                                                                                                  




jeffrey.craig,

There is the requirement that M2PA send repeated PROVING-NORMAL or
PROVING-EMERGENCY messages while running T3 for this to work.

--brian

jeffrey.craig@tekelec.com wrote:
(Wed, 12 Feb 2003 15:25:35)
>
> Hello Brian,
>
> This thread has gotten too long. Let's try and conclude it with
> a summary of the recommended changes to the M2PA draft.
> The intent is to match Q.703.
>
> Regards,
>
> Jeff
>
> ----------
> Figure 7 is updated to show a transition from INITIAL_ALIGNMENT
> to PROVING upon reception of either a Link Status Alignment or
> Link Status Proving Normal or Link Status Proving Emergency
> message. (Q.703 figure 4, figure 9 sheet 2)
>
>
> ----------
> OLD TEXT 4.1.3
> After the association is established, M2PA SHALL send a Link Status
> Out of Service message to its peer.
>
> NEW TEXT 4.1.3
> After the association is established, M2PA SHALL send a Link Status
> Out of Service message to its peer and, prior to receiving a Start
> Request from MTP3, MAY periodically send Link Status Out of
> Service messages. During the Out of Service State M2PA SHOULD
> respond to received messages other than Link Status Out of Service
> by sending Link Status Out of Service.
>
>
> ----------
> OLD TEXT 4.1.3
> M2PA stops timer T2 when it has received the Link Status Alignment
> message from its peer.
>
> NEW TEXT 4.1.3
> M2PA stops timer T2 when it has received a Link Status Alignment
> message or Link Status Proving Normal or Link Status Emergency
> message from its peer.
>
>
> ----------
> OLD TEXT 4.1.3
> Note: Between the time M2PA sends Link Status Alignment to its peer
> and receives Link Status Alignment from its peer, M2PA may receive
> Link Status Out of Service from its peer. This message is
> ignored. After receiving Link Status Alignment from the peer, receipt
> of a Link Status Out of Service message causes M2PA to send Out of
> Service to MTP3 and return to the Out of Service state.
>
> NEW TEXT 4.1.3
> Note: Between the time M2PA sends Link Status Alignment to its peer
> and receives either Link Status Alignment or Link Status Proving
> Normal or Link Status Proving Emergency from its peer, M2PA may
> receive Link Status Out of Service from its peer. This message is
> ignored. After receiving Link Status Alignment or Link Status
> Proving Normal or Link Status Proving Emergency from its peer,
> receipt of a Link Status Out of Service message causes M2PA to
> send Out of Service to MTP3 and return to the Out of Service state.
>
> ----------
> OLD TEXT 4.1.3
>
> When M2PA has both sent and received the Link Status Alignment
> message, it has completed alignment.  M2PA starts the aligned timer
> T3 and moves to the proving state.
>
> M2PA stops timer T3 when it receives a Proving Normal or Proving
> Emergency message and starts proving period timer T4.
>
> NEW TEXT 4.1.3
>
> When M2PA has sent Link Status Alignment and received either a Link
> Status Alignment or a Link Status Proving Normal or a Link Status
> Proving Emergency message, it has completed alignment. M2PA starts
> the aligned timer T3 and moves to the proving state. Note that
> timer T3 is started even if reception of a proving message caused
> the transition to the proving state.
>
> M2PA stops timer T3 when it receives a Link Status Proving Normal
> or Link Status Proving Emergency message and starts proving period
> timer T4.
>

--
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 09:41:30 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA13119
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 09:41:30 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DEpPt24997
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 09:51:25 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DEp9p24980;
	Thu, 13 Feb 2003 09:51:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DEoop24958
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 09:50:50 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA13078
	for <sigtran@ietf.org>; Thu, 13 Feb 2003 09:40:23 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:birS3sWnNujzzG0jxOf44ky27oIKImy2@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1DEi7D13205;
	Thu, 13 Feb 2003 07:44:07 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1DEi7L17712;
	Thu, 13 Feb 2003 07:44:07 -0700
Date: Thu, 13 Feb 2003 07:44:06 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
Cc: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030213074406.A15571@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>,
	Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030212141714.D1284@openss7.org> <20030212220925.31809.qmail@web12805.mail.yahoo.com> <20030212200002.C17731@openss7.org> <3E4B49D9.2D602CE1@axes-mach01.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E4B49D9.2D602CE1@axes-mach01.usa.alcatel.com>; from ksrini@axes-mach01.usa.alcatel.com on Thu, Feb 13, 2003 at 01:01:37PM +0530
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Srini,

The decision to restart while pending is a different decision again
from the start or stop decision.  If a brief period of time has
elapsed, the network conditions at the time of restart are the same
as the network conditions immediately before the restart.  IMO from
the AS-PENDING state, the SG should move back to AS-ACTIVE with the
first ASP coming active, as in the RFC.

How the SG controls the restart decision is by setting the pending
timer.  It was made clear that the T(pending) timer can be 0.

For M3UA, time-controlled changeover and controlled rerouting buffer
messages for about a second.  T(pending) timer values of 1 or 2 seconds
are very usable, but the operator is welcome to set the timer to 0
and force the start decision immediately.

I think by controlling the timeout value one can get whatever behavior
the operator wants.

The ASP, on the other hand, cannot really know what T(pending) timeout
value is set at the SG and should treat the situation as though it is
initially starting.  That is, it should wait for NTFY(AS-ACTIVE) before
starting traffic.  That way, if the SG is peforming the start deicision
instead of the restart decision, and waiting for n ASPs, it will withhold
NTFY(AS-ACTIVE) until n are up, and things should work correctly.

Sound good?

--brian


ksrinivas wrote:                                                                 (Thu, 13 Feb 2003 13:01:37)
> Brian,
> 
>     I agree with your comments. An AS is moved from AS-PENDING to AS-ACTIVE on the first ASP becoming Active. Am I
> right?. In that case, Should we send all the bufferred or queued traffic at SG to the first ASP that becomes Active
> immediately or wait until n number of ASPs become Active?.
> 
> --Srini
> 
> "Brian F. G. Bidulock" wrote:
> 
> > Tolga,
> >
> > Tolga Asveren wrote:                                                                 (Wed, 12 Feb 2003 14:09:25)
> >
> > > [TOLGA]Is there a place in RFC/IG, which indicates
> > > that ASPs SHOULD wait for NTFY(AS Active) to start
> > > send/receive traffic? To my knowledge there isn't and
> > > this depends just on receiving ASPAC-ACK and on the
> > > number of ACTIVE ASPs -based on the above paragraph-
> > > IMO the not nice thing with those two statements is
> > > that ASPs SHOULD wait for n ACTIVE ASPs to start
> > > sending traffic, whereas SGP MAY withhold traffic.
> > > There should be a consistency between them.
> >
> > Also, section 4.3.4.3 middle of page 82, top of page 83 (just after
> > the sections you quoted):
> >
> > Loadshare:
> >
> >    An SGP or IPSP, upon reception of an ASP Active message for the first
> >    ASP in a Loadshare AS, MAY choose not to direct traffic to a newly
> >    active ASP until it determines that there are sufficient resources to
> >    handle the expected load (e.g., until there are "n" ASPs in state
> >    ASP-ACTIVE in the AS).  In this case, the SGP or IPSP SHOULD withhold
> >    the Notify (AS-ACTIVE) until there are sufficient resources.
> >
> > Broadcast:
> >
> >    An SGP or IPSP, upon reception of an ASP Active message for the first
> >    ASP in a Broadcast AS, MAY choose not to direct traffic to a newly
> >    active ASP until it determines that there are sufficient resources to
> >    handle the expected load (e.g., until there are "n" ASPs in state
> >    ASP-ACTIVE in the AS).  In this case, the SGP or IPSP SHOULD withhold
> >    the Notify (AS-ACTIVE) until there are sufficient resources.
> >
> > As the SGP SHOULD withhold the Notify (AS-ACTIVE), the ASP can rely on
> > its being withheld as an indication that the SGP is not directing traffic
> > to to the active ASPs for the AS because insufficient ASPs have been
> > activated.  Because the ASP SHOULD withhold traffic in the case of
> > insufficient ASPs, the ASP SHOULD NOT start traffic until it receives
> > Notify (AS-ACTIVE) from the SGP.  That seems pretty straightforward to me.
> >
> > Also, see example 5.1.4.
> >
> > > [TOLGA]If you think that for start case it is best to
> > > wait for n, then SG SHOULD withhold traffic instead of
> > > MAY withhold traffic(like ASP SHOULD wait for n ACTIVE
> > > ASPs) In fact as far as I recall the request to change
> > > this MAY to SHOULD was the reason, why the debate
> > > about n+k started. IMO how ASP and SGP behave
> > > regarding n should be the same.
> >
> > I disagree with SHOULD.  Directing traffic on activation requires
> > and exchange between the SG and the SS7 network.  These exchanges
> > are outside the scope of the RFC and the RFC cannot place requirements
> > on the SS7 network.  Also, in the case of SI granularity AS, the
> > SG cannot "direct traffic" to the AS.  It will take the expiry of UFC
> > timers at the remote signalling points before any traffic will be
> > restarted or directed.  By saying MAY, we are saying that the SG is
> > graciously permitted to do what it needs to do with respect to the
> > SS7 network.
> >
> > However, we do then say that if the SG does not direct traffic that
> > it SHOULD withold the Notify (AS-ACTIVE) to let the ASP know what it
> > is doing.  This is SG/ASP interaction and is within the scope of the
> > RFC.
> >
> > Also, what the ASPs do with regard to the SG is directly within the
> > scope of the RFC and the RFC can place the SHOULD requirement on the
> > ASPs.  The ASPs should not restart traffic until it knows that n ASPs
> > are up.  And how does it tell this?  Well the RFC alludes to some
> > inter-ASP coordination which is also outside thes scope of the RFC
> > (APS to ASP protocol).  The only requirement in the RFC that allows
> > the ASP to determine that n ASPs are active is that the SG has sent
> > Notify (AS-ACTIVE) for the AS.
> >
> > If you want to clarify "SHOULD wait until n ASPs are active" to
> > "SHOULD wait for Notify (AS-ACTIVE)", feel free, as it seems very
> > apparent to me anyway (and was the result of the thread just after
> > last call on the subject, too).
> >
> > > If this MAY is changed to SHOULD,
> >
> > MAY cannot change to SHOULD because it is an action towards the
> > SS7 network upon which M3UA cannot place requirements.
> >
> > > I would think still
> > > an update is necessary to AS state machine for
> > > AS-INACTIVE to AS-ACTIVE transition (nth ASP goes
> > > ACTIVE instead of 1st ASP goes ACTIVE and we don't
> > > touch transition from ACTIVE to PENDING)
> >
> > No, the state transition is still valid, it is just that the SGP
> > SHOULD withhold the Notify (AS-ACTIVE) until the nth ASP is active
> > iff it withholds traffic.
> >
> > The current state machine in the RFC is fine.
> >
> > > If nobody thinks that there is a realistic scenario,
> > > where n>1 is mandatory, i.e. less than n ACTIVE ASP
> > > will change AS state to PENDING-, then I think the
> > > approach you explained makes sense.
> >
> > Putting perfectly good processors out of service by blocking all
> > traffic to them and creating a complete service outage where one is
> > not necessary could not possibly be construed as realistic by any
> > operator.  I'd hate to be in front of Congress explaning that one!
> >
> > --brian
> >
> > --
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 09:56:20 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA14002
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 09:56:20 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DF6F325774
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 10:06:15 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DF5pp25720;
	Thu, 13 Feb 2003 10:05:52 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DF43p25648
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 10:04:03 -0500
Received: from domino2.nc.tekelec.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA13807
	for <sigtran@ietf.org>; Thu, 13 Feb 2003 09:53:36 -0500 (EST)
From: jeffrey.craig@tekelec.com
Received: from domino4.nc.tekelec.com ([198.89.40.9])
          by domino2.nc.tekelec.com (Lotus Domino Release 5.0.10)
          with ESMTP id 2003021309510283:34359 ;
          Thu, 13 Feb 2003 09:51:02 -0500 
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.7  March 21, 2001
Message-ID: <OF1B9BFAE2.55D43239-ON85256CCC.0050C2A3@nc.tekelec.com>
Date: Thu, 13 Feb 2003 09:53:06 -0500
MIME-Version: 1.0
X-MIMETrack: Serialize by Router on Domino4/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/13/2003 09:53:08 AM,
	Itemize by SMTP Server on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/13/2003 09:51:02 AM,
	Serialize by Router on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/13/2003 09:51:04 AM,
	Serialize complete at 02/13/2003 09:51:04 AM
Content-type: text/plain; charset=us-ascii
Subject: [Sigtran] M2PA draft needs to note that timer T1 differs from Q.703 timer T1
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Hello All,

The M2PA draft should be revised to note that its implementation
of timer T1 does not conform to timer T1 as described by Q.703
(SDL, Figure 8, sheet 5). In Q.703 this timer is started upon entry
to the Initial Alignment state and has a duration that accounts for
up to four normal alignment sequences.

M2PA timer T1 is started upon entry to either ALIGNED_READY
or ALIGNED_NOT_READY states and has a recommended
duration much shorter than that of four normal alignment sequences.
I have no problem with this treatment, but it is an exception from
the general conformance to Q.703.

Regards,

Jeff


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 09:58:54 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA14175
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 09:58:54 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DF8nM26608
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 10:08:49 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DF8Gp26584;
	Thu, 13 Feb 2003 10:08:16 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DF7wp26541
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 10:07:58 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA14101
	for <sigtran@ietf.org>; Thu, 13 Feb 2003 09:57:30 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:D5t5+sZ5AOIoeg5JIVStLV7id/YIbzhu@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1DF1CD13766;
	Thu, 13 Feb 2003 08:01:12 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1DF1Bd20144;
	Thu, 13 Feb 2003 08:01:11 -0700
Date: Thu, 13 Feb 2003 08:01:11 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tolga Asveren <tolga_asveren@yahoo.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030213080111.B15571@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030212200002.C17731@openss7.org> <20030213122623.77496.qmail@web12808.mail.yahoo.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20030213122623.77496.qmail@web12808.mail.yahoo.com>; from tolga_asveren@yahoo.com on Thu, Feb 13, 2003 at 04:26:23AM -0800
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Tolga,

Tolga Asveren wrote:                                                                 (Thu, 13 Feb 2003 04:26:23)
> > As the SGP SHOULD withhold the Notify (AS-ACTIVE),
> > the ASP can rely on
> > its being withheld as an indication that the SGP is
> > not directing traffic
> > to to the active ASPs for the AS because
> > insufficient ASPs have been
> > activated.  Because the ASP SHOULD withhold traffic
> > in the case of
> > insufficient ASPs, the ASP SHOULD NOT start traffic
> > until it receives
> > Notify (AS-ACTIVE) from the SGP.  That seems pretty
> > straightforward to me.
> [TOLGA]This is not straightforward at all to me. NTFY
> is used only as an advisory message. Probably you also
> recall that I proposed to use NTFY(AS-ACTIVE) like you
> want to use above (just before the text was frozen)
> but we end up with the sentence saying that ASP SHOULD
> coordinate among themselves number of ACTIVE ASPs and
> SHOULD start sending traffic based on that.

The ASP is not prohibited from using the NTFY messages
to deteremine when to do things.

> [TOLGA]IMO, Unless necessary changes are done to make
> ASP start traffic based on NTFY, if MAY is not changed
> to SHOULD, there is the possibility of inconsistent
> behavior between ASP and SGP. Let's say n is 3 and SG
> decides to send traffic with the first ACTIVE ASP,
> ASPs won't send traffic based on the current M3UA RFC.

SHOULDN'T and won't are two different things.

> If we decide to use NTFY to make ASPs aware when to
> start traffic, then still I would think concept of n+k
> is not necessary on procedures between ASp/SGP. SGP
> decides when traffic will start and will inform ASPs
> about that.

Understand that the SGP does not decide when traffic
will start because it is not generating any traffic.
All the SG can do is inform the SS7 network that the
AS is accessible and in the SI case can't even do that.
You cannot make MAY a SHOULD because the for SI AS
the SG cannot even do anything to "direct traffic".

> [TOLGA]Nowhere it is stated that NTFY is to be
> interpreted so stronglt by ASP and till now we always
> interpreted NTFY only as an advisory message.

It is advisory, but because a SHOULD is placed on the SG,
the ASP can reasonably rely on the NTFY being sent when
"n" ASPs are active.

> > No, the state transition is still valid, it is just
> > that the SGP
> > SHOULD withhold the Notify (AS-ACTIVE) until the nth
> > ASP is active
> > iff it withholds traffic.
> [TOLGA] Something not clear to me, let's assume SG
> wants to withhold traffic till 3 ASPs are ACTIVE. Will
> it advertise PC of AS available to SS7 network with
> the first ACTIVE ASP? 

In the case of SI AS there is nothing to advertise.

> It is strange to me thinking that AS state changes to
> ACTIVE and NTFY(AS ACTIVE) is withhold. I would think,
> AS chnages to ACTIVE and NTFY is sent. But when AS
> changes to ACTIVE state can be decided by SG itself.

That's just the way we stated it in the RFC.

> > Putting perfectly good processors out of service by
> > blocking all
> > traffic to them and creating a complete service
> > outage where one is
> > not necessary could not possibly be construed as
> > realistic by any
> > operator.  I'd hate to be in front of Congress
> > explaning that one!
> [TOLGA]I was thinking on the cases, where one is sure
> that with less than n ACTIVE ASP, congestion will
> happen soon, so just to act a bit in advance.

That is a bad assumption: that congestion will happen soon.

During normal operation a telephony system normally handles less
than 1/10th of its maximum load.  During a large part of the day
telephony platforms handle trivial traffic.  The platforms are
engineered for peak high-day operation (for ISUP platforms, this
is Mother's Day busy hour).  So, perhaps your assumption might
hold at 9:00 am on Mother's Day if the platform is close to
being expanded, but not during the other 8,550 hours of the year,
or if the platform has capacity left in its planning cycle.

Decisions to stop cash flow on revenue affecting platforms are
best left manual.

--brian


-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 10:02:44 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA14406
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 10:02:44 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DFCej26802
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 10:12:40 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DFC7p26779;
	Thu, 13 Feb 2003 10:12:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DFBXp26734
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 10:11:33 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA14311;
	Thu, 13 Feb 2003 10:01:05 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:R7A3s5eIQOBTy0VUq8nYvas6/qcf71bX@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1DF4ID13799;
	Thu, 13 Feb 2003 08:04:18 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1DF4Hc20537;
	Thu, 13 Feb 2003 08:04:17 -0700
Date: Thu, 13 Feb 2003 08:04:17 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: jeffrey.craig@tekelec.com
Cc: sigtran@ietf.org, sigtran-admin@ietf.org,
        Tom George <Tom.George@alcatel.com>, wayne n taylor <waynet@cisco.com>
Subject: Re: [Sigtran] M2PA draft 7 question -- alignment conclusion
Message-ID: <20030213080417.C15571@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: jeffrey.craig@tekelec.com, sigtran@ietf.org,
	sigtran-admin@ietf.org, Tom George <Tom.George@alcatel.com>,
	wayne n taylor <waynet@cisco.com>
References: <OF2BD7F60F.80D04655-ON85256CCC.004D9BA1@nc.tekelec.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF2BD7F60F.80D04655-ON85256CCC.004D9BA1@nc.tekelec.com>; from jeffrey.craig@tekelec.com on Thu, Feb 13, 2003 at 09:26:41AM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Jeffrey,

Q.703 ignores receipt of SIOS after sending SIO and while
running T2.  M2PA should as well ignore LSO after sending LSA
while running T2.  Then there is no problem.

--brian

jeffrey.craig@tekelec.com wrote:                                             (Thu, 13 Feb 2003 09:26:41)
> 
> Hello Brian,
> 
> I should know better than put 'conclusion' in the title of
> a SIGTRAN post.
> 
> >There is the requirement that M2PA send repeated PROVING-NORMAL or
> >PROVING-EMERGENCY messages while running T3 for this to work.
> 
> So what do you recommend? Leave M2PA alignment as it is?
> Make it match Q.703 with some noted exceptions? I don't
> see mention in Q.703 of repeated SIN/SIE during T3. Perhaps
> you could explain.
> 
> I now think that it is a bad idea to change M2PA such that
> during OOS it responds to received LSA with LSO.
> I think this introduces a failed alignment loop, which was the
> original topic of this thread. M2PA should continue to receive
> LSA during OOS and record the reception so that it can be acted
> upon when MTP3 start is subsequently received.
> 
> Case where M2PA during OOS responds to received
> LSA by sending LSO. Results in continuous failure
> to align.
> 
> 
> MTP3     |    M2PA     |    M2PA     |   MTP3
>          |             |             |
>          |(clear LSA)  |(clear LSA)  |
>          | OOS/Estab   | OOS/Estab   |
>          |             |             |
> start -->+             |             |
>          | INIT ALIGN  |             |
>          |      LSA -->+             |
>          | T2 Start    |             |
>          |             +<-- LSO      |
>          |             |             |
>          +<-- OSI      |             |
>          | T2 Stop     |             |
>          |(clear LSA)  |             |
>          | OOS/Estab   |             |
>          |             |             |
>          |             |             +<-- start
>          |             | INIT ALIGN  |
>          |             +<-- LSA      |
>          |             | T2 Start    |
>          |      LSO -->+             |
>          |             |             |
>          |             |     OSI --->+
>          |             | T2 Stop     |
>          |             |(clear LSA)  |
>          |             | OOS/Estab   |
>          |             |             |
> start -->+             |             |
>          /         <repeats>         /
> 
> 
> ----------
> OLD TEXT 4.1.3
> After the association is established, M2PA SHALL send a Link Status
> Out of Service message to its peer.
> 
> NEW TEXT 4.1.3 (revision #2)
> After the association is established, M2PA SHALL send a Link Status
> Out of Service message to its peer and, prior to receiving a Start
> Request from MTP3, MAY periodically send Link Status Out of
> Service messages. During the Out of Service State M2PA SHOULD
> discard all received messages except Link Status Out of Service and
> Link Status Alignment and respond to each discarded message by sending
> a Link Status Out of Service message.
> 
> Regards,
> 
> Jeff
> 
> 
> 
>                                                                                                                                   
>                       "Brian F. G.                                                                                                
>                       Bidulock"                To:      jeffrey.craig@tekelec.com                                                 
>                       <bidulock@openss         cc:      wayne n taylor <waynet@cisco.com>, Tom George <Tom.George@alcatel.com>,   
>                       7.org>                   sigtran@ietf.org                                                                   
>                       Sent by:                 Subject: Re: [Sigtran] M2PA draft 7 question -- alignment conclusion               
>                       sigtran-admin@ie                                                                                            
>                       tf.org                                                                                                      
>                                                                                                                                   
>                                                                                                                                   
>                       02/12/03 03:46                                                                                              
>                       PM                                                                                                          
>                       Please respond                                                                                              
>                       to bidulock                                                                                                 
>                                                                                                                                   
>                                                                                                                                   
> 
> 
> 
> 
> jeffrey.craig,
> 
> There is the requirement that M2PA send repeated PROVING-NORMAL or
> PROVING-EMERGENCY messages while running T3 for this to work.
> 
> --brian
> 
> jeffrey.craig@tekelec.com wrote:
> (Wed, 12 Feb 2003 15:25:35)
> >
> > Hello Brian,
> >
> > This thread has gotten too long. Let's try and conclude it with
> > a summary of the recommended changes to the M2PA draft.
> > The intent is to match Q.703.
> >
> > Regards,
> >
> > Jeff
> >
> > ----------
> > Figure 7 is updated to show a transition from INITIAL_ALIGNMENT
> > to PROVING upon reception of either a Link Status Alignment or
> > Link Status Proving Normal or Link Status Proving Emergency
> > message. (Q.703 figure 4, figure 9 sheet 2)
> >
> >
> > ----------
> > OLD TEXT 4.1.3
> > After the association is established, M2PA SHALL send a Link Status
> > Out of Service message to its peer.
> >
> > NEW TEXT 4.1.3
> > After the association is established, M2PA SHALL send a Link Status
> > Out of Service message to its peer and, prior to receiving a Start
> > Request from MTP3, MAY periodically send Link Status Out of
> > Service messages. During the Out of Service State M2PA SHOULD
> > respond to received messages other than Link Status Out of Service
> > by sending Link Status Out of Service.
> >
> >
> > ----------
> > OLD TEXT 4.1.3
> > M2PA stops timer T2 when it has received the Link Status Alignment
> > message from its peer.
> >
> > NEW TEXT 4.1.3
> > M2PA stops timer T2 when it has received a Link Status Alignment
> > message or Link Status Proving Normal or Link Status Emergency
> > message from its peer.
> >
> >
> > ----------
> > OLD TEXT 4.1.3
> > Note: Between the time M2PA sends Link Status Alignment to its peer
> > and receives Link Status Alignment from its peer, M2PA may receive
> > Link Status Out of Service from its peer. This message is
> > ignored. After receiving Link Status Alignment from the peer, receipt
> > of a Link Status Out of Service message causes M2PA to send Out of
> > Service to MTP3 and return to the Out of Service state.
> >
> > NEW TEXT 4.1.3
> > Note: Between the time M2PA sends Link Status Alignment to its peer
> > and receives either Link Status Alignment or Link Status Proving
> > Normal or Link Status Proving Emergency from its peer, M2PA may
> > receive Link Status Out of Service from its peer. This message is
> > ignored. After receiving Link Status Alignment or Link Status
> > Proving Normal or Link Status Proving Emergency from its peer,
> > receipt of a Link Status Out of Service message causes M2PA to
> > send Out of Service to MTP3 and return to the Out of Service state.
> >
> > ----------
> > OLD TEXT 4.1.3
> >
> > When M2PA has both sent and received the Link Status Alignment
> > message, it has completed alignment.  M2PA starts the aligned timer
> > T3 and moves to the proving state.
> >
> > M2PA stops timer T3 when it receives a Proving Normal or Proving
> > Emergency message and starts proving period timer T4.
> >
> > NEW TEXT 4.1.3
> >
> > When M2PA has sent Link Status Alignment and received either a Link
> > Status Alignment or a Link Status Proving Normal or a Link Status
> > Proving Emergency message, it has completed alignment. M2PA starts
> > the aligned timer T3 and moves to the proving state. Note that
> > timer T3 is started even if reception of a proving message caused
> > the transition to the proving state.
> >
> > M2PA stops timer T3 when it receives a Link Status Proving Normal
> > or Link Status Proving Emergency message and starts proving period
> > timer T4.
> >
> 
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 
> 
> 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 10:03:28 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA14471
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 10:03:28 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DFDNc26862
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 10:13:23 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DFD6p26837;
	Thu, 13 Feb 2003 10:13:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DFChp26815
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 10:12:43 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA14390
	for <sigtran@ietf.org>; Thu, 13 Feb 2003 10:02:16 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:pP/RWxdhsX+wmWxGAUInwAeQNiesEAIy@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1DF60D13830;
	Thu, 13 Feb 2003 08:06:00 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1DF60R20772;
	Thu, 13 Feb 2003 08:06:00 -0700
Date: Thu, 13 Feb 2003 08:05:59 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: jeffrey.craig@tekelec.com
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft needs to note that timer T1 differs from Q.703 timer T1
Message-ID: <20030213080559.D15571@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: jeffrey.craig@tekelec.com, sigtran@ietf.org
References: <OF1B9BFAE2.55D43239-ON85256CCC.0050C2A3@nc.tekelec.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF1B9BFAE2.55D43239-ON85256CCC.0050C2A3@nc.tekelec.com>; from jeffrey.craig@tekelec.com on Thu, Feb 13, 2003 at 09:53:06AM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Jeffrey,

I would prefer to have all timers match.  I prefer this because
then network operators can use experience from Q.703 to set timer
values for M2PA.

--brian

jeffrey.craig@tekelec.com wrote:                                             (Thu, 13 Feb 2003 09:53:06)
> Hello All,
> 
> The M2PA draft should be revised to note that its implementation
> of timer T1 does not conform to timer T1 as described by Q.703
> (SDL, Figure 8, sheet 5). In Q.703 this timer is started upon entry
> to the Initial Alignment state and has a duration that accounts for
> up to four normal alignment sequences.
> 
> M2PA timer T1 is started upon entry to either ALIGNED_READY
> or ALIGNED_NOT_READY states and has a recommended
> duration much shorter than that of four normal alignment sequences.
> I have no problem with this treatment, but it is an exception from
> the general conformance to Q.703.
> 
> Regards,
> 
> Jeff
> 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 10:53:31 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA16196
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 10:53:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DG3RN29957
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 11:03:27 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DG3Cp29939;
	Thu, 13 Feb 2003 11:03:13 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DG2op29908
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 11:02:50 -0500
Received: from domino2.nc.tekelec.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA16159;
	Thu, 13 Feb 2003 10:52:22 -0500 (EST)
From: jeffrey.craig@tekelec.com
Received: from domino4.nc.tekelec.com ([198.89.40.9])
          by domino2.nc.tekelec.com (Lotus Domino Release 5.0.10)
          with ESMTP id 2003021310494771:34549 ;
          Thu, 13 Feb 2003 10:49:47 -0500 
Subject: Re: [Sigtran] M2PA draft 7 question -- alignment conclusion
To: bidulock@openss7.org
Cc: sigtran@ietf.org, sigtran-admin@ietf.org,
        Tom George <Tom.George@alcatel.com>, wayne n taylor <waynet@cisco.com>
X-Mailer: Lotus Notes Release 5.0.7  March 21, 2001
Message-ID: <OF3321D4DD.5FE14298-ON85256CCC.005403BA@nc.tekelec.com>
Date: Thu, 13 Feb 2003 10:51:52 -0500
MIME-Version: 1.0
X-MIMETrack: Serialize by Router on Domino4/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/13/2003 10:51:53 AM,
	Itemize by SMTP Server on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/13/2003 10:49:47 AM,
	Serialize by Router on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/13/2003 10:49:50 AM,
	Serialize complete at 02/13/2003 10:49:50 AM
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hello Brian,

>Q.703 ignores receipt of SIOS after sending SIO and while
>running T2.  M2PA should as well ignore LSO after sending LSA
>while running T2.  Then there is no problem.

Yep, my figure was wrong.

As far as the repeated SIN/SIE during T3 goes, it seems to me
that M2PA statisfies this (figure below).


> > ----------
> > Figure 7 is updated to show a transition from INITIAL_ALIGNMENT
> > to PROVING upon reception of either a Link Status Alignment or
> > Link Status Proving Normal or Link Status Proving Emergency
> > message. (Q.703 figure 4, figure 9 sheet 2)
> >
> >

----------
OLD TEXT 4.1.3
After the association is established, M2PA SHALL send a Link Status
Out of Service message to its peer.

NEW TEXT 4.1.3 (revision #3)
After the association is established, M2PA SHALL send a Link Status
Out of Service message to its peer and, prior to receiving a Start
Request from MTP3, MAY periodically send Link Status Out of
Service messages. During the Out of Service State M2PA SHOULD
ignore received Link Status Out of Service messages and
SHOULD respond to all other received messages by sending a Link
Status Out of Service message.


> >
> > ----------
> > OLD TEXT 4.1.3
> > M2PA stops timer T2 when it has received the Link Status Alignment
> > message from its peer.
> >
> > NEW TEXT 4.1.3
> > M2PA stops timer T2 when it has received a Link Status Alignment
> > message or Link Status Proving Normal or Link Status Emergency
> > message from its peer.
> >
> >
> > ----------
> > OLD TEXT 4.1.3
> > Note: Between the time M2PA sends Link Status Alignment to its peer
> > and receives Link Status Alignment from its peer, M2PA may receive
> > Link Status Out of Service from its peer. This message is
> > ignored. After receiving Link Status Alignment from the peer, receipt
> > of a Link Status Out of Service message causes M2PA to send Out of
> > Service to MTP3 and return to the Out of Service state.
> >
> > NEW TEXT 4.1.3
> > Note: Between the time M2PA sends Link Status Alignment to its peer
> > and receives either Link Status Alignment or Link Status Proving
> > Normal or Link Status Proving Emergency from its peer, M2PA may
> > receive Link Status Out of Service from its peer. This message is
> > ignored. After receiving Link Status Alignment or Link Status
> > Proving Normal or Link Status Proving Emergency from its peer,
> > receipt of a Link Status Out of Service message causes M2PA to
> > send Out of Service to MTP3 and return to the Out of Service state.
> >
> > ----------
> > OLD TEXT 4.1.3
> >
> > When M2PA has both sent and received the Link Status Alignment
> > message, it has completed alignment.  M2PA starts the aligned timer
> > T3 and moves to the proving state.
> >
> > M2PA stops timer T3 when it receives a Proving Normal or Proving
> > Emergency message and starts proving period timer T4.
> >
> > NEW TEXT 4.1.3
> >
> > When M2PA has sent Link Status Alignment and received either a Link
> > Status Alignment or a Link Status Proving Normal or a Link Status
> > Proving Emergency message, it has completed alignment. M2PA starts
> > the aligned timer T3 and moves to the proving state. Note that
> > timer T3 is started even if reception of a proving message caused
> > the transition to the proving state.
> >
> > M2PA stops timer T3 when it receives a Link Status Proving Normal
> > or Link Status Proving Emergency message and starts proving period
> > timer T4.
> >

MTP3     |    M2PA     |    M2PA     |   MTP3
         |             |             |
         | (clear LSA) | (clear LSA) |
         | (clear LSPx)| (clear LSPx)|
         | OOS/Estab   | OOS/Estab   |
         |             |             |
         |      LSO -->+             |
         |             |  (Discard)  |
         |             |             |
         |             +<-- LSO      |
         |  (Discard)  |             |
         |             |             |
start -->+             |             |
         | INIT ALIGN  |             |
         |      LSA -->+             |
         | T2 Start    |             |
         |             |  (Discard)  |
         |             +<-- LSO      |
         |  (Discard)  |             |
         |             |             |
         |             |             +<-- start
         |             | INIT ALIGN  |
         |             +<-- LSA      |
         |             | T2 Start    |
         |             |             |
         | (mark LSA)  |             |
         | T2 Stop     |             |
         | PROVING     |             |
         | T3 Start    |             |
         |     LSPx -->+             |
         |             |             |
         |             | (mark LSA*) |
         |             | T2 Stop     |
         |             | PROVING     |
         |             | T3 Start    |
         |             +<-- LSPx     |
         |             |             |
         | (mark LSPx) |             |
         | T3 Stop     |             |
         | T4 Start    |             |
         |     LSPx -->|             |
         |             | (mark LSPx) |
         |             | T3 Stop     |
         |             | T4 Start    |
         |             +<-- LSPx     |
         /             /              /

Regards,

Jeff




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 11:01:05 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA16335
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 11:01:05 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DGB2v31117
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 11:11:02 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DGAep31084;
	Thu, 13 Feb 2003 11:10:42 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DG73p30282
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 11:07:03 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA16247;
	Thu, 13 Feb 2003 10:56:34 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:/razR/bob4QeLlv/A1X6Bp/IRA85UZLx@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1DFxCD14655;
	Thu, 13 Feb 2003 08:59:12 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1DFxCF28249;
	Thu, 13 Feb 2003 08:59:12 -0700
Date: Thu, 13 Feb 2003 08:59:12 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: jeffrey.craig@tekelec.com
Cc: sigtran@ietf.org, sigtran-admin@ietf.org,
        Tom George <Tom.George@alcatel.com>, wayne n taylor <waynet@cisco.com>
Subject: Re: [Sigtran] M2PA draft 7 question -- alignment conclusion
Message-ID: <20030213085912.A27831@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: jeffrey.craig@tekelec.com, sigtran@ietf.org,
	sigtran-admin@ietf.org, Tom George <Tom.George@alcatel.com>,
	wayne n taylor <waynet@cisco.com>
References: <OF3321D4DD.5FE14298-ON85256CCC.005403BA@nc.tekelec.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF3321D4DD.5FE14298-ON85256CCC.005403BA@nc.tekelec.com>; from jeffrey.craig@tekelec.com on Thu, Feb 13, 2003 at 10:51:52AM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Jeff,

That's it.

--brian

jeffrey.craig@tekelec.com wrote:                                             (Thu, 13 Feb 2003 10:51:52)
> 
> Hello Brian,
> 
> >Q.703 ignores receipt of SIOS after sending SIO and while
> >running T2.  M2PA should as well ignore LSO after sending LSA
> >while running T2.  Then there is no problem.
> 
> Yep, my figure was wrong.
> 
> As far as the repeated SIN/SIE during T3 goes, it seems to me
> that M2PA statisfies this (figure below).
> 
> 
> > > ----------
> > > Figure 7 is updated to show a transition from INITIAL_ALIGNMENT
> > > to PROVING upon reception of either a Link Status Alignment or
> > > Link Status Proving Normal or Link Status Proving Emergency
> > > message. (Q.703 figure 4, figure 9 sheet 2)
> > >
> > >
> 
> ----------
> OLD TEXT 4.1.3
> After the association is established, M2PA SHALL send a Link Status
> Out of Service message to its peer.
> 
> NEW TEXT 4.1.3 (revision #3)
> After the association is established, M2PA SHALL send a Link Status
> Out of Service message to its peer and, prior to receiving a Start
> Request from MTP3, MAY periodically send Link Status Out of
> Service messages. During the Out of Service State M2PA SHOULD
> ignore received Link Status Out of Service messages and
> SHOULD respond to all other received messages by sending a Link
> Status Out of Service message.
> 
> 
> > >
> > > ----------
> > > OLD TEXT 4.1.3
> > > M2PA stops timer T2 when it has received the Link Status Alignment
> > > message from its peer.
> > >
> > > NEW TEXT 4.1.3
> > > M2PA stops timer T2 when it has received a Link Status Alignment
> > > message or Link Status Proving Normal or Link Status Emergency
> > > message from its peer.
> > >
> > >
> > > ----------
> > > OLD TEXT 4.1.3
> > > Note: Between the time M2PA sends Link Status Alignment to its peer
> > > and receives Link Status Alignment from its peer, M2PA may receive
> > > Link Status Out of Service from its peer. This message is
> > > ignored. After receiving Link Status Alignment from the peer, receipt
> > > of a Link Status Out of Service message causes M2PA to send Out of
> > > Service to MTP3 and return to the Out of Service state.
> > >
> > > NEW TEXT 4.1.3
> > > Note: Between the time M2PA sends Link Status Alignment to its peer
> > > and receives either Link Status Alignment or Link Status Proving
> > > Normal or Link Status Proving Emergency from its peer, M2PA may
> > > receive Link Status Out of Service from its peer. This message is
> > > ignored. After receiving Link Status Alignment or Link Status
> > > Proving Normal or Link Status Proving Emergency from its peer,
> > > receipt of a Link Status Out of Service message causes M2PA to
> > > send Out of Service to MTP3 and return to the Out of Service state.
> > >
> > > ----------
> > > OLD TEXT 4.1.3
> > >
> > > When M2PA has both sent and received the Link Status Alignment
> > > message, it has completed alignment.  M2PA starts the aligned timer
> > > T3 and moves to the proving state.
> > >
> > > M2PA stops timer T3 when it receives a Proving Normal or Proving
> > > Emergency message and starts proving period timer T4.
> > >
> > > NEW TEXT 4.1.3
> > >
> > > When M2PA has sent Link Status Alignment and received either a Link
> > > Status Alignment or a Link Status Proving Normal or a Link Status
> > > Proving Emergency message, it has completed alignment. M2PA starts
> > > the aligned timer T3 and moves to the proving state. Note that
> > > timer T3 is started even if reception of a proving message caused
> > > the transition to the proving state.
> > >
> > > M2PA stops timer T3 when it receives a Link Status Proving Normal
> > > or Link Status Proving Emergency message and starts proving period
> > > timer T4.
> > >
> 
> MTP3     |    M2PA     |    M2PA     |   MTP3
>          |             |             |
>          | (clear LSA) | (clear LSA) |
>          | (clear LSPx)| (clear LSPx)|
>          | OOS/Estab   | OOS/Estab   |
>          |             |             |
>          |      LSO -->+             |
>          |             |  (Discard)  |
>          |             |             |
>          |             +<-- LSO      |
>          |  (Discard)  |             |
>          |             |             |
> start -->+             |             |
>          | INIT ALIGN  |             |
>          |      LSA -->+             |
>          | T2 Start    |             |
>          |             |  (Discard)  |
>          |             +<-- LSO      |
>          |  (Discard)  |             |
>          |             |             |
>          |             |             +<-- start
>          |             | INIT ALIGN  |
>          |             +<-- LSA      |
>          |             | T2 Start    |
>          |             |             |
>          | (mark LSA)  |             |
>          | T2 Stop     |             |
>          | PROVING     |             |
>          | T3 Start    |             |
>          |     LSPx -->+             |
>          |             |             |
>          |             | (mark LSA*) |
>          |             | T2 Stop     |
>          |             | PROVING     |
>          |             | T3 Start    |
>          |             +<-- LSPx     |
>          |             |             |
>          | (mark LSPx) |             |
>          | T3 Stop     |             |
>          | T4 Start    |             |
>          |     LSPx -->|             |
>          |             | (mark LSPx) |
>          |             | T3 Stop     |
>          |             | T4 Start    |
>          |             +<-- LSPx     |
>          /             /              /
> 
> Regards,
> 
> Jeff
> 
> 
> 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 11:54:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA18390
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 11:54:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DH4rI02513
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 12:04:53 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DH4Vp02489;
	Thu, 13 Feb 2003 12:04:31 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DH0lp01970
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 12:00:47 -0500
Received: from mailrelay01.wind.it (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA18317
	for <sigtran@ietf.org>; Thu, 13 Feb 2003 11:50:19 -0500 (EST)
Received: from mailrelay01.wind.it (MAILRELAY01 [212.141.48.59]) by mailrelay01.wind.it with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2654.52)
	id 1VDPRPXL; Thu, 13 Feb 2003 17:45:26 +0100
Received: by wexrm2co.wind with Internet Mail Service (5.5.2653.19)
	id <1VDDCCDW>; Thu, 13 Feb 2003 17:53:55 +0100
Message-ID: <AEA4C681309E384EB558453487484F1020CC40@wexrm3a.wind>
From: Troiano Marco <Marco.Troiano@mail.wind.it>
To: sigtran@ietf.org
Date: Thu, 13 Feb 2003 17:53:54 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: multipart/mixed;
	boundary="------------InterScan_NT_MIME_Boundary"
Subject: [Sigtran] I: sctp
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

--------------InterScan_NT_MIME_Boundary
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C2D380.57E76AC4"

------_=_NextPart_001_01C2D380.57E76AC4
Content-Type: text/plain;
	charset="iso-8859-1"

> Hi all.
> My name is Marco Troiano and I am a Telecommunication Enginner. I am
> studying  SCTP and I would like to ask you some questions about RFC 2960:
> 
> *	what is the length of the following fields:
> *	"Host Name address" parameter for INIT chunk (see page 24)?
> *	"State Cookie" parameter for INIT ACK chunk (see page 32)?
> *	"Sender-specific Heartbeat info" parameter for HEARTBEAT chunk (see
> page 36)?
> *	 "Cookie" parameter for COOKIE chunk (see page 46)?
> *	what are the possible values for Heartbeat timers?
> *	how can i choose the periodicity for sending Heartbeat chunks?
> *	what is the worst case for the length of an SCTP control chunk?
> 
> Thanks in advance
> 
> Marco
> 
> 
> 
> 

------_=_NextPart_001_01C2D380.57E76AC4
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Diso-8859-1">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
5.5.2655.35">
<TITLE>I: sctp</TITLE>
</HEAD>
<BODY>

<P><FONT SIZE=3D2 FACE=3D"Arial">Hi all.</FONT>
<BR><FONT SIZE=3D2 FACE=3D"Arial">My name is Marco Troiano and I am a =
Telecommunication Enginner. I am studying&nbsp; SCTP and I would like =
to ask you some questions about RFC 2960:</FONT></P>

<UL><LI><FONT SIZE=3D2 FACE=3D"Arial">what is the length of the =
following fields:</FONT></LI>
<UL>
<UL><LI><FONT SIZE=3D2 FACE=3D"Arial">&quot;Host Name address&quot; =
parameter for INIT chunk (see page 24)?</FONT></LI>
<LI><FONT SIZE=3D2 FACE=3D"Arial">&quot;State Cookie&quot; parameter =
for INIT ACK chunk (see page 32)?</FONT></LI>
<LI><FONT SIZE=3D2 FACE=3D"Arial">&quot;Sender-specific Heartbeat =
info&quot; parameter for HEARTBEAT chunk (see page 36)?</FONT></LI>
<LI><FONT SIZE=3D2 FACE=3D"Arial">&nbsp;&quot;Cookie&quot; parameter =
for COOKIE chunk (see page 46)?</FONT></LI>
</UL></UL><LI><FONT SIZE=3D2 FACE=3D"Arial">what are the possible =
values for Heartbeat timers?</FONT></LI>
<LI><FONT SIZE=3D2 FACE=3D"Arial">how can i choose the periodicity for =
sending Heartbeat chunks?</FONT></LI>
<LI><FONT SIZE=3D2 FACE=3D"Arial">what is the worst case for the length =
of an SCTP control chunk?</FONT></LI>
<BR>
</UL>
<P><FONT SIZE=3D2 FACE=3D"Arial">Thanks in advance</FONT>
</P>

<P><FONT SIZE=3D2 FACE=3D"Arial">Marco</FONT>
</P>
<BR>
<BR>
<BR>

</BODY>
</HTML>
------_=_NextPart_001_01C2D380.57E76AC4--

--------------InterScan_NT_MIME_Boundary--

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 14:11:50 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA22433
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 14:11:50 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DJF7212319
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 14:15:07 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DJEmp12279;
	Thu, 13 Feb 2003 14:14:50 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DJDTp12213
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 14:13:29 -0500
Received: from rtp-core-2.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA22387;
	Thu, 13 Feb 2003 14:09:41 -0500 (EST)
Received: from dingdong.cisco.com (IDENT:mirapoint@dingdong.cisco.com [64.102.17.16])
	by rtp-core-2.cisco.com (8.12.6/8.12.6) with ESMTP id h1DJDMNh005264;
	Thu, 13 Feb 2003 14:13:22 -0500 (EST)
Received: from cisco.com (dhcp-64-102-222-129.cisco.com [64.102.222.129])
	by dingdong.cisco.com (Mirapoint Messaging Server MOS 3.2.1-GA)
	with ESMTP id ABL11475;
	Thu, 13 Feb 2003 14:13:20 -0500 (EST)
Message-ID: <3E4BEE4F.B4B2B40D@cisco.com>
Date: Thu, 13 Feb 2003 14:13:19 -0500
From: wayne n taylor <waynet@cisco.com>
X-Mailer: Mozilla 4.79 [en]C-CCK-MCD   (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: jeffrey.craig@tekelec.com
CC: bidulock@openss7.org, sigtran@ietf.org, sigtran-admin@ietf.org,
        Tom George <Tom.George@alcatel.com>
Subject: Re: [Sigtran] M2PA draft 7 question -- alignment conclusion
References: <OF3321D4DD.5FE14298-ON85256CCC.005403BA@nc.tekelec.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Guys,
one comment about the flows in the figure....

<snip>

>          | T2 Stop     |             |
>          | PROVING     |             |
>          | T3 Start    |             |
>          |     LSPx -->+             |
>          |             |             |
>          |             | (mark LSA*) |
>          |             | T2 Stop     |
>          |             | PROVING     |
>          |             | T3 Start    |
>          |             +<-- LSPx     |
>          |             |             |
>          | (mark LSPx) |             |
>          | T3 Stop     |             |
>          | T4 Start    |             |
>          |     LSPx -->|             |
is it acceptable to bank on getting this LSPx within the T3 timer
duration (1-1.5 sec) ?. i reviewed the Q.703 SDLs and it was not clear
how both nodes transition from the "ALIGNED" state unless it does
something similar to what you show. obviously it is expected that T3
could timeout, but maybe T3 is little tight for IP networks. maybe the
T3 timer range should be expanded, say 1-5 seconds ??
>          |             | (mark LSPx) |
>          |             | T3 Stop     |
>          |             | T4 Start    |
>          |             +<-- LSPx     |

regards,

wayne

jeffrey.craig@tekelec.com wrote:
> 
> Hello Brian,
> 
> >Q.703 ignores receipt of SIOS after sending SIO and while
> >running T2.  M2PA should as well ignore LSO after sending LSA
> >while running T2.  Then there is no problem.
> 
> Yep, my figure was wrong.
> 
> As far as the repeated SIN/SIE during T3 goes, it seems to me
> that M2PA statisfies this (figure below).
> 
> > > ----------
> > > Figure 7 is updated to show a transition from INITIAL_ALIGNMENT
> > > to PROVING upon reception of either a Link Status Alignment or
> > > Link Status Proving Normal or Link Status Proving Emergency
> > > message. (Q.703 figure 4, figure 9 sheet 2)
> > >
> > >
> 
> ----------
> OLD TEXT 4.1.3
> After the association is established, M2PA SHALL send a Link Status
> Out of Service message to its peer.
> 
> NEW TEXT 4.1.3 (revision #3)
> After the association is established, M2PA SHALL send a Link Status
> Out of Service message to its peer and, prior to receiving a Start
> Request from MTP3, MAY periodically send Link Status Out of
> Service messages. During the Out of Service State M2PA SHOULD
> ignore received Link Status Out of Service messages and
> SHOULD respond to all other received messages by sending a Link
> Status Out of Service message.
> 
> > >
> > > ----------
> > > OLD TEXT 4.1.3
> > > M2PA stops timer T2 when it has received the Link Status Alignment
> > > message from its peer.
> > >
> > > NEW TEXT 4.1.3
> > > M2PA stops timer T2 when it has received a Link Status Alignment
> > > message or Link Status Proving Normal or Link Status Emergency
> > > message from its peer.
> > >
> > >
> > > ----------
> > > OLD TEXT 4.1.3
> > > Note: Between the time M2PA sends Link Status Alignment to its peer
> > > and receives Link Status Alignment from its peer, M2PA may receive
> > > Link Status Out of Service from its peer. This message is
> > > ignored. After receiving Link Status Alignment from the peer, receipt
> > > of a Link Status Out of Service message causes M2PA to send Out of
> > > Service to MTP3 and return to the Out of Service state.
> > >
> > > NEW TEXT 4.1.3
> > > Note: Between the time M2PA sends Link Status Alignment to its peer
> > > and receives either Link Status Alignment or Link Status Proving
> > > Normal or Link Status Proving Emergency from its peer, M2PA may
> > > receive Link Status Out of Service from its peer. This message is
> > > ignored. After receiving Link Status Alignment or Link Status
> > > Proving Normal or Link Status Proving Emergency from its peer,
> > > receipt of a Link Status Out of Service message causes M2PA to
> > > send Out of Service to MTP3 and return to the Out of Service state.
> > >
> > > ----------
> > > OLD TEXT 4.1.3
> > >
> > > When M2PA has both sent and received the Link Status Alignment
> > > message, it has completed alignment.  M2PA starts the aligned timer
> > > T3 and moves to the proving state.
> > >
> > > M2PA stops timer T3 when it receives a Proving Normal or Proving
> > > Emergency message and starts proving period timer T4.
> > >
> > > NEW TEXT 4.1.3
> > >
> > > When M2PA has sent Link Status Alignment and received either a Link
> > > Status Alignment or a Link Status Proving Normal or a Link Status
> > > Proving Emergency message, it has completed alignment. M2PA starts
> > > the aligned timer T3 and moves to the proving state. Note that
> > > timer T3 is started even if reception of a proving message caused
> > > the transition to the proving state.
> > >
> > > M2PA stops timer T3 when it receives a Link Status Proving Normal
> > > or Link Status Proving Emergency message and starts proving period
> > > timer T4.
> > >
> 
> MTP3     |    M2PA     |    M2PA     |   MTP3
>          |             |             |
>          | (clear LSA) | (clear LSA) |
>          | (clear LSPx)| (clear LSPx)|
>          | OOS/Estab   | OOS/Estab   |
>          |             |             |
>          |      LSO -->+             |
>          |             |  (Discard)  |
>          |             |             |
>          |             +<-- LSO      |
>          |  (Discard)  |             |
>          |             |             |
> start -->+             |             |
>          | INIT ALIGN  |             |
>          |      LSA -->+             |
>          | T2 Start    |             |
>          |             |  (Discard)  |
>          |             +<-- LSO      |
>          |  (Discard)  |             |
>          |             |             |
>          |             |             +<-- start
>          |             | INIT ALIGN  |
>          |             +<-- LSA      |
>          |             | T2 Start    |
>          |             |             |
>          | (mark LSA)  |             |
>          | T2 Stop     |             |
>          | PROVING     |             |
>          | T3 Start    |             |
>          |     LSPx -->+             |
>          |             |             |
>          |             | (mark LSA*) |
>          |             | T2 Stop     |
>          |             | PROVING     |
>          |             | T3 Start    |
>          |             +<-- LSPx     |
>          |             |             |
>          | (mark LSPx) |             |
>          | T3 Stop     |             |
>          | T4 Start    |             |
>          |     LSPx -->|             |
>          |             | (mark LSPx) |
>          |             | T3 Stop     |
>          |             | T4 Start    |
>          |             +<-- LSPx     |
>          /             /              /
> 
> Regards,
> 
> Jeff
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 14:17:07 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA22582
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 14:17:07 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DJKNA12582
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 14:20:23 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DJK7p12561;
	Thu, 13 Feb 2003 14:20:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DJJSp12510
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 14:19:28 -0500
Received: from ilsa.franken.de (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA22547
	for <sigtran@ietf.org>; Thu, 13 Feb 2003 14:15:40 -0500 (EST)
Received: from lurchi.franken.de (pD958F54A.dip.t-dialin.net [217.88.245.74])
	by ilsa.franken.de (Postfix) with ESMTP
	id 95B9D24613; Thu, 13 Feb 2003 20:17:46 +0100 (CET)
Date: Thu, 13 Feb 2003 20:14:40 +0100
Subject: Re: [Sigtran] I: sctp
Content-Type: text/plain; charset=WINDOWS-1252; format=flowed
Mime-Version: 1.0 (Apple Message framework v551)
Cc: sigtran@ietf.org
To: Troiano Marco <Marco.Troiano@mail.wind.it>
From: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>
In-Reply-To: <AEA4C681309E384EB558453487484F1020CC40@wexrm3a.wind>
Message-Id: <668FB4CB-3F87-11D7-95FD-0030654C1AB6@lurchi.franken.de>
X-Mailer: Apple Mail (2.551)
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h1DJJSp12511
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Hi Marco,
see my comments below.
Best regards
Michael
On Thursday, Feb 13, 2003, at 17:53 Europe/Berlin, Troiano Marco wrote:

> Hi all.
> My name is Marco Troiano and I am a Telecommunication Enginner. I am 
> studying  SCTP and I would like to ask you some questions about RFC 
> 2960:
>
> • 	what is the length of the following fields:
> • 	"Host Name address" parameter for INIT chunk (see page 24)?
Length of the parameter header (= 4) + length of the hostname.
This depends on the hostname.
> • 	"State Cookie" parameter for INIT ACK chunk (see page 32)?
Length of the parameter header (= 4) + length of the cookie information.
This depends on the cookie information.
> • 	"Sender-specific Heartbeat info" parameter for HEARTBEAT chunk (see 
> page 36)?
Length of the parameter header (= 4) + length of the heartbeat 
information.
This depends on the heartbeat information.
> • 	 "Cookie" parameter for COOKIE chunk (see page 46)?
Length of the chunk header (= 4) + length of the cookie information.
This depends on the cookie information.
> • 	what are the possible values for Heartbeat timers?
The time between the heartbeats is RTO + HBInterval.
RTO is betweeen RTO.Min and RTO.Max and HB.Interval >= 0.
Which values can be chosen and how to do it is implementation dependent.
> • 	how can i choose the periodicity for sending Heartbeat chunks?
This is implementation dependent, but normally you are able to
configure RTO.Min, RTO.Max, HB.Interval.
> • 	what is the worst case for the length of an SCTP control chunk?
Hmm,
64K - IPV4 Header Length (20 Bytes)
     - SCTP Common Header Length (12 Bytes)
So you can send a HEARTBEAT chunk of 65504 bytes length. Not that it is
useful, but of maximum length.
>
> Thanks in advance
>
> Marco
>
>
>
>

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 14:48:08 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA23378
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 14:48:08 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DJpPd14812
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 14:51:25 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DJp8p14767;
	Thu, 13 Feb 2003 14:51:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DJo1p14682
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 14:50:01 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA23314;
	Thu, 13 Feb 2003 14:46:11 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:Zeg5chhyKd7sDeJFuaBnlv7LkxHiE1hF@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1DJniD19053;
	Thu, 13 Feb 2003 12:49:44 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1DJnho28516;
	Thu, 13 Feb 2003 12:49:43 -0700
Date: Thu, 13 Feb 2003 12:49:43 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: wayne n taylor <waynet@cisco.com>
Cc: jeffrey.craig@tekelec.com, sigtran@ietf.org, sigtran-admin@ietf.org,
        Tom George <Tom.George@alcatel.com>
Subject: Re: [Sigtran] M2PA draft 7 question -- alignment conclusion
Message-ID: <20030213124943.A27917@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: wayne n taylor <waynet@cisco.com>,
	jeffrey.craig@tekelec.com, sigtran@ietf.org, sigtran-admin@ietf.org,
	Tom George <Tom.George@alcatel.com>
References: <OF3321D4DD.5FE14298-ON85256CCC.005403BA@nc.tekelec.com> <3E4BEE4F.B4B2B40D@cisco.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E4BEE4F.B4B2B40D@cisco.com>; from waynet@cisco.com on Thu, Feb 13, 2003 at 02:13:19PM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

wayne,

wayne n taylor wrote:                                                                (Thu, 13 Feb 2003 14:13:19)
> Guys,
> one comment about the flows in the figure....
> 
> <snip>
> 
> >          | T2 Stop     |             |
> >          | PROVING     |             |
> >          | T3 Start    |             |
> >          |     LSPx -->+             |
> >          |             |             |
> >          |             | (mark LSA*) |
> >          |             | T2 Stop     |
> >          |             | PROVING     |
> >          |             | T3 Start    |
> >          |             +<-- LSPx     |
> >          |             |             |
> >          | (mark LSPx) |             |
> >          | T3 Stop     |             |
> >          | T4 Start    |             |
> >          |     LSPx -->|             |
> is it acceptable to bank on getting this LSPx within the T3 timer
> duration (1-1.5 sec) ?. i reviewed the Q.703 SDLs and it was not clear
> how both nodes transition from the "ALIGNED" state unless it does
> something similar to what you show. obviously it is expected that T3
> could timeout, but maybe T3 is little tight for IP networks. maybe the
> T3 timer range should be expanded, say 1-5 seconds ??

You will probably need a better network than that.

--brian

> >          |             | (mark LSPx) |
> >          |             | T3 Stop     |
> >          |             | T4 Start    |
> >          |             +<-- LSPx     |
> 
> regards,
> 
> wayne
> 
> jeffrey.craig@tekelec.com wrote:
> > 
> > Hello Brian,
> > 
> > >Q.703 ignores receipt of SIOS after sending SIO and while
> > >running T2.  M2PA should as well ignore LSO after sending LSA
> > >while running T2.  Then there is no problem.
> > 
> > Yep, my figure was wrong.
> > 
> > As far as the repeated SIN/SIE during T3 goes, it seems to me
> > that M2PA statisfies this (figure below).
> > 
> > > > ----------
> > > > Figure 7 is updated to show a transition from INITIAL_ALIGNMENT
> > > > to PROVING upon reception of either a Link Status Alignment or
> > > > Link Status Proving Normal or Link Status Proving Emergency
> > > > message. (Q.703 figure 4, figure 9 sheet 2)
> > > >
> > > >
> > 
> > ----------
> > OLD TEXT 4.1.3
> > After the association is established, M2PA SHALL send a Link Status
> > Out of Service message to its peer.
> > 
> > NEW TEXT 4.1.3 (revision #3)
> > After the association is established, M2PA SHALL send a Link Status
> > Out of Service message to its peer and, prior to receiving a Start
> > Request from MTP3, MAY periodically send Link Status Out of
> > Service messages. During the Out of Service State M2PA SHOULD
> > ignore received Link Status Out of Service messages and
> > SHOULD respond to all other received messages by sending a Link
> > Status Out of Service message.
> > 
> > > >
> > > > ----------
> > > > OLD TEXT 4.1.3
> > > > M2PA stops timer T2 when it has received the Link Status Alignment
> > > > message from its peer.
> > > >
> > > > NEW TEXT 4.1.3
> > > > M2PA stops timer T2 when it has received a Link Status Alignment
> > > > message or Link Status Proving Normal or Link Status Emergency
> > > > message from its peer.
> > > >
> > > >
> > > > ----------
> > > > OLD TEXT 4.1.3
> > > > Note: Between the time M2PA sends Link Status Alignment to its peer
> > > > and receives Link Status Alignment from its peer, M2PA may receive
> > > > Link Status Out of Service from its peer. This message is
> > > > ignored. After receiving Link Status Alignment from the peer, receipt
> > > > of a Link Status Out of Service message causes M2PA to send Out of
> > > > Service to MTP3 and return to the Out of Service state.
> > > >
> > > > NEW TEXT 4.1.3
> > > > Note: Between the time M2PA sends Link Status Alignment to its peer
> > > > and receives either Link Status Alignment or Link Status Proving
> > > > Normal or Link Status Proving Emergency from its peer, M2PA may
> > > > receive Link Status Out of Service from its peer. This message is
> > > > ignored. After receiving Link Status Alignment or Link Status
> > > > Proving Normal or Link Status Proving Emergency from its peer,
> > > > receipt of a Link Status Out of Service message causes M2PA to
> > > > send Out of Service to MTP3 and return to the Out of Service state.
> > > >
> > > > ----------
> > > > OLD TEXT 4.1.3
> > > >
> > > > When M2PA has both sent and received the Link Status Alignment
> > > > message, it has completed alignment.  M2PA starts the aligned timer
> > > > T3 and moves to the proving state.
> > > >
> > > > M2PA stops timer T3 when it receives a Proving Normal or Proving
> > > > Emergency message and starts proving period timer T4.
> > > >
> > > > NEW TEXT 4.1.3
> > > >
> > > > When M2PA has sent Link Status Alignment and received either a Link
> > > > Status Alignment or a Link Status Proving Normal or a Link Status
> > > > Proving Emergency message, it has completed alignment. M2PA starts
> > > > the aligned timer T3 and moves to the proving state. Note that
> > > > timer T3 is started even if reception of a proving message caused
> > > > the transition to the proving state.
> > > >
> > > > M2PA stops timer T3 when it receives a Link Status Proving Normal
> > > > or Link Status Proving Emergency message and starts proving period
> > > > timer T4.
> > > >
> > 
> > MTP3     |    M2PA     |    M2PA     |   MTP3
> >          |             |             |
> >          | (clear LSA) | (clear LSA) |
> >          | (clear LSPx)| (clear LSPx)|
> >          | OOS/Estab   | OOS/Estab   |
> >          |             |             |
> >          |      LSO -->+             |
> >          |             |  (Discard)  |
> >          |             |             |
> >          |             +<-- LSO      |
> >          |  (Discard)  |             |
> >          |             |             |
> > start -->+             |             |
> >          | INIT ALIGN  |             |
> >          |      LSA -->+             |
> >          | T2 Start    |             |
> >          |             |  (Discard)  |
> >          |             +<-- LSO      |
> >          |  (Discard)  |             |
> >          |             |             |
> >          |             |             +<-- start
> >          |             | INIT ALIGN  |
> >          |             +<-- LSA      |
> >          |             | T2 Start    |
> >          |             |             |
> >          | (mark LSA)  |             |
> >          | T2 Stop     |             |
> >          | PROVING     |             |
> >          | T3 Start    |             |
> >          |     LSPx -->+             |
> >          |             |             |
> >          |             | (mark LSA*) |
> >          |             | T2 Stop     |
> >          |             | PROVING     |
> >          |             | T3 Start    |
> >          |             +<-- LSPx     |
> >          |             |             |
> >          | (mark LSPx) |             |
> >          | T3 Stop     |             |
> >          | T4 Start    |             |
> >          |     LSPx -->|             |
> >          |             | (mark LSPx) |
> >          |             | T3 Stop     |
> >          |             | T4 Start    |
> >          |             +<-- LSPx     |
> >          /             /              /
> > 
> > Regards,
> > 
> > Jeff
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 15:57:35 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA29734
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 15:57:35 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DL0st19638
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 16:00:54 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DL0ap19607;
	Thu, 13 Feb 2003 16:00:36 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DKxtp19545
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 15:59:55 -0500
Received: from rtp-core-1.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA29677;
	Thu, 13 Feb 2003 15:56:04 -0500 (EST)
Received: from dingdong.cisco.com (IDENT:mirapoint@dingdong.cisco.com [64.102.17.16])
	by rtp-core-1.cisco.com (8.12.6/8.12.6) with ESMTP id h1DKxkJR010418;
	Thu, 13 Feb 2003 15:59:46 -0500 (EST)
Received: from cisco.com (dhcp-64-102-222-129.cisco.com [64.102.222.129])
	by dingdong.cisco.com (Mirapoint Messaging Server MOS 3.2.1-GA)
	with ESMTP id ABL15232;
	Thu, 13 Feb 2003 15:59:44 -0500 (EST)
Message-ID: <3E4C073F.4C073EB0@cisco.com>
Date: Thu, 13 Feb 2003 15:59:43 -0500
From: wayne n taylor <waynet@cisco.com>
X-Mailer: Mozilla 4.79 [en]C-CCK-MCD   (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: bidulock@openss7.org
CC: jeffrey.craig@tekelec.com, sigtran@ietf.org, sigtran-admin@ietf.org,
        Tom George <Tom.George@alcatel.com>
Subject: Re: [Sigtran] M2PA draft 7 question -- alignment conclusion
References: <OF3321D4DD.5FE14298-ON85256CCC.005403BA@nc.tekelec.com> <3E4BEE4F.B4B2B40D@cisco.com> <20030213124943.A27917@openss7.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Brian,
you are probably right !. but there are some folks who will want to run
this over satellite ...

wayne

"Brian F. G. Bidulock" wrote:
> 
> wayne,
> 
> wayne n taylor wrote:                                                                (Thu, 13 Feb 2003 14:13:19)
> > Guys,
> > one comment about the flows in the figure....
> >
> > <snip>
> >
> > >          | T2 Stop     |             |
> > >          | PROVING     |             |
> > >          | T3 Start    |             |
> > >          |     LSPx -->+             |
> > >          |             |             |
> > >          |             | (mark LSA*) |
> > >          |             | T2 Stop     |
> > >          |             | PROVING     |
> > >          |             | T3 Start    |
> > >          |             +<-- LSPx     |
> > >          |             |             |
> > >          | (mark LSPx) |             |
> > >          | T3 Stop     |             |
> > >          | T4 Start    |             |
> > >          |     LSPx -->|             |
> > is it acceptable to bank on getting this LSPx within the T3 timer
> > duration (1-1.5 sec) ?. i reviewed the Q.703 SDLs and it was not clear
> > how both nodes transition from the "ALIGNED" state unless it does
> > something similar to what you show. obviously it is expected that T3
> > could timeout, but maybe T3 is little tight for IP networks. maybe the
> > T3 timer range should be expanded, say 1-5 seconds ??
> 
> You will probably need a better network than that.
> 
> --brian
> 
> > >          |             | (mark LSPx) |
> > >          |             | T3 Stop     |
> > >          |             | T4 Start    |
> > >          |             +<-- LSPx     |
> >
> > regards,
> >
> > wayne
> >
> > jeffrey.craig@tekelec.com wrote:
> > >
> > > Hello Brian,
> > >
> > > >Q.703 ignores receipt of SIOS after sending SIO and while
> > > >running T2.  M2PA should as well ignore LSO after sending LSA
> > > >while running T2.  Then there is no problem.
> > >
> > > Yep, my figure was wrong.
> > >
> > > As far as the repeated SIN/SIE during T3 goes, it seems to me
> > > that M2PA statisfies this (figure below).
> > >
> > > > > ----------
> > > > > Figure 7 is updated to show a transition from INITIAL_ALIGNMENT
> > > > > to PROVING upon reception of either a Link Status Alignment or
> > > > > Link Status Proving Normal or Link Status Proving Emergency
> > > > > message. (Q.703 figure 4, figure 9 sheet 2)
> > > > >
> > > > >
> > >
> > > ----------
> > > OLD TEXT 4.1.3
> > > After the association is established, M2PA SHALL send a Link Status
> > > Out of Service message to its peer.
> > >
> > > NEW TEXT 4.1.3 (revision #3)
> > > After the association is established, M2PA SHALL send a Link Status
> > > Out of Service message to its peer and, prior to receiving a Start
> > > Request from MTP3, MAY periodically send Link Status Out of
> > > Service messages. During the Out of Service State M2PA SHOULD
> > > ignore received Link Status Out of Service messages and
> > > SHOULD respond to all other received messages by sending a Link
> > > Status Out of Service message.
> > >
> > > > >
> > > > > ----------
> > > > > OLD TEXT 4.1.3
> > > > > M2PA stops timer T2 when it has received the Link Status Alignment
> > > > > message from its peer.
> > > > >
> > > > > NEW TEXT 4.1.3
> > > > > M2PA stops timer T2 when it has received a Link Status Alignment
> > > > > message or Link Status Proving Normal or Link Status Emergency
> > > > > message from its peer.
> > > > >
> > > > >
> > > > > ----------
> > > > > OLD TEXT 4.1.3
> > > > > Note: Between the time M2PA sends Link Status Alignment to its peer
> > > > > and receives Link Status Alignment from its peer, M2PA may receive
> > > > > Link Status Out of Service from its peer. This message is
> > > > > ignored. After receiving Link Status Alignment from the peer, receipt
> > > > > of a Link Status Out of Service message causes M2PA to send Out of
> > > > > Service to MTP3 and return to the Out of Service state.
> > > > >
> > > > > NEW TEXT 4.1.3
> > > > > Note: Between the time M2PA sends Link Status Alignment to its peer
> > > > > and receives either Link Status Alignment or Link Status Proving
> > > > > Normal or Link Status Proving Emergency from its peer, M2PA may
> > > > > receive Link Status Out of Service from its peer. This message is
> > > > > ignored. After receiving Link Status Alignment or Link Status
> > > > > Proving Normal or Link Status Proving Emergency from its peer,
> > > > > receipt of a Link Status Out of Service message causes M2PA to
> > > > > send Out of Service to MTP3 and return to the Out of Service state.
> > > > >
> > > > > ----------
> > > > > OLD TEXT 4.1.3
> > > > >
> > > > > When M2PA has both sent and received the Link Status Alignment
> > > > > message, it has completed alignment.  M2PA starts the aligned timer
> > > > > T3 and moves to the proving state.
> > > > >
> > > > > M2PA stops timer T3 when it receives a Proving Normal or Proving
> > > > > Emergency message and starts proving period timer T4.
> > > > >
> > > > > NEW TEXT 4.1.3
> > > > >
> > > > > When M2PA has sent Link Status Alignment and received either a Link
> > > > > Status Alignment or a Link Status Proving Normal or a Link Status
> > > > > Proving Emergency message, it has completed alignment. M2PA starts
> > > > > the aligned timer T3 and moves to the proving state. Note that
> > > > > timer T3 is started even if reception of a proving message caused
> > > > > the transition to the proving state.
> > > > >
> > > > > M2PA stops timer T3 when it receives a Link Status Proving Normal
> > > > > or Link Status Proving Emergency message and starts proving period
> > > > > timer T4.
> > > > >
> > >
> > > MTP3     |    M2PA     |    M2PA     |   MTP3
> > >          |             |             |
> > >          | (clear LSA) | (clear LSA) |
> > >          | (clear LSPx)| (clear LSPx)|
> > >          | OOS/Estab   | OOS/Estab   |
> > >          |             |             |
> > >          |      LSO -->+             |
> > >          |             |  (Discard)  |
> > >          |             |             |
> > >          |             +<-- LSO      |
> > >          |  (Discard)  |             |
> > >          |             |             |
> > > start -->+             |             |
> > >          | INIT ALIGN  |             |
> > >          |      LSA -->+             |
> > >          | T2 Start    |             |
> > >          |             |  (Discard)  |
> > >          |             +<-- LSO      |
> > >          |  (Discard)  |             |
> > >          |             |             |
> > >          |             |             +<-- start
> > >          |             | INIT ALIGN  |
> > >          |             +<-- LSA      |
> > >          |             | T2 Start    |
> > >          |             |             |
> > >          | (mark LSA)  |             |
> > >          | T2 Stop     |             |
> > >          | PROVING     |             |
> > >          | T3 Start    |             |
> > >          |     LSPx -->+             |
> > >          |             |             |
> > >          |             | (mark LSA*) |
> > >          |             | T2 Stop     |
> > >          |             | PROVING     |
> > >          |             | T3 Start    |
> > >          |             +<-- LSPx     |
> > >          |             |             |
> > >          | (mark LSPx) |             |
> > >          | T3 Stop     |             |
> > >          | T4 Start    |             |
> > >          |     LSPx -->|             |
> > >          |             | (mark LSPx) |
> > >          |             | T3 Stop     |
> > >          |             | T4 Start    |
> > >          |             +<-- LSPx     |
> > >          /             /              /
> > >
> > > Regards,
> > >
> > > Jeff
> > >
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 17:50:27 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA16031
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 17:50:27 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DMrlD27403
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 17:53:47 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DMrSp27389;
	Thu, 13 Feb 2003 17:53:28 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DMqZp27331
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 17:52:35 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA15816
	for <sigtran@ietf.org>; Thu, 13 Feb 2003 17:48:44 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1DMqReW014210;
	Thu, 13 Feb 2003 16:52:28 -0600 (CST)
Message-ID: <3E4C21AB.4F810AD4@alcatel.com>
Date: Thu, 13 Feb 2003 16:52:27 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft 7 question -- alignment conclusion
References: <OF3321D4DD.5FE14298-ON85256CCC.005403BA@nc.tekelec.com> <3E4BEE4F.B4B2B40D@cisco.com> <20030213124943.A27917@openss7.org> <3E4C073F.4C073EB0@cisco.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

If someone does want to use this over satellite, there will probably be
a number of changes that will be necessary. I would rather not try to
account for satellite transmission in this draft. What do the rest of
you think?

tomG.

wayne n taylor wrote:
> 
> Brian,
> you are probably right !. but there are some folks who will want to run
> this over satellite ...
> 
> wayne
  [...]

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 17:52:55 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA16423
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 17:52:55 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1DMuFK27524
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 17:56:15 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DMu4p27506;
	Thu, 13 Feb 2003 17:56:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DMt9p27453
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 17:55:09 -0500
Received: from web12802.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id RAA16122
	for <sigtran@ietf.org>; Thu, 13 Feb 2003 17:51:17 -0500 (EST)
Message-ID: <20030213225501.37884.qmail@web12802.mail.yahoo.com>
Received: from [62.29.66.73] by web12802.mail.yahoo.com via HTTP; Thu, 13 Feb 2003 14:55:01 PST
Date: Thu, 13 Feb 2003 14:55:01 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
In-Reply-To: <20030213080111.B15571@openss7.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Brian,
[..snip...]
> The ASP is not prohibited from using the NTFY
> messages
> to deteremine when to do things.
[TOLGA]I think it is clear that NTFY as of now has
only advisory meaning. According to RFC/IG ASP will
start traffic based on the paragraph I referred, i.e.
when ASP knows that n ASPs are ACTIVE, not based on
NTFY(AS ACTIVE). 

[..snip..]
> Understand that the SGP does not decide when traffic
> will start because it is not generating any traffic.
> All the SG can do is inform the SS7 network that the
> AS is accessible and in the SI case can't even do
> that.
> You cannot make MAY a SHOULD because the for SI AS
> the SG cannot even do anything to "direct traffic".
[TOLGA]If n is not interpreted as mandatory, there is
no need to change MAY to SHOULD, otherwise there is.
When I say starting traffic, I refer to the
communication between ASP and SGP, so I don't believe
there is anything to understand special from my point
of view. Also for SI RKs, there is no problem, SG can
reply with UPU for traffic destined to that AS and can
drop any message destined for that AS, when it is
INACTIVE.
If n is to be interpreted as the engineered value for
high taffic case (which also sounds more realistic to
me), then I see changing the meaning of NTFY as the
only solution, because which value SG is going to
decide as enoguh at a given time can't be known by
ASPs. Basically this will make the concept of n
disappear on ASP side.

[..snip..]
> It is advisory, but because a SHOULD is placed on
> the SG,
> the ASP can reasonably rely on the NTFY being sent
> when
> "n" ASPs are active.
[TOLGA]But unless this is stated in RFC/IG, we can't
rely that ASPs will work that way.
> 
> > > No, the state transition is still valid, it is
> just
> > > that the SGP
> > > SHOULD withhold the Notify (AS-ACTIVE) until the
> nth
> > > ASP is active
> > > iff it withholds traffic.
> > [TOLGA] Something not clear to me, let's assume SG
> > wants to withhold traffic till 3 ASPs are ACTIVE.
> Will
> > it advertise PC of AS available to SS7 network
> with
> > the first ACTIVE ASP? 
> 
> In the case of SI AS there is nothing to advertise.
[TOLGA]If I need to be more precise for SI replying
with UPU and for PC advertising availability, still
the real question is, when will SG decide that
SPMC/User Part is available, I would think only then,
when it decides to send NTFY(ACTIVE)?
> 
> > It is strange to me thinking that AS state changes
> to
> > ACTIVE and NTFY(AS ACTIVE) is withhold. I would
> think,
> > AS chnages to ACTIVE and NTFY is sent. But when AS
> > changes to ACTIVE state can be decided by SG
> itself.
> 
> That's just the way we stated it in the RFC.
[TOLGA]Now, in fact using NTFY on ASP to decide to
start traffic will introduce a new state to the state
machine(unless transition from INACTIVE to ACTIVE is
modified in the AS state machine with something like
determined by SG based on number of ACTIVE ASPs),
ACTIVE but not allowed to send traffic. What about
stopping traffic? Will SG have the rigth to send
NTFY(INACTIVE) and for ASPs to stop traffic and keep
them still in ACTIVE state?

> That is a bad assumption: that congestion will
> happen soon.
[TOLGA]Unless somebody comes up with an
counter-example, I also would think that n>1 as
mandatory is not necessary.

__________________________________________________
Do you Yahoo!?
Yahoo! Shopping - Send Flowers for Valentine's Day
http://shopping.yahoo.com
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 20:43:31 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA25618
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 20:43:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1E1kt005041
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 20:46:55 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E1kfp05016;
	Thu, 13 Feb 2003 20:46:41 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E1fEp04762
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 20:41:14 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA25441
	for <sigtran@ietf.org>; Thu, 13 Feb 2003 20:37:17 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:IhRRScAAs0PsyT4bGjBt7EcXrJcL+X4V@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1E1f2D25665;
	Thu, 13 Feb 2003 18:41:02 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1E1f1l12998;
	Thu, 13 Feb 2003 18:41:01 -0700
Date: Thu, 13 Feb 2003 18:41:01 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tolga Asveren <tolga_asveren@yahoo.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030213184101.A28760@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030213080111.B15571@openss7.org> <20030213225501.37884.qmail@web12802.mail.yahoo.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20030213225501.37884.qmail@web12802.mail.yahoo.com>; from tolga_asveren@yahoo.com on Thu, Feb 13, 2003 at 02:55:01PM -0800
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Tolga,

Tolga Asveren wrote:                                                                 (Thu, 13 Feb 2003 14:55:01)
> Brian,
> [..snip...]
> > The ASP is not prohibited from using the NTFY
> > messages
> > to deteremine when to do things.
> [TOLGA]I think it is clear that NTFY as of now has
> only advisory meaning. According to RFC/IG ASP will
> start traffic based on the paragraph I referred, i.e.
> when ASP knows that n ASPs are ACTIVE, not based on
> NTFY(AS ACTIVE). 
> 
> [..snip..]
> > Understand that the SGP does not decide when traffic
> > will start because it is not generating any traffic.
> > All the SG can do is inform the SS7 network that the
> > AS is accessible and in the SI case can't even do
> > that.
> > You cannot make MAY a SHOULD because the for SI AS
> > the SG cannot even do anything to "direct traffic".
> [TOLGA]If n is not interpreted as mandatory, there is
> no need to change MAY to SHOULD, otherwise there is.
> When I say starting traffic, I refer to the
> communication between ASP and SGP, so I don't believe
> there is anything to understand special from my point
> of view. Also for SI RKs, there is no problem, SG can
> reply with UPU for traffic destined to that AS and can
> drop any message destined for that AS, when it is
> INACTIVE.

There you go saying what the SG will do towards the
SS7 network.  The RFC does not place requirements on what
the SG does towards the SS7 network.  It is in the SG's
discretion to manage the SPMC and MTP-SAP as it sees fit.

> If n is to be interpreted as the engineered value for
> high taffic case (which also sounds more realistic to
> me), then I see changing the meaning of NTFY as the
> only solution, because which value SG is going to
> decide as enoguh at a given time can't be known by
> ASPs.

The ASP can always be configured with the same number
as the SG.

The ASP can always know when the SG sends traffic
(because the ASP receives traffic when the SG sends it.)

> Basically this will make the concept of n
> disappear on ASP side.

It does not disappear at all.  The ASPs need to know
how many of them are required to meet n+k redundnacy
requirements, to meet the engineered load, when a
standby should be activated into the pool or an excess
active ASP removed from the pool.  On start the ASPs
need to know how many of them need to activate.

The need to know these things namely because the
SG MAY withhold traffic until "n" of them are up and
they SHOULD not send traffic until "n" of them are up.

> 
> [..snip..]
> > It is advisory, but because a SHOULD is placed on
> > the SG,
> > the ASP can reasonably rely on the NTFY being sent
> > when
> > "n" ASPs are active.
> [TOLGA]But unless this is stated in RFC/IG, we can't
> rely that ASPs will work that way.

ASPs don't have to work that way.  You should know this
about M3UA by now: there are a number of ways to peel
the same onion.  The ASP designer can rely on this SHOULD
or not.  Just don't say that there is no way for the
ASP to tell what the SG is doing: it can by what
notifications it receives.  What it decides to do with
those notifications are up to the ASP.

But it still stands that an ASP SHOULD NOT send traffic
until "n" of them are up for the start scenario.

Perhaps if we change the word "send" to "initiate" that
would satisfy you.

As in: ...the ASP SHOULD NOT initiate traffic until it
knows that "n" ASPs are active...

> > 
> > > > No, the state transition is still valid, it is
> > just
> > > > that the SGP
> > > > SHOULD withhold the Notify (AS-ACTIVE) until the
> > nth
> > > > ASP is active
> > > > iff it withholds traffic.
> > > [TOLGA] Something not clear to me, let's assume SG
> > > wants to withhold traffic till 3 ASPs are ACTIVE.
> > Will
> > > it advertise PC of AS available to SS7 network
> > with
> > > the first ACTIVE ASP? 
> > 
> > In the case of SI AS there is nothing to advertise.
> [TOLGA]If I need to be more precise for SI replying
> with UPU and for PC advertising availability, still
> the real question is, when will SG decide that
> SPMC/User Part is available, I would think only then,
> when it decides to send NTFY(ACTIVE)?

The RFC does not place requirements on how the SG
manages the SPMC/MTP-SAP and it does not specify the NIF.
This is why we say MAY to directing traffic towards
ASPs.  When will the SG do this?  When it's good and
ready.  The requirement is on the SGP/ASP exchange: that,
IF the SG waits to initiate traffic then it also waits
to send NTFY(AS-ACTIVE).


> > 
> > > It is strange to me thinking that AS state changes
> > to
> > > ACTIVE and NTFY(AS ACTIVE) is withhold. I would
> > think,
> > > AS chnages to ACTIVE and NTFY is sent. But when AS
> > > changes to ACTIVE state can be decided by SG
> > itself.
> > 
> > That's just the way we stated it in the RFC.
> [TOLGA]Now, in fact using NTFY on ASP to decide to
> start traffic will introduce a new state to the state
> machine(unless transition from INACTIVE to ACTIVE is
> modified in the AS state machine with something like
> determined by SG based on number of ACTIVE ASPs),
> ACTIVE but not allowed to send traffic. What about
> stopping traffic? Will SG have the rigth to send
> NTFY(INACTIVE) and for ASPs to stop traffic and keep
> them still in ACTIVE state?

Your being way to pendantic.  The state diagram is
an overview.  There are about eight states missing,
such as:  ASP Active sent, waiting for ASP Active Ack,
ASP Inactive sent, waiting for ASP Inactive Ack,
ASP Down sent, waiting for ASP Down Ack.

You see the ASP does not send traffic after sending
ASP Inactive, but it can expect to receive traffic
until it receives ASP Inactive Ack, etc, etc,

It is an overview diagram.  If you want to crack out
the SDLs you will have 50 pages before your done.

> > That is a bad assumption: that congestion will
> > happen soon.
> [TOLGA]Unless somebody comes up with an
> counter-example, I also would think that n>1 as
> mandatory is not necessary.

Good.  Then we strike section 3.15 of the IG, right?

--brian


-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 20:53:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA25765
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 20:53:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1E1vMa05479
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 20:57:22 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E1v9p05466;
	Thu, 13 Feb 2003 20:57:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E1urp05446
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 20:56:53 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA25754
	for <sigtran@ietf.org>; Thu, 13 Feb 2003 20:52:57 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:K1r4XvFxkELO4m2ZK+vyZlhx9BlzxdCl@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1E1ugD25823;
	Thu, 13 Feb 2003 18:56:42 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1E1ugZ14927;
	Thu, 13 Feb 2003 18:56:42 -0700
Date: Thu, 13 Feb 2003 18:56:42 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tom George <Tom.George@alcatel.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft 7 question -- alignment conclusion
Message-ID: <20030213185642.B28760@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tom George <Tom.George@alcatel.com>, sigtran@ietf.org
References: <OF3321D4DD.5FE14298-ON85256CCC.005403BA@nc.tekelec.com> <3E4BEE4F.B4B2B40D@cisco.com> <20030213124943.A27917@openss7.org> <3E4C073F.4C073EB0@cisco.com> <3E4C21AB.4F810AD4@alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E4C21AB.4F810AD4@alcatel.com>; from Tom.George@alcatel.com on Thu, Feb 13, 2003 at 04:52:27PM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Tom,

As M2PA (and SCTP for that matter) have nothing close to Preventative
Cyclic Retransmission to reduce delays with forward error correction,
and because FEC is about the only approach that works for satellite
(high bandwidth, long delay), one would either need to reduce delay
with FEC beneath (or within) SCTP or give up on running M2PA over
satellite.

There is a way to get SCTP to perform FEC, that is to set RTO MAX
beneath RTT.  For fixed bandwidth channels, nother approach would
be that if there is no new DATA chunks to be transmitted, to fill
the remainder of the cwnd with retranmissions of already transmitted
TSNs.

It think this would be better addressed with a Satellite-SCTP or
a completely different transport than at M2PA.

--brian

Tom George wrote:                                  (Thu, 13 Feb 2003 16:52:27)
> If someone does want to use this over satellite, there will probably be
> a number of changes that will be necessary. I would rather not try to
> account for satellite transmission in this draft. What do the rest of
> you think?
> 
> tomG.
> 
> wayne n taylor wrote:
> > 
> > Brian,
> > you are probably right !. but there are some folks who will want to run
> > this over satellite ...
> > 
> > wayne
>   [...]
> 
> -- 
> * Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
> * 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
> * Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
> * I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
> * to advertise or represent its position,  operations, or activities. *
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 13 23:52:29 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA00192
	for <sigtran-archive@odin.ietf.org>; Thu, 13 Feb 2003 23:52:29 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1E4twU15385
	for sigtran-archive@odin.ietf.org; Thu, 13 Feb 2003 23:55:58 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E4tep15374;
	Thu, 13 Feb 2003 23:55:40 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E4sZp15334
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 23:54:35 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA00178
	for <sigtran@ietf.org>; Thu, 13 Feb 2003 23:50:33 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1E4sFeW019719;
	Thu, 13 Feb 2003 22:54:17 -0600 (CST)
Received: from axes-mach01.usa.alcatel.com by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id JAA18651; Fri, 14 Feb 2003 09:23:09 -0500 (GMT)
Message-ID: <3E4C769A.DAE063E0@axes-mach01.usa.alcatel.com>
Date: Fri, 14 Feb 2003 10:24:50 +0530
From: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: bidulock@openss7.org
CC: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
References: <20030212141714.D1284@openss7.org> <20030212220925.31809.qmail@web12805.mail.yahoo.com> <20030212200002.C17731@openss7.org> <3E4B49D9.2D602CE1@axes-mach01.usa.alcatel.com> <20030213074406.A15571@openss7.org>
Content-Type: text/plain; charset=x-user-defined
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Brian,

   1. Is the T(pending) different from T(r)?. (If Yes, RFC / IG does not include this).
   2. Should we start it after An AS is moved from AS-PENDING to AS-ACTIVE i.e., on the first ASP becoming Active?.
   3. On timeout of T(pending) or after n number of ASPs becoming Active, should we send Notify(AS-Active)?.
   4. ASP should wait for NTFY(AS-ACTIVE) before starting traffic.

Is my understanding right?.

--Srini

"Brian F. G. Bidulock" wrote:

> Srini,
>
> The decision to restart while pending is a different decision again
> from the start or stop decision.  If a brief period of time has
> elapsed, the network conditions at the time of restart are the same
> as the network conditions immediately before the restart.  IMO from
> the AS-PENDING state, the SG should move back to AS-ACTIVE with the
> first ASP coming active, as in the RFC.
>
> How the SG controls the restart decision is by setting the pending
> timer.  It was made clear that the T(pending) timer can be 0.
>
> For M3UA, time-controlled changeover and controlled rerouting buffer
> messages for about a second.  T(pending) timer values of 1 or 2 seconds
> are very usable, but the operator is welcome to set the timer to 0
> and force the start decision immediately.
>
> I think by controlling the timeout value one can get whatever behavior
> the operator wants.
>
> The ASP, on the other hand, cannot really know what T(pending) timeout
> value is set at the SG and should treat the situation as though it is
> initially starting.  That is, it should wait for NTFY(AS-ACTIVE) before
> starting traffic.  That way, if the SG is peforming the start deicision
> instead of the restart decision, and waiting for n ASPs, it will withhold
> NTFY(AS-ACTIVE) until n are up, and things should work correctly.
>
> Sound good?
>
> --brian
>
> ksrinivas wrote:                                                                 (Thu, 13 Feb 2003 13:01:37)
> > Brian,
> >
> >     I agree with your comments. An AS is moved from AS-PENDING to AS-ACTIVE on the first ASP becoming Active. Am I
> > right?. In that case, Should we send all the bufferred or queued traffic at SG to the first ASP that becomes Active
> > immediately or wait until n number of ASPs become Active?.
> >
> > --Srini
> >
> > "Brian F. G. Bidulock" wrote:
> >
> > > Tolga,
> > >
> > > Tolga Asveren wrote:                                                                 (Wed, 12 Feb 2003 14:09:25)
> > >
> > > > [TOLGA]Is there a place in RFC/IG, which indicates
> > > > that ASPs SHOULD wait for NTFY(AS Active) to start
> > > > send/receive traffic? To my knowledge there isn't and
> > > > this depends just on receiving ASPAC-ACK and on the
> > > > number of ACTIVE ASPs -based on the above paragraph-
> > > > IMO the not nice thing with those two statements is
> > > > that ASPs SHOULD wait for n ACTIVE ASPs to start
> > > > sending traffic, whereas SGP MAY withhold traffic.
> > > > There should be a consistency between them.
> > >
> > > Also, section 4.3.4.3 middle of page 82, top of page 83 (just after
> > > the sections you quoted):
> > >
> > > Loadshare:
> > >
> > >    An SGP or IPSP, upon reception of an ASP Active message for the first
> > >    ASP in a Loadshare AS, MAY choose not to direct traffic to a newly
> > >    active ASP until it determines that there are sufficient resources to
> > >    handle the expected load (e.g., until there are "n" ASPs in state
> > >    ASP-ACTIVE in the AS).  In this case, the SGP or IPSP SHOULD withhold
> > >    the Notify (AS-ACTIVE) until there are sufficient resources.
> > >
> > > Broadcast:
> > >
> > >    An SGP or IPSP, upon reception of an ASP Active message for the first
> > >    ASP in a Broadcast AS, MAY choose not to direct traffic to a newly
> > >    active ASP until it determines that there are sufficient resources to
> > >    handle the expected load (e.g., until there are "n" ASPs in state
> > >    ASP-ACTIVE in the AS).  In this case, the SGP or IPSP SHOULD withhold
> > >    the Notify (AS-ACTIVE) until there are sufficient resources.
> > >
> > > As the SGP SHOULD withhold the Notify (AS-ACTIVE), the ASP can rely on
> > > its being withheld as an indication that the SGP is not directing traffic
> > > to to the active ASPs for the AS because insufficient ASPs have been
> > > activated.  Because the ASP SHOULD withhold traffic in the case of
> > > insufficient ASPs, the ASP SHOULD NOT start traffic until it receives
> > > Notify (AS-ACTIVE) from the SGP.  That seems pretty straightforward to me.
> > >
> > > Also, see example 5.1.4.
> > >
> > > > [TOLGA]If you think that for start case it is best to
> > > > wait for n, then SG SHOULD withhold traffic instead of
> > > > MAY withhold traffic(like ASP SHOULD wait for n ACTIVE
> > > > ASPs) In fact as far as I recall the request to change
> > > > this MAY to SHOULD was the reason, why the debate
> > > > about n+k started. IMO how ASP and SGP behave
> > > > regarding n should be the same.
> > >
> > > I disagree with SHOULD.  Directing traffic on activation requires
> > > and exchange between the SG and the SS7 network.  These exchanges
> > > are outside the scope of the RFC and the RFC cannot place requirements
> > > on the SS7 network.  Also, in the case of SI granularity AS, the
> > > SG cannot "direct traffic" to the AS.  It will take the expiry of UFC
> > > timers at the remote signalling points before any traffic will be
> > > restarted or directed.  By saying MAY, we are saying that the SG is
> > > graciously permitted to do what it needs to do with respect to the
> > > SS7 network.
> > >
> > > However, we do then say that if the SG does not direct traffic that
> > > it SHOULD withold the Notify (AS-ACTIVE) to let the ASP know what it
> > > is doing.  This is SG/ASP interaction and is within the scope of the
> > > RFC.
> > >
> > > Also, what the ASPs do with regard to the SG is directly within the
> > > scope of the RFC and the RFC can place the SHOULD requirement on the
> > > ASPs.  The ASPs should not restart traffic until it knows that n ASPs
> > > are up.  And how does it tell this?  Well the RFC alludes to some
> > > inter-ASP coordination which is also outside thes scope of the RFC
> > > (APS to ASP protocol).  The only requirement in the RFC that allows
> > > the ASP to determine that n ASPs are active is that the SG has sent
> > > Notify (AS-ACTIVE) for the AS.
> > >
> > > If you want to clarify "SHOULD wait until n ASPs are active" to
> > > "SHOULD wait for Notify (AS-ACTIVE)", feel free, as it seems very
> > > apparent to me anyway (and was the result of the thread just after
> > > last call on the subject, too).
> > >
> > > > If this MAY is changed to SHOULD,
> > >
> > > MAY cannot change to SHOULD because it is an action towards the
> > > SS7 network upon which M3UA cannot place requirements.
> > >
> > > > I would think still
> > > > an update is necessary to AS state machine for
> > > > AS-INACTIVE to AS-ACTIVE transition (nth ASP goes
> > > > ACTIVE instead of 1st ASP goes ACTIVE and we don't
> > > > touch transition from ACTIVE to PENDING)
> > >
> > > No, the state transition is still valid, it is just that the SGP
> > > SHOULD withhold the Notify (AS-ACTIVE) until the nth ASP is active
> > > iff it withholds traffic.
> > >
> > > The current state machine in the RFC is fine.
> > >
> > > > If nobody thinks that there is a realistic scenario,
> > > > where n>1 is mandatory, i.e. less than n ACTIVE ASP
> > > > will change AS state to PENDING-, then I think the
> > > > approach you explained makes sense.
> > >
> > > Putting perfectly good processors out of service by blocking all
> > > traffic to them and creating a complete service outage where one is
> > > not necessary could not possibly be construed as realistic by any
> > > operator.  I'd hate to be in front of Congress explaning that one!
> > >
> > > --brian
> > >
> > > --
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
>
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 14 01:05:23 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA01302
	for <sigtran-archive@odin.ietf.org>; Fri, 14 Feb 2003 01:05:23 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1E68qR19558
	for sigtran-archive@odin.ietf.org; Fri, 14 Feb 2003 01:08:52 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E68Xp19546;
	Fri, 14 Feb 2003 01:08:35 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E67jp19516
	for <sigtran@optimus.ietf.org>; Fri, 14 Feb 2003 01:07:45 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA01293
	for <sigtran@ietf.org>; Fri, 14 Feb 2003 01:03:43 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:ECPBApgB4exa8UOwyZ9u8vZDMlXnuGgj@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1E67RD30393;
	Thu, 13 Feb 2003 23:07:27 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1E67R814632;
	Thu, 13 Feb 2003 23:07:27 -0700
Date: Thu, 13 Feb 2003 23:07:27 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
Cc: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030213230727.A14023@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>,
	Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030212141714.D1284@openss7.org> <20030212220925.31809.qmail@web12805.mail.yahoo.com> <20030212200002.C17731@openss7.org> <3E4B49D9.2D602CE1@axes-mach01.usa.alcatel.com> <20030213074406.A15571@openss7.org> <3E4C769A.DAE063E0@axes-mach01.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E4C769A.DAE063E0@axes-mach01.usa.alcatel.com>; from ksrini@axes-mach01.usa.alcatel.com on Fri, Feb 14, 2003 at 10:24:50AM +0530
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

ksrinivas,

ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 10:24:50)
> Brian,
> 
>    1. Is the T(pending) different from T(r)?. (If Yes, RFC / IG does not
>    include this).

Sorry, I meant T(r).

>    2. Should we start it after An AS is moved from AS-PENDING to AS-ACTIVE
>    i.e., on the first ASP becoming Active?.

No, T(r) is started when moving to AS-PENDING stae.

>    3. On timeout of T(pending) or after n number of ASPs becoming Active,
>    should we send Notify(AS-Active)?.

On timeout of T(r) move to AS-INACTIVE and send Notify(AS-Inactive).

>    4. ASP should wait for NTFY(AS-ACTIVE) before starting traffic.

The ASP could, but it is not required or recommended to do so in the RFC.

--brian

> 
> Is my understanding right?.
> 
> --Srini
> 
> "Brian F. G. Bidulock" wrote:
> 
> > Srini,
> >
> > The decision to restart while pending is a different decision again
> > from the start or stop decision.  If a brief period of time has
> > elapsed, the network conditions at the time of restart are the same
> > as the network conditions immediately before the restart.  IMO from
> > the AS-PENDING state, the SG should move back to AS-ACTIVE with the
> > first ASP coming active, as in the RFC.
> >
> > How the SG controls the restart decision is by setting the pending
> > timer.  It was made clear that the T(pending) timer can be 0.
> >
> > For M3UA, time-controlled changeover and controlled rerouting buffer
> > messages for about a second.  T(pending) timer values of 1 or 2 seconds
> > are very usable, but the operator is welcome to set the timer to 0
> > and force the start decision immediately.
> >
> > I think by controlling the timeout value one can get whatever behavior
> > the operator wants.
> >
> > The ASP, on the other hand, cannot really know what T(pending) timeout
> > value is set at the SG and should treat the situation as though it is
> > initially starting.  That is, it should wait for NTFY(AS-ACTIVE) before
> > starting traffic.  That way, if the SG is peforming the start deicision
> > instead of the restart decision, and waiting for n ASPs, it will withhold
> > NTFY(AS-ACTIVE) until n are up, and things should work correctly.
> >
> > Sound good?
> >
> > --brian
> >
> > ksrinivas wrote:                                                                 (Thu, 13 Feb 2003 13:01:37)
> > > Brian,
> > >
> > >     I agree with your comments. An AS is moved from AS-PENDING to AS-ACTIVE on the first ASP becoming Active. Am I
> > > right?. In that case, Should we send all the bufferred or queued traffic at SG to the first ASP that becomes Active
> > > immediately or wait until n number of ASPs become Active?.
> > >
> > > --Srini
> > >
> > > "Brian F. G. Bidulock" wrote:
> > >
> > > > Tolga,
> > > >
> > > > Tolga Asveren wrote:                                                                 (Wed, 12 Feb 2003 14:09:25)
> > > >
> > > > > [TOLGA]Is there a place in RFC/IG, which indicates
> > > > > that ASPs SHOULD wait for NTFY(AS Active) to start
> > > > > send/receive traffic? To my knowledge there isn't and
> > > > > this depends just on receiving ASPAC-ACK and on the
> > > > > number of ACTIVE ASPs -based on the above paragraph-
> > > > > IMO the not nice thing with those two statements is
> > > > > that ASPs SHOULD wait for n ACTIVE ASPs to start
> > > > > sending traffic, whereas SGP MAY withhold traffic.
> > > > > There should be a consistency between them.
> > > >
> > > > Also, section 4.3.4.3 middle of page 82, top of page 83 (just after
> > > > the sections you quoted):
> > > >
> > > > Loadshare:
> > > >
> > > >    An SGP or IPSP, upon reception of an ASP Active message for the first
> > > >    ASP in a Loadshare AS, MAY choose not to direct traffic to a newly
> > > >    active ASP until it determines that there are sufficient resources to
> > > >    handle the expected load (e.g., until there are "n" ASPs in state
> > > >    ASP-ACTIVE in the AS).  In this case, the SGP or IPSP SHOULD withhold
> > > >    the Notify (AS-ACTIVE) until there are sufficient resources.
> > > >
> > > > Broadcast:
> > > >
> > > >    An SGP or IPSP, upon reception of an ASP Active message for the first
> > > >    ASP in a Broadcast AS, MAY choose not to direct traffic to a newly
> > > >    active ASP until it determines that there are sufficient resources to
> > > >    handle the expected load (e.g., until there are "n" ASPs in state
> > > >    ASP-ACTIVE in the AS).  In this case, the SGP or IPSP SHOULD withhold
> > > >    the Notify (AS-ACTIVE) until there are sufficient resources.
> > > >
> > > > As the SGP SHOULD withhold the Notify (AS-ACTIVE), the ASP can rely on
> > > > its being withheld as an indication that the SGP is not directing traffic
> > > > to to the active ASPs for the AS because insufficient ASPs have been
> > > > activated.  Because the ASP SHOULD withhold traffic in the case of
> > > > insufficient ASPs, the ASP SHOULD NOT start traffic until it receives
> > > > Notify (AS-ACTIVE) from the SGP.  That seems pretty straightforward to me.
> > > >
> > > > Also, see example 5.1.4.
> > > >
> > > > > [TOLGA]If you think that for start case it is best to
> > > > > wait for n, then SG SHOULD withhold traffic instead of
> > > > > MAY withhold traffic(like ASP SHOULD wait for n ACTIVE
> > > > > ASPs) In fact as far as I recall the request to change
> > > > > this MAY to SHOULD was the reason, why the debate
> > > > > about n+k started. IMO how ASP and SGP behave
> > > > > regarding n should be the same.
> > > >
> > > > I disagree with SHOULD.  Directing traffic on activation requires
> > > > and exchange between the SG and the SS7 network.  These exchanges
> > > > are outside the scope of the RFC and the RFC cannot place requirements
> > > > on the SS7 network.  Also, in the case of SI granularity AS, the
> > > > SG cannot "direct traffic" to the AS.  It will take the expiry of UFC
> > > > timers at the remote signalling points before any traffic will be
> > > > restarted or directed.  By saying MAY, we are saying that the SG is
> > > > graciously permitted to do what it needs to do with respect to the
> > > > SS7 network.
> > > >
> > > > However, we do then say that if the SG does not direct traffic that
> > > > it SHOULD withold the Notify (AS-ACTIVE) to let the ASP know what it
> > > > is doing.  This is SG/ASP interaction and is within the scope of the
> > > > RFC.
> > > >
> > > > Also, what the ASPs do with regard to the SG is directly within the
> > > > scope of the RFC and the RFC can place the SHOULD requirement on the
> > > > ASPs.  The ASPs should not restart traffic until it knows that n ASPs
> > > > are up.  And how does it tell this?  Well the RFC alludes to some
> > > > inter-ASP coordination which is also outside thes scope of the RFC
> > > > (APS to ASP protocol).  The only requirement in the RFC that allows
> > > > the ASP to determine that n ASPs are active is that the SG has sent
> > > > Notify (AS-ACTIVE) for the AS.
> > > >
> > > > If you want to clarify "SHOULD wait until n ASPs are active" to
> > > > "SHOULD wait for Notify (AS-ACTIVE)", feel free, as it seems very
> > > > apparent to me anyway (and was the result of the thread just after
> > > > last call on the subject, too).
> > > >
> > > > > If this MAY is changed to SHOULD,
> > > >
> > > > MAY cannot change to SHOULD because it is an action towards the
> > > > SS7 network upon which M3UA cannot place requirements.
> > > >
> > > > > I would think still
> > > > > an update is necessary to AS state machine for
> > > > > AS-INACTIVE to AS-ACTIVE transition (nth ASP goes
> > > > > ACTIVE instead of 1st ASP goes ACTIVE and we don't
> > > > > touch transition from ACTIVE to PENDING)
> > > >
> > > > No, the state transition is still valid, it is just that the SGP
> > > > SHOULD withhold the Notify (AS-ACTIVE) until the nth ASP is active
> > > > iff it withholds traffic.
> > > >
> > > > The current state machine in the RFC is fine.
> > > >
> > > > > If nobody thinks that there is a realistic scenario,
> > > > > where n>1 is mandatory, i.e. less than n ACTIVE ASP
> > > > > will change AS state to PENDING-, then I think the
> > > > > approach you explained makes sense.
> > > >
> > > > Putting perfectly good processors out of service by blocking all
> > > > traffic to them and creating a complete service outage where one is
> > > > not necessary could not possibly be construed as realistic by any
> > > > operator.  I'd hate to be in front of Congress explaning that one!
> > > >
> > > > --brian
> > > >
> > > > --
> > > > Brian F. G. Bidulock
> > > > bidulock@openss7.org
> > > > http://www.openss7.org/
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> > --
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 14 01:16:56 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA01472
	for <sigtran-archive@odin.ietf.org>; Fri, 14 Feb 2003 01:16:55 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1E6KPI19923
	for sigtran-archive@odin.ietf.org; Fri, 14 Feb 2003 01:20:25 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E6K7p19906;
	Fri, 14 Feb 2003 01:20:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E6JCp19834
	for <sigtran@optimus.ietf.org>; Fri, 14 Feb 2003 01:19:12 -0500
Received: from india.mercurykr.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA01401
	for <sigtran@ietf.org>; Fri, 14 Feb 2003 01:14:54 -0500 (EST)
Received: from srivatsp (ptil-33-162-ind.primus-india.net [203.196.162.33] (may be forged))
	by india.mercurykr.com (8.11.2/8.11.2) with SMTP id h1EBO6Z04069
	for <sigtran@ietf.org>; Fri, 14 Feb 2003 16:54:06 +0530
Message-Id: <200302141124.h1EBO6Z04069@india.mercurykr.com>
Date: Fri, 14 Feb 2003 11:48:50 +0500
From: "Srivats P." <srivats@india.mercurykr.com>
Reply-To: srivatsp@email.com
To: sigtran <sigtran@ietf.org>
Organization: Mercury Corporation, India
X-mailer: FoxMail 2.1 [en]
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] M3UA: Inter-AS communication via SG
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

All,

Does the SIGTRAN/M3UA Architecture allow inter-AS
communication via the SG? (ASes are at point code
granularity). e.g. if AS1 is 1-1-1, AS2 is 2-2-2,
and SG is 3-3-3 ... can 1-1-1 talk to 2-2-2 via the
SG ?

If yes, when AS1 becomes inaccessible to SG, does
the SG send a DUNA(1-1-1) to 2-2-2 alongwith the
TFPs it sends into the SS7 network ?

If yes, is it correct to send a DUNA(*) to all ASPs
on SS7 network isolation. When I say "SS7 Isolation",
I mean no SS7 SP/STP is accessible to the SG, but
AS(es) at point code granularity are.

Regards,
Srivats P.

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 14 01:20:48 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA01518
	for <sigtran-archive@odin.ietf.org>; Fri, 14 Feb 2003 01:20:48 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1E6OHa20025
	for sigtran-archive@odin.ietf.org; Fri, 14 Feb 2003 01:24:17 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E6O4p20015;
	Fri, 14 Feb 2003 01:24:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E6NXp19987
	for <sigtran@optimus.ietf.org>; Fri, 14 Feb 2003 01:23:33 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA01507
	for <sigtran@ietf.org>; Fri, 14 Feb 2003 01:19:31 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1E6NCJT018581;
	Fri, 14 Feb 2003 00:23:13 -0600 (CST)
Received: from axes-mach01.usa.alcatel.com by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id KAA20933; Fri, 14 Feb 2003 10:52:08 -0500 (GMT)
Message-ID: <3E4C8B75.9726C9DB@axes-mach01.usa.alcatel.com>
Date: Fri, 14 Feb 2003 11:53:49 +0530
From: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: bidulock@openss7.org
CC: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
References: <20030212141714.D1284@openss7.org> <20030212220925.31809.qmail@web12805.mail.yahoo.com> <20030212200002.C17731@openss7.org> <3E4B49D9.2D602CE1@axes-mach01.usa.alcatel.com> <20030213074406.A15571@openss7.org> <3E4C769A.DAE063E0@axes-mach01.usa.alcatel.com> <20030213230727.A14023@openss7.org>
Content-Type: text/plain; charset=x-user-defined
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Brian,

"Brian F. G. Bidulock" wrote:

> ksrinivas,
>
> ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 10:24:50)
> > Brian,
> >
> >    1. Is the T(pending) different from T(r)?. (If Yes, RFC / IG does not
> >    include this).
>
> Sorry, I meant T(r).
>
> >    2. Should we start it after An AS is moved from AS-PENDING to AS-ACTIVE
> >    i.e., on the first ASP becoming Active?.
>
> No, T(r) is started when moving to AS-PENDING stae.

An AS is moved from AS-PENDING to AS-ACTIVE i.e., on the first ASP becoming Active. Should we wait till T(r) expires or n
number of ASP become Active in the AS-Active state itself?.
On T(r) expiry should we not start sending traffic to ASP, by sending Notify(AS-Active) and if required Notify(Insuff. ASP
resources)?.

>
>
> >    3. On timeout of T(pending) or after n number of ASPs becoming Active,
> >    should we send Notify(AS-Active)?.
>
> On timeout of T(r) move to AS-INACTIVE and send Notify(AS-Inactive).
>
> >    4. ASP should wait for NTFY(AS-ACTIVE) before starting traffic.
>
> The ASP could, but it is not required or recommended to do so in the RFC.
>
> --brian
>
> >
> > Is my understanding right?.
> >
> > --Srini
> >
> > "Brian F. G. Bidulock" wrote:
> >
> > > Srini,
> > >
> > > The decision to restart while pending is a different decision again
> > > from the start or stop decision.  If a brief period of time has
> > > elapsed, the network conditions at the time of restart are the same
> > > as the network conditions immediately before the restart.  IMO from
> > > the AS-PENDING state, the SG should move back to AS-ACTIVE with the
> > > first ASP coming active, as in the RFC.
> > >
> > > How the SG controls the restart decision is by setting the pending
> > > timer.  It was made clear that the T(pending) timer can be 0.
> > >
> > > For M3UA, time-controlled changeover and controlled rerouting buffer
> > > messages for about a second.  T(pending) timer values of 1 or 2 seconds
> > > are very usable, but the operator is welcome to set the timer to 0
> > > and force the start decision immediately.
> > >
> > > I think by controlling the timeout value one can get whatever behavior
> > > the operator wants.
> > >
> > > The ASP, on the other hand, cannot really know what T(pending) timeout
> > > value is set at the SG and should treat the situation as though it is
> > > initially starting.  That is, it should wait for NTFY(AS-ACTIVE) before
> > > starting traffic.  That way, if the SG is peforming the start deicision
> > > instead of the restart decision, and waiting for n ASPs, it will withhold
> > > NTFY(AS-ACTIVE) until n are up, and things should work correctly.
> > >
> > > Sound good?
> > >
> > > --brian
> > >
> > > ksrinivas wrote:                                                                 (Thu, 13 Feb 2003 13:01:37)
> > > > Brian,
> > > >
> > > >     I agree with your comments. An AS is moved from AS-PENDING to AS-ACTIVE on the first ASP becoming Active. Am I
> > > > right?. In that case, Should we send all the bufferred or queued traffic at SG to the first ASP that becomes Active
> > > > immediately or wait until n number of ASPs become Active?.
> > > >
> > > > --Srini
> > > >
> > > > "Brian F. G. Bidulock" wrote:
> > > >
> > > > > Tolga,
> > > > >
> > > > > Tolga Asveren wrote:                                                                 (Wed, 12 Feb 2003 14:09:25)
> > > > >
> > > > > > [TOLGA]Is there a place in RFC/IG, which indicates
> > > > > > that ASPs SHOULD wait for NTFY(AS Active) to start
> > > > > > send/receive traffic? To my knowledge there isn't and
> > > > > > this depends just on receiving ASPAC-ACK and on the
> > > > > > number of ACTIVE ASPs -based on the above paragraph-
> > > > > > IMO the not nice thing with those two statements is
> > > > > > that ASPs SHOULD wait for n ACTIVE ASPs to start
> > > > > > sending traffic, whereas SGP MAY withhold traffic.
> > > > > > There should be a consistency between them.
> > > > >
> > > > > Also, section 4.3.4.3 middle of page 82, top of page 83 (just after
> > > > > the sections you quoted):
> > > > >
> > > > > Loadshare:
> > > > >
> > > > >    An SGP or IPSP, upon reception of an ASP Active message for the first
> > > > >    ASP in a Loadshare AS, MAY choose not to direct traffic to a newly
> > > > >    active ASP until it determines that there are sufficient resources to
> > > > >    handle the expected load (e.g., until there are "n" ASPs in state
> > > > >    ASP-ACTIVE in the AS).  In this case, the SGP or IPSP SHOULD withhold
> > > > >    the Notify (AS-ACTIVE) until there are sufficient resources.
> > > > >
> > > > > Broadcast:
> > > > >
> > > > >    An SGP or IPSP, upon reception of an ASP Active message for the first
> > > > >    ASP in a Broadcast AS, MAY choose not to direct traffic to a newly
> > > > >    active ASP until it determines that there are sufficient resources to
> > > > >    handle the expected load (e.g., until there are "n" ASPs in state
> > > > >    ASP-ACTIVE in the AS).  In this case, the SGP or IPSP SHOULD withhold
> > > > >    the Notify (AS-ACTIVE) until there are sufficient resources.
> > > > >
> > > > > As the SGP SHOULD withhold the Notify (AS-ACTIVE), the ASP can rely on
> > > > > its being withheld as an indication that the SGP is not directing traffic
> > > > > to to the active ASPs for the AS because insufficient ASPs have been
> > > > > activated.  Because the ASP SHOULD withhold traffic in the case of
> > > > > insufficient ASPs, the ASP SHOULD NOT start traffic until it receives
> > > > > Notify (AS-ACTIVE) from the SGP.  That seems pretty straightforward to me.
> > > > >
> > > > > Also, see example 5.1.4.
> > > > >
> > > > > > [TOLGA]If you think that for start case it is best to
> > > > > > wait for n, then SG SHOULD withhold traffic instead of
> > > > > > MAY withhold traffic(like ASP SHOULD wait for n ACTIVE
> > > > > > ASPs) In fact as far as I recall the request to change
> > > > > > this MAY to SHOULD was the reason, why the debate
> > > > > > about n+k started. IMO how ASP and SGP behave
> > > > > > regarding n should be the same.
> > > > >
> > > > > I disagree with SHOULD.  Directing traffic on activation requires
> > > > > and exchange between the SG and the SS7 network.  These exchanges
> > > > > are outside the scope of the RFC and the RFC cannot place requirements
> > > > > on the SS7 network.  Also, in the case of SI granularity AS, the
> > > > > SG cannot "direct traffic" to the AS.  It will take the expiry of UFC
> > > > > timers at the remote signalling points before any traffic will be
> > > > > restarted or directed.  By saying MAY, we are saying that the SG is
> > > > > graciously permitted to do what it needs to do with respect to the
> > > > > SS7 network.
> > > > >
> > > > > However, we do then say that if the SG does not direct traffic that
> > > > > it SHOULD withold the Notify (AS-ACTIVE) to let the ASP know what it
> > > > > is doing.  This is SG/ASP interaction and is within the scope of the
> > > > > RFC.
> > > > >
> > > > > Also, what the ASPs do with regard to the SG is directly within the
> > > > > scope of the RFC and the RFC can place the SHOULD requirement on the
> > > > > ASPs.  The ASPs should not restart traffic until it knows that n ASPs
> > > > > are up.  And how does it tell this?  Well the RFC alludes to some
> > > > > inter-ASP coordination which is also outside thes scope of the RFC
> > > > > (APS to ASP protocol).  The only requirement in the RFC that allows
> > > > > the ASP to determine that n ASPs are active is that the SG has sent
> > > > > Notify (AS-ACTIVE) for the AS.
> > > > >
> > > > > If you want to clarify "SHOULD wait until n ASPs are active" to
> > > > > "SHOULD wait for Notify (AS-ACTIVE)", feel free, as it seems very
> > > > > apparent to me anyway (and was the result of the thread just after
> > > > > last call on the subject, too).
> > > > >
> > > > > > If this MAY is changed to SHOULD,
> > > > >
> > > > > MAY cannot change to SHOULD because it is an action towards the
> > > > > SS7 network upon which M3UA cannot place requirements.
> > > > >
> > > > > > I would think still
> > > > > > an update is necessary to AS state machine for
> > > > > > AS-INACTIVE to AS-ACTIVE transition (nth ASP goes
> > > > > > ACTIVE instead of 1st ASP goes ACTIVE and we don't
> > > > > > touch transition from ACTIVE to PENDING)
> > > > >
> > > > > No, the state transition is still valid, it is just that the SGP
> > > > > SHOULD withhold the Notify (AS-ACTIVE) until the nth ASP is active
> > > > > iff it withholds traffic.
> > > > >
> > > > > The current state machine in the RFC is fine.
> > > > >
> > > > > > If nobody thinks that there is a realistic scenario,
> > > > > > where n>1 is mandatory, i.e. less than n ACTIVE ASP
> > > > > > will change AS state to PENDING-, then I think the
> > > > > > approach you explained makes sense.
> > > > >
> > > > > Putting perfectly good processors out of service by blocking all
> > > > > traffic to them and creating a complete service outage where one is
> > > > > not necessary could not possibly be construed as realistic by any
> > > > > operator.  I'd hate to be in front of Congress explaning that one!
> > > > >
> > > > > --brian
> > > > >
> > > > > --
> > > > > Brian F. G. Bidulock
> > > > > bidulock@openss7.org
> > > > > http://www.openss7.org/
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >
> > > --
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
>
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 14 01:45:40 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA01949
	for <sigtran-archive@odin.ietf.org>; Fri, 14 Feb 2003 01:45:40 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1E6nA121723
	for sigtran-archive@odin.ietf.org; Fri, 14 Feb 2003 01:49:10 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E6jgp21623;
	Fri, 14 Feb 2003 01:45:42 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E6fxp21479
	for <sigtran@optimus.ietf.org>; Fri, 14 Feb 2003 01:42:05 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA01862
	for <sigtran@ietf.org>; Fri, 14 Feb 2003 01:37:57 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1E6fceW028271;
	Fri, 14 Feb 2003 00:41:39 -0600 (CST)
Received: from axes-mach01.usa.alcatel.com by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id LAA21622; Fri, 14 Feb 2003 11:10:34 -0500 (GMT)
Message-ID: <3E4C8FC8.774D5838@axes-mach01.usa.alcatel.com>
Date: Fri, 14 Feb 2003 12:12:16 +0530
From: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: bidulock@openss7.org
CC: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
References: <20030212141714.D1284@openss7.org> <20030212220925.31809.qmail@web12805.mail.yahoo.com> <20030212200002.C17731@openss7.org> <3E4B49D9.2D602CE1@axes-mach01.usa.alcatel.com> <20030213074406.A15571@openss7.org> <3E4C769A.DAE063E0@axes-mach01.usa.alcatel.com> <20030213230727.A14023@openss7.org>
Content-Type: text/plain; charset=x-user-defined
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Brian,

"Brian F. G. Bidulock" wrote:

> ksrinivas,
>
> ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 10:24:50)
> > Brian,
> >
> >    1. Is the T(pending) different from T(r)?. (If Yes, RFC / IG does not
> >    include this).
>
> Sorry, I meant T(r).
>
> >    2. Should we start it after An AS is moved from AS-PENDING to AS-ACTIVE
> >    i.e., on the first ASP becoming Active?.
>
> No, T(r) is started when moving to AS-PENDING stae.
>
> >    3. On timeout of T(pending) or after n number of ASPs becoming Active,
> >    should we send Notify(AS-Active)?.
>
> On timeout of T(r) move to AS-INACTIVE and send Notify(AS-Inactive).

If one ASP is in ASP-Active state for n > 1 and T(r) expires, Should we not send Notify(AS-Active) and Notify(Insuff. ASP
Resources) to all the Inactive ASPs and send existing(bufferred) traffic by initiating the congestion procedures?.

This will be similar to the case of, an ASP going ASP-Inactive, causing lesser than n number of  ASPs to be in ASP-Active
state.

>
>
> >    4. ASP should wait for NTFY(AS-ACTIVE) before starting traffic.
>
> The ASP could, but it is not required or recommended to do so in the RFC.
>
> --brian
>
> >
> > Is my understanding right?.
> >
> > --Srini
> >
> > "Brian F. G. Bidulock" wrote:
> >
> > > Srini,
> > >
> > > The decision to restart while pending is a different decision again
> > > from the start or stop decision.  If a brief period of time has
> > > elapsed, the network conditions at the time of restart are the same
> > > as the network conditions immediately before the restart.  IMO from
> > > the AS-PENDING state, the SG should move back to AS-ACTIVE with the
> > > first ASP coming active, as in the RFC.
> > >
> > > How the SG controls the restart decision is by setting the pending
> > > timer.  It was made clear that the T(pending) timer can be 0.
> > >
> > > For M3UA, time-controlled changeover and controlled rerouting buffer
> > > messages for about a second.  T(pending) timer values of 1 or 2 seconds
> > > are very usable, but the operator is welcome to set the timer to 0
> > > and force the start decision immediately.
> > >
> > > I think by controlling the timeout value one can get whatever behavior
> > > the operator wants.
> > >
> > > The ASP, on the other hand, cannot really know what T(pending) timeout
> > > value is set at the SG and should treat the situation as though it is
> > > initially starting.  That is, it should wait for NTFY(AS-ACTIVE) before
> > > starting traffic.  That way, if the SG is peforming the start deicision
> > > instead of the restart decision, and waiting for n ASPs, it will withhold
> > > NTFY(AS-ACTIVE) until n are up, and things should work correctly.
> > >
> > > Sound good?
> > >
> > > --brian
> > >
> > > ksrinivas wrote:                                                                 (Thu, 13 Feb 2003 13:01:37)
> > > > Brian,
> > > >
> > > >     I agree with your comments. An AS is moved from AS-PENDING to AS-ACTIVE on the first ASP becoming Active. Am I
> > > > right?. In that case, Should we send all the bufferred or queued traffic at SG to the first ASP that becomes Active
> > > > immediately or wait until n number of ASPs become Active?.
> > > >
> > > > --Srini
> > > >
> > > > "Brian F. G. Bidulock" wrote:
> > > >
> > > > > Tolga,
> > > > >
> > > > > Tolga Asveren wrote:                                                                 (Wed, 12 Feb 2003 14:09:25)
> > > > >
> > > > > > [TOLGA]Is there a place in RFC/IG, which indicates
> > > > > > that ASPs SHOULD wait for NTFY(AS Active) to start
> > > > > > send/receive traffic? To my knowledge there isn't and
> > > > > > this depends just on receiving ASPAC-ACK and on the
> > > > > > number of ACTIVE ASPs -based on the above paragraph-
> > > > > > IMO the not nice thing with those two statements is
> > > > > > that ASPs SHOULD wait for n ACTIVE ASPs to start
> > > > > > sending traffic, whereas SGP MAY withhold traffic.
> > > > > > There should be a consistency between them.
> > > > >
> > > > > Also, section 4.3.4.3 middle of page 82, top of page 83 (just after
> > > > > the sections you quoted):
> > > > >
> > > > > Loadshare:
> > > > >
> > > > >    An SGP or IPSP, upon reception of an ASP Active message for the first
> > > > >    ASP in a Loadshare AS, MAY choose not to direct traffic to a newly
> > > > >    active ASP until it determines that there are sufficient resources to
> > > > >    handle the expected load (e.g., until there are "n" ASPs in state
> > > > >    ASP-ACTIVE in the AS).  In this case, the SGP or IPSP SHOULD withhold
> > > > >    the Notify (AS-ACTIVE) until there are sufficient resources.
> > > > >
> > > > > Broadcast:
> > > > >
> > > > >    An SGP or IPSP, upon reception of an ASP Active message for the first
> > > > >    ASP in a Broadcast AS, MAY choose not to direct traffic to a newly
> > > > >    active ASP until it determines that there are sufficient resources to
> > > > >    handle the expected load (e.g., until there are "n" ASPs in state
> > > > >    ASP-ACTIVE in the AS).  In this case, the SGP or IPSP SHOULD withhold
> > > > >    the Notify (AS-ACTIVE) until there are sufficient resources.
> > > > >
> > > > > As the SGP SHOULD withhold the Notify (AS-ACTIVE), the ASP can rely on
> > > > > its being withheld as an indication that the SGP is not directing traffic
> > > > > to to the active ASPs for the AS because insufficient ASPs have been
> > > > > activated.  Because the ASP SHOULD withhold traffic in the case of
> > > > > insufficient ASPs, the ASP SHOULD NOT start traffic until it receives
> > > > > Notify (AS-ACTIVE) from the SGP.  That seems pretty straightforward to me.
> > > > >
> > > > > Also, see example 5.1.4.
> > > > >
> > > > > > [TOLGA]If you think that for start case it is best to
> > > > > > wait for n, then SG SHOULD withhold traffic instead of
> > > > > > MAY withhold traffic(like ASP SHOULD wait for n ACTIVE
> > > > > > ASPs) In fact as far as I recall the request to change
> > > > > > this MAY to SHOULD was the reason, why the debate
> > > > > > about n+k started. IMO how ASP and SGP behave
> > > > > > regarding n should be the same.
> > > > >
> > > > > I disagree with SHOULD.  Directing traffic on activation requires
> > > > > and exchange between the SG and the SS7 network.  These exchanges
> > > > > are outside the scope of the RFC and the RFC cannot place requirements
> > > > > on the SS7 network.  Also, in the case of SI granularity AS, the
> > > > > SG cannot "direct traffic" to the AS.  It will take the expiry of UFC
> > > > > timers at the remote signalling points before any traffic will be
> > > > > restarted or directed.  By saying MAY, we are saying that the SG is
> > > > > graciously permitted to do what it needs to do with respect to the
> > > > > SS7 network.
> > > > >
> > > > > However, we do then say that if the SG does not direct traffic that
> > > > > it SHOULD withold the Notify (AS-ACTIVE) to let the ASP know what it
> > > > > is doing.  This is SG/ASP interaction and is within the scope of the
> > > > > RFC.
> > > > >
> > > > > Also, what the ASPs do with regard to the SG is directly within the
> > > > > scope of the RFC and the RFC can place the SHOULD requirement on the
> > > > > ASPs.  The ASPs should not restart traffic until it knows that n ASPs
> > > > > are up.  And how does it tell this?  Well the RFC alludes to some
> > > > > inter-ASP coordination which is also outside thes scope of the RFC
> > > > > (APS to ASP protocol).  The only requirement in the RFC that allows
> > > > > the ASP to determine that n ASPs are active is that the SG has sent
> > > > > Notify (AS-ACTIVE) for the AS.
> > > > >
> > > > > If you want to clarify "SHOULD wait until n ASPs are active" to
> > > > > "SHOULD wait for Notify (AS-ACTIVE)", feel free, as it seems very
> > > > > apparent to me anyway (and was the result of the thread just after
> > > > > last call on the subject, too).
> > > > >
> > > > > > If this MAY is changed to SHOULD,
> > > > >
> > > > > MAY cannot change to SHOULD because it is an action towards the
> > > > > SS7 network upon which M3UA cannot place requirements.
> > > > >
> > > > > > I would think still
> > > > > > an update is necessary to AS state machine for
> > > > > > AS-INACTIVE to AS-ACTIVE transition (nth ASP goes
> > > > > > ACTIVE instead of 1st ASP goes ACTIVE and we don't
> > > > > > touch transition from ACTIVE to PENDING)
> > > > >
> > > > > No, the state transition is still valid, it is just that the SGP
> > > > > SHOULD withhold the Notify (AS-ACTIVE) until the nth ASP is active
> > > > > iff it withholds traffic.
> > > > >
> > > > > The current state machine in the RFC is fine.
> > > > >
> > > > > > If nobody thinks that there is a realistic scenario,
> > > > > > where n>1 is mandatory, i.e. less than n ACTIVE ASP
> > > > > > will change AS state to PENDING-, then I think the
> > > > > > approach you explained makes sense.
> > > > >
> > > > > Putting perfectly good processors out of service by blocking all
> > > > > traffic to them and creating a complete service outage where one is
> > > > > not necessary could not possibly be construed as realistic by any
> > > > > operator.  I'd hate to be in front of Congress explaning that one!
> > > > >
> > > > > --brian
> > > > >
> > > > > --
> > > > > Brian F. G. Bidulock
> > > > > bidulock@openss7.org
> > > > > http://www.openss7.org/
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >
> > > --
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
>
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 14 01:52:13 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA02064
	for <sigtran-archive@odin.ietf.org>; Fri, 14 Feb 2003 01:52:13 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1E6thl21954
	for sigtran-archive@odin.ietf.org; Fri, 14 Feb 2003 01:55:43 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E6tMp21924;
	Fri, 14 Feb 2003 01:55:22 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E6shp21892
	for <sigtran@optimus.ietf.org>; Fri, 14 Feb 2003 01:54:43 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA02034
	for <sigtran@ietf.org>; Fri, 14 Feb 2003 01:50:40 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:DcD8cOEzz/JCYf9AxG3ktLaLuXb6gLNT@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1E6sPD31084;
	Thu, 13 Feb 2003 23:54:25 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1E6sOU20962;
	Thu, 13 Feb 2003 23:54:24 -0700
Date: Thu, 13 Feb 2003 23:54:24 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
Cc: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030213235424.A20602@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>,
	Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030212141714.D1284@openss7.org> <20030212220925.31809.qmail@web12805.mail.yahoo.com> <20030212200002.C17731@openss7.org> <3E4B49D9.2D602CE1@axes-mach01.usa.alcatel.com> <20030213074406.A15571@openss7.org> <3E4C769A.DAE063E0@axes-mach01.usa.alcatel.com> <20030213230727.A14023@openss7.org> <3E4C8FC8.774D5838@axes-mach01.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E4C8FC8.774D5838@axes-mach01.usa.alcatel.com>; from ksrini@axes-mach01.usa.alcatel.com on Fri, Feb 14, 2003 at 12:12:16PM +0530
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

ksrinivas,

ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 12:12:16)
> Brian,
> 
> "Brian F. G. Bidulock" wrote:
> 
> > ksrinivas,
> >
> > ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 10:24:50)
> > > Brian,
> > >
> > >    1. Is the T(pending) different from T(r)?. (If Yes, RFC / IG does not
> > >    include this).
> >
> > Sorry, I meant T(r).
> >
> > >    2. Should we start it after An AS is moved from AS-PENDING to AS-ACTIVE
> > >    i.e., on the first ASP becoming Active?.
> >
> > No, T(r) is started when moving to AS-PENDING stae.
> >
> > >    3. On timeout of T(pending) or after n number of ASPs becoming Active,
> > >    should we send Notify(AS-Active)?.
> >
> > On timeout of T(r) move to AS-INACTIVE and send Notify(AS-Inactive).
> 
> If one ASP is in ASP-Active state for n > 1 and T(r) expires, Should we not
> send Notify(AS-Active) and Notify(Insuff. ASP Resources) to all the Inactive
> ASPs and send existing(bufferred) traffic by initiating the congestion
> procedures?.
> 
> This will be similar to the case of, an ASP going ASP-Inactive, causing
> lesser than n number of  ASPs to be in ASP-Active state.

One does not move to the AS-PENDING state until all the ASPs are ASP-INACTIVE
for the AS.  When T(r) expires, move to the AS-INACTIVE state as no ASPs are
active for the AS and the SG has waited long enough.

--brian

> 
> >
> >
> > >    4. ASP should wait for NTFY(AS-ACTIVE) before starting traffic.
> >
> > The ASP could, but it is not required or recommended to do so in the RFC.
> >
> > --brian
> >
> > >
> > > Is my understanding right?.
> > >
> > > --Srini
> > >
> > > "Brian F. G. Bidulock" wrote:
> > >
> > > > Srini,
> > > >
> > > > The decision to restart while pending is a different decision again
> > > > from the start or stop decision.  If a brief period of time has
> > > > elapsed, the network conditions at the time of restart are the same
> > > > as the network conditions immediately before the restart.  IMO from
> > > > the AS-PENDING state, the SG should move back to AS-ACTIVE with the
> > > > first ASP coming active, as in the RFC.
> > > >
> > > > How the SG controls the restart decision is by setting the pending
> > > > timer.  It was made clear that the T(pending) timer can be 0.
> > > >
> > > > For M3UA, time-controlled changeover and controlled rerouting buffer
> > > > messages for about a second.  T(pending) timer values of 1 or 2 seconds
> > > > are very usable, but the operator is welcome to set the timer to 0
> > > > and force the start decision immediately.
> > > >
> > > > I think by controlling the timeout value one can get whatever behavior
> > > > the operator wants.
> > > >
> > > > The ASP, on the other hand, cannot really know what T(pending) timeout
> > > > value is set at the SG and should treat the situation as though it is
> > > > initially starting.  That is, it should wait for NTFY(AS-ACTIVE) before
> > > > starting traffic.  That way, if the SG is peforming the start deicision
> > > > instead of the restart decision, and waiting for n ASPs, it will withhold
> > > > NTFY(AS-ACTIVE) until n are up, and things should work correctly.
> > > >
> > > > Sound good?
> > > >
> > > > --brian
> > > >
> > > > ksrinivas wrote:                                                                 (Thu, 13 Feb 2003 13:01:37)
> > > > > Brian,
> > > > >
> > > > >     I agree with your comments. An AS is moved from AS-PENDING to AS-ACTIVE on the first ASP becoming Active. Am I
> > > > > right?. In that case, Should we send all the bufferred or queued traffic at SG to the first ASP that becomes Active
> > > > > immediately or wait until n number of ASPs become Active?.
> > > > >
> > > > > --Srini
> > > > >
> > > > > "Brian F. G. Bidulock" wrote:
> > > > >
> > > > > > Tolga,
> > > > > >
> > > > > > Tolga Asveren wrote:                                                                 (Wed, 12 Feb 2003 14:09:25)
> > > > > >
> > > > > > > [TOLGA]Is there a place in RFC/IG, which indicates
> > > > > > > that ASPs SHOULD wait for NTFY(AS Active) to start
> > > > > > > send/receive traffic? To my knowledge there isn't and
> > > > > > > this depends just on receiving ASPAC-ACK and on the
> > > > > > > number of ACTIVE ASPs -based on the above paragraph-
> > > > > > > IMO the not nice thing with those two statements is
> > > > > > > that ASPs SHOULD wait for n ACTIVE ASPs to start
> > > > > > > sending traffic, whereas SGP MAY withhold traffic.
> > > > > > > There should be a consistency between them.
> > > > > >
> > > > > > Also, section 4.3.4.3 middle of page 82, top of page 83 (just after
> > > > > > the sections you quoted):
> > > > > >
> > > > > > Loadshare:
> > > > > >
> > > > > >    An SGP or IPSP, upon reception of an ASP Active message for the first
> > > > > >    ASP in a Loadshare AS, MAY choose not to direct traffic to a newly
> > > > > >    active ASP until it determines that there are sufficient resources to
> > > > > >    handle the expected load (e.g., until there are "n" ASPs in state
> > > > > >    ASP-ACTIVE in the AS).  In this case, the SGP or IPSP SHOULD withhold
> > > > > >    the Notify (AS-ACTIVE) until there are sufficient resources.
> > > > > >
> > > > > > Broadcast:
> > > > > >
> > > > > >    An SGP or IPSP, upon reception of an ASP Active message for the first
> > > > > >    ASP in a Broadcast AS, MAY choose not to direct traffic to a newly
> > > > > >    active ASP until it determines that there are sufficient resources to
> > > > > >    handle the expected load (e.g., until there are "n" ASPs in state
> > > > > >    ASP-ACTIVE in the AS).  In this case, the SGP or IPSP SHOULD withhold
> > > > > >    the Notify (AS-ACTIVE) until there are sufficient resources.
> > > > > >
> > > > > > As the SGP SHOULD withhold the Notify (AS-ACTIVE), the ASP can rely on
> > > > > > its being withheld as an indication that the SGP is not directing traffic
> > > > > > to to the active ASPs for the AS because insufficient ASPs have been
> > > > > > activated.  Because the ASP SHOULD withhold traffic in the case of
> > > > > > insufficient ASPs, the ASP SHOULD NOT start traffic until it receives
> > > > > > Notify (AS-ACTIVE) from the SGP.  That seems pretty straightforward to me.
> > > > > >
> > > > > > Also, see example 5.1.4.
> > > > > >
> > > > > > > [TOLGA]If you think that for start case it is best to
> > > > > > > wait for n, then SG SHOULD withhold traffic instead of
> > > > > > > MAY withhold traffic(like ASP SHOULD wait for n ACTIVE
> > > > > > > ASPs) In fact as far as I recall the request to change
> > > > > > > this MAY to SHOULD was the reason, why the debate
> > > > > > > about n+k started. IMO how ASP and SGP behave
> > > > > > > regarding n should be the same.
> > > > > >
> > > > > > I disagree with SHOULD.  Directing traffic on activation requires
> > > > > > and exchange between the SG and the SS7 network.  These exchanges
> > > > > > are outside the scope of the RFC and the RFC cannot place requirements
> > > > > > on the SS7 network.  Also, in the case of SI granularity AS, the
> > > > > > SG cannot "direct traffic" to the AS.  It will take the expiry of UFC
> > > > > > timers at the remote signalling points before any traffic will be
> > > > > > restarted or directed.  By saying MAY, we are saying that the SG is
> > > > > > graciously permitted to do what it needs to do with respect to the
> > > > > > SS7 network.
> > > > > >
> > > > > > However, we do then say that if the SG does not direct traffic that
> > > > > > it SHOULD withold the Notify (AS-ACTIVE) to let the ASP know what it
> > > > > > is doing.  This is SG/ASP interaction and is within the scope of the
> > > > > > RFC.
> > > > > >
> > > > > > Also, what the ASPs do with regard to the SG is directly within the
> > > > > > scope of the RFC and the RFC can place the SHOULD requirement on the
> > > > > > ASPs.  The ASPs should not restart traffic until it knows that n ASPs
> > > > > > are up.  And how does it tell this?  Well the RFC alludes to some
> > > > > > inter-ASP coordination which is also outside thes scope of the RFC
> > > > > > (APS to ASP protocol).  The only requirement in the RFC that allows
> > > > > > the ASP to determine that n ASPs are active is that the SG has sent
> > > > > > Notify (AS-ACTIVE) for the AS.
> > > > > >
> > > > > > If you want to clarify "SHOULD wait until n ASPs are active" to
> > > > > > "SHOULD wait for Notify (AS-ACTIVE)", feel free, as it seems very
> > > > > > apparent to me anyway (and was the result of the thread just after
> > > > > > last call on the subject, too).
> > > > > >
> > > > > > > If this MAY is changed to SHOULD,
> > > > > >
> > > > > > MAY cannot change to SHOULD because it is an action towards the
> > > > > > SS7 network upon which M3UA cannot place requirements.
> > > > > >
> > > > > > > I would think still
> > > > > > > an update is necessary to AS state machine for
> > > > > > > AS-INACTIVE to AS-ACTIVE transition (nth ASP goes
> > > > > > > ACTIVE instead of 1st ASP goes ACTIVE and we don't
> > > > > > > touch transition from ACTIVE to PENDING)
> > > > > >
> > > > > > No, the state transition is still valid, it is just that the SGP
> > > > > > SHOULD withhold the Notify (AS-ACTIVE) until the nth ASP is active
> > > > > > iff it withholds traffic.
> > > > > >
> > > > > > The current state machine in the RFC is fine.
> > > > > >
> > > > > > > If nobody thinks that there is a realistic scenario,
> > > > > > > where n>1 is mandatory, i.e. less than n ACTIVE ASP
> > > > > > > will change AS state to PENDING-, then I think the
> > > > > > > approach you explained makes sense.
> > > > > >
> > > > > > Putting perfectly good processors out of service by blocking all
> > > > > > traffic to them and creating a complete service outage where one is
> > > > > > not necessary could not possibly be construed as realistic by any
> > > > > > operator.  I'd hate to be in front of Congress explaning that one!
> > > > > >
> > > > > > --brian
> > > > > >
> > > > > > --
> > > > > > Brian F. G. Bidulock
> > > > > > bidulock@openss7.org
> > > > > > http://www.openss7.org/
> > > > > > _______________________________________________
> > > > > > Sigtran mailing list
> > > > > > Sigtran@ietf.org
> > > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > >
> > > > --
> > > > Brian F. G. Bidulock
> > > > bidulock@openss7.org
> > > > http://www.openss7.org/
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> > --
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 14 02:13:27 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA12304
	for <sigtran-archive@odin.ietf.org>; Fri, 14 Feb 2003 02:13:27 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1E7GwA00874
	for sigtran-archive@odin.ietf.org; Fri, 14 Feb 2003 02:16:58 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E7Gdp00866;
	Fri, 14 Feb 2003 02:16:40 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E7Fgp00814
	for <sigtran@optimus.ietf.org>; Fri, 14 Feb 2003 02:15:42 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA12277
	for <sigtran@ietf.org>; Fri, 14 Feb 2003 02:11:40 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:DA0+4zE1VkmhM5bXvqLe8pbaE7rao6rT@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1E7FOD31586;
	Fri, 14 Feb 2003 00:15:24 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1E7FOs23712;
	Fri, 14 Feb 2003 00:15:24 -0700
Date: Fri, 14 Feb 2003 00:15:24 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Srivats P." <srivats@india.mercurykr.com>
Cc: sigtran <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA: Inter-AS communication via SG
Message-ID: <20030214001524.B20602@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Srivats P." <srivats@india.mercurykr.com>,
	sigtran <sigtran@ietf.org>
References: <200302141124.h1EBO6Z04069@india.mercurykr.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <200302141124.h1EBO6Z04069@india.mercurykr.com>; from srivats@india.mercurykr.com on Fri, Feb 14, 2003 at 11:48:50AM +0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Srivats,

Srivats P. wrote:                                                                  (Fri, 14 Feb 2003 11:48:50)
> All,
> 
> Does the SIGTRAN/M3UA Architecture allow inter-AS
> communication via the SG? (ASes are at point code
> granularity). e.g. if AS1 is 1-1-1, AS2 is 2-2-2,
> and SG is 3-3-3 ... can 1-1-1 talk to 2-2-2 via the
> SG ?

Yes.

> 
> If yes, when AS1 becomes inaccessible to SG, does
> the SG send a DUNA(1-1-1) to 2-2-2 alongwith the
> TFPs it sends into the SS7 network ?

That would depend on whether the SG supports such routing keys and such
routes.  Although an SS7 platform will often support multiple point
codes, it will not necessarily support MTP transfer or MTP management
between them.  But such a thing is sometimes done, although it doesn't
have much practical use in a real network.

> 
> If yes, is it correct to send a DUNA(*) to all ASPs
> on SS7 network isolation. When I say "SS7 Isolation",
> I mean no SS7 SP/STP is accessible to the SG, but
> AS(es) at point code granularity are.

It is almost never correct to send DUNA(*) to all ASPs when using point
code granularity routing keys when the SG acts as an STP and one intends
to carry SCCP over M3UA.  This is because the STP has its own point code
(and possibly alias point code) which are still available to all active
AS.  Of importance would be the GTT function at the STP which is
addressable by applications at the AS.  Although one doesn't normally
route traffic between point code on a single platform, transferring
traffic between SCCP users via SCCP is required.

Instead the SG should send DUNA for each of the unavailable destinations
as they become unavailable.  Also, DUNA corresponds to MTP-PAUSE.  There
is no MTP-PAUSE(*).  Individual MTP-PAUSE primitives are delivered.

In the backhaul case, where the AS Point Code is the same as the SG's
point code (SEP instead of STP) then DUNA(*) is usable.

In section 3.4.1 it says:

                 ...  A mask value equal (or greater than) the
   number of bits in the PC indicates that the entire network
   appearance is affected - this is used to indicate network
   isolation to the ASP.

But in section 4.6 second paragraph it says:

   When the SG discovers SS7 network isolation, the SGPs send an
   indication to all concerned available ASPs (i.e., ASPs in the ASP-
   ACTIVE state) using DUNA messages for the concerned destinations.

So, I think its clear that you send DUNAs for the affected destinations
to concerned ASPs rather than just sending DUNA(*) whenever a partition
occurs.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 14 02:59:00 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA12907
	for <sigtran-archive@odin.ietf.org>; Fri, 14 Feb 2003 02:59:00 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1E82Wq04003
	for sigtran-archive@odin.ietf.org; Fri, 14 Feb 2003 03:02:32 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E82Cp03980;
	Fri, 14 Feb 2003 03:02:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E81Op03959
	for <sigtran@optimus.ietf.org>; Fri, 14 Feb 2003 03:01:24 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA12893
	for <sigtran@ietf.org>; Fri, 14 Feb 2003 02:57:20 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1E811qe011477;
	Fri, 14 Feb 2003 02:01:02 -0600 (CST)
Received: from axes-mach01.usa.alcatel.com by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id MAA23575; Fri, 14 Feb 2003 12:29:57 -0500 (GMT)
Message-ID: <3E4CA263.220FF34D@axes-mach01.usa.alcatel.com>
Date: Fri, 14 Feb 2003 13:31:39 +0530
From: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: bidulock@openss7.org
CC: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
References: <20030212141714.D1284@openss7.org> <20030212220925.31809.qmail@web12805.mail.yahoo.com> <20030212200002.C17731@openss7.org> <3E4B49D9.2D602CE1@axes-mach01.usa.alcatel.com> <20030213074406.A15571@openss7.org> <3E4C769A.DAE063E0@axes-mach01.usa.alcatel.com> <20030213230727.A14023@openss7.org> <3E4C8FC8.774D5838@axes-mach01.usa.alcatel.com> <20030213235424.A20602@openss7.org>
Content-Type: text/plain; charset=x-user-defined
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Brian,

"Brian F. G. Bidulock" wrote:

> ksrinivas,
>
> ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 12:12:16)
> > Brian,
> >
> > "Brian F. G. Bidulock" wrote:
> >
> > > ksrinivas,
> > >
> > > ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 10:24:50)
> > > > Brian,
> > > >
> > > >    1. Is the T(pending) different from T(r)?. (If Yes, RFC / IG does not
> > > >    include this).
> > >
> > > Sorry, I meant T(r).
> > >
> > > >    2. Should we start it after An AS is moved from AS-PENDING to AS-ACTIVE
> > > >    i.e., on the first ASP becoming Active?.
> > >
> > > No, T(r) is started when moving to AS-PENDING stae.
> > >
> > > >    3. On timeout of T(pending) or after n number of ASPs becoming Active,
> > > >    should we send Notify(AS-Active)?.
> > >
> > > On timeout of T(r) move to AS-INACTIVE and send Notify(AS-Inactive).
> >
> > If one ASP is in ASP-Active state for n > 1 and T(r) expires, Should we not
> > send Notify(AS-Active) and Notify(Insuff. ASP Resources) to all the Inactive
> > ASPs and send existing(bufferred) traffic by initiating the congestion
> > procedures?.
> >
> > This will be similar to the case of, an ASP going ASP-Inactive, causing
> > lesser than n number of  ASPs to be in ASP-Active state.
>
> One does not move to the AS-PENDING state until all the ASPs are ASP-INACTIVE
> for the AS.  When T(r) expires, move to the AS-INACTIVE state as no ASPs are
> active for the AS and the SG has waited long enough.

My question is, While AS is in AS-Pending state, one of the ASP has sent ASP Active for n > 1 and timer has expired, i.e., number
of Active ASP are still less than n(but not 0). Do we still have to move to AS-INACTIVE?. IMO, We have to move AS state to Active
and send Notify(AS-Active) and Notify(Insuff. ASP Resources) to all the Inactive ASPs.

>
>
> --brian
>
> >
> > >
> > >
> > > >    4. ASP should wait for NTFY(AS-ACTIVE) before starting traffic.
> > >
> > > The ASP could, but it is not required or recommended to do so in the RFC.
> > >
> > > --brian
> > >
> > > >
> > > > Is my understanding right?.
> > > >
> > > > --Srini
> > > >
> > > > "Brian F. G. Bidulock" wrote:
> > > >
> > > > > Srini,
> > > > >
> > > > > The decision to restart while pending is a different decision again
> > > > > from the start or stop decision.  If a brief period of time has
> > > > > elapsed, the network conditions at the time of restart are the same
> > > > > as the network conditions immediately before the restart.  IMO from
> > > > > the AS-PENDING state, the SG should move back to AS-ACTIVE with the
> > > > > first ASP coming active, as in the RFC.
> > > > >
> > > > > How the SG controls the restart decision is by setting the pending
> > > > > timer.  It was made clear that the T(pending) timer can be 0.
> > > > >
> > > > > For M3UA, time-controlled changeover and controlled rerouting buffer
> > > > > messages for about a second.  T(pending) timer values of 1 or 2 seconds
> > > > > are very usable, but the operator is welcome to set the timer to 0
> > > > > and force the start decision immediately.
> > > > >
> > > > > I think by controlling the timeout value one can get whatever behavior
> > > > > the operator wants.
> > > > >
> > > > > The ASP, on the other hand, cannot really know what T(pending) timeout
> > > > > value is set at the SG and should treat the situation as though it is
> > > > > initially starting.  That is, it should wait for NTFY(AS-ACTIVE) before
> > > > > starting traffic.  That way, if the SG is peforming the start deicision
> > > > > instead of the restart decision, and waiting for n ASPs, it will withhold
> > > > > NTFY(AS-ACTIVE) until n are up, and things should work correctly.
> > > > >
> > > > > Sound good?
> > > > >
> > > > > --brian
> > > > >
> > > > > ksrinivas wrote:                                                                 (Thu, 13 Feb 2003 13:01:37)
> > > > > > Brian,
> > > > > >
> > > > > >     I agree with your comments. An AS is moved from AS-PENDING to AS-ACTIVE on the first ASP becoming Active. Am I
> > > > > > right?. In that case, Should we send all the bufferred or queued traffic at SG to the first ASP that becomes Active
> > > > > > immediately or wait until n number of ASPs become Active?.
> > > > > >
> > > > > > --Srini
> > > > > >
> > > > > > "Brian F. G. Bidulock" wrote:
> > > > > >
> > > > > > > Tolga,
> > > > > > >
> > > > > > > Tolga Asveren wrote:                                                                 (Wed, 12 Feb 2003 14:09:25)
> > > > > > >
> > > > > > > > [TOLGA]Is there a place in RFC/IG, which indicates
> > > > > > > > that ASPs SHOULD wait for NTFY(AS Active) to start
> > > > > > > > send/receive traffic? To my knowledge there isn't and
> > > > > > > > this depends just on receiving ASPAC-ACK and on the
> > > > > > > > number of ACTIVE ASPs -based on the above paragraph-
> > > > > > > > IMO the not nice thing with those two statements is
> > > > > > > > that ASPs SHOULD wait for n ACTIVE ASPs to start
> > > > > > > > sending traffic, whereas SGP MAY withhold traffic.
> > > > > > > > There should be a consistency between them.
> > > > > > >
> > > > > > > Also, section 4.3.4.3 middle of page 82, top of page 83 (just after
> > > > > > > the sections you quoted):
> > > > > > >
> > > > > > > Loadshare:
> > > > > > >
> > > > > > >    An SGP or IPSP, upon reception of an ASP Active message for the first
> > > > > > >    ASP in a Loadshare AS, MAY choose not to direct traffic to a newly
> > > > > > >    active ASP until it determines that there are sufficient resources to
> > > > > > >    handle the expected load (e.g., until there are "n" ASPs in state
> > > > > > >    ASP-ACTIVE in the AS).  In this case, the SGP or IPSP SHOULD withhold
> > > > > > >    the Notify (AS-ACTIVE) until there are sufficient resources.
> > > > > > >
> > > > > > > Broadcast:
> > > > > > >
> > > > > > >    An SGP or IPSP, upon reception of an ASP Active message for the first
> > > > > > >    ASP in a Broadcast AS, MAY choose not to direct traffic to a newly
> > > > > > >    active ASP until it determines that there are sufficient resources to
> > > > > > >    handle the expected load (e.g., until there are "n" ASPs in state
> > > > > > >    ASP-ACTIVE in the AS).  In this case, the SGP or IPSP SHOULD withhold
> > > > > > >    the Notify (AS-ACTIVE) until there are sufficient resources.
> > > > > > >
> > > > > > > As the SGP SHOULD withhold the Notify (AS-ACTIVE), the ASP can rely on
> > > > > > > its being withheld as an indication that the SGP is not directing traffic
> > > > > > > to to the active ASPs for the AS because insufficient ASPs have been
> > > > > > > activated.  Because the ASP SHOULD withhold traffic in the case of
> > > > > > > insufficient ASPs, the ASP SHOULD NOT start traffic until it receives
> > > > > > > Notify (AS-ACTIVE) from the SGP.  That seems pretty straightforward to me.
> > > > > > >
> > > > > > > Also, see example 5.1.4.
> > > > > > >
> > > > > > > > [TOLGA]If you think that for start case it is best to
> > > > > > > > wait for n, then SG SHOULD withhold traffic instead of
> > > > > > > > MAY withhold traffic(like ASP SHOULD wait for n ACTIVE
> > > > > > > > ASPs) In fact as far as I recall the request to change
> > > > > > > > this MAY to SHOULD was the reason, why the debate
> > > > > > > > about n+k started. IMO how ASP and SGP behave
> > > > > > > > regarding n should be the same.
> > > > > > >
> > > > > > > I disagree with SHOULD.  Directing traffic on activation requires
> > > > > > > and exchange between the SG and the SS7 network.  These exchanges
> > > > > > > are outside the scope of the RFC and the RFC cannot place requirements
> > > > > > > on the SS7 network.  Also, in the case of SI granularity AS, the
> > > > > > > SG cannot "direct traffic" to the AS.  It will take the expiry of UFC
> > > > > > > timers at the remote signalling points before any traffic will be
> > > > > > > restarted or directed.  By saying MAY, we are saying that the SG is
> > > > > > > graciously permitted to do what it needs to do with respect to the
> > > > > > > SS7 network.
> > > > > > >
> > > > > > > However, we do then say that if the SG does not direct traffic that
> > > > > > > it SHOULD withold the Notify (AS-ACTIVE) to let the ASP know what it
> > > > > > > is doing.  This is SG/ASP interaction and is within the scope of the
> > > > > > > RFC.
> > > > > > >
> > > > > > > Also, what the ASPs do with regard to the SG is directly within the
> > > > > > > scope of the RFC and the RFC can place the SHOULD requirement on the
> > > > > > > ASPs.  The ASPs should not restart traffic until it knows that n ASPs
> > > > > > > are up.  And how does it tell this?  Well the RFC alludes to some
> > > > > > > inter-ASP coordination which is also outside thes scope of the RFC
> > > > > > > (APS to ASP protocol).  The only requirement in the RFC that allows
> > > > > > > the ASP to determine that n ASPs are active is that the SG has sent
> > > > > > > Notify (AS-ACTIVE) for the AS.
> > > > > > >
> > > > > > > If you want to clarify "SHOULD wait until n ASPs are active" to
> > > > > > > "SHOULD wait for Notify (AS-ACTIVE)", feel free, as it seems very
> > > > > > > apparent to me anyway (and was the result of the thread just after
> > > > > > > last call on the subject, too).
> > > > > > >
> > > > > > > > If this MAY is changed to SHOULD,
> > > > > > >
> > > > > > > MAY cannot change to SHOULD because it is an action towards the
> > > > > > > SS7 network upon which M3UA cannot place requirements.
> > > > > > >
> > > > > > > > I would think still
> > > > > > > > an update is necessary to AS state machine for
> > > > > > > > AS-INACTIVE to AS-ACTIVE transition (nth ASP goes
> > > > > > > > ACTIVE instead of 1st ASP goes ACTIVE and we don't
> > > > > > > > touch transition from ACTIVE to PENDING)
> > > > > > >
> > > > > > > No, the state transition is still valid, it is just that the SGP
> > > > > > > SHOULD withhold the Notify (AS-ACTIVE) until the nth ASP is active
> > > > > > > iff it withholds traffic.
> > > > > > >
> > > > > > > The current state machine in the RFC is fine.
> > > > > > >
> > > > > > > > If nobody thinks that there is a realistic scenario,
> > > > > > > > where n>1 is mandatory, i.e. less than n ACTIVE ASP
> > > > > > > > will change AS state to PENDING-, then I think the
> > > > > > > > approach you explained makes sense.
> > > > > > >
> > > > > > > Putting perfectly good processors out of service by blocking all
> > > > > > > traffic to them and creating a complete service outage where one is
> > > > > > > not necessary could not possibly be construed as realistic by any
> > > > > > > operator.  I'd hate to be in front of Congress explaning that one!
> > > > > > >
> > > > > > > --brian
> > > > > > >
> > > > > > > --
> > > > > > > Brian F. G. Bidulock
> > > > > > > bidulock@openss7.org
> > > > > > > http://www.openss7.org/
> > > > > > > _______________________________________________
> > > > > > > Sigtran mailing list
> > > > > > > Sigtran@ietf.org
> > > > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > >
> > > > > --
> > > > > Brian F. G. Bidulock
> > > > > bidulock@openss7.org
> > > > > http://www.openss7.org/
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >
> > > --
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
>
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 14 03:34:33 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA13599
	for <sigtran-archive@odin.ietf.org>; Fri, 14 Feb 2003 03:34:33 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1E8c6606705
	for sigtran-archive@odin.ietf.org; Fri, 14 Feb 2003 03:38:06 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E8bpp06687;
	Fri, 14 Feb 2003 03:37:52 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E8aHp05931
	for <sigtran@optimus.ietf.org>; Fri, 14 Feb 2003 03:36:17 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA13586
	for <sigtran@ietf.org>; Fri, 14 Feb 2003 03:32:13 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:8PmPMbLI3JLMXpNYpkVVEFB7GGBmOH0o@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1E8ZwD00386;
	Fri, 14 Feb 2003 01:35:58 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1E8Zv601431;
	Fri, 14 Feb 2003 01:35:57 -0700
Date: Fri, 14 Feb 2003 01:35:57 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
Cc: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030214013557.A25186@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>,
	Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030212141714.D1284@openss7.org> <20030212220925.31809.qmail@web12805.mail.yahoo.com> <20030212200002.C17731@openss7.org> <3E4B49D9.2D602CE1@axes-mach01.usa.alcatel.com> <20030213074406.A15571@openss7.org> <3E4C769A.DAE063E0@axes-mach01.usa.alcatel.com> <20030213230727.A14023@openss7.org> <3E4C8FC8.774D5838@axes-mach01.usa.alcatel.com> <20030213235424.A20602@openss7.org> <3E4CA263.220FF34D@axes-mach01.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E4CA263.220FF34D@axes-mach01.usa.alcatel.com>; from ksrini@axes-mach01.usa.alcatel.com on Fri, Feb 14, 2003 at 01:31:39PM +0530
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

ksrinivas,

ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 13:31:39)
> Brian,
> 
> "Brian F. G. Bidulock" wrote:
> 
> > ksrinivas,
> >
> > ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 12:12:16)
> > > Brian,
> > >
> > > "Brian F. G. Bidulock" wrote:
> > >
> > > > ksrinivas,
> > > >
> > > > ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 10:24:50)
> > > > > Brian,
> > > > >
> > > > >    1. Is the T(pending) different from T(r)?. (If Yes, RFC / IG does not
> > > > >    include this).
> > > >
> > > > Sorry, I meant T(r).
> > > >
> > > > >    2. Should we start it after An AS is moved from AS-PENDING to AS-ACTIVE
> > > > >    i.e., on the first ASP becoming Active?.
> > > >
> > > > No, T(r) is started when moving to AS-PENDING stae.
> > > >
> > > > >    3. On timeout of T(pending) or after n number of ASPs becoming Active,
> > > > >    should we send Notify(AS-Active)?.
> > > >
> > > > On timeout of T(r) move to AS-INACTIVE and send Notify(AS-Inactive).
> > >
> > > If one ASP is in ASP-Active state for n > 1 and T(r) expires, Should we not
> > > send Notify(AS-Active) and Notify(Insuff. ASP Resources) to all the Inactive
> > > ASPs and send existing(bufferred) traffic by initiating the congestion
> > > procedures?.
> > >
> > > This will be similar to the case of, an ASP going ASP-Inactive, causing
> > > lesser than n number of  ASPs to be in ASP-Active state.
> >
> > One does not move to the AS-PENDING state until all the ASPs are ASP-INACTIVE
> > for the AS.  When T(r) expires, move to the AS-INACTIVE state as no ASPs are
> > active for the AS and the SG has waited long enough.
> 
> My question is, While AS is in AS-Pending state, one of the ASP has sent ASP
> Active for n > 1 and timer has expired, i.e., number of Active ASP are still
> less than n(but not 0). Do we still have to move to AS-INACTIVE?. IMO, We
> have to move AS state to Active and send Notify(AS-Active) and
> Notify(Insuff. ASP Resources) to all the Inactive ASPs.

If the timer expired before the ASPAC, you are already in the AS-INACTIVE
state when the ASPAC arrived.  Then you would do the normal procedure from
the AS-INACTIVE state.  If the ASPAC arrives before the timer expires, the
timer is stopped and the AS-ACTIVE state entered.  Understand?

--brian


-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 14 03:55:54 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA14001
	for <sigtran-archive@odin.ietf.org>; Fri, 14 Feb 2003 03:55:54 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1E8xS507558
	for sigtran-archive@odin.ietf.org; Fri, 14 Feb 2003 03:59:28 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E8xAp07540;
	Fri, 14 Feb 2003 03:59:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E8wup07516
	for <sigtran@optimus.ietf.org>; Fri, 14 Feb 2003 03:58:56 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA13987
	for <sigtran@ietf.org>; Fri, 14 Feb 2003 03:54:51 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1E8wXJT014972;
	Fri, 14 Feb 2003 02:58:34 -0600 (CST)
Received: from axes-mach01.usa.alcatel.com by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id NAA25223; Fri, 14 Feb 2003 13:27:29 -0500 (GMT)
Message-ID: <3E4CAFE0.FC545136@axes-mach01.usa.alcatel.com>
Date: Fri, 14 Feb 2003 14:29:13 +0530
From: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: bidulock@openss7.org
CC: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
References: <20030212141714.D1284@openss7.org> <20030212220925.31809.qmail@web12805.mail.yahoo.com> <20030212200002.C17731@openss7.org> <3E4B49D9.2D602CE1@axes-mach01.usa.alcatel.com> <20030213074406.A15571@openss7.org> <3E4C769A.DAE063E0@axes-mach01.usa.alcatel.com> <20030213230727.A14023@openss7.org> <3E4C8FC8.774D5838@axes-mach01.usa.alcatel.com> <20030213235424.A20602@openss7.org> <3E4CA263.220FF34D@axes-mach01.usa.alcatel.com> <20030214013557.A25186@openss7.org>
Content-Type: text/plain; charset=x-user-defined
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Brian,

>>If the ASPAC arrives before the timer expires, the timer is stopped and the AS-ACTIVE state >>entered.  Understand?

When Notify(AS-Active) has to be sent?. for the restart case also, do we have to wait for n number of ASPs to become
Active and then only resume the traffic by sending Notify(AS-Active)?.

--Srini

"Brian F. G. Bidulock" wrote:

> ksrinivas,
>
> ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 13:31:39)
> > Brian,
> >
> > "Brian F. G. Bidulock" wrote:
> >
> > > ksrinivas,
> > >
> > > ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 12:12:16)
> > > > Brian,
> > > >
> > > > "Brian F. G. Bidulock" wrote:
> > > >
> > > > > ksrinivas,
> > > > >
> > > > > ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 10:24:50)
> > > > > > Brian,
> > > > > >
> > > > > >    1. Is the T(pending) different from T(r)?. (If Yes, RFC / IG does not
> > > > > >    include this).
> > > > >
> > > > > Sorry, I meant T(r).
> > > > >
> > > > > >    2. Should we start it after An AS is moved from AS-PENDING to AS-ACTIVE
> > > > > >    i.e., on the first ASP becoming Active?.
> > > > >
> > > > > No, T(r) is started when moving to AS-PENDING stae.
> > > > >
> > > > > >    3. On timeout of T(pending) or after n number of ASPs becoming Active,
> > > > > >    should we send Notify(AS-Active)?.
> > > > >
> > > > > On timeout of T(r) move to AS-INACTIVE and send Notify(AS-Inactive).
> > > >
> > > > If one ASP is in ASP-Active state for n > 1 and T(r) expires, Should we not
> > > > send Notify(AS-Active) and Notify(Insuff. ASP Resources) to all the Inactive
> > > > ASPs and send existing(bufferred) traffic by initiating the congestion
> > > > procedures?.
> > > >
> > > > This will be similar to the case of, an ASP going ASP-Inactive, causing
> > > > lesser than n number of  ASPs to be in ASP-Active state.
> > >
> > > One does not move to the AS-PENDING state until all the ASPs are ASP-INACTIVE
> > > for the AS.  When T(r) expires, move to the AS-INACTIVE state as no ASPs are
> > > active for the AS and the SG has waited long enough.
> >
> > My question is, While AS is in AS-Pending state, one of the ASP has sent ASP
> > Active for n > 1 and timer has expired, i.e., number of Active ASP are still
> > less than n(but not 0). Do we still have to move to AS-INACTIVE?. IMO, We
> > have to move AS state to Active and send Notify(AS-Active) and
> > Notify(Insuff. ASP Resources) to all the Inactive ASPs.
>
> If the timer expired before the ASPAC, you are already in the AS-INACTIVE
> state when the ASPAC arrived.  Then you would do the normal procedure from
> the AS-INACTIVE state.  If the ASPAC arrives before the timer expires, the
> timer is stopped and the AS-ACTIVE state entered.  Understand?
>
> --brian
>
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 14 04:03:44 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA14166
	for <sigtran-archive@odin.ietf.org>; Fri, 14 Feb 2003 04:03:44 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1E97Ia08579
	for sigtran-archive@odin.ietf.org; Fri, 14 Feb 2003 04:07:18 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E975p08276;
	Fri, 14 Feb 2003 04:07:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E96Ip08124
	for <sigtran@optimus.ietf.org>; Fri, 14 Feb 2003 04:06:18 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA14122
	for <sigtran@ietf.org>; Fri, 14 Feb 2003 04:02:13 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:i9jlzYSumDsP59CBFfacv1DAdVPNBtEc@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1E95wD00774;
	Fri, 14 Feb 2003 02:05:58 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1E95wa04841;
	Fri, 14 Feb 2003 02:05:58 -0700
Date: Fri, 14 Feb 2003 02:05:57 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
Cc: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030214020557.A4640@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>,
	Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030212200002.C17731@openss7.org> <3E4B49D9.2D602CE1@axes-mach01.usa.alcatel.com> <20030213074406.A15571@openss7.org> <3E4C769A.DAE063E0@axes-mach01.usa.alcatel.com> <20030213230727.A14023@openss7.org> <3E4C8FC8.774D5838@axes-mach01.usa.alcatel.com> <20030213235424.A20602@openss7.org> <3E4CA263.220FF34D@axes-mach01.usa.alcatel.com> <20030214013557.A25186@openss7.org> <3E4CAFE0.FC545136@axes-mach01.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E4CAFE0.FC545136@axes-mach01.usa.alcatel.com>; from ksrini@axes-mach01.usa.alcatel.com on Fri, Feb 14, 2003 at 02:29:13PM +0530
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

ksrinivas,

ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 14:29:13)
> Brian,
> 
> >>If the ASPAC arrives before the timer expires, the timer is stopped and
> >>the AS-ACTIVE state >>entered.  Understand?
> 
> When Notify(AS-Active) has to be sent?. for the restart case also, do we
> have to wait for n number of ASPs to become Active and then only resume the
> traffic by sending Notify(AS-Active)?.

No, I think the RFC says to do that only from the AS-INACTIVE state, it doesn't
say to do that from the AS-PENDING state.  Moving from AS-PENDING to AS-ACTIVE
the Notify(AS-Active) should IMO be sent immediately.

If one wants to treat restart like start, set T(r) to 0 seconds.  Then the
AS-INACTIVE state is entered immediately and AS-PENDING could even be skipped.

--brian

> 
> --Srini
> 
> "Brian F. G. Bidulock" wrote:
> 
> > ksrinivas,
> >
> > ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 13:31:39)
> > > Brian,
> > >
> > > "Brian F. G. Bidulock" wrote:
> > >
> > > > ksrinivas,
> > > >
> > > > ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 12:12:16)
> > > > > Brian,
> > > > >
> > > > > "Brian F. G. Bidulock" wrote:
> > > > >
> > > > > > ksrinivas,
> > > > > >
> > > > > > ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 10:24:50)
> > > > > > > Brian,
> > > > > > >
> > > > > > >    1. Is the T(pending) different from T(r)?. (If Yes, RFC / IG does not
> > > > > > >    include this).
> > > > > >
> > > > > > Sorry, I meant T(r).
> > > > > >
> > > > > > >    2. Should we start it after An AS is moved from AS-PENDING to AS-ACTIVE
> > > > > > >    i.e., on the first ASP becoming Active?.
> > > > > >
> > > > > > No, T(r) is started when moving to AS-PENDING stae.
> > > > > >
> > > > > > >    3. On timeout of T(pending) or after n number of ASPs becoming Active,
> > > > > > >    should we send Notify(AS-Active)?.
> > > > > >
> > > > > > On timeout of T(r) move to AS-INACTIVE and send Notify(AS-Inactive).
> > > > >
> > > > > If one ASP is in ASP-Active state for n > 1 and T(r) expires, Should we not
> > > > > send Notify(AS-Active) and Notify(Insuff. ASP Resources) to all the Inactive
> > > > > ASPs and send existing(bufferred) traffic by initiating the congestion
> > > > > procedures?.
> > > > >
> > > > > This will be similar to the case of, an ASP going ASP-Inactive, causing
> > > > > lesser than n number of  ASPs to be in ASP-Active state.
> > > >
> > > > One does not move to the AS-PENDING state until all the ASPs are ASP-INACTIVE
> > > > for the AS.  When T(r) expires, move to the AS-INACTIVE state as no ASPs are
> > > > active for the AS and the SG has waited long enough.
> > >
> > > My question is, While AS is in AS-Pending state, one of the ASP has sent ASP
> > > Active for n > 1 and timer has expired, i.e., number of Active ASP are still
> > > less than n(but not 0). Do we still have to move to AS-INACTIVE?. IMO, We
> > > have to move AS state to Active and send Notify(AS-Active) and
> > > Notify(Insuff. ASP Resources) to all the Inactive ASPs.
> >
> > If the timer expired before the ASPAC, you are already in the AS-INACTIVE
> > state when the ASPAC arrived.  Then you would do the normal procedure from
> > the AS-INACTIVE state.  If the ASPAC arrives before the timer expires, the
> > timer is stopped and the AS-ACTIVE state entered.  Understand?
> >
> > --brian
> >
> > --
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 14 04:12:49 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA14331
	for <sigtran-archive@odin.ietf.org>; Fri, 14 Feb 2003 04:12:49 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1E9GNs09243
	for sigtran-archive@odin.ietf.org; Fri, 14 Feb 2003 04:16:23 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E9G8p09235;
	Fri, 14 Feb 2003 04:16:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E9FJp09206
	for <sigtran@optimus.ietf.org>; Fri, 14 Feb 2003 04:15:19 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA14314
	for <sigtran@ietf.org>; Fri, 14 Feb 2003 04:11:13 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1E9EtJT016687;
	Fri, 14 Feb 2003 03:14:57 -0600 (CST)
Received: from axes-mach01.usa.alcatel.com by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id NAA25773; Fri, 14 Feb 2003 13:43:51 -0500 (GMT)
Message-ID: <3E4CB3B6.3285FA6E@axes-mach01.usa.alcatel.com>
Date: Fri, 14 Feb 2003 14:45:35 +0530
From: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: bidulock@openss7.org
CC: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
References: <20030212200002.C17731@openss7.org> <3E4B49D9.2D602CE1@axes-mach01.usa.alcatel.com> <20030213074406.A15571@openss7.org> <3E4C769A.DAE063E0@axes-mach01.usa.alcatel.com> <20030213230727.A14023@openss7.org> <3E4C8FC8.774D5838@axes-mach01.usa.alcatel.com> <20030213235424.A20602@openss7.org> <3E4CA263.220FF34D@axes-mach01.usa.alcatel.com> <20030214013557.A25186@openss7.org> <3E4CAFE0.FC545136@axes-mach01.usa.alcatel.com> <20030214020557.A4640@openss7.org>
Content-Type: text/plain; charset=x-user-defined
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Brian,

  I feel the things are pretty clear now. So what has to be done to section 3.15 in IG?.

--Srini.

"Brian F. G. Bidulock" wrote:

> ksrinivas,
>
> ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 14:29:13)
> > Brian,
> >
> > >>If the ASPAC arrives before the timer expires, the timer is stopped and
> > >>the AS-ACTIVE state >>entered.  Understand?
> >
> > When Notify(AS-Active) has to be sent?. for the restart case also, do we
> > have to wait for n number of ASPs to become Active and then only resume the
> > traffic by sending Notify(AS-Active)?.
>
> No, I think the RFC says to do that only from the AS-INACTIVE state, it doesn't
> say to do that from the AS-PENDING state.  Moving from AS-PENDING to AS-ACTIVE
> the Notify(AS-Active) should IMO be sent immediately.
>
> If one wants to treat restart like start, set T(r) to 0 seconds.  Then the
> AS-INACTIVE state is entered immediately and AS-PENDING could even be skipped.
>
> --brian
>
> >
> > --Srini
> >
> > "Brian F. G. Bidulock" wrote:
> >
> > > ksrinivas,
> > >
> > > ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 13:31:39)
> > > > Brian,
> > > >
> > > > "Brian F. G. Bidulock" wrote:
> > > >
> > > > > ksrinivas,
> > > > >
> > > > > ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 12:12:16)
> > > > > > Brian,
> > > > > >
> > > > > > "Brian F. G. Bidulock" wrote:
> > > > > >
> > > > > > > ksrinivas,
> > > > > > >
> > > > > > > ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 10:24:50)
> > > > > > > > Brian,
> > > > > > > >
> > > > > > > >    1. Is the T(pending) different from T(r)?. (If Yes, RFC / IG does not
> > > > > > > >    include this).
> > > > > > >
> > > > > > > Sorry, I meant T(r).
> > > > > > >
> > > > > > > >    2. Should we start it after An AS is moved from AS-PENDING to AS-ACTIVE
> > > > > > > >    i.e., on the first ASP becoming Active?.
> > > > > > >
> > > > > > > No, T(r) is started when moving to AS-PENDING stae.
> > > > > > >
> > > > > > > >    3. On timeout of T(pending) or after n number of ASPs becoming Active,
> > > > > > > >    should we send Notify(AS-Active)?.
> > > > > > >
> > > > > > > On timeout of T(r) move to AS-INACTIVE and send Notify(AS-Inactive).
> > > > > >
> > > > > > If one ASP is in ASP-Active state for n > 1 and T(r) expires, Should we not
> > > > > > send Notify(AS-Active) and Notify(Insuff. ASP Resources) to all the Inactive
> > > > > > ASPs and send existing(bufferred) traffic by initiating the congestion
> > > > > > procedures?.
> > > > > >
> > > > > > This will be similar to the case of, an ASP going ASP-Inactive, causing
> > > > > > lesser than n number of  ASPs to be in ASP-Active state.
> > > > >
> > > > > One does not move to the AS-PENDING state until all the ASPs are ASP-INACTIVE
> > > > > for the AS.  When T(r) expires, move to the AS-INACTIVE state as no ASPs are
> > > > > active for the AS and the SG has waited long enough.
> > > >
> > > > My question is, While AS is in AS-Pending state, one of the ASP has sent ASP
> > > > Active for n > 1 and timer has expired, i.e., number of Active ASP are still
> > > > less than n(but not 0). Do we still have to move to AS-INACTIVE?. IMO, We
> > > > have to move AS state to Active and send Notify(AS-Active) and
> > > > Notify(Insuff. ASP Resources) to all the Inactive ASPs.
> > >
> > > If the timer expired before the ASPAC, you are already in the AS-INACTIVE
> > > state when the ASPAC arrived.  Then you would do the normal procedure from
> > > the AS-INACTIVE state.  If the ASPAC arrives before the timer expires, the
> > > timer is stopped and the AS-ACTIVE state entered.  Understand?
> > >
> > > --brian
> > >
> > > --
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
>
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 14 04:13:43 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA14370
	for <sigtran-archive@odin.ietf.org>; Fri, 14 Feb 2003 04:13:43 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1E9HHJ09316
	for sigtran-archive@odin.ietf.org; Fri, 14 Feb 2003 04:17:17 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E9H6p09305;
	Fri, 14 Feb 2003 04:17:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1E9Ggp09266
	for <sigtran@optimus.ietf.org>; Fri, 14 Feb 2003 04:16:42 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA14326
	for <sigtran@ietf.org>; Fri, 14 Feb 2003 04:12:37 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:y+nMcOzsq8eXVcLpteHVm957iFrGQDWV@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1E9GND00918;
	Fri, 14 Feb 2003 02:16:23 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1E9GM005283;
	Fri, 14 Feb 2003 02:16:22 -0700
Date: Fri, 14 Feb 2003 02:16:22 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
Cc: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030214021622.A5112@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>,
	Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030213074406.A15571@openss7.org> <3E4C769A.DAE063E0@axes-mach01.usa.alcatel.com> <20030213230727.A14023@openss7.org> <3E4C8FC8.774D5838@axes-mach01.usa.alcatel.com> <20030213235424.A20602@openss7.org> <3E4CA263.220FF34D@axes-mach01.usa.alcatel.com> <20030214013557.A25186@openss7.org> <3E4CAFE0.FC545136@axes-mach01.usa.alcatel.com> <20030214020557.A4640@openss7.org> <3E4CB3B6.3285FA6E@axes-mach01.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E4CB3B6.3285FA6E@axes-mach01.usa.alcatel.com>; from ksrini@axes-mach01.usa.alcatel.com on Fri, Feb 14, 2003 at 02:45:35PM +0530
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

ksrinivas,

IMO section 3.15 of the IG just needs to be deleted.

--brian

ksrinivas wrote:                                                                 (Fri, 14 Feb 2003 14:45:35)
> Brian,
> 
>   I feel the things are pretty clear now. So what has to be done to section 3.15 in IG?.
> 
> --Srini.
> 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 14 08:08:37 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA21658
	for <sigtran-archive@odin.ietf.org>; Fri, 14 Feb 2003 08:08:37 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1EDCFB25355
	for sigtran-archive@odin.ietf.org; Fri, 14 Feb 2003 08:12:15 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1EDC0p25339;
	Fri, 14 Feb 2003 08:12:00 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1EDAqp25251
	for <sigtran@optimus.ietf.org>; Fri, 14 Feb 2003 08:10:52 -0500
Received: from domino2.nc.tekelec.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA21588
	for <sigtran@ietf.org>; Fri, 14 Feb 2003 08:06:43 -0500 (EST)
From: jeffrey.craig@tekelec.com
Received: from domino4.nc.tekelec.com ([198.89.40.9])
          by domino2.nc.tekelec.com (Lotus Domino Release 5.0.10)
          with ESMTP id 2003021408040757:36056 ;
          Fri, 14 Feb 2003 08:04:07 -0500 
Subject: Re: [Sigtran] M2PA draft 7 question -- alignment conclusion
To: Tom George <Tom.George@alcatel.com>
Cc: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.7  March 21, 2001
Message-ID: <OFF0C14EC4.26E78D94-ON85256CCD.0047CE69@nc.tekelec.com>
Date: Fri, 14 Feb 2003 08:06:12 -0500
MIME-Version: 1.0
X-MIMETrack: Serialize by Router on Domino4/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/14/2003 08:06:13 AM,
	Itemize by SMTP Server on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/14/2003 08:04:07 AM,
	Serialize by Router on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/14/2003 08:04:10 AM,
	Serialize complete at 02/14/2003 08:04:10 AM
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hello Tom,

I agree that the treatment of high latency and lossy networks is better
addressed by the reliable tranport protocol, i.e. not M2PA.

Besides, the M2PA timer durations listed in the draft are recommendations,
not requirements.

Regards,

Jeff



                                                                                                                                  
                      Tom George                                                                                                  
                      <Tom.George@alca         To:                                                                                
                      tel.com>                 cc:      sigtran@ietf.org                                                          
                      Sent by:                 Subject: Re: [Sigtran] M2PA draft 7 question -- alignment conclusion               
                      sigtran-admin@ie                                                                                            
                      tf.org                                                                                                      
                                                                                                                                  
                                                                                                                                  
                      02/13/03 05:52                                                                                              
                      PM                                                                                                          
                                                                                                                                  
                                                                                                                                  




If someone does want to use this over satellite, there will probably be
a number of changes that will be necessary. I would rather not try to
account for satellite transmission in this draft. What do the rest of
you think?

tomG.

wayne n taylor wrote:
>
> Brian,
> you are probably right !. but there are some folks who will want to run
> this over satellite ...
>
> wayne
  [...]

--
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 14 13:02:39 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA29119
	for <sigtran-archive@odin.ietf.org>; Fri, 14 Feb 2003 13:02:38 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1EI6M611795
	for sigtran-archive@odin.ietf.org; Fri, 14 Feb 2003 13:06:22 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1EI5vp11770;
	Fri, 14 Feb 2003 13:05:57 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1DDlgp20953
	for <sigtran@optimus.ietf.org>; Thu, 13 Feb 2003 08:47:42 -0500
Received: from beamer.mchh.siemens.de (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA11994
	for <sigtran@ietf.org>; Thu, 13 Feb 2003 08:37:16 -0500 (EST)
Received: from blues.mchh.siemens.de ([139.21.204.206])
	by beamer.mchh.siemens.de (8.9.3/8.9.3) with ESMTP id OAA20444;
	Thu, 13 Feb 2003 14:41:00 +0100 (MET)
Received: from mchh253e.mchh.siemens.de (mchh253e.mchh.siemens.de [139.21.200.63])
	by blues.mchh.siemens.de (8.9.1/8.9.1) with ESMTP id OAA09351;
	Thu, 13 Feb 2003 14:40:33 +0100 (MET)
Received: by MCHH253E with Internet Mail Service (5.5.2653.19)
	id <D131S4SS>; Thu, 13 Feb 2003 14:40:09 +0100
Message-ID: <3D72C18DF140D311A54F0008C7DBAC8D02917441@MCHH252E>
From: Gaiser Franz <franz.gaiser@siemens.com>
To: "'jeffrey.craig@tekelec.com'" <jeffrey.craig@tekelec.com>,
        Tom George
	 <Tom.George@alcatel.com>
Cc: sigtran@ietf.org
Subject: AW: [Sigtran] M2PA draft 07 comment: LPO
Date: Thu, 13 Feb 2003 14:39:44 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h1DDlgp20954
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Hello jeff,

an remark to your M2PA state event machine. In the case 'Proving' (T4 expired) is finished and local Processor Outage is occurred, I missing an arrow from state 'PROVING' to state 'ALIGNED_NOT_READY'.

regards, franz


-----Ursprüngliche Nachricht-----
Von: jeffrey.craig@tekelec.com [mailto:jeffrey.craig@tekelec.com]
Gesendet: Donnerstag, 13. Februar 2003 13:58
An: Tom George
Cc: sigtran@ietf.org
Betreff: Re: [Sigtran] M2PA draft 07 comment: LPO


Hello Tom,

The updated M2PA state machine figure still does not
match the Q.703 SDLs.

Here follows my attempt to get a closer match....

Regards,

Jeff

                   +-----------+
                   |   POWER   |
                   |    OFF    |
                   +-----------+
                         |
                         | Power On
                         |
       Flush Buffers     |  +-----------------------------------------+
       OR Retrieval      |  |                                         |
              +-------+  |  |                                         |
              |       |  V  V                                         |
              |    +-----------+                                      |
              +--->|  OUT OF   |                                      |
  +--------------->|  SERVICE  |<--+                                  |
  |                +-----------+   | Link Configured                  |
  |                      |   |     | (Associate)                      |
  |                      |   +-----+                                  |
  |     Receive LS OOS   |                                            |
  |           +-------+  | MTP3 Start                                 |
  |           |       |  V                                            |
  |           |    +-----------+                                      |
  |           +--->|    NOT    |                                      |
  +<---------------|  ALIGNED  |------------------------------------->+
  |     MTP3 Stop  +-----------+     SCTP Comm Error                  |
  |  OR T2 Expiry        |        OR SCTP Comm Lost                   |
  |                      |                                            |
  |                      |    Receive LS Align                        |
  |                      | OR Receive LS Prov Norm                    |
  |     Receive LS Align | OR Receive LS Prov Emer                    |
  |           +-------+  |                                            |
  |           |       |  V                                            |
  |           |    +-----------+                                      |
  |           +--->|  ALIGNED  |                                      |
  +<---------------|           |------------------------------------->+
  |     MTP3 Stop  +-----------+     SCTP Comm Error                  |
  |  OR T3 Expiry      ^    |     OR SCTP Comm Lost                   |
  |  OR Receive LS OOS |    |                                         |
  |                    |    |    Receive LS Prov Norm                 |
  |                    |    | OR Receive LS Prov Emer                 |
  |                    |    |                                         |
  |   Receive LS Align |    |                                         |
  |                    |    |                                         |
  |                    |    |                                         |
  | Receive LS Prov X  |    |                                         |
  |           +------+ |    |                                         |
  |           |      | |    V                                         |
  |           |    +-----------+                                      |
  |           +--->|  PROVING  |                                      |
  |<---------------|           |------------------------------------->+
  |     MTP3 Stop  +-----------+     SCTP Comm Error                  |
  |  OR Receive LS OOS   |        OR SCTP Comm Lost                   |
  |                      |                                            |
  |                      | T4 Expiry                                  |
  |                      |                                            |
  |                      |             MTP3 Stop            +-------->+
  |                      |             OR Receive LS OOS    |         |
  |                      |             OR SCTP Comm Error   |         |
  |                      |             OR SCTP Comm Lost    |         |
  |                      |             OR T1 Expiry         |         |
  |                      |             OR M2PA Link Failure |         |
  |                      |             OR Receive LS Align  |         |
  |                      V                                  |         |
  |                +-----------+                   +-----------+      |
  |                |  ALIGNED  |                   |  ALIGNED  |      |
  +<---------------|   READY   |<----------------->| NOT READY |      |
  |     MTP3 Stop  +-----------+     LPO/LPR       +-----------+      |
  |  OR T1 Expiry        |   |                         | Rcv LS Ready |
  |  OR Receive LS OOS   |   |                         | OR Rcv Data  |
  |  OR Receive LS Align |   +------------------------>+ OR Rcv LS PO |
  |  OR SCTP Comm Error  |   Receive LS PO             |              |
  |  OR SCTP Comm Lost   |                             |              |
  |                      |    Receive LS Ready         |              |
  |                      | OR Receive User Data        |              |
  |                      V                             V              |
  |                +-----------+ LPO or Rcv LS PO  +-----------+      |
  |                |    IN     |------------------>| PROCESSOR |      |
  +<---------------|  SERVICE  |<------------------|  OUTAGE   |      |
                   +-----------+    MTP3 Flush     +-----------+      |
       MTP3 Stop                 OR MTP3 Continue           |         |
    OR Receive LS OOS                                       |         |
    OR SCTP Comm Error                    MTP3 Stop         |         |
|
    OR SCTP Comm Lost                  OR Receive LS OOS    |         |
    OR T6 Expiry                       OR SCTP Comm Error   |         |
    OR T7 Expiry                       OR SCTP Comm Lost    |         |
    OR M2PA Link Failure               OR M2PA Link Failure |         |
    OR Receive LS Align                OR Receive LS Align  +-------->+

























                                                                                                                                  
                      Tom George                                                                                                  
                      <Tom.George@alca         To:                                                                                
                      tel.com>                 cc:      sigtran@ietf.org                                                          
                      Sent by:                 Subject: Re: [Sigtran] M2PA draft 07 comment: LPO                                  
                      sigtran-admin@ie                                                                                            
                      tf.org                                                                                                      
                                                                                                                                  
                                                                                                                                  
                      02/12/03 06:52                                                                                              
                      PM                                                                                                          
                                                                                                                                  
                                                                                                                                  




Franz,

I updated the diagram because of another discussion. I put an excerpt
from it further down in this message. I think it shows the transition
you asked about.

But there is nothing in the text that says this. I have this paragraph:

======
If there is a local processor outage condition during the alignment
procedure, M2PA sends Link Status Processor Outage to its peer. When
the local processor outage condition ends, then M2PA SHALL send Link
Status Processor Outage Ended to its peer. M2PA SHALL attempt to
complete the alignment procedure during the local processor outage
condition.
======

Perhaps I should add something like this: "If the local processor outage
condition exists when M2PA completes Proving, M2PA shall enter the
ALIGNED NOT READY state."


======
  |                       |   +-----+                  |
  |                       |   |     | LPO/LPR          |
  |                       V   V     |                  |
  |                 +-----------+   |                  |
  |                 |  PROVING  |---+                  |
  |<----------------|           |--------------------->+
  |      MTP3 Stop  +-----------+     SCTP Comm Error  |
  |   OR T3 Expiry        |  |     OR SCTP Comm Lost   |
  |   OR Receive LS OOS   |  |                         |
  |                       |  +-------------------------|--------+
  |                       |     T4 Expiry              |        |
  |                       |     Send LS PO             |        |
  |                       |                            |        |
  |                       | T4 Expiry                  |        |
  |                       | Send LS Ready              |        |
  |                       |                            |        |
  |                       |                            |        |
  |                       V           SCTP Comm Error  |        V
  |                 +-----------+  OR SCTP Comm Lost   |  +-----------+
  |                 |  ALIGNED  |--------------------->+  |  ALIGNED  |
  +<----------------|   READY   |<----------------------->| NOT READY |
  |      MTP3 Stop  +-----------+     LPO/LPR             +-----------+
  |   OR T1 Expiry        |        OR Receive LS PO/PR       |       |
  |   OR Receive LS OOS   |                                  |   T1  |
  |   OR Receive LS Align | Receive LS Ready      Receive    | Expiry|
  |                       | OR Receive User Data    LS Ready | OR Rcv|
  |                       |                                  |    LS |
  |                       |                                  |  Align|
======


tomG.

"Brian F. G. Bidulock" wrote:
>
> Gaiser,
>
> Gaiser Franz wrote:     (Tue, 11 Feb 2003 11:09:58)
> > Hi all,
> >
> > I have an additional comment/remark to the description of handling
Processor
> > Outage in draft 07of M2PA. The handling of the following situation is
not
> > clear to me: - Proving Period completed (T4 expired) and Local
Processor
> > Outage is occured. In the case without LPO the m2pa would start sending
LS
> > Ready to the peer and moves to state ALIGNED READY.
>
> > For the above situation (with LPO) the handling is not clearly stated.
Is LS
> > Ready and/or LS PO to be send ?According to Q.703  the m2pa should send
LS
> > PO message instead of LS Ready and moves to state ALIGNED NOT READY.
>
> Then that's what M2PA should do.
>
> --brian
>
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 14 13:08:41 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA29387
	for <sigtran-archive@odin.ietf.org>; Fri, 14 Feb 2003 13:08:41 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1EICPm12847
	for sigtran-archive@odin.ietf.org; Fri, 14 Feb 2003 13:12:25 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1EICCp12839;
	Fri, 14 Feb 2003 13:12:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1EIBKp12776
	for <sigtran@optimus.ietf.org>; Fri, 14 Feb 2003 13:11:20 -0500
Received: from w2ksjexg01.ciena.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA29262
	for <sigtran@ietf.org>; Fri, 14 Feb 2003 13:07:00 -0500 (EST)
Received: from wntcsdexg01.csd.ciena.com ([10.34.31.31]) by w2ksjexg01.ciena.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2653.13)
	id 1H7HL54N; Fri, 14 Feb 2003 10:10:28 -0800
Received: by webdev-llnt.oni.com with Internet Mail Service (5.5.2653.19)
	id <Y5W9MYB8>; Fri, 14 Feb 2003 10:10:43 -0800
Message-ID: <2135200C183FD5119588009027DE572302836C94@webdev-owa.oni.com>
From: "Ong, Lyndon" <LyOng@ciena.com>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Date: Fri, 14 Feb 2003 10:10:43 -0800
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Sigtran] meeting scheduled in San Francisco
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Hi Folks,

Since there was interest in meeting last time in at Atlanta, I've asked for
a meeting slot 
at the San Francisco meeting.  In addition to some review of our status and
discussion
of future items, I have also been informed that there is a liaison coming in
from ITU-T
SG 15 that we should have a look at (not sure if it has been published yet).

We currently have 2 hours on the schedule, although I would expect that we
might only
take 1.

Cheers,

L. Ong
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 14 17:35:22 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA07808
	for <sigtran-archive@odin.ietf.org>; Fri, 14 Feb 2003 17:35:21 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1EMdCg32150
	for sigtran-archive@odin.ietf.org; Fri, 14 Feb 2003 17:39:12 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1EMcrp32089;
	Fri, 14 Feb 2003 17:38:53 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1EMZRp31201
	for <sigtran@optimus.ietf.org>; Fri, 14 Feb 2003 17:35:27 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA07641
	for <sigtran@ietf.org>; Fri, 14 Feb 2003 17:31:06 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1EMYpJT019763;
	Fri, 14 Feb 2003 16:34:51 -0600 (CST)
Message-ID: <3E4D6F0B.943CC91D@alcatel.com>
Date: Fri, 14 Feb 2003 16:34:51 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft 7 question -- alignment conclusion
References: <OFF0C14EC4.26E78D94-ON85256CCD.0047CE69@nc.tekelec.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Then I think we can conclude this "conclusions" thread? 8-)

I am OK with the changes Jeff gave, including the "revision 3" of the
comments on section 4.1.3. I may make some minor editorial changes so
that it fits better with the new OOS section.

tomG.

jeffrey.craig@tekelec.com wrote:
> 
> Hello Tom,
> 
> I agree that the treatment of high latency and lossy networks is better
> addressed by the reliable tranport protocol, i.e. not M2PA.
> 
> Besides, the M2PA timer durations listed in the draft are recommendations,
> not requirements.
> 
> Regards,
> 
> Jeff
  [...]

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Sat Feb 15 13:23:48 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA05796
	for <sigtran-archive@odin.ietf.org>; Sat, 15 Feb 2003 13:23:48 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1FIS3i06526
	for sigtran-archive@odin.ietf.org; Sat, 15 Feb 2003 13:28:03 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1FIRkp06500;
	Sat, 15 Feb 2003 13:27:47 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1FIN2p06395
	for <sigtran@optimus.ietf.org>; Sat, 15 Feb 2003 13:23:02 -0500
Received: from web12805.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id NAA05757
	for <sigtran@ietf.org>; Sat, 15 Feb 2003 13:18:15 -0500 (EST)
Message-ID: <20030215182202.45051.qmail@web12805.mail.yahoo.com>
Received: from [62.29.76.176] by web12805.mail.yahoo.com via HTTP; Sat, 15 Feb 2003 10:22:02 PST
Date: Sat, 15 Feb 2003 10:22:02 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
In-Reply-To: <20030214021622.A5112@openss7.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Hi Everybody,
> ksrinivas,
> 
> IMO section 3.15 of the IG just needs to be deleted.
> 
> --brian
Just to briefly summarize my opinion about n+k debate:
a)If nobody sees a need for n as mandatory, let's make
necessary changes(i.e. return to the old state
machine)

b)IMO, if n is to be interpreted as the engineered
value (but no mandatory) and the decision is to be
made by SG when traffic can flow between ASP/SGP, it
should be stated somewhere that NTFY(AS ACTIVE) is to
be interpreted by ASP to start traffic, because ASPs
can't decide when SG is going to decide that there are
enough ACTIVE ASPs for an AS.

c)IMO, it is unacceptable to have the statement that
ASPs SHOULD wait for n ACTIVE ASPs to start traffic,
if n is to be interpreted as in b). Because in this
case, SG can decide to start traffic before n ASPs are
ACTIVE, but ASPs will wait for n ACTIVE ASPs, so they
won't send traffic.

d)I would think, the most clean way is still to update
the transition from INACTIVE to ACTIVE in the very
first version of AS state machine by saying that SG
decides based on traffic conditions and on the number
of ACTIVE ASPs when this transition happens.

e)I don't know, whether there might be cases, where it
makes sense that ACTIVE to PENDING transition is also
decided by SG based on number of ACTIVE ASPs and
traffic conditions(intead of the last ASP going
INACTIVE)

    Tolga

__________________________________________________
Do you Yahoo!?
Yahoo! Shopping - Send Flowers for Valentine's Day
http://shopping.yahoo.com
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Sat Feb 15 13:36:45 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA05972
	for <sigtran-archive@odin.ietf.org>; Sat, 15 Feb 2003 13:36:45 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1FIf1Q07540
	for sigtran-archive@odin.ietf.org; Sat, 15 Feb 2003 13:41:01 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1FIelp07531;
	Sat, 15 Feb 2003 13:40:47 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1FIbgp07446
	for <sigtran@optimus.ietf.org>; Sat, 15 Feb 2003 13:37:42 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA05937
	for <sigtran@ietf.org>; Sat, 15 Feb 2003 13:32:55 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:K336bJNNn3UMzlNLCBsoOOm4SaWUNpxR@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1FIagD27267;
	Sat, 15 Feb 2003 11:36:42 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1FIag818040;
	Sat, 15 Feb 2003 11:36:42 -0700
Date: Sat, 15 Feb 2003 11:36:41 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tolga Asveren <tolga_asveren@yahoo.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030215113641.A17855@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030214021622.A5112@openss7.org> <20030215182202.45051.qmail@web12805.mail.yahoo.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20030215182202.45051.qmail@web12805.mail.yahoo.com>; from tolga_asveren@yahoo.com on Sat, Feb 15, 2003 at 10:22:02AM -0800
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Tolga,

I have explained on this thread already the stop, start and
restart decisions.  Also, 'n' has always meant the number
required for engineered maximum load, you have simply
just mis-interpreted it.  Do not update the inactive to
active transition: it is adequately described in the RFC.

Strike IG section 3.15 an let the RFC be as it stands.

--brian

Tolga Asveren wrote:                     (Sat, 15 Feb 2003 10:22:02)
> Hi Everybody,
> > ksrinivas,
> > 
> > IMO section 3.15 of the IG just needs to be deleted.
> > 
> > --brian
> Just to briefly summarize my opinion about n+k debate:
> a)If nobody sees a need for n as mandatory, let's make
> necessary changes(i.e. return to the old state
> machine)
> 
> b)IMO, if n is to be interpreted as the engineered
> value (but no mandatory) and the decision is to be
> made by SG when traffic can flow between ASP/SGP, it
> should be stated somewhere that NTFY(AS ACTIVE) is to
> be interpreted by ASP to start traffic, because ASPs
> can't decide when SG is going to decide that there are
> enough ACTIVE ASPs for an AS.
> 
> c)IMO, it is unacceptable to have the statement that
> ASPs SHOULD wait for n ACTIVE ASPs to start traffic,
> if n is to be interpreted as in b). Because in this
> case, SG can decide to start traffic before n ASPs are
> ACTIVE, but ASPs will wait for n ACTIVE ASPs, so they
> won't send traffic.
> 
> d)I would think, the most clean way is still to update
> the transition from INACTIVE to ACTIVE in the very
> first version of AS state machine by saying that SG
> decides based on traffic conditions and on the number
> of ACTIVE ASPs when this transition happens.
> 
> e)I don't know, whether there might be cases, where it
> makes sense that ACTIVE to PENDING transition is also
> decided by SG based on number of ACTIVE ASPs and
> traffic conditions(intead of the last ASP going
> INACTIVE)
> 
>     Tolga
> 
> __________________________________________________
> Do you Yahoo!?
> Yahoo! Shopping - Send Flowers for Valentine's Day
> http://shopping.yahoo.com
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 17 05:51:30 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA19684
	for <sigtran-archive@odin.ietf.org>; Mon, 17 Feb 2003 05:51:29 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1HAuXb17472
	for sigtran-archive@odin.ietf.org; Mon, 17 Feb 2003 05:56:33 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1HAuBp17464;
	Mon, 17 Feb 2003 05:56:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1HAstp17378
	for <sigtran@optimus.ietf.org>; Mon, 17 Feb 2003 05:54:55 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA19612
	for <sigtran@ietf.org>; Mon, 17 Feb 2003 05:49:15 -0500 (EST)
Subject: Re: [Sigtran] meeting scheduled in San Francisco
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF6A7A322B.B316B233-ONC2256CD0.00393130@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Mon, 17 Feb 2003 12:49:32 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/17/2003
 12:49:42 PM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hi Lyndon,

I hope this time there will be some decisions about future work items. As
you can guess easily I am mostly interested about SG-SG draft. Last time in
Atlanta while talking with people after the unofficial SIGTRAN meeting (so
kind of a more unofficial small meeting) I had the impression that they are
not thinking negative about it. In fact as a lot of people might recall,
during initial IPSP discussions (just before the text was frozen and when
there was more involvement to the discussions) there was a tendency in the
WG to go after SG-SG communication, after the M3UA document is
finished/corrected.

Afterwards several times I tried to bring up this topic, but unfortunately
there weren't many comments (like there weren't many comments for other
topics either). Now I am really wondering, how any progress can be made
about it (does not necessarily mean to have a WG document for SG-SG
communication but to have a WG opinion about this topic).

For example should no disagreement be interpreted as agreement (like for
GR-303 document, just as clarification, I wouldn't think so as SG-SG
communication is certainly a bigger change, and I believe there should be
more people in the WG apart from Ken expressing their opinions)?

As the WG chair what do you think about how to proceed?

(BTW, I am sorry for the formatting problem with the current draft, already
preparing a new version including Ken's comments/questions and some more
clarifications)

    Tolga



                                                                                                           
                    "Ong, Lyndon"                                                                          
                    <LyOng@ciena.c       To:     "'sigtran@ietf.org'" <sigtran@ietf.org>                   
                    om>                  cc:                                                               
                    Sent by:             Subject:     [Sigtran] meeting scheduled in San Francisco         
                    sigtran-admin@                                                                         
                    ietf.org                                                                               
                                                                                                           
                                                                                                           
                    14.02.2003                                                                             
                    20:10                                                                                  
                                                                                                           
                                                                                                           




Hi Folks,

Since there was interest in meeting last time in at Atlanta, I've asked for
a meeting slot
at the San Francisco meeting.  In addition to some review of our status and
discussion
of future items, I have also been informed that there is a liaison coming
in
from ITU-T
SG 15 that we should have a look at (not sure if it has been published
yet).

We currently have 2 hours on the schedule, although I would expect that we
might only
take 1.

Cheers,

L. Ong
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 17 10:00:51 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA23407
	for <sigtran-archive@odin.ietf.org>; Mon, 17 Feb 2003 10:00:51 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1HF5wE31548
	for sigtran-archive@odin.ietf.org; Mon, 17 Feb 2003 10:05:58 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1HF5fp31536;
	Mon, 17 Feb 2003 10:05:43 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1HF3wp31420
	for <sigtran@optimus.ietf.org>; Mon, 17 Feb 2003 10:03:58 -0500
Received: from penguin.wise.edt.ericsson.se (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA23329
	for <sigtran@ietf.org>; Mon, 17 Feb 2003 09:58:18 -0500 (EST)
Received: from esealnt611.al.sw.ericsson.se (esealnt611.al.sw.ericsson.se [153.88.254.68])
	by penguin.wise.edt.ericsson.se (8.12.1/8.12.1/WIREfire-1.4) with ESMTP id h1HF24Av015392;
	Mon, 17 Feb 2003 16:02:05 +0100 (MET)
Received: from ESEALNT745.al.sw.ericsson.se ([153.88.251.5]) by esealnt611.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55)
	id FDV2PQ9C; Mon, 17 Feb 2003 16:02:04 +0100
Received: by ESEALNT745.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <DW9DD29W>; Mon, 17 Feb 2003 16:02:04 +0100
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65965@EESTQNT105.es.eu.ericsson.se>
X-Sybari-Trust: 71befe30 9ffcebbb a5ee123c 00000138
From: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>,
        Tolga Asveren
	 <tolga_asveren@yahoo.com>
Cc: sigtran@ietf.org
Subject: RE: [Sigtran] Re: M3UA Notify message
Date: Mon, 17 Feb 2003 16:02:02 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hi Brian, 

Why do you say that: "'n' has always meant the number required for engineered maximum load" when the RFC says:

   The failover model supports an "n+k" redundancy model, where "n" ASPs
   is the minimum number of redundant ASPs required to handle traffic
   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
   and "k" ASPs are available to take over for a failed or unavailable
   ASP.

(??)


Cheers // Javier.



> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Saturday, February 15, 2003 7:37 PM
> To: Tolga Asveren
> Cc: sigtran@ietf.org
> Subject: Re: [Sigtran] Re: M3UA Notify message
> 
> 
> Tolga,
> 
> I have explained on this thread already the stop, start and
> restart decisions.  Also, 'n' has always meant the number
> required for engineered maximum load, you have simply
> just mis-interpreted it.  Do not update the inactive to
> active transition: it is adequately described in the RFC.
> 
> Strike IG section 3.15 an let the RFC be as it stands.
> 
> --brian
> 
> Tolga Asveren wrote:                     (Sat, 15 Feb 2003 10:22:02)
> > Hi Everybody,
> > > ksrinivas,
> > > 
> > > IMO section 3.15 of the IG just needs to be deleted.
> > > 
> > > --brian
> > Just to briefly summarize my opinion about n+k debate:
> > a)If nobody sees a need for n as mandatory, let's make
> > necessary changes(i.e. return to the old state
> > machine)
> > 
> > b)IMO, if n is to be interpreted as the engineered
> > value (but no mandatory) and the decision is to be
> > made by SG when traffic can flow between ASP/SGP, it
> > should be stated somewhere that NTFY(AS ACTIVE) is to
> > be interpreted by ASP to start traffic, because ASPs
> > can't decide when SG is going to decide that there are
> > enough ACTIVE ASPs for an AS.
> > 
> > c)IMO, it is unacceptable to have the statement that
> > ASPs SHOULD wait for n ACTIVE ASPs to start traffic,
> > if n is to be interpreted as in b). Because in this
> > case, SG can decide to start traffic before n ASPs are
> > ACTIVE, but ASPs will wait for n ACTIVE ASPs, so they
> > won't send traffic.
> > 
> > d)I would think, the most clean way is still to update
> > the transition from INACTIVE to ACTIVE in the very
> > first version of AS state machine by saying that SG
> > decides based on traffic conditions and on the number
> > of ACTIVE ASPs when this transition happens.
> > 
> > e)I don't know, whether there might be cases, where it
> > makes sense that ACTIVE to PENDING transition is also
> > decided by SG based on number of ACTIVE ASPs and
> > traffic conditions(intead of the last ASP going
> > INACTIVE)
> > 
> >     Tolga
> > 
> > __________________________________________________
> > Do you Yahoo!?
> > Yahoo! Shopping - Send Flowers for Valentine's Day
> > http://shopping.yahoo.com
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 17 10:25:10 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA23795
	for <sigtran-archive@odin.ietf.org>; Mon, 17 Feb 2003 10:25:10 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1HFUIA00550
	for sigtran-archive@odin.ietf.org; Mon, 17 Feb 2003 10:30:18 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1HFU6p00539;
	Mon, 17 Feb 2003 10:30:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1HFTJp00494
	for <sigtran@optimus.ietf.org>; Mon, 17 Feb 2003 10:29:19 -0500
Received: from penguin.wise.edt.ericsson.se (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA23755
	for <sigtran@ietf.org>; Mon, 17 Feb 2003 10:23:40 -0500 (EST)
Received: from esealnt612.al.sw.ericsson.se (esealnt612.al.sw.ericsson.se [153.88.254.71])
	by penguin.wise.edt.ericsson.se (8.12.1/8.12.1/WIREfire-1.4) with ESMTP id h1HFRQAv021554;
	Mon, 17 Feb 2003 16:27:26 +0100 (MET)
Received: from ESEALNT745.al.sw.ericsson.se ([153.88.251.5]) by esealnt612.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55)
	id FDVGPQGT; Mon, 17 Feb 2003 16:27:26 +0100
Received: by ESEALNT745.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <DW9DDJ3A>; Mon, 17 Feb 2003 16:27:26 +0100
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65969@EESTQNT105.es.eu.ericsson.se>
X-Sybari-Trust: 52b6999b 9ffcebbb 7a95d2f4 00000138
From: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>
To: "'Ong, Lyndon'" <LyOng@ciena.com>,
        "'sigtran@ietf.org'"
	 <sigtran@ietf.org>
Subject: RE: [Sigtran] meeting scheduled in San Francisco
Date: Mon, 17 Feb 2003 16:17:19 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Any draft-agenda at this time?

Thanks // Javier.


> -----Original Message-----
> From: Ong, Lyndon [mailto:LyOng@ciena.com]
> Sent: Friday, February 14, 2003 7:11 PM
> To: 'sigtran@ietf.org'
> Subject: [Sigtran] meeting scheduled in San Francisco
> 
> 
> Hi Folks,
> 
> Since there was interest in meeting last time in at Atlanta, 
> I've asked for
> a meeting slot 
> at the San Francisco meeting.  In addition to some review of 
> our status and
> discussion
> of future items, I have also been informed that there is a 
> liaison coming in
> from ITU-T
> SG 15 that we should have a look at (not sure if it has been 
> published yet).
> 
> We currently have 2 hours on the schedule, although I would 
> expect that we
> might only
> take 1.
> 
> Cheers,
> 
> L. Ong
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 17 15:26:41 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA00525
	for <sigtran-archive@odin.ietf.org>; Mon, 17 Feb 2003 15:26:41 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1HKVvO20365
	for sigtran-archive@odin.ietf.org; Mon, 17 Feb 2003 15:31:57 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1HKVep20351;
	Mon, 17 Feb 2003 15:31:41 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1HKUIp20268
	for <sigtran@optimus.ietf.org>; Mon, 17 Feb 2003 15:30:18 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA00467
	for <sigtran@ietf.org>; Mon, 17 Feb 2003 15:24:29 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:JikzQMXIcnYSpX/pO+wMYV3ze/bTcXS1@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1HKSHD12475;
	Mon, 17 Feb 2003 13:28:17 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1HKSHM17363;
	Mon, 17 Feb 2003 13:28:17 -0700
Date: Mon, 17 Feb 2003 13:28:16 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>
Cc: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030217132816.A16039@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>,
	Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65965@EESTQNT105.es.eu.ericsson.se>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <1AB3D30B989BF141BBD5C70057B2EF7C01B65965@EESTQNT105.es.eu.ericsson.se>; from javier.pastor-balbas@ece.ericsson.se on Mon, Feb 17, 2003 at 04:02:02PM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Javier,

Same thing.

What did you think one would use to determine the "traffic" that "n" ASPs
are required to handle, other than their engineered maximum traffic
handling capability?

--brian

Javier Pastor-Balbas (ECE) wrote:                                                   (Mon, 17 Feb 2003 16:02:02)
> 
> Hi Brian, 
> 
> Why do you say that: "'n' has always meant the number required for engineered maximum load" when the RFC says:
> 
>    The failover model supports an "n+k" redundancy model, where "n" ASPs
>    is the minimum number of redundant ASPs required to handle traffic
>    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>    and "k" ASPs are available to take over for a failed or unavailable
>    ASP.
> 
> (??)
> 
> 
> Cheers // Javier.
> 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 17 18:01:21 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA03443
	for <sigtran-archive@odin.ietf.org>; Mon, 17 Feb 2003 18:01:21 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1HN6d030394
	for sigtran-archive@odin.ietf.org; Mon, 17 Feb 2003 18:06:39 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1HN6Jp30381;
	Mon, 17 Feb 2003 18:06:21 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1HN3fp30310
	for <sigtran@optimus.ietf.org>; Mon, 17 Feb 2003 18:03:41 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA03383
	for <sigtran@ietf.org>; Mon, 17 Feb 2003 17:57:52 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1HN1aqe024848;
	Mon, 17 Feb 2003 17:01:37 -0600 (CST)
Message-ID: <3E5169D0.7D28817B@alcatel.com>
Date: Mon, 17 Feb 2003 17:01:36 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Security section (was: m2pa comment: author list)
References: <4.3.2.7.2.20030212094830.04b6cf50@cia.cisco.com> <3E4A6A9A.8747D276@alcatel.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Here is my 2nd attempt at revising the security section.

tomG.

==========================================
6. Security

M2PA is designed to carry signaling messages for telephony
services. As such, M2PA MUST involve the security needs of several
parties: 

   - the end users of the services

   - the network providers

   - the applications involved

Additional requirements MAY come from local regulation.  

While these parties may have some overlapping security needs, their
needs may not be identical. Any security solution SHOULD fulfill all
of the different parties' needs.

6.1 Security Requirements

Consult [Security] for a discussion of security requirements and for
guidance on the use of security protocols.

When M2PA is running in professionally managed corporate or service
provider network, it is reasonable to expect that this network
includes an appropriate security policy framework. The "Site Security
Handbook" [RFC2196] SHOULD be consulted for guidance.

6.2 Protecting Confidentiality

Particularly for wireless users, the requirement for confidentiality
MAY include the masking of IP addresses and ports. In this case
application-level encryption is not sufficient. IPSec ESP SHOULD be
used instead [RFC2401].  Regardless of which level performs the
encryption, the IPSec ISAKMP service SHOULD be used for key
management.


{Add a normative reference:}

  [Security]
       J. Loughney, M. Tuexen, and J. Pastor-Balbas, "Security
       Considerations for SIGTRAN Protocols,"
       draft-ietf-sigtran-security-01.txt (January 2003).

==========================================


-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 17 18:20:07 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA03772
	for <sigtran-archive@odin.ietf.org>; Mon, 17 Feb 2003 18:20:07 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1HNPQ031804
	for sigtran-archive@odin.ietf.org; Mon, 17 Feb 2003 18:25:26 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1HNPAp31793;
	Mon, 17 Feb 2003 18:25:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1HNOWp31725
	for <sigtran@optimus.ietf.org>; Mon, 17 Feb 2003 18:24:32 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA03754
	for <sigtran@ietf.org>; Mon, 17 Feb 2003 18:18:43 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1HNMTqe027792;
	Mon, 17 Feb 2003 17:22:30 -0600 (CST)
Message-ID: <3E516EB5.FA8A033E@alcatel.com>
Date: Mon, 17 Feb 2003 17:22:29 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft needs to note that timer T1 differs from Q.703 
 timer T1
References: <OF1B9BFAE2.55D43239-ON85256CCC.0050C2A3@nc.tekelec.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Jeff,

M2PA timer T1 seems to follow the description of the Q.703 timer T1 at
the end of Q.703 section 7.3 "Initial Alignment Procedure". I used the
Q.703 range and recommended value of T1. 

I don't see what you are pointing to in Fig 8 sheet 5. T1 is started as
LSC goes to Aligned/ready. Please explain more.

Thanks,
tomG.

jeffrey.craig@tekelec.com wrote:
> 
> Hello All,
> 
> The M2PA draft should be revised to note that its implementation
> of timer T1 does not conform to timer T1 as described by Q.703
> (SDL, Figure 8, sheet 5). In Q.703 this timer is started upon entry
> to the Initial Alignment state and has a duration that accounts for
> up to four normal alignment sequences.
> 
> M2PA timer T1 is started upon entry to either ALIGNED_READY
> or ALIGNED_NOT_READY states and has a recommended
> duration much shorter than that of four normal alignment sequences.
> I have no problem with this treatment, but it is an exception from
> the general conformance to Q.703.
> 
> Regards,
> 
> Jeff

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 18 00:36:39 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA09647
	for <sigtran-archive@odin.ietf.org>; Tue, 18 Feb 2003 00:36:39 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1I5g5219743
	for sigtran-archive@odin.ietf.org; Tue, 18 Feb 2003 00:42:05 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1I5fap19717;
	Tue, 18 Feb 2003 00:41:37 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1I5d6p19633
	for <sigtran@optimus.ietf.org>; Tue, 18 Feb 2003 00:39:06 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA09601
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 00:33:08 -0500 (EST)
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OFFC2918A0.D2C5B93D-ONC2256CD1.001D9A4C@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Tue, 18 Feb 2003 07:33:27 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/18/2003
 07:33:30 AM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,
--- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
> Javier,
>
> Same thing.
>
> What did you think one would use to determine the
> "traffic" that "n" ASPs
> are required to handle, other than their engineered
> maximum traffic
> handling capability?
[TOLGA]IMO, the sentence Javier is referring to is more likeliy to be
interpreted as n mandatory, i.e. as a value, which is decided by just
configuration rather than configuration+traffic load at a given time. If
there is consensus that using n as mandatory is not usefull, I would think
this sentence should be modified to make this clear.
Furthermore for your opinion that no more clarifications are necessary in
RFC to make n+k work I disagree, because for me from the clearness of RFC
point of view important is what it stands in RFC/IG, not the
interpretations of any individual, including myself. So, when we reach a
consensus how it is supposed to work (I think there we don't have big
problems) this better should be documented in IG.
>
> --brian
>
> Javier Pastor-Balbas (ECE) wrote:
>                                (Mon, 17 Feb 2003
> 16:02:02)
> >
> > Hi Brian,
> >
> > Why do you say that: "'n' has always meant the
> number required for engineered maximum load" when
> the RFC says:
> >
> >    The failover model supports an "n+k" redundancy
> model, where "n" ASPs
> >    is the minimum number of redundant ASPs
> required to handle traffic
> >
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> >    and "k" ASPs are available to take over for a
> failed or unavailable
> >    ASP.
> >
> > (??)
> >
> >
> > Cheers // Javier.

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 18 01:53:10 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA10468
	for <sigtran-archive@odin.ietf.org>; Tue, 18 Feb 2003 01:53:10 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1I6wc123288
	for sigtran-archive@odin.ietf.org; Tue, 18 Feb 2003 01:58:38 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1I6wOp23276;
	Tue, 18 Feb 2003 01:58:25 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1I6vRp23239
	for <sigtran@optimus.ietf.org>; Tue, 18 Feb 2003 01:57:27 -0500
Received: from penguin.wise.edt.ericsson.se (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA10447
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 01:51:27 -0500 (EST)
Received: from esealnt611.al.sw.ericsson.se (esealnt611.al.sw.ericsson.se [153.88.254.68])
	by penguin.wise.edt.ericsson.se (8.12.1/8.12.1/WIREfire-1.4) with ESMTP id h1I6tFAv015006
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 07:55:15 +0100 (MET)
Received: from ESEALNT744.al.sw.ericsson.se ([153.88.251.4]) by esealnt611.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55)
	id FDV2TWPK; Tue, 18 Feb 2003 07:55:15 +0100
Received: by ESEALNT744.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <1ZX15LZB>; Tue, 18 Feb 2003 07:55:15 +0100
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65970@EESTQNT105.es.eu.ericsson.se>
X-Sybari-Trust: c0c79288 9ffcebbb a5ee123c 00000138
From: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>
To: sigtran@ietf.org
Subject: RE: [Sigtran] Re: M3UA Notify message
Date: Tue, 18 Feb 2003 07:57:23 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hi Brian, 

According to the RFC "n" is required to handle traffic with minimum losses, what is expected from the SS7 point of view in an interworking scenario. From that point of view, you don't start traffic till you have the enough capacity to handle it, i.e. "n" ASPs. 

I completely agree with Tolga.

Cheers // Javier.

> -----Original Message-----
> From: TOLGA ASVEREN [mailto:TOLGA.ASVEREN@o2.com.tr]
> Sent: Tuesday, February 18, 2003 6:33 AM
> To: sigtran@ietf.org
> Subject: Re: [Sigtran] Re: M3UA Notify message
> 
> 
> 
> Brian,
> --- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
> > Javier,
> >
> > Same thing.
> >
> > What did you think one would use to determine the
> > "traffic" that "n" ASPs
> > are required to handle, other than their engineered
> > maximum traffic
> > handling capability?
> [TOLGA]IMO, the sentence Javier is referring to is more likeliy to be
> interpreted as n mandatory, i.e. as a value, which is decided by just
> configuration rather than configuration+traffic load at a 
> given time. If
> there is consensus that using n as mandatory is not usefull, 
> I would think
> this sentence should be modified to make this clear.
> Furthermore for your opinion that no more clarifications are 
> necessary in
> RFC to make n+k work I disagree, because for me from the 
> clearness of RFC
> point of view important is what it stands in RFC/IG, not the
> interpretations of any individual, including myself. So, when 
> we reach a
> consensus how it is supposed to work (I think there we don't have big
> problems) this better should be documented in IG.
> >
> > --brian
> >
> > Javier Pastor-Balbas (ECE) wrote:
> >                                (Mon, 17 Feb 2003
> > 16:02:02)
> > >
> > > Hi Brian,
> > >
> > > Why do you say that: "'n' has always meant the
> > number required for engineered maximum load" when
> > the RFC says:
> > >
> > >    The failover model supports an "n+k" redundancy
> > model, where "n" ASPs
> > >    is the minimum number of redundant ASPs
> > required to handle traffic
> > >
> > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> > >    and "k" ASPs are available to take over for a
> > failed or unavailable
> > >    ASP.
> > >
> > > (??)
> > >
> > >
> > > Cheers // Javier.
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 18 02:27:52 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA20931
	for <sigtran-archive@odin.ietf.org>; Tue, 18 Feb 2003 02:27:52 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1I7XLx03457
	for sigtran-archive@odin.ietf.org; Tue, 18 Feb 2003 02:33:21 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1I7X8p03449;
	Tue, 18 Feb 2003 02:33:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1I7Wfp03418
	for <sigtran@optimus.ietf.org>; Tue, 18 Feb 2003 02:32:41 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA20918
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 02:26:41 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:kiuZB8H4pLAglaqrSp8h+Z257WqH5VSp@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1I7USD24187;
	Tue, 18 Feb 2003 00:30:29 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1I7USY31203;
	Tue, 18 Feb 2003 00:30:28 -0700
Date: Tue, 18 Feb 2003 00:30:28 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030218003028.A27117@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OFFC2918A0.D2C5B93D-ONC2256CD1.001D9A4C@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OFFC2918A0.D2C5B93D-ONC2256CD1.001D9A4C@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Tue, Feb 18, 2003 at 07:33:27AM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

TOLGA,

TOLGA ASVEREN wrote:                                                                 (Tue, 18 Feb 2003 07:33:27)
> [TOLGA]IMO, the sentence Javier is referring to is more likeliy to be
> interpreted as n mandatory, i.e. as a value, which is decided by just
> configuration rather than configuration+traffic load at a given time.

Then why does is say "traffic"?

> If
> there is consensus that using n as mandatory is not usefull, I would think
> this sentence should be modified to make this clear.

It says "traffic" which is clear.

> Furthermore for your opinion that no more clarifications are necessary in
> RFC to make n+k work I disagree, because for me from the clearness of RFC
> point of view important is what it stands in RFC/IG, not the
> interpretations of any individual, including myself. So, when we reach a
> consensus how it is supposed to work (I think there we don't have big
> problems) this better should be documented in IG.

Please put forth a proposal, but strike section 3.15.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 18 02:29:45 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA20965
	for <sigtran-archive@odin.ietf.org>; Tue, 18 Feb 2003 02:29:45 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1I7ZEj03570
	for sigtran-archive@odin.ietf.org; Tue, 18 Feb 2003 02:35:14 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1I7Z3p03548;
	Tue, 18 Feb 2003 02:35:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1I7YDp03511
	for <sigtran@optimus.ietf.org>; Tue, 18 Feb 2003 02:34:13 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA20946
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 02:28:13 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:3VZ//SQe6PQlokwhR6sxRjhLivCniv7d@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1I7W1D24208;
	Tue, 18 Feb 2003 00:32:01 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1I7W1m31232;
	Tue, 18 Feb 2003 00:32:01 -0700
Date: Tue, 18 Feb 2003 00:32:01 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030218003201.B27117@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>,
	sigtran@ietf.org
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65970@EESTQNT105.es.eu.ericsson.se>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <1AB3D30B989BF141BBD5C70057B2EF7C01B65970@EESTQNT105.es.eu.ericsson.se>; from javier.pastor-balbas@ece.ericsson.se on Tue, Feb 18, 2003 at 07:57:23AM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Javier,

Javier Pastor-Balbas (ECE) wrote:                                                   (Tue, 18 Feb 2003 07:57:23)
> 
> Hi Brian, 
> 
> According to the RFC "n" is required to handle traffic with minimum losses,
> what is expected from the SS7 point of view in an interworking scenario.

That would be the numbe required to handle maximum engineered load, which is
the same thing.

> From that point of view, you don't start traffic till you have the enough
> capacity to handle it, i.e. "n" ASPs. 

The SG is permitted to do this by the RFC.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 18 05:48:29 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA24847
	for <sigtran-archive@odin.ietf.org>; Tue, 18 Feb 2003 05:48:29 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1IAs1m17076
	for sigtran-archive@odin.ietf.org; Tue, 18 Feb 2003 05:54:01 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1IArkp17047;
	Tue, 18 Feb 2003 05:53:47 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1IAqwp17021
	for <sigtran@optimus.ietf.org>; Tue, 18 Feb 2003 05:52:58 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA24837
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 05:46:54 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1IAocqe017337;
	Tue, 18 Feb 2003 04:50:39 -0600 (CST)
Received: from axsppc39 by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id PAA05545; Tue, 18 Feb 2003 15:19:33 -0500 (GMT)
Message-ID: <00ec01c2d73c$aea73b00$5901ff0a@axsppc39>
Reply-To: "bijoy" <bijoy@axes-mach01.usa.alcatel.com>
From: "bijoy" <bijoy@axes-mach01.usa.alcatel.com>
To: <sigtran@ietf.org>, "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
References: <OFFC2918A0.D2C5B93D-ONC2256CD1.001D9A4C@LocalDomain>
Subject: Re: [Sigtran] Re: M3UA Notify message
Date: Tue, 18 Feb 2003 16:28:14 +0530
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----=_NextPart_000_00E7_01C2D76A.BC7A1DA0"
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 5.50.4133.2400
X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

This is a multi-part message in MIME format.

------=_NextPart_000_00E7_01C2D76A.BC7A1DA0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

=20
Hi Tolga,

RFC at 1.4.4.1 says

The failover model supports an "n+k" redundancy model, where "n" ASPs

is the minimum number of redundant ASPs required to handle traffic

and "k" ASPs are available to take over for a failed or unavailable

ASP.=20

     I agree to what brein says because k ASP's are only used to take =
over

when one or more configured ASP's(=3Dn) are unavailable . So the

traffic mentioned in the RFC 1.4.4.1 actually has to be the maximum =
configured

traffic.



thanks

bijoy/sibi

  ----- Original Message -----=20
  From: TOLGA ASVEREN=20
  To: sigtran@ietf.org=20
  Sent: Tuesday, February 18, 2003 11:03 AM
  Subject: Re: [Sigtran] Re: M3UA Notify message



  Brian,
  --- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
  > Javier,
  >
  > Same thing.
  >
  > What did you think one would use to determine the
  > "traffic" that "n" ASPs
  > are required to handle, other than their engineered
  > maximum traffic
  > handling capability?
  [TOLGA]IMO, the sentence Javier is referring to is more likeliy to be
  interpreted as n mandatory, i.e. as a value, which is decided by just
  configuration rather than configuration+traffic load at a given time. =
If
  there is consensus that using n as mandatory is not usefull, I would =
think
  this sentence should be modified to make this clear.
  Furthermore for your opinion that no more clarifications are necessary =
in
  RFC to make n+k work I disagree, because for me from the clearness of =
RFC
  point of view important is what it stands in RFC/IG, not the
  interpretations of any individual, including myself. So, when we reach =
a
  consensus how it is supposed to work (I think there we don't have big
  problems) this better should be documented in IG.
  >
  > --brian
  >
  > Javier Pastor-Balbas (ECE) wrote:
  >                                (Mon, 17 Feb 2003
  > 16:02:02)
  > >
  > > Hi Brian,
  > >
  > > Why do you say that: "'n' has always meant the
  > number required for engineered maximum load" when
  > the RFC says:
  > >
  > >    The failover model supports an "n+k" redundancy
  > model, where "n" ASPs
  > >    is the minimum number of redundant ASPs
  > required to handle traffic
  > >
  > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  > >    and "k" ASPs are available to take over for a
  > failed or unavailable
  > >    ASP.
  > >
  > > (??)
  > >
  > >
  > > Cheers // Javier.

  _______________________________________________
  Sigtran mailing list
  Sigtran@ietf.org
  https://www1.ietf.org/mailman/listinfo/sigtran


------=_NextPart_000_00E7_01C2D76A.BC7A1DA0
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content=3D"text/html; charset=3Diso-8859-1" =
http-equiv=3DContent-Type>
<META content=3D"MSHTML 5.00.3103.1000" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3D"Bookman Old Style"><FONT size=3D2>&nbsp;
<P>Hi Tolga,</P>
<P>RFC at 1.4.4.1 says</P>
<P>The failover model supports an "n+k" redundancy model, where "n" =
ASPs</P>
<P>is the minimum number of redundant ASPs required to handle =
traffic</P>
<P>and "<STRONG>k" ASPs are available to take over for a failed or=20
unavailable</STRONG></P>
<P><STRONG>ASP. </STRONG></P>
<P>&nbsp;&nbsp;&nbsp;&nbsp; I agree to what brein says because k ASP's =
are only=20
used to take over</P>
<P>when&nbsp;one or more configured ASP's(=3Dn) are unavailable . So =
the</P>
<P>traffic mentioned in the RFC 1.4.4.1 actually has to be the maximum=20
configured</P>
<P>traffic.</P>
<P>&nbsp;</P>
<P>thanks</P>
<P>bijoy/sibi</P></FONT></FONT></DIV>
<BLOCKQUOTE=20
style=3D"BORDER-LEFT: #000000 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: =
0px; PADDING-LEFT: 5px; PADDING-RIGHT: 0px">
  <DIV style=3D"FONT: 10pt arial">----- Original Message ----- </DIV>
  <DIV=20
  style=3D"BACKGROUND: #e4e4e4; FONT: 10pt arial; font-color: =
black"><B>From:</B>=20
  <A href=3D"mailto:TOLGA.ASVEREN@o2.com.tr" =
title=3DTOLGA.ASVEREN@o2.com.tr>TOLGA=20
  ASVEREN</A> </DIV>
  <DIV style=3D"FONT: 10pt arial"><B>To:</B> <A =
href=3D"mailto:sigtran@ietf.org"=20
  title=3Dsigtran@ietf.org>sigtran@ietf.org</A> </DIV>
  <DIV style=3D"FONT: 10pt arial"><B>Sent:</B> Tuesday, February 18, =
2003 11:03=20
  AM</DIV>
  <DIV style=3D"FONT: 10pt arial"><B>Subject:</B> Re: [Sigtran] Re: M3UA =
Notify=20
  message</DIV>
  <DIV><BR></DIV><BR>Brian,<BR>--- "Brian F. G. Bidulock" &lt;<A=20
  href=3D"mailto:bidulock@openss7.org">bidulock@openss7.org</A>&gt; =
wrote:<BR>&gt;=20
  Javier,<BR>&gt;<BR>&gt; Same thing.<BR>&gt;<BR>&gt; What did you think =
one=20
  would use to determine the<BR>&gt; "traffic" that "n" ASPs<BR>&gt; are =

  required to handle, other than their engineered<BR>&gt; maximum=20
  traffic<BR>&gt; handling capability?<BR>[TOLGA]IMO, the sentence =
Javier is=20
  referring to is more likeliy to be<BR>interpreted as n mandatory, i.e. =
as a=20
  value, which is decided by just<BR>configuration rather than=20
  configuration+traffic load at a given time. If<BR>there is consensus =
that=20
  using n as mandatory is not usefull, I would think<BR>this sentence =
should be=20
  modified to make this clear.<BR>Furthermore for your opinion that no =
more=20
  clarifications are necessary in<BR>RFC to make n+k work I disagree, =
because=20
  for me from the clearness of RFC<BR>point of view important is what it =
stands=20
  in RFC/IG, not the<BR>interpretations of any individual, including =
myself. So,=20
  when we reach a<BR>consensus how it is supposed to work (I think there =
we=20
  don't have big<BR>problems) this better should be documented in=20
  IG.<BR>&gt;<BR>&gt; --brian<BR>&gt;<BR>&gt; Javier Pastor-Balbas (ECE) =

  =
wrote:<BR>&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=20
  (Mon, 17 Feb 2003<BR>&gt; 16:02:02)<BR>&gt; &gt;<BR>&gt; &gt; Hi=20
  Brian,<BR>&gt; &gt;<BR>&gt; &gt; Why do you say that: "'n' has always =
meant=20
  the<BR>&gt; number required for engineered maximum load" when<BR>&gt; =
the RFC=20
  says:<BR>&gt; &gt;<BR>&gt; &gt;&nbsp;&nbsp;&nbsp; The failover model =
supports=20
  an "n+k" redundancy<BR>&gt; model, where "n" ASPs<BR>&gt;=20
  &gt;&nbsp;&nbsp;&nbsp; is the minimum number of redundant ASPs<BR>&gt; =

  required to handle traffic<BR>&gt; &gt;<BR>&gt;=20
  =
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^<BR>&gt=
;=20
  &gt;&nbsp;&nbsp;&nbsp; and "k" ASPs are available to take over for =
a<BR>&gt;=20
  failed or unavailable<BR>&gt; &gt;&nbsp;&nbsp;&nbsp; ASP.<BR>&gt; =
&gt;<BR>&gt;=20
  &gt; (??)<BR>&gt; &gt;<BR>&gt; &gt;<BR>&gt; &gt; Cheers //=20
  =
Javier.<BR><BR>_______________________________________________<BR>Sigtran=
=20
  mailing list<BR><A =
href=3D"mailto:Sigtran@ietf.org">Sigtran@ietf.org</A><BR><A=20
  =
href=3D"https://www1.ietf.org/mailman/listinfo/sigtran">https://www1.ietf=
.org/mailman/listinfo/sigtran</A><BR></BLOCKQUOTE></BODY></HTML>

------=_NextPart_000_00E7_01C2D76A.BC7A1DA0--

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 18 06:16:24 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA25308
	for <sigtran-archive@odin.ietf.org>; Tue, 18 Feb 2003 06:16:24 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1IBLvZ18751
	for sigtran-archive@odin.ietf.org; Tue, 18 Feb 2003 06:21:57 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1IBLUp18719;
	Tue, 18 Feb 2003 06:21:30 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1IBKHp18672
	for <sigtran@optimus.ietf.org>; Tue, 18 Feb 2003 06:20:17 -0500
Received: from albatross.wise.edt.ericsson.se (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA25278
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 06:14:13 -0500 (EST)
Received: from esealnt612.al.sw.ericsson.se (esealnt612.al.sw.ericsson.se [153.88.254.71])
	by albatross.wise.edt.ericsson.se (8.12.1/8.12.1/WIREfire-1.4) with ESMTP id h1IBI0KV025740
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 12:18:00 +0100 (MET)
Received: from ESEALNT744.al.sw.ericsson.se ([153.88.251.4]) by esealnt612.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55)
	id FDVGWTBN; Tue, 18 Feb 2003 12:17:56 +0100
Received: by ESEALNT744.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <1ZX15QC8>; Tue, 18 Feb 2003 12:17:56 +0100
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B6597B@EESTQNT105.es.eu.ericsson.se>
X-Sybari-Trust: 479d82b0 9ffcebbb 7a95d2f4 00000138
From: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>
To: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Date: Tue, 18 Feb 2003 12:17:54 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian, 


> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Tuesday, February 18, 2003 8:32 AM
> To: Javier Pastor-Balbas (ECE)
> Cc: sigtran@ietf.org
> Subject: Re: [Sigtran] Re: M3UA Notify message
> 
> 
> Javier,
> 
> Javier Pastor-Balbas (ECE) wrote:                             
>                       (Tue, 18 Feb 2003 07:57:23)
> > 
> > Hi Brian, 
> > 
> > According to the RFC "n" is required to handle traffic with 
> minimum losses,
> > what is expected from the SS7 point of view in an 
> interworking scenario.
> 
> That would be the numbe required to handle maximum engineered 
> load, which is
> the same thing.
> 

If you refer to the compromise load engineered to be available 99% (aprox.) of the time, then I agree. But this is a statically value fed at configuration time. And the system won't work till n number is reach since the system will likely have quality contrains.


Cheers // Javier. 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 18 06:47:03 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA25667
	for <sigtran-archive@odin.ietf.org>; Tue, 18 Feb 2003 06:47:02 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1IBqah20382
	for sigtran-archive@odin.ietf.org; Tue, 18 Feb 2003 06:52:36 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1IBqLp20374;
	Tue, 18 Feb 2003 06:52:21 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1IBp8p20330
	for <sigtran@optimus.ietf.org>; Tue, 18 Feb 2003 06:51:08 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA25646
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 06:45:03 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:kmA6LAcf6iZVaSYMDlM/zQtIMxcsCEC5@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1IBmpD27965;
	Tue, 18 Feb 2003 04:48:51 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1IBmog04054;
	Tue, 18 Feb 2003 04:48:50 -0700
Date: Tue, 18 Feb 2003 04:48:50 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030218044850.A3693@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>,
	sigtran@ietf.org
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B6597B@EESTQNT105.es.eu.ericsson.se>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <1AB3D30B989BF141BBD5C70057B2EF7C01B6597B@EESTQNT105.es.eu.ericsson.se>; from javier.pastor-balbas@ece.ericsson.se on Tue, Feb 18, 2003 at 12:17:54PM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Javier,

Javier Pastor-Balbas (ECE) wrote:             (Tue, 18 Feb 2003 12:17:54)
> > Javier Pastor-Balbas (ECE) wrote:             (Tue, 18 Feb 2003 07:57:23)
> > > 
> > > Hi Brian, 
> > > 
> > > According to the RFC "n" is required to handle traffic with minimum
> > > losses, what is expected from the SS7 point of view in an interworking
> > > scenario.

Please cite this passage, because I cannot find it.  I find no mention of
minimum number of ASPs and "message loss".  The only reference to message loss
that I know of is unrelated.

> > 
> > That would be the numbe required to handle maximum engineered load, which
> > is the same thing.
> > 
> 
> If you refer to the compromise load engineered to be available 99% (aprox.)
> of the time, then I agree. But this is a statically value fed at
> configuration time. And the system won't work till n number is reach since
> the system will likely have quality contrains.

The maximum engineered load is a figure established at configuration time.

Carrier-grade systems must always work with 1 processor.  The system will just
not necessarily be able to handle the offered load.  (But it might:
particularly during quiet periods, off-peak, normal-day, etc.)

The offered load cannot be established at configuration time and varies widely
for telephony systems.

You must also realized that the system might, at any time,  be offered a load
greater than the maximum engineered load (mass calling, busy periods, etc.):
so for carrier-grade systems even "n" processors must implement overload
controls (and they could not claim SS7 compliance if they didn't).

Look back in this thread and you will see the factors contributing to the
anticipated offered load for the start, restart and stop decision.

The decisions are not based on less than "n" processors not allowing the
system to work (where n > 1).  They are based on the negative effects of
delivering pent up traffic to processors that will likely just congest.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 18 07:31:05 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA26498
	for <sigtran-archive@odin.ietf.org>; Tue, 18 Feb 2003 07:31:05 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1ICaem22683
	for sigtran-archive@odin.ietf.org; Tue, 18 Feb 2003 07:36:40 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1ICaQp22675;
	Tue, 18 Feb 2003 07:36:27 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1ICZ7p22630
	for <sigtran@optimus.ietf.org>; Tue, 18 Feb 2003 07:35:07 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA26450
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 07:29:00 -0500 (EST)
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OFAD225262.DB03AD02-ONC2256CD1.004193B7@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Tue, 18 Feb 2003 14:29:16 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/18/2003
 02:29:23 PM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hi Bijoy,

IMO the sentence you have quoted from the RFC is not inline with what Brian
says (I think it refers to n as mandatory). But if the consensus is that n
> 1 as mandatory is not usefull, then I would think this sentence might be
modified.

     Tolga




Hi Tolga,


RFC at 1.4.4.1 says


The failover model supports an "n+k" redundancy model, where "n" ASPs


is the minimum number of redundant ASPs required to handle traffic


and "k" ASPs are available to take over for a failed or unavailable


ASP.


     I agree to what brein says because k ASP's are only used to take over


when one or more configured ASP's(=n) are unavailable . So the


traffic mentioned in the RFC 1.4.4.1 actually has to be the maximum
configured


traffic.





thanks


bijoy/sibi


 ----- Original Message -----
 From: TOLGA ASVEREN
 To: sigtran@ietf.org
 Sent: Tuesday, February 18, 2003 11:03 AM
 Subject: Re: [Sigtran] Re: M3UA Notify message


 Brian,
 --- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
 > Javier,
 >
 > Same thing.
 >
 > What did you think one would use to determine the
 > "traffic" that "n" ASPs
 > are required to handle, other than their engineered
 > maximum traffic
 > handling capability?
 [TOLGA]IMO, the sentence Javier is referring to is more likeliy to be
 interpreted as n mandatory, i.e. as a value, which is decided by just
 configuration rather than configuration+traffic load at a given time. If
 there is consensus that using n as mandatory is not usefull, I would think
 this sentence should be modified to make this clear.
 Furthermore for your opinion that no more clarifications are necessary in
 RFC to make n+k work I disagree, because for me from the clearness of RFC
 point of view important is what it stands in RFC/IG, not the
 interpretations of any individual, including myself. So, when we reach a
 consensus how it is supposed to work (I think there we don't have big
 problems) this better should be documented in IG.
 >
 > --brian
 >
 > Javier Pastor-Balbas (ECE) wrote:
 >                                (Mon, 17 Feb 2003
 > 16:02:02)
 > >
 > > Hi Brian,
 > >
 > > Why do you say that: "'n' has always meant the
 > number required for engineered maximum load" when
 > the RFC says:
 > >
 > >    The failover model supports an "n+k" redundancy
 > model, where "n" ASPs
 > >    is the minimum number of redundant ASPs
 > required to handle traffic
 > >
 > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 > >    and "k" ASPs are available to take over for a
 > failed or unavailable
 > >    ASP.
 > >
 > > (??)
 > >
 > >
 > > Cheers // Javier.

 _______________________________________________
 Sigtran mailing list
 Sigtran@ietf.org
 https://www1.ietf.org/mailman/listinfo/sigtran



_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 18 07:58:47 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA27085
	for <sigtran-archive@odin.ietf.org>; Tue, 18 Feb 2003 07:58:47 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1ID4MO24603
	for sigtran-archive@odin.ietf.org; Tue, 18 Feb 2003 08:04:22 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1ID47p24595;
	Tue, 18 Feb 2003 08:04:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1ID35p24548
	for <sigtran@optimus.ietf.org>; Tue, 18 Feb 2003 08:03:05 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA27060
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 07:56:58 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:iBOFsqocNduzmrpmZKPSs3Ltu2iIpFtu@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1ID0kD28932;
	Tue, 18 Feb 2003 06:00:46 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1ID0kC05764;
	Tue, 18 Feb 2003 06:00:46 -0700
Date: Tue, 18 Feb 2003 06:00:45 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030218060045.A5588@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OFAD225262.DB03AD02-ONC2256CD1.004193B7@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OFAD225262.DB03AD02-ONC2256CD1.004193B7@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Tue, Feb 18, 2003 at 02:29:16PM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

TOLGA,

I think Bijoy's point was that as "k" are not active when "n" are,
"n" must be the number required to handle the maximum engineered load.
Therefore "traffic == maximum engineered load" in section 1.4.4.1.

The sentence does not really need modification.

--brian

TOLGA ASVEREN wrote:                                                                 (Tue, 18 Feb 2003 14:29:16)
> 
> Hi Bijoy,
> 
> IMO the sentence you have quoted from the RFC is not inline with what Brian
> says (I think it refers to n as mandatory). But if the consensus is that n
> > 1 as mandatory is not usefull, then I would think this sentence might be
> modified.
> 
>      Tolga
> 
> 
> 
> 
> Hi Tolga,
> 
> 
> RFC at 1.4.4.1 says
> 
> 
> The failover model supports an "n+k" redundancy model, where "n" ASPs
> 
> 
> is the minimum number of redundant ASPs required to handle traffic
> 
> 
> and "k" ASPs are available to take over for a failed or unavailable
> 
> 
> ASP.
> 
> 
>      I agree to what brein says because k ASP's are only used to take over
> 
> 
> when one or more configured ASP's(=n) are unavailable . So the
> 
> 
> traffic mentioned in the RFC 1.4.4.1 actually has to be the maximum
> configured
> 
> 
> traffic.
> 
> 
> 
> 
> 
> thanks
> 
> 
> bijoy/sibi
> 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 18 08:43:15 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA28076
	for <sigtran-archive@odin.ietf.org>; Tue, 18 Feb 2003 08:43:15 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1IDmpG28340
	for sigtran-archive@odin.ietf.org; Tue, 18 Feb 2003 08:48:51 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1IDmZp28329;
	Tue, 18 Feb 2003 08:48:36 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1IDlhp28291
	for <sigtran@optimus.ietf.org>; Tue, 18 Feb 2003 08:47:43 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA28062
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 08:41:35 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1IDjKqe003894;
	Tue, 18 Feb 2003 07:45:21 -0600 (CST)
Received: from axsppc39 by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id SAA08997; Tue, 18 Feb 2003 18:14:11 -0500 (GMT)
Message-ID: <034901c2d755$130609b0$5901ff0a@axsppc39>
Reply-To: "bijoy" <bijoy@axes-mach01.usa.alcatel.com>
From: "bijoy" <bijoy@axes-mach01.usa.alcatel.com>
To: <bidulock@openss7.org>, "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Cc: <sigtran@ietf.org>
References: <OFAD225262.DB03AD02-ONC2256CD1.004193B7@LocalDomain> <20030218060045.A5588@openss7.org>
Subject: Re: [Sigtran] Re: M3UA Notify message
Date: Tue, 18 Feb 2003 19:22:19 +0530
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----=_NextPart_000_0340_01C2D783.0E29E730"
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 5.50.4133.2400
X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

This is a multi-part message in MIME format.

------=_NextPart_000_0340_01C2D783.0E29E730
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

hi brian,
  that is what i meant....
  ----- Original Message -----=20
  From: Brian F. G. Bidulock=20
  To: TOLGA ASVEREN=20
  Cc: sigtran@ietf.org=20
  Sent: Tuesday, February 18, 2003 6:30 PM
  Subject: Re: [Sigtran] Re: M3UA Notify message


  TOLGA,

  I think Bijoy's point was that as "k" are not active when "n" are,
  "n" must be the number required to handle the maximum engineered load.
  Therefore "traffic =3D=3D maximum engineered load" in section 1.4.4.1.

  The sentence does not really need modification.

  --brian

  TOLGA ASVEREN wrote:                                                   =
              (Tue, 18 Feb 2003 14:29:16)
  >=20
  > Hi Bijoy,
  >=20
  > IMO the sentence you have quoted from the RFC is not inline with =
what Brian
  > says (I think it refers to n as mandatory). But if the consensus is =
that n
  > > 1 as mandatory is not usefull, then I would think this sentence =
might be
  > modified.
  >=20
  >      Tolga
  >=20
  >=20
  >=20
  >=20
  > Hi Tolga,
  >=20
  >=20
  > RFC at 1.4.4.1 says
  >=20
  >=20
  > The failover model supports an "n+k" redundancy model, where "n" =
ASPs
  >=20
  >=20
  > is the minimum number of redundant ASPs required to handle traffic
  >=20
  >=20
  > and "k" ASPs are available to take over for a failed or unavailable
  >=20
  >=20
  > ASP.
  >=20
  >=20
  >      I agree to what brein says because k ASP's are only used to =
take over
  >=20
  >=20
  > when one or more configured ASP's(=3Dn) are unavailable . So the
  >=20
  >=20
  > traffic mentioned in the RFC 1.4.4.1 actually has to be the maximum
  > configured
  >=20
  >=20
  > traffic.
  >=20
  >=20
  >=20
  >=20
  >=20
  > thanks
  >=20
  >=20
  > bijoy/sibi
  >=20

  --=20
  Brian F. G. Bidulock
  bidulock@openss7.org
  http://www.openss7.org/
  _______________________________________________
  Sigtran mailing list
  Sigtran@ietf.org
  https://www1.ietf.org/mailman/listinfo/sigtran


------=_NextPart_000_0340_01C2D783.0E29E730
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content=3D"text/html; charset=3Diso-8859-1" =
http-equiv=3DContent-Type>
<META content=3D"MSHTML 5.00.3103.1000" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3D"Bookman Old Style">hi brian,</FONT></DIV>
<DIV><FONT face=3D"Bookman Old Style">&nbsp; that is what i =
meant....</FONT></DIV>
<BLOCKQUOTE=20
style=3D"BORDER-LEFT: #000000 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: =
0px; PADDING-LEFT: 5px; PADDING-RIGHT: 0px">
  <DIV style=3D"FONT: 10pt arial">----- Original Message ----- </DIV>
  <DIV=20
  style=3D"BACKGROUND: #e4e4e4; FONT: 10pt arial; font-color: =
black"><B>From:</B>=20
  <A href=3D"mailto:bidulock@openss7.org" =
title=3Dbidulock@openss7.org>Brian F. G.=20
  Bidulock</A> </DIV>
  <DIV style=3D"FONT: 10pt arial"><B>To:</B> <A=20
  href=3D"mailto:TOLGA.ASVEREN@o2.com.tr" =
title=3DTOLGA.ASVEREN@o2.com.tr>TOLGA=20
  ASVEREN</A> </DIV>
  <DIV style=3D"FONT: 10pt arial"><B>Cc:</B> <A =
href=3D"mailto:sigtran@ietf.org"=20
  title=3Dsigtran@ietf.org>sigtran@ietf.org</A> </DIV>
  <DIV style=3D"FONT: 10pt arial"><B>Sent:</B> Tuesday, February 18, =
2003 6:30=20
  PM</DIV>
  <DIV style=3D"FONT: 10pt arial"><B>Subject:</B> Re: [Sigtran] Re: M3UA =
Notify=20
  message</DIV>
  <DIV><BR></DIV>TOLGA,<BR><BR>I think Bijoy's point was that as "k" are =
not=20
  active when "n" are,<BR>"n" must be the number required to handle the =
maximum=20
  engineered load.<BR>Therefore "traffic =3D=3D maximum engineered load" =
in section=20
  1.4.4.1.<BR><BR>The sentence does not really need=20
  modification.<BR><BR>--brian<BR><BR>TOLGA ASVEREN=20
  =
wrote:&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=
bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;&nbsp;=20
  (Tue, 18 Feb 2003 14:29:16)<BR>&gt; <BR>&gt; Hi Bijoy,<BR>&gt; =
<BR>&gt; IMO=20
  the sentence you have quoted from the RFC is not inline with what=20
  Brian<BR>&gt; says (I think it refers to n as mandatory). But if the =
consensus=20
  is that n<BR>&gt; &gt; 1 as mandatory is not usefull, then I would =
think this=20
  sentence might be<BR>&gt; modified.<BR>&gt;=20
  <BR>&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Tolga<BR>&gt; <BR>&gt; <BR>&gt; =

  <BR>&gt; <BR>&gt; Hi Tolga,<BR>&gt; <BR>&gt; <BR>&gt; RFC at 1.4.4.1=20
  says<BR>&gt; <BR>&gt; <BR>&gt; The failover model supports an "n+k" =
redundancy=20
  model, where "n" ASPs<BR>&gt; <BR>&gt; <BR>&gt; is the minimum number =
of=20
  redundant ASPs required to handle traffic<BR>&gt; <BR>&gt; <BR>&gt; =
and "k"=20
  ASPs are available to take over for a failed or unavailable<BR>&gt; =
<BR>&gt;=20
  <BR>&gt; ASP.<BR>&gt; <BR>&gt; <BR>&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =
I agree=20
  to what brein says because k ASP's are only used to take over<BR>&gt; =
<BR>&gt;=20
  <BR>&gt; when one or more configured ASP's(=3Dn) are unavailable . So=20
  the<BR>&gt; <BR>&gt; <BR>&gt; traffic mentioned in the RFC 1.4.4.1 =
actually=20
  has to be the maximum<BR>&gt; configured<BR>&gt; <BR>&gt; <BR>&gt;=20
  traffic.<BR>&gt; <BR>&gt; <BR>&gt; <BR>&gt; <BR>&gt; <BR>&gt; =
thanks<BR>&gt;=20
  <BR>&gt; <BR>&gt; bijoy/sibi<BR>&gt; <BR><BR>-- <BR>Brian F. G. =
Bidulock<BR><A=20
  href=3D"mailto:bidulock@openss7.org">bidulock@openss7.org</A><BR><A=20
  =
href=3D"http://www.openss7.org/">http://www.openss7.org/</A><BR>_________=
______________________________________<BR>Sigtran=20
  mailing list<BR><A =
href=3D"mailto:Sigtran@ietf.org">Sigtran@ietf.org</A><BR><A=20
  =
href=3D"https://www1.ietf.org/mailman/listinfo/sigtran">https://www1.ietf=
.org/mailman/listinfo/sigtran</A><BR></BLOCKQUOTE></BODY></HTML>

------=_NextPart_000_0340_01C2D783.0E29E730--

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 18 08:59:00 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA28302
	for <sigtran-archive@odin.ietf.org>; Tue, 18 Feb 2003 08:59:00 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1IE4bY28875
	for sigtran-archive@odin.ietf.org; Tue, 18 Feb 2003 09:04:37 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1IE4Lp28860;
	Tue, 18 Feb 2003 09:04:21 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1IE3Dp28821
	for <sigtran@optimus.ietf.org>; Tue, 18 Feb 2003 09:03:13 -0500
Received: from albatross.wise.edt.ericsson.se (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA28289
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 08:57:04 -0500 (EST)
Received: from esealnt611.al.sw.ericsson.se (esealnt611.al.sw.ericsson.se [153.88.254.68])
	by albatross.wise.edt.ericsson.se (8.12.1/8.12.1/WIREfire-1.4) with ESMTP id h1IE0qKW013515;
	Tue, 18 Feb 2003 15:00:52 +0100 (MET)
Received: from ESEALNT744.al.sw.ericsson.se ([153.88.251.4]) by esealnt611.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55)
	id FDV2XZ6T; Tue, 18 Feb 2003 15:00:52 +0100
Received: by ESEALNT744.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <1ZX15SVG>; Tue, 18 Feb 2003 15:00:52 +0100
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65981@EESTQNT105.es.eu.ericsson.se>
X-Sybari-Trust: b4a2ffe6 9ffcebbb a5ee123c 00000138
From: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>
Cc: sigtran@ietf.org
Subject: RE: [Sigtran] Re: M3UA Notify message
Date: Tue, 18 Feb 2003 15:00:50 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Brian, 

I agree 

RFC simply says: 

	"n ASPs is the minimum number of redundant ASPs required to handle traffic"

As this is the minimum number, it can be inferred that for #ASPs < n no traffic can be handled. Nothing is mentioned in the RFC about engineered load although it may be thought that n value calculation is based on that.  

I would say that the current model does not support the use of extra ASPs for loads bigger than the engineered nor the use of less than n for smaller loads... The message is clear saying that #ASPs<n implies no traffic, i.e. whenever you can not support the engineered load, stop the system. This last sentence is not in the RFC but it can be deduced as the engineered load for n can... 

Cheers // Javier. 

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Tuesday, February 18, 2003 12:49 PM
> To: Javier Pastor-Balbas (ECE)
> Cc: sigtran@ietf.org
> Subject: Re: [Sigtran] Re: M3UA Notify message
> 
> 
> Javier,
> 
> Javier Pastor-Balbas (ECE) wrote:             (Tue, 18 Feb 
> 2003 12:17:54)
> > > Javier Pastor-Balbas (ECE) wrote:             (Tue, 18 
> Feb 2003 07:57:23)
> > > > 
> > > > Hi Brian, 
> > > > 
> > > > According to the RFC "n" is required to handle traffic 
> with minimum
> > > > losses, what is expected from the SS7 point of view in 
> an interworking
> > > > scenario.
> 
> Please cite this passage, because I cannot find it.  I find 
> no mention of
> minimum number of ASPs and "message loss".  The only 
> reference to message loss
> that I know of is unrelated.
> 



> > > 
> > > That would be the numbe required to handle maximum 
> engineered load, which
> > > is the same thing.
> > > 
> > 
> > If you refer to the compromise load engineered to be 
> available 99% (aprox.)
> > of the time, then I agree. But this is a statically value fed at
> > configuration time. And the system won't work till n number 
> is reach since
> > the system will likely have quality contrains.
> 
> The maximum engineered load is a figure established at 
> configuration time.
> 
> Carrier-grade systems must always work with 1 processor.  The 
> system will just
> not necessarily be able to handle the offered load.  (But it might:
> particularly during quiet periods, off-peak, normal-day, etc.)
> 
> The offered load cannot be established at configuration time 
> and varies widely
> for telephony systems.
> 
> You must also realized that the system might, at any time,  
> be offered a load
> greater than the maximum engineered load (mass calling, busy 
> periods, etc.):
> so for carrier-grade systems even "n" processors must 
> implement overload
> controls (and they could not claim SS7 compliance if they didn't).
> 
> Look back in this thread and you will see the factors 
> contributing to the
> anticipated offered load for the start, restart and stop decision.
> 
> The decisions are not based on less than "n" processors not 
> allowing the
> system to work (where n > 1).  They are based on the negative 
> effects of
> delivering pent up traffic to processors that will likely 
> just congest.
> 
> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 18 09:13:50 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA28492
	for <sigtran-archive@odin.ietf.org>; Tue, 18 Feb 2003 09:13:50 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1IEJSc30114
	for sigtran-archive@odin.ietf.org; Tue, 18 Feb 2003 09:19:28 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1IEJ7p30099;
	Tue, 18 Feb 2003 09:19:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1IEIfp30074
	for <sigtran@optimus.ietf.org>; Tue, 18 Feb 2003 09:18:41 -0500
Received: from domino2.nc.tekelec.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA28480
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 09:12:32 -0500 (EST)
From: jeffrey.craig@tekelec.com
Received: from domino4.nc.tekelec.com ([198.89.40.9])
          by domino2.nc.tekelec.com (Lotus Domino Release 5.0.10)
          with ESMTP id 2003021809095812:38616 ;
          Tue, 18 Feb 2003 09:09:58 -0500 
Subject: Re: [Sigtran] M2PA draft needs to note that timer T1 differs from Q.703
  timer T1
To: Tom George <Tom.George@alcatel.com>
Cc: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.7  March 21, 2001
Message-ID: <OFFE4EB91B.0CA46D09-ON85256CD1.004D90C5@nc.tekelec.com>
Date: Tue, 18 Feb 2003 09:12:05 -0500
MIME-Version: 1.0
X-MIMETrack: Serialize by Router on Domino4/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/18/2003 09:12:06 AM,
	Itemize by SMTP Server on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/18/2003 09:09:58 AM,
	Serialize by Router on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/18/2003 09:10:00 AM,
	Serialize complete at 02/18/2003 09:10:00 AM
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hello tom,

>M2PA timer T1 seems to follow the description of the Q.703 timer T1 at
>the end of Q.703 section 7.3 "Initial Alignment Procedure". I used the
>Q.703 range and recommended value of T1.
>
>I don't see what you are pointing to in Fig 8 sheet 5. T1 is started as
>LSC goes to Aligned/ready. Please explain more.

I take it back. I must have been looking at M2PA v6, since the v7 default
value for T1 looks fine. Starting T1 on entry to either ALIGNED_READY or
ALIGNED_NOT_READY and stopping it on entry to either IN_SERVICE,
PROCESSOR_OUTAGE, or OUT_OF_SERVICE is also fine. So in short, there is
no problem. Sorry for the confusion.

Regards,

Jeff





                                                                                                                
                      Tom George                                                                                
                      <Tom.George@alca         To:                                                              
                      tel.com>                 cc:      sigtran@ietf.org                                        
                      Sent by:                 Subject: Re: [Sigtran] M2PA draft needs to note that timer T1    
                      sigtran-admin@ie         differs from Q.703  timer T1                                     
                      tf.org                                                                                    
                                                                                                                
                                                                                                                
                      02/17/03 06:22                                                                            
                      PM                                                                                        
                                                                                                                
                                                                                                                




Jeff,

M2PA timer T1 seems to follow the description of the Q.703 timer T1 at
the end of Q.703 section 7.3 "Initial Alignment Procedure". I used the
Q.703 range and recommended value of T1.

I don't see what you are pointing to in Fig 8 sheet 5. T1 is started as
LSC goes to Aligned/ready. Please explain more.

Thanks,
tomG.

jeffrey.craig@tekelec.com wrote:
>
> Hello All,
>
> The M2PA draft should be revised to note that its implementation
> of timer T1 does not conform to timer T1 as described by Q.703
> (SDL, Figure 8, sheet 5). In Q.703 this timer is started upon entry
> to the Initial Alignment state and has a duration that accounts for
> up to four normal alignment sequences.
>
> M2PA timer T1 is started upon entry to either ALIGNED_READY
> or ALIGNED_NOT_READY states and has a recommended
> duration much shorter than that of four normal alignment sequences.
> I have no problem with this treatment, but it is an exception from
> the general conformance to Q.703.
>
> Regards,
>
> Jeff

--
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 18 09:27:44 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA28668
	for <sigtran-archive@odin.ietf.org>; Tue, 18 Feb 2003 09:27:44 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1IEXM930655
	for sigtran-archive@odin.ietf.org; Tue, 18 Feb 2003 09:33:22 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1IEX6p30634;
	Tue, 18 Feb 2003 09:33:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1IEWwp30611
	for <sigtran@optimus.ietf.org>; Tue, 18 Feb 2003 09:32:58 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA28647
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 09:26:49 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:O+BpZNFcSNskEOs1HwSz4lVnxB851tzT@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1IEUbD30173;
	Tue, 18 Feb 2003 07:30:37 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1IEUbM08142;
	Tue, 18 Feb 2003 07:30:37 -0700
Date: Tue, 18 Feb 2003 07:30:37 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030218073037.A7661@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>,
	sigtran@ietf.org
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65981@EESTQNT105.es.eu.ericsson.se>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <1AB3D30B989BF141BBD5C70057B2EF7C01B65981@EESTQNT105.es.eu.ericsson.se>; from javier.pastor-balbas@ece.ericsson.se on Tue, Feb 18, 2003 at 03:00:50PM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Javier,

Javier Pastor-Balbas (ECE) wrote:                                                   (Tue, 18 Feb 2003 15:00:50)
> Brian, 
> 
> I agree 
> 
> RFC simply says: 
> 
> 	"n ASPs is the minimum number of redundant ASPs required to handle traffic"
> 
> As this is the minimum number, it can be inferred that for #ASPs < n no
> traffic can be handled. Nothing is mentioned in the RFC about engineered
> load although it may be thought that n value calculation is based on that.  
> 
> I would say that the current model does not support the use of extra ASPs
> for loads bigger than the engineered nor the use of less than n for smaller
> loads... The message is clear saying that #ASPs<n implies no traffic, i.e.
> whenever you can not support the engineered load, stop the system. This last
> sentence is not in the RFC but it can be deduced as the engineered load for
> n can... 

It does not say that.

It says:

   The failover model supports an "n+k" redundancy model, where "n" ASPs
   is the minimum number of redundant ASPs required to handle traffic
   and "k" ASPs are available to take over for a failed or unavailable
   ASP.  A "1+1" active/backup redundancy is a subset of this model. A
   simplex "1+0" model is also supported as a subset, with no ASP
   redundancy.

And quite frankly this passage was written when we still had "n+k" active/
standby in the document.

Directly with respect to loadsharing modes, it says:

   An SGP or IPSP, upon reception of an ASP Active message for the first
   ASP in a Loadshare AS, MAY choose not to direct traffic to a newly
   active ASP until it determines that there are sufficient resources to
   handle the expected load (e.g., until there are "n" ASPs in state
   ASP-ACTIVE in the AS).  In this case, the SGP or IPSP SHOULD withhold
   the Notify (AS-ACTIVE) until there are sufficient resources.

   For the n+k redundancy case, ASPs which are in that AS should
   coordinate among themselves the number of active ASPs in the AS, and
   should start sending traffic only after n ASPs are active.

   All ASPs within a loadsharing mode AS must be able to process any
   Data message received for the AS, to accommodate any potential
   failover or rebalancing of the offered load.


Note especially the passages "sufficient resources to handle the expected
load" and "to accomodate any potential failover or rebalancing of the offered
load".  Note also, the "must" on being able to process any Data message
received for the AS.

Similar statements are made for broadcast mode.

It is clear from the later that "traffic" in section 1.4.4.1 for the
case of loadshare and broadcast AS relates to "expected load" and
"offered load", the only configurable figure of which is "engineered load".

This is particularly true as the interpretation of IG 3.15 has no
practical use.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 18 15:27:08 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA07423
	for <sigtran-archive@odin.ietf.org>; Tue, 18 Feb 2003 15:27:08 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1IKWr523665
	for sigtran-archive@odin.ietf.org; Tue, 18 Feb 2003 15:32:53 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1IKWap23649;
	Tue, 18 Feb 2003 15:32:37 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1IKVWp23597
	for <sigtran@optimus.ietf.org>; Tue, 18 Feb 2003 15:31:32 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA07395
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 15:25:16 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1IKT4eW001466;
	Tue, 18 Feb 2003 14:29:05 -0600 (CST)
Message-ID: <3E529790.9E2F6EE7@alcatel.com>
Date: Tue, 18 Feb 2003 14:29:04 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft needs to note that timer T1 differs from 
 Q.703timer T1
References: <OFFE4EB91B.0CA46D09-ON85256CD1.004D90C5@nc.tekelec.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

No apology necessary. This is the kind of comment I like! 8-)

jeffrey.craig@tekelec.com wrote:
> 
> Hello tom,
> 
> >M2PA timer T1 seems to follow the description of the Q.703 timer T1 at
> >the end of Q.703 section 7.3 "Initial Alignment Procedure". I used the
> >Q.703 range and recommended value of T1.
> >
> >I don't see what you are pointing to in Fig 8 sheet 5. T1 is started as
> >LSC goes to Aligned/ready. Please explain more.
> 
> I take it back. I must have been looking at M2PA v6, since the v7 default
> value for T1 looks fine. Starting T1 on entry to either ALIGNED_READY or
> ALIGNED_NOT_READY and stopping it on entry to either IN_SERVICE,
> PROCESSOR_OUTAGE, or OUT_OF_SERVICE is also fine. So in short, there is
> no problem. Sorry for the confusion.
> 
> Regards,
> 
> Jeff
  [...]

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 18 17:58:58 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA10651
	for <sigtran-archive@odin.ietf.org>; Tue, 18 Feb 2003 17:58:58 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1IN4jK01227
	for sigtran-archive@odin.ietf.org; Tue, 18 Feb 2003 18:04:45 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1IN4Fp01203;
	Tue, 18 Feb 2003 18:04:16 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1IN29p01128
	for <sigtran@optimus.ietf.org>; Tue, 18 Feb 2003 18:02:09 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA10628
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 17:55:50 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1IMxaeW028845;
	Tue, 18 Feb 2003 16:59:36 -0600 (CST)
Message-ID: <3E52BAD8.270FA4AC@alcatel.com>
Date: Tue, 18 Feb 2003 16:59:36 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: AW: [Sigtran] M2PA draft 07 comment: LPO
References: <OFB928FB8F.3D1D3FC6-ON85256CCC.004BF844@nc.tekelec.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Jeff,

One thing that confuses me about this diagram is the distinction between
ALIGNED and PROVING. It looks like both ends would get to ALIGNED, start
T3 and wait for the other end to send LS Proving. But they don't send LS
Proving until they go to PROVING, so it looks like both ends would just
sit and let their T3s expire.

So I have the sending and receiving of LS Proving messages going on in
the PROVING state. 


BTW, I would be very happy to reference Q.703 wherever possible and am
open to suggestions on how to do it. Unfortunately, there are parts of
the SDLs that do not apply to M2PA (e.g., sending FISUs), so I have been
stumped on how to do this. 

tomG.


jeffrey.craig@tekelec.com wrote:
> 
> Hello Franz,
> 
> >an remark to your M2PA state event machine. In the case
> >'Proving' (T4 expired) is finished and local Processor
> >Outage is occurred, I missing an arrow from state
> >'PROVING' to state 'ALIGNED_NOT_READY'.
> 
> You are correct, it's an error. Here is another try....
> (It would be a lot easier if M2PA simply referred to Q.703 SDLs.)
> 
> Regards,
> 
> Jeff
> 
>                    +-----------+
>                    |   POWER   |
>                    |    OFF    |
>                    +-----------+
>                          |
>                          | Power On
>                          |
>        Flush Buffers     |  +-----------------------------------------+
>        OR Retrieval      |  |                                         |
>               +-------+  |  |                                         |
>               |       |  V  V                                         |
>               |    +-----------+                                      |
>               +--->|  OUT OF   |                                      |
>   +--------------->|  SERVICE  |<--+                                  |
>   |                +-----------+   | Link Configured                  |
>   |                      |   |     | (Associate)                      |
>   |                      |   +-----+                                  |
>   |     Receive LS OOS   |                                            |
>   |           +-------+  | MTP3 Start                                 |
>   |           |       |  V                                            |
>   |           |    +-----------+                                      |
>   |           +--->|    NOT    |                                      |
>   +<---------------|  ALIGNED  |------------------------------------->+
>   |     MTP3 Stop  +-----------+     SCTP Comm Error                  |
>   |  OR T2 Expiry        |        OR SCTP Comm Lost                   |
>   |                      |                                            |
>   |                      |    Receive LS Align                        |
>   |                      | OR Receive LS Prov Norm                    |
>   |     Receive LS Align | OR Receive LS Prov Emer                    |
>   |           +-------+  |                                            |
>   |           |       |  V                                            |
>   |           |    +-----------+                                      |
>   |           +--->|  ALIGNED  |                                      |
>   +<---------------|           |------------------------------------->+
>   |     MTP3 Stop  +-----------+     SCTP Comm Error                  |
>   |  OR T3 Expiry      ^    |     OR SCTP Comm Lost                   |
>   |  OR Receive LS OOS |    |                                         |
>   |                    |    |    Receive LS Prov Norm                 |
>   |                    |    | OR Receive LS Prov Emer                 |
>   |                    |    |                                         |
>   |   Receive LS Align |    |                                         |
>   |                    |    |                                         |
>   |                    |    |                                         |
>   | Receive LS Prov X  |    |                                         |
>   |           +------+ |    |                                         |
>   |           |      | |    V                                         |
>   |           |    +-----------+                                      |
>   |           +--->|  PROVING  |------------------------------------->+
>   |<---------------|           |------+          SCTP Comm Error      |
>   |     MTP3 Stop  +-----------+      |         OR SCTP Comm Lost     |
>   |  OR Receive LS OOS   |            |                               |
>   |                      |  T4 Expiry |                               |
>   |                      |   (LPO)    |                               |
>   |                      |            |                               |
>   |                      |            | MTP3 Stop            +------->+
>   |                      |            | OR Receive LS OOS    |        |
>   |                      |            | OR SCTP Comm Error   |        |
>   |                      |            | OR SCTP Comm Lost    |        |
>   |                      |            | OR T1 Expiry         |        |
>   |           T4 Expiry  |            | OR M2PA Link Failure |        |
>   |            (No LPO)  |            | OR Receive LS Align  |        |
>   |                      V            |                      |        |
>   |                +-----------+ LPO  V            +-----------+      |
>   |                |  ALIGNED  |------+----------->|  ALIGNED  |      |
>   +<---------------|   READY   |<------------------| NOT READY |      |
>   |     MTP3 Stop  +-----------+       LPR         +-----------+      |
>   |  OR T1 Expiry        |   |                         | Rcv LS Ready |
>   |  OR Receive LS OOS   |   |                         | OR Rcv Data  |
>   |  OR Receive LS Align |   +------------------------>+ OR Rcv LS PO |
>   |  OR SCTP Comm Error  |   Receive LS PO             |              |
>   |  OR SCTP Comm Lost   |                             |              |
>   |                      |    Receive LS Ready         |              |
>   |                      | OR Receive User Data        |              |
>   |                      V                             V              |
>   |                +-----------+ LPO or Rcv LS PO  +-----------+      |
>   |                |    IN     |------------------>| PROCESSOR |      |
>   +<---------------|  SERVICE  |<------------------|  OUTAGE   |      |
>                    +-----------+    MTP3 Flush     +-----------+      |
>        MTP3 Stop                 OR MTP3 Continue            |        |
>     OR Receive LS OOS                                        |        |
>     OR SCTP Comm Error                     MTP3 Stop         |        |
>     OR SCTP Comm Lost                   OR Receive LS OOS    |        |
>     OR T6 Expiry                        OR SCTP Comm Error   |        |
>     OR T7 Expiry                        OR SCTP Comm Lost    |        |
>     OR M2PA Link Failure                OR M2PA Link Failure |        |
>     OR Receive LS Align                 OR Receive LS Align  +------->+

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 18 19:31:32 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA12701
	for <sigtran-archive@odin.ietf.org>; Tue, 18 Feb 2003 19:31:32 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1J0bLQ08131
	for sigtran-archive@odin.ietf.org; Tue, 18 Feb 2003 19:37:21 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1J0avp07439;
	Tue, 18 Feb 2003 19:36:59 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1J0Znp07391
	for <sigtran@optimus.ietf.org>; Tue, 18 Feb 2003 19:35:49 -0500
Received: from w2ksjexg01.ciena.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA12633
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 19:29:29 -0500 (EST)
Received: from wntcsdexg01.csd.ciena.com ([10.34.31.31]) by w2ksjexg01.ciena.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2653.13)
	id FG7X17JL; Tue, 18 Feb 2003 16:31:00 -0800
Received: by webdev-llnt.oni.com with Internet Mail Service (5.5.2653.19)
	id <Y5W9M6GP>; Tue, 18 Feb 2003 14:46:57 -0800
Message-ID: <2135200C183FD5119588009027DE572302836CBC@webdev-owa.oni.com>
From: "Ong, Lyndon" <LyOng@ciena.com>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Date: Tue, 18 Feb 2003 14:46:56 -0800
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Sigtran] agenda for San Francisco
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Hi Folks,

Please let me know if you have any items you would like to discuss on
the agenda in San Francisco (and especially if you are planning to 
submit a document and/or present information).

Thanks,

L. Ong
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 18 19:31:53 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA12719
	for <sigtran-archive@odin.ietf.org>; Tue, 18 Feb 2003 19:31:53 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1J0bgd08345
	for sigtran-archive@odin.ietf.org; Tue, 18 Feb 2003 19:37:42 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1J0bEp07930;
	Tue, 18 Feb 2003 19:37:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1J0Zop07395
	for <sigtran@optimus.ietf.org>; Tue, 18 Feb 2003 19:35:50 -0500
Received: from w2ksjexg01.ciena.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA12635
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 19:29:30 -0500 (EST)
Received: from wntcsdexg01.csd.ciena.com ([10.34.31.31]) by w2ksjexg01.ciena.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2653.13)
	id FG7X17K3; Tue, 18 Feb 2003 16:31:03 -0800
Received: by webdev-llnt.oni.com with Internet Mail Service (5.5.2653.19)
	id <Y5W9M6GS>; Tue, 18 Feb 2003 14:50:50 -0800
Message-ID: <2135200C183FD5119588009027DE572302836CBD@webdev-owa.oni.com>
From: "Ong, Lyndon" <LyOng@ciena.com>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Date: Tue, 18 Feb 2003 14:50:49 -0800
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Sigtran] Last Call on signaling applicability statement
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Hi Folks,

The Last Call on the security framework has now closed.  Last Call on
M2PA actually closed last week, although I saw a few more comments
dribble in.

I've had a request now to start Last Call on the Signaling Applicability
Statement,
draft-ietf-sigtran-signalling-over-sctp-applic-07.txt

Please send in your comments starting today and ending Last Call on Tuesday,
March 4th.

Thanks,

L. Ong

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 18 19:32:04 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA12732
	for <sigtran-archive@odin.ietf.org>; Tue, 18 Feb 2003 19:32:04 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1J0bsS08363
	for sigtran-archive@odin.ietf.org; Tue, 18 Feb 2003 19:37:54 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1J0bQp08210;
	Tue, 18 Feb 2003 19:37:26 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1J0Zsp07399
	for <sigtran@optimus.ietf.org>; Tue, 18 Feb 2003 19:35:54 -0500
Received: from w2ksjexg01.ciena.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA12637
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 19:29:31 -0500 (EST)
Received: from wntcsdexg01.csd.ciena.com ([10.34.31.31]) by w2ksjexg01.ciena.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2653.13)
	id FG7X17R4; Tue, 18 Feb 2003 16:31:14 -0800
Received: by webdev-llnt.oni.com with Internet Mail Service (5.5.2653.19)
	id <Y5W9M6J9>; Tue, 18 Feb 2003 16:08:08 -0800
Message-ID: <2135200C183FD5119588009027DE572302836CC1@webdev-owa.oni.com>
From: "Ong, Lyndon" <LyOng@ciena.com>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Date: Tue, 18 Feb 2003 16:08:08 -0800
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Sigtran] liaison statement from ITU-T SG 15
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Hi Folks,

One item on the agenda for San Francisco is a liaison statement sent to us
from ITU-T SG 15.  The liaison can be found at
http://ietf.org/IESG/LIAISON/ls15-15.htm.

Any help from someone familiar with this liaison would be appreciated :o)

Cheers,

L. Ong
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 18 22:07:08 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA16112
	for <sigtran-archive@odin.ietf.org>; Tue, 18 Feb 2003 22:07:08 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1J3D1t18092
	for sigtran-archive@odin.ietf.org; Tue, 18 Feb 2003 22:13:01 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1J3CCp18017;
	Tue, 18 Feb 2003 22:12:13 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1J3AZp17965
	for <sigtran@optimus.ietf.org>; Tue, 18 Feb 2003 22:10:35 -0500
Received: from pt.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA16078
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 22:04:09 -0500 (EST)
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id WAA03610
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 22:02:54 -0500
Message-ID: <3E52F4FC.1030303@pt.com>
Date: Tue, 18 Feb 2003 22:07:40 -0500
From: Andrew Booth <abooth@pt.com>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.1) Gecko/20020826
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: sigtran@ietf.org
Subject: [Sigtran] Repost - More M2PA 7 questions]
Content-Type: multipart/mixed;
 boundary="------------020905030506040206030901"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


This is a multi-part message in MIME format.
--------------020905030506040206030901
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Hi,

I posted these a while ago and haven't seen any responses.  Question (2) 
is really the only relevant one.

Thanks for any info,
Andrew

-- 
Andrew Booth, Software Developer
Signaling Systems Group, Performance Technologies
613-237-4344

Visit www.pt.com/mailing.html to subscribe to our Packets & Signals eLetter.


--------------020905030506040206030901
Content-Type: message/rfc822;
 name="[Sigtran] More M2PA 7 questions"
Content-Disposition: inline;
 filename="[Sigtran] More M2PA 7 questions"

Return-Path: <sigtran-admin@ietf.org>
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id MAA16598;
	Thu, 6 Feb 2003 12:33:50 -0500
Received: from pt.com (localhost [127.0.0.1])
	by pt.com (8.9.3+Sun/8.9.1) with ESMTP id MAA18071;
	Thu, 6 Feb 2003 12:37:39 -0500 (EST)
Received: from www1.ietf.org (ietf.org [132.151.1.19])
	by pt.com (8.9.3+Sun/8.9.3) with ESMTP id MAA18063;
	Thu, 6 Feb 2003 12:37:36 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16Ha8p17588;
	Thu, 6 Feb 2003 12:36:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h16HZMp17532
	for <sigtran@optimus.ietf.org>; Thu, 6 Feb 2003 12:35:22 -0500
Received: from pt.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA26949
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 12:28:18 -0500 (EST)
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id MAA11584
	for <sigtran@ietf.org>; Thu, 6 Feb 2003 12:28:11 -0500
Message-ID: <3E429BFD.70407@pt.com>
Date: Thu, 06 Feb 2003 12:31:41 -0500
From: Andrew Booth <abooth@pt.com>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.1) Gecko/20020826
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: sigtran@ietf.org
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Subject: [Sigtran] More M2PA 7 questions
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Hi All,

A few more questions on M2PA draft 7.

1) Does every User Data received from the peer increases the expected FSN
for the subsequent message, even in cases where the received message is 
later
discarded as part of a Flush Buffers or Changeover procedure?

2) Suppose that an M2PA implementation has received messages from a peer
and that some of the messages are buffered at the M2PA layer (perhaps it is
in processor outage).  Suppose now the link transitions to OOS and the
MTP3 sends a Retrieve BSNT to M2PA.

According to 4.2.6,

   "M2PA determines the Forward Sequence Number of the last User Data
   message received from the peer.  This is the BSNT."

In the M2PA situation it seems more appropriate to respond with the FSN
of the last message which was delivered to MTP3.
Am I missing something?

3) How should T7 be handled when remote processor outage abates?  Should 
we immediately check for excessive ack delay, or reset T7 somehow?

4) In section 5.1 we transition to proving without starting T3 and to 
aligned ready without receiving LS Proving.  Are these omissions 
intentional?

5) In section 5.8 there is some introductory text :

"Note that in this example, the sequence numbers and messages requested
by MTP3 are sent from SCTP to M2PA in the Communication Lost
primitive. In general, the retrieval of sequence numbers and messages
is implementation dependent."

Why are messages retrieved from SCTP?  Is this text indicating that the 
M2PA retransmit buffer is implemented in the SCTP?  Just a little confused.

6) In section 4.2.6, is the following correct or should it be rephrased 
in the context of a circular buffer?
"any transmitted User Data messages beginning with the first
       unacknowledged message with FSN greater than FSNC." 


Thanks for any info,
Andrew

-- 
Andrew Booth, Software Developer
Signaling Systems Group, Performance Technologies
613-237-4344

Visit www.pt.com/mailing.html to subscribe to our Packets & Signals eLetter.


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran


--------------020905030506040206030901--

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 18 22:23:18 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA16318
	for <sigtran-archive@odin.ietf.org>; Tue, 18 Feb 2003 22:23:17 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1J3TBF18791
	for sigtran-archive@odin.ietf.org; Tue, 18 Feb 2003 22:29:11 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1J3Sap18780;
	Tue, 18 Feb 2003 22:28:36 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1J3RSp18747
	for <sigtran@optimus.ietf.org>; Tue, 18 Feb 2003 22:27:28 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA16291
	for <sigtran@ietf.org>; Tue, 18 Feb 2003 22:21:01 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:8uBUD4DqACqdB7zwubtFhCMS0QTJ3+Zo@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1J3OnD09944;
	Tue, 18 Feb 2003 20:24:49 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1J3OnK30705;
	Tue, 18 Feb 2003 20:24:49 -0700
Date: Tue, 18 Feb 2003 20:24:49 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Andrew Booth <abooth@pt.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
Message-ID: <20030218202449.A30468@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Andrew Booth <abooth@pt.com>, sigtran@ietf.org
References: <3E52F4FC.1030303@pt.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E52F4FC.1030303@pt.com>; from abooth@pt.com on Tue, Feb 18, 2003 at 10:07:40PM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Andrew,

The draft is incorrect.

The BSNT is the last acknowledged received DATA (data should not be
acknowledged unless it is locally "deliverable" to MTP3).

Otherwise there is a slim chance that message loss will occur on changeover
resulting from processor outage or busy conditions.

The statement in section 4.2.6 should be changed to reflect this.

--brian


Andrew Booth wrote:                                                                 (Tue, 18 Feb 2003 22:07:40)
> Hi,
> 
> I posted these a while ago and haven't seen any responses.  Question (2) 
> is really the only relevant one.
> 
> Thanks for any info,
> Andrew
> 
> -- 
> Andrew Booth, Software Developer
> Signaling Systems Group, Performance Technologies
> 613-237-4344
> 
> Visit www.pt.com/mailing.html to subscribe to our Packets & Signals eLetter.
> 

> From: Andrew Booth <abooth@pt.com>
> To: sigtran@ietf.org
> Date: Thu, 06 Feb 2003 12:31:41 -0500
> Subject: [Sigtran] More M2PA 7 questions
> 
> Hi All,
> 
> A few more questions on M2PA draft 7.
> 
> 1) Does every User Data received from the peer increases the expected FSN
> for the subsequent message, even in cases where the received message is 
> later
> discarded as part of a Flush Buffers or Changeover procedure?
> 
> 2) Suppose that an M2PA implementation has received messages from a peer
> and that some of the messages are buffered at the M2PA layer (perhaps it is
> in processor outage).  Suppose now the link transitions to OOS and the
> MTP3 sends a Retrieve BSNT to M2PA.
> 
> According to 4.2.6,
> 
>    "M2PA determines the Forward Sequence Number of the last User Data
>    message received from the peer.  This is the BSNT."
> 
> In the M2PA situation it seems more appropriate to respond with the FSN
> of the last message which was delivered to MTP3.
> Am I missing something?
> 
> 3) How should T7 be handled when remote processor outage abates?  Should 
> we immediately check for excessive ack delay, or reset T7 somehow?
> 
> 4) In section 5.1 we transition to proving without starting T3 and to 
> aligned ready without receiving LS Proving.  Are these omissions 
> intentional?
> 
> 5) In section 5.8 there is some introductory text :
> 
> "Note that in this example, the sequence numbers and messages requested
> by MTP3 are sent from SCTP to M2PA in the Communication Lost
> primitive. In general, the retrieval of sequence numbers and messages
> is implementation dependent."
> 
> Why are messages retrieved from SCTP?  Is this text indicating that the 
> M2PA retransmit buffer is implemented in the SCTP?  Just a little confused.
> 
> 6) In section 4.2.6, is the following correct or should it be rephrased 
> in the context of a circular buffer?
> "any transmitted User Data messages beginning with the first
>        unacknowledged message with FSN greater than FSNC." 
> 
> 
> Thanks for any info,
> Andrew
> 
> -- 
> Andrew Booth, Software Developer
> Signaling Systems Group, Performance Technologies
> 613-237-4344
> 
> Visit www.pt.com/mailing.html to subscribe to our Packets & Signals eLetter.
> 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 


-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 19 04:29:01 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA16883
	for <sigtran-archive@odin.ietf.org>; Wed, 19 Feb 2003 04:29:01 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1J9Z2g22829
	for sigtran-archive@odin.ietf.org; Wed, 19 Feb 2003 04:35:02 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1J9Yjp22744;
	Wed, 19 Feb 2003 04:34:45 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1J9Xwp22626
	for <sigtran@optimus.ietf.org>; Wed, 19 Feb 2003 04:33:58 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA16839
	for <sigtran@ietf.org>; Wed, 19 Feb 2003 04:27:25 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1J9V0qi007750;
	Wed, 19 Feb 2003 03:31:05 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h1J9UWF15270;
	Wed, 19 Feb 2003 03:30:32 -0600 (CST)
Date: Wed, 19 Feb 2003 03:30:32 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>
cc: "'bidulock@openss7.org'" <bidulock@openss7.org>, sigtran@ietf.org
Subject: RE: [Sigtran] Re: M3UA Notify message
In-Reply-To: <1AB3D30B989BF141BBD5C70057B2EF7C01B65981@EESTQNT105.es.eu.ericsson.se>
Message-ID: <Pine.SOL.4.10.10302190313420.8070-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hi,

I beleive 'n' is minimum number of ASPs to handle the 'best traffic T'.
To handle the 'worst traffic T+1', SG may require 'n+1' ASPs.

'n'is the derived value based on the traffic report given by Traffic
Metering and Measurement.

So 'n' is fed during configuration of stack.When traffic increases from
'T' to 'T+1' or # APSs < n - this case SG is required to send NOTIFY to
'k' ASP.

And one more thing when # APSs < n, SG still can handle the traffic with
single ASP.

When 'n' ASPs are active
------------------------

SLS-X  ---->ASP1
SLS-Y  ---->ASP2

When ASP1 goes down
-------------------

SLS-(Y,X)  ----->ASP2

Instead of banning the complete traffic ,it can be handled (not 100%) by
ASP2.


--Samuel

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Tue, 18 Feb 2003, Javier Pastor-Balbas (ECE) wrote:

> Brian, 
> 
> I agree 
> 
> RFC simply says: 
> 
> 	"n ASPs is the minimum number of redundant ASPs required to handle traffic"
> 
> As this is the minimum number, it can be inferred that for #ASPs < n no traffic can be handled. Nothing is mentioned in the RFC about engineered load although it may be thought that n value calculation is based on that.  
> 
> I would say that the current model does not support the use of extra ASPs for loads bigger than the engineered nor the use of less than n for smaller loads... The message is clear saying that #ASPs<n implies no traffic, i.e. whenever you can not support the engineered load, stop the system. This last sentence is not in the RFC but it can be deduced as the engineered load for n can... 
> 
> Cheers // Javier. 
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: Tuesday, February 18, 2003 12:49 PM
> > To: Javier Pastor-Balbas (ECE)
> > Cc: sigtran@ietf.org
> > Subject: Re: [Sigtran] Re: M3UA Notify message
> > 
> > 
> > Javier,
> > 
> > Javier Pastor-Balbas (ECE) wrote:             (Tue, 18 Feb 
> > 2003 12:17:54)
> > > > Javier Pastor-Balbas (ECE) wrote:             (Tue, 18 
> > Feb 2003 07:57:23)
> > > > > 
> > > > > Hi Brian, 
> > > > > 
> > > > > According to the RFC "n" is required to handle traffic 
> > with minimum
> > > > > losses, what is expected from the SS7 point of view in 
> > an interworking
> > > > > scenario.
> > 
> > Please cite this passage, because I cannot find it.  I find 
> > no mention of
> > minimum number of ASPs and "message loss".  The only 
> > reference to message loss
> > that I know of is unrelated.
> > 
> 
> 
> 
> > > > 
> > > > That would be the numbe required to handle maximum 
> > engineered load, which
> > > > is the same thing.
> > > > 
> > > 
> > > If you refer to the compromise load engineered to be 
> > available 99% (aprox.)
> > > of the time, then I agree. But this is a statically value fed at
> > > configuration time. And the system won't work till n number 
> > is reach since
> > > the system will likely have quality contrains.
> > 
> > The maximum engineered load is a figure established at 
> > configuration time.
> > 
> > Carrier-grade systems must always work with 1 processor.  The 
> > system will just
> > not necessarily be able to handle the offered load.  (But it might:
> > particularly during quiet periods, off-peak, normal-day, etc.)
> > 
> > The offered load cannot be established at configuration time 
> > and varies widely
> > for telephony systems.
> > 
> > You must also realized that the system might, at any time,  
> > be offered a load
> > greater than the maximum engineered load (mass calling, busy 
> > periods, etc.):
> > so for carrier-grade systems even "n" processors must 
> > implement overload
> > controls (and they could not claim SS7 compliance if they didn't).
> > 
> > Look back in this thread and you will see the factors 
> > contributing to the
> > anticipated offered load for the start, restart and stop decision.
> > 
> > The decisions are not based on less than "n" processors not 
> > allowing the
> > system to work (where n > 1).  They are based on the negative 
> > effects of
> > delivering pent up traffic to processors that will likely 
> > just congest.
> > 
> > --brian
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 19 04:55:32 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA17353
	for <sigtran-archive@odin.ietf.org>; Wed, 19 Feb 2003 04:55:32 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1JA1YQ24953
	for sigtran-archive@odin.ietf.org; Wed, 19 Feb 2003 05:01:34 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1JA1Dp24893;
	Wed, 19 Feb 2003 05:01:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1J9xpp24719
	for <sigtran@optimus.ietf.org>; Wed, 19 Feb 2003 04:59:51 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA17193
	for <sigtran@ietf.org>; Wed, 19 Feb 2003 04:53:16 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:X8G7bkQyj9MdXO3q/HZeSKwZaCRxHpcY@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1J9v5D14964;
	Wed, 19 Feb 2003 02:57:05 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1J9v5Q05320;
	Wed, 19 Feb 2003 02:57:05 -0700
Date: Wed, 19 Feb 2003 02:57:04 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Coene Lode <Lode.Coene@siemens.atea.be>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-signalling-over-sctp- applic-04.txt
Message-ID: <20030219025704.A5002@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Coene Lode <Lode.Coene@siemens.atea.be>, sigtran@ietf.org
References: <E76F715C0429D5118F2100508BB9EDEE037F59B3@hrtades7.atea.be>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <E76F715C0429D5118F2100508BB9EDEE037F59B3@hrtades7.atea.be>; from Lode.Coene@siemens.atea.be on Thu, Mar 07, 2002 at 09:10:05AM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: 8bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Coene,

Although many of these comments were handled between draft-04 and draft-07,
there are a number that have not been addressed.

Please consider the unresolved comments in this list as comments for LC.

--brian

Coene Lode wrote:                                                                    (Thu, 07 Mar 2002 09:10:05)
> Brian,
> 
> Thank you for the detailed list.
> I will not answer them point for point as the document itself was/is not
> fully baked yet. 
> With your input It looks like we'll get much nearer that goal.
> We will take into account all the points you mentioned during our editing in
> Minneapolis.
> I hope that after a new release all your points will have taken care of.
> 
> yours sincerly,
> Lode
> 
> Siemens atea
> atealaan 34          2200 Herentals, Belgium
> E-mail: lode.coene@siemens.atea.be
> Tel: +32-14-252081
> Fax: +32-14-253212
> 
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: Wednesday, March 06, 2002 11:47 AM
> > To: Coene Lode
> > Cc: sigtran@ietf.org
> > Subject: Re: [Sigtran] I-D
> > ACTION:draft-ietf-sigtran-signalling-over-sctp-applic-04.txt
> > 
> > 
> > Lode,
> > 
> > A few comments on this draft:
> > 
> > Section 2:
> > 
> > >   ...            Message boundaries are preserved during data
> > >   transport and so no message delineation is needed. ...
> > 
> > Yet, each UA message has a common header which contains a label
> > and a length which indicates the length of the message.  This is
> > a message delimitation.  Why is it present in each UA if it is
> > not required?
> > 
> > >                                                  ... The 
> > user data can
> > >   be delivered by the order of transmission within a stream(in
> > >   sequence delivery) or the order of arrival.
> > 
> > Unordered delivery in SCTP and "order of arrival" delivery 
> > are different
> > things.  SCTP is permitted to place a priority on out of order packets
> > from various streams.  Preference for delivery of stream "0" unordered
> > packets ahead of other stream unordered packets is of 
> > significant advantage
> > to most UAs (regarless of order of arrival).  Please use the term
> > "unordered delivery" instead of "order of arrival".
> > 
> > >   SCTP can be used to provide redundancy and fault tolerance at the
> > >   transport layer and below. Telephony applications needing 
> > this level
> > >   of fault tolerance can make use of SCTP's multi-homing support.
> > 
> > "Fault tolerance" usually has a different connotation when used in
> > telephony related computing.  It indicates a level of reliability
> > (greater than 5 9's).  Please reword as follows:
> > 
> > <   SCTP can be used to provide redundancy at the transport layer and
> > <   below. Telephony applications needing this level of redundancy can
> > <   make use of SCTP's multi-homing support.
> > 
> > Section 3.1:
> > 
> > >                                               ...  Additionally,
> > >   telephony applications may use SCTP stream based flow control
> > >   [SCTPFLOW].
> > 
> > I believe that stream based flow controls were dropped (in favor of
> > application level flow controls).  Please remove this sentence.
> > 
> > Section 3.2:
> > 
> > >   Telephony systems often must achieve high availability in 
> > operation.
> > 
> > Again, "high availability" has the connotation of a specific level
> > of reliability in telephony related computing (in this case less than
> > 5 9's).  Please reword as:
> > 
> > <   Telephony systems often must achieve redundancy in operation.
> > 
> > or
> > 
> > <   Telephony systems often must have no single point of failure in
> > <   operation.
> > 
> > >   transport layer and below must be detected very fast so that the
> > 
> > grammar, reword as:
> > 
> > <   transport layer and below must be detected quickly so that the
> > 
> > >   For example, they are often required to be able to preserve stable
> > >   calls during a component failure. Therefore error 
> > situations at the
> > >   transport layer and below must be detected very fast so that the
> > >   application can take approriate steps to recover and preserve the
> > >   stable calls. ...
> > 
> > This example is a poor one and shows an utter lack of understanding
> > for telephony.  Stable calls have *no* signalling associated 
> > with them.
> > It is only a call in transition that has signalling.  Signalling
> > transport need not worry about affecting stable calls.  Only redundant
> > systems maintaining call state need worry about this.  While this is
> > a system requirement for an MG or MGC, it has nothing to do with
> > signalling components.  Also, sharing of call state between redundant
> > systems is explicitly excluded from the scope of the UAs.  Please
> > remove this example: it makes the applicability statement 
> > sound foolish.
> > 
> > On the other hand, telephony applications pose requirements on the
> > availability of signalling paths to meet service availability and
> > performance requirements.  SS7 has such signalling relation 
> > availability
> > requirements.  These are the telephony requirements which impose on
> > SCTP's ability to detect failures.  (Again, nothing to do with call
> > state.)  Please replace this example in section 3.2 with words to this
> > effect.
> > 
> > Section 3.2.2:
> > 
> > >   For faster detection of (un)availability of idle paths, the
> > >   telephony application may consider lowering the SCTP parameter
> > >   HB.interval. It should be noted this will result in a 
> > higher traffic
> > >   load.
> > 
> > It is not always true that lowered HB.interval will incresae traffic
> > load.  Please change "will result" to "might result".
> > 
> > Section 3.2.3:
> > 
> > >   Setting Path.Max.Retrans and Association.Max.Retrans SCTP 
> > parameters
> > >   to lower values will speed up both destination address and peer
> > >   failure detection. However, if these values are set too low, the
> > >   probability of false detections will increase.
> > 
> > What false detection?  A retransmission is a retransmission.  There is
> > no "false" retransmission or "false detection" mentioned anywhere.
> > In tests we have shown that Path.Max.Retrans of zero (0) (changing
> > destination address on every retransmission) works quite 
> > fine.  I don't
> > understand what you are trying to warn against here...
> > 
> > Section 3.3:
> > 
> > >   Telephony applications often require short end-to-end message
> > >   delays.  The methods described in section 3.2.1 on 
> > lowering RTO and
> > >   not using delayed SACK may be considered.
> > 
> > Not using delayed SACK does nothing to shorten end-to-end message
> > delay.  Please remove it.
> > 
> > Section 4:
> > 
> > >   Users Adaptation Layers are defined to substitute the telephony
> > >   signaling protocol that is below of the telephony 
> > signaling protocol
> > >   to be relayed.
> > 
> > "that is below of the"????  This sentence is so malformed I don't know
> > what it means.  Please correct.
> > 
> > Section 4:
> > 
> > >      - Support management of SCTP transport associations and traffic
> > >         between SGs and ISEPs or two ISEPs
> > 
> > It is a sore point that several UAs do *not* adequately support
> > management between "two ISEPs."
> > 
> > Section 4:
> > 
> > >                 SP:   Signaling Protocol
> > 
> > What is shown is not a signalling protocol, but a signalling protocol
> > user.  Please change to SPU (Signalling Protocol User).
> > 
> > >                 TTST: Traditional Telephony Signaling Transport
> > 
> > The only traditional SS7 signalling that qualifies as a transport
> > is TCAP.  I suggest you use "Protocol" instead of "Transport".
> > Please replace TTST with TTSP (Traditional Telephony Signalling
> > Protocol).
> > 
> > >   It is also referred as SG to AS communication. AS I the name that
> > 
> > Typo: change "AS I the name" to "AS is the name"
> > 
> > >   It is also referred as IPSP communication. IPSP is the name of the
> > >   role that an IP-based node plays UAL usually gives to the ISEP
> > >   nodes. It stands for IP Signaling Point.
> > 
> > English usage.  Please reword this entire paragraph (especially the
> > middle line "IP-based node plays UAL usually").
> > 
> > Section 4.1:
> > 
> > >                                             ... The default 
> > orientation
> > >   would be for the SG to take on the role of server while 
> > the ISEP is
> > >   the client. ...
> > 
> > Grammar: change "would be for" to "is for".
> > 
> > >   Examples of the upper layers to be transported would be Q.931 and
> > >   QSIG.
> > 
> > Grammar: change "would be" to "are".
> > 
> > >   communication where the ISEP role is typically played by a node
> > >   called MGC defined in [RFC2719].
> > 
> > Grammar: change "called MGC defined" to "called an MGC, as defined".
> > 
> > >   The value assigned by IANA for the Payload Protocol 
> > Identifier in the
> > >   SCTP Payload Data chunk is ô1ö
> > 
> > Non-ascii characters at the end of this sentence.
> > 
> > Section 4.2:
> > 
> > >                                                   ... It is 
> > basically
> > >   implemented in an interworking scenario with SG.
> > 
> > Usage: change "basically" to "typically".
> > 
> > >           AN    û Access Network
> > >           NIF   û Nodal Interworking Function
> > >           LAPV5 û Link Access Protocol for the V5 channel
> > 
> > Non-ascii characters in text.
> > 
> > Section 4.3:
> > 
> > >   The DUA is built on top of IUA defining the necessary 
> > extensions to
> > >   IUA for a DPNSS/DASS2 transport. ...
> > 
> > grammar: change "top of IAU defining" to "top of IUA and defines"
> > 
> > >                                ... DPNSS stands for Digital Private
> > >   Network Signaling System and DASS2 for Digital Access Signaling
> > >   System No 2
> > 
> > Missing period at end of sentence.
> > 
> > >   The value assigned by IANA for the Payload Protocol 
> > Identifier in the
> > >   SCTP Payload Data chunk is ôTBDö.
> > 
> > Non-ascii text at end of sentence.
> > 
> > Does not DUA have a PPI assigned yet?
> > 
> > Section 4.4:
> > 
> > >   This protocol would be mainly used between a Signaling 
> > Gateway (SG)
> > >   and Media Gateway Controler (MGC). ...
> > 
> > Grammar: change "would be" to "is"; change "mainly" to "typically" or
> > "largely" or "usually".
> > 
> > >   The only SS7 MTP2 User is MTP3 that is the protocol transported by
> > >   this UAL.
> > 
> > Really bad sentence formation:  change to:
> > 
> > <   MTP3 is the only MTP2-User that is transport by this UAL.
> > 
> > Besides, the statement is incorrect: MTP3b could also be transported.
> > 
> > >   The SG provides a interworking of transport functions with the IP
> > >   transport, in order to transfer the MTP2-User signaling 
> > messages to
> > >   and from an Application Server (e.g. MGC) where the peer MTP2-
> > >   User protocol layer exists.
> > 
> > This is a relly awkward paragraph.  I think you mean "interworking of
> > SS7 transport functions with IP transport functions".  Change 
> > "in order
> > to" to "to", change "the MTP2-User" to "MTP2-User"; change 
> > "to and from"
> > to "with"; change "Application Server (e.g., MGC) where the peer
> > MTP2-User protocol layer exists" to "MTP2-User at an 
> > Application Server
> > (e.g., MGC)."
> > 
> > >   The value assigned by IANA for the Payload Protocol 
> > Identifier in the
> > >   SCTP Payload Data chunk is ô2ö
> > 
> > Non-ascii text at end of sentence.
> > 
> > Section 4.5:
> > 
> > >   This protocol is used between SS7 Signaling Points 
> > employing the MTP
> > >   Level 3 protocol. The SS7 Signaling Points may also 
> > employ standard
> > >   SS7 links using the SS7 MTP Layer 2 to provide transport 
> > of MTP Layer
> > >   3 signaling messages.
> > 
> > Word usage: change "employing" to "using" and "employ" to "use".
> > 
> > There is no such thing as "MTP Layer 2" or "MTP Layer 3"; in SS7 it is
> > referred to as "MTP Level 2" and "MTP Level 3".
> > 
> > >   Both configurations: intercommunication of SS7 and IP with SG and
> > >   communication between ISEPs are possible.
> > 
> > There is no such word as "intercommunication".  "Communication" is
> > already between two things ("inter").
> > 
> > Besides, M2PA only provides for communication between SPs.  There is
> > no need for the two diagrams for M2PA.  There is no 
> > distinction between
> > the first diagram and the right-hand side of the second diagram for
> > M2PA.
> > 
> > >   These figures are only an example. Other configurations 
> > are possible.
> > >   For example, IPSPs without traditional SS7 links could use the
> > >   protocol layers MTP3/M2PA/SCTP/IP to route SS7 messages 
> > in a network
> > >   with all IP links.
> > >
> > >   Another example is that two SGs could be connected over 
> > an IP network
> > >   to form an SG mated pair similar to the way STPs are 
> > provisioned in
> > >   traditional SS7 networks.
> > 
> > These statements sound foolish.  M2PA is really just a different sort
> > of SS7 link.  Plesae remove.
> > 
> > >   The value assigned by IANA for the Payload Protocol 
> > Identifier in the
> > >   SCTP Payload Data chunk is TBD
> > 
> > Does not M2PA have a PPI yet?
> > 
> > >      b. M2PA: SG-IPSP connection is an SS7 link.
> > >         M2UA: SG-MGC connection is not an SS7 link. It is an
> > >               extension of MTP to a remote entity.
> > 
> > There is nothing stopping M2UA from being used in a peer-to-peer
> > fashion.
> > 
> > >      c. M2PA: SG is an SS7 node with a point code.
> > >         M2UA: SG is not an SS7 node and has no point code.
> > 
> > This is not necessarily a correct statement.  There is nothing
> > restricting an M2UA SG from having a point code or being an
> > SS7 node in its own right.
> > 
> > >      d. M2PA: SG can have upper SS7 layers, e.g., SCCP.
> > >         M2UA: SG does not have upper SS7 layers since it 
> > has no MTP3.
> > 
> > Same is true here.  The above statement is false.
> > 
> > >      e. M2PA: relies on MTP3 for management procedures.
> > >         M2UA: uses M2UA management procedures.
> > 
> > If need explanation of the absurdity of this statement, consider
> > this SG:
> >                      ____
> >                     |
> >                     |
> >                 +------+
> >                 | M2UA |
> >                 +------+
> >                 | M2PA |
> >                 +------+
> >                     |
> >              _______|
> > 
> > Which is referred to in the M2UA document.
> > 
> > This SG is an M2PA SG, but it has no point code, has no local MTP3,
> > has no local upper layers (SCCP), and relies on M2UA for management
> > procedures.
> > 
> > Section 4.6:
> > 
> > >      - Interconnection of SS7 and IP nodes
> > 
> > Grammar: replace "Interconection" with "Connection".
> > 
> > 
> > >   It works using the client-server philosophy. ...
> > 
> > This might be ok in German, but is awkward in English.  Reword as:
> > 
> > <   M3UA uses a client-server architecture.
> > 
> > >                                            ... ISEP is 
> > recommended to
> > >   be client when talking with a SG.
> > 
> > <   It is recommended that the ISEP act as the client and initiate
> > <   SCTP associations with the SG.
> > 
> > >                                 ... The reserved port by 
> > IANA is 2905
> > >   to listen to possible client connections.
> > 
> > <   The port reserved by IANA is port number 2905.  This is the port
> > <   upon which the SG should listen for client connections.
> > 
> > >   The assigned payload protocol identifier for the SCTP 
> > DATA chunks is
> > >   ô3ö.
> > 
> > Non-ascii text at end of sentence.
> > 
> > 
> > Section 4.7:
> > 
> > >                 ... SUA can support only non-call related signaling.
> > 
> > Almost all SUA signalling is call related.  I think you mean
> > non-call-associated signalling; but, then, *all* SS7 signalling is
> > non-call-associated signalling.  Please strike this sentence.
> > 
> > >   SUA does not pose stringent timing constraints on SCTP due to the
> > >   fact that SUA applications have broad timing requirement 
> > (from 10 of
> > >   seconds to hours) which the applications guard themselves and the
> > >   timing supervision of the application is end-to-end, not hop-by-
> > >   hop(as with ISUP).
> > 
> > This statement is false and misleading.  SCCP, particularly for the
> > cited applications: MAP, INAP, SMS, BSSAP, RANAP, has very stringent
> > timing requirements, often more stringent than ISUP due to the fact
> > that these represent transaction which are made during ISUP 
> > call setup.
> > 
> > Tens of seconds to hours is false; and atypical.  Transaction timers
> > an performance requirements for INAP, for example, are 1 
> > second maximum
> > with 100 ms nominal query/response times.  Please remove this 
> > paragraph.
> > 
> > >      - Interconnection of SS7 and IP
> > 
> > Change "Interconnection" to "Connection".
> > 
> > >   connections. The payload protocol identifier for the SCTP 
> > DATA chunks
> > >   is ô4ö.
> > 
> > Non-ascii text.
> > 
> > 
> > General:
> > 
> > There is a mixture of "signalling" (two l's British/Canadian spelling)
> > and "signaling" (American spelling).  As the title contains signalling
> > with two l's, please use signalling with two l's throughout.
> > 
> > --brian
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 19 11:21:50 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA00356
	for <sigtran-archive@odin.ietf.org>; Wed, 19 Feb 2003 11:21:50 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1JGS0i23392
	for sigtran-archive@odin.ietf.org; Wed, 19 Feb 2003 11:28:00 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1JGRXp23365;
	Wed, 19 Feb 2003 11:27:34 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1JGPop23237
	for <sigtran@optimus.ietf.org>; Wed, 19 Feb 2003 11:25:50 -0500
Received: from domino2.nc.tekelec.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA00277
	for <sigtran@ietf.org>; Wed, 19 Feb 2003 11:19:06 -0500 (EST)
From: jeffrey.craig@tekelec.com
Received: from domino4.nc.tekelec.com ([198.89.40.9])
          by domino2.nc.tekelec.com (Lotus Domino Release 5.0.10)
          with ESMTP id 2003021911163241:2939 ;
          Wed, 19 Feb 2003 11:16:32 -0500 
Subject: Re: AW: [Sigtran] M2PA draft 07 comment: LPO
To: Tom George <Tom.George@alcatel.com>
Cc: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.7  March 21, 2001
Message-ID: <OF71910696.3AE98EF8-ON85256CD2.00448044@nc.tekelec.com>
Date: Wed, 19 Feb 2003 11:18:39 -0500
MIME-Version: 1.0
X-MIMETrack: Serialize by Router on Domino4/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/19/2003 11:18:40 AM,
	Itemize by SMTP Server on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/19/2003 11:16:32 AM,
	Serialize by Router on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/19/2003 11:16:35 AM,
	Serialize complete at 02/19/2003 11:16:35 AM
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hello Tom,

My lengthy responses are below.

Regards,

Jeff




                                                                                                                                  
                      Tom George                                                                                                  
                      <Tom.George@alca         To:                                                                                
                      tel.com>                 cc:      sigtran@ietf.org                                                          
                      Sent by:                 Subject: Re: AW: [Sigtran] M2PA draft 07 comment: LPO                              
                      sigtran-admin@ie                                                                                            
                      tf.org                                                                                                      
                                                                                                                                  
                                                                                                                                  
                      02/18/03 05:59                                                                                              
                      PM                                                                                                          
                                                                                                                                  
                                                                                                                                  




>Jeff,
>
>One thing that confuses me about this diagram is the distinction between
>ALIGNED and PROVING. It looks like both ends would get to ALIGNED, start
>T3 and wait for the other end to send LS Proving. But they don't send LS
>Proving until they go to PROVING, so it looks like both ends would just
>sit and let their T3s expire.

Multiple-paged SDLs are such fun, aren't they? I obviously have joined
you in confusion, considering my previous timer T1 post.

Q.703 figure 9 sheet 2 shows that either LSPx should be sent
upon transition to ALIGNED from NOT_ALIGNED. LSPx is repeatedly sent
during PROVING. This ensures that during successful alignment a peer
will receive LSPx during ALIGNED, will disable T3, and move to
PROVING (shown on figure 9 sheet 3).

The ALIGNED state is used to ensure that a LSPx is received.
The PROVING state is used to ensure the viability of the link under
greater load, greater message size. It may also be used for MTU
discovery, opening SCTP windows, etc.

>
>So I have the sending and receiving of LS Proving messages going on in
>the PROVING state.
>

M2PA PROVING procedure matches Q.703, but INITIAL_ALIGNMENT
does not currently match, leading to the confusion over when
to send/receive LSA and LSPx. That is why I suggest that the
NOT_ALIGNED and ALIGNED states be used in place of the singular
INITIAL_ALIGNMENT state.

>
>BTW, I would be very happy to reference Q.703 wherever possible and am
>open to suggestions on how to do it. Unfortunately, there are parts of
>the SDLs that do not apply to M2PA (e.g., sending FISUs), so I have been
>stumped on how to do this.

Yes, I understand that it is a challenge. In section "4.1.3 Link
Alignment", I think it would be helpful to reference the Q.703 SDLs
as a guide and list the basic equivalence between the M2PA and
MTP2 primitives:

LSA  == SIO
LSB  == SIB
LSO  == SIOS
LSPN == SIN
LSPE == SIE
LSPO == SIPO
Data == MSU

For each paragraph in section 4.1.3, one could reference the applicable
SDL figure and sheet. This may be excessive, but I think it would add
clarity. Here follows my attempt to revise the draft based on the
recent alignment and LPO discussions:

----------------
OLD TEXT 4.1.3
After the association is established, M2PA SHALL send a Link Status
Out of Service message to its peer.

NEW TEXT 4.1.3
After the association is established, M2PA SHALL send a Link Status
Out of Service message to its peer and, prior to receiving a Start
Request from MTP3, MAY periodically send Link Status Out of
Service messages ([Q.703] Figure 4 and Figure 8 Sheet 1). During
the Out of Service State M2PA SHOULD ignore received Link Status
Out of Service messages and SHOULD respond to all other received
messages by sending a Link Status Out of Service message."


----------------
OLD TEXT 4.1.3

Once the association is established and M2PA has received a Start
Request from MTP3, M2PA sends the Link Status Alignment message to
its peer. If M2PA has not already received the Link Status Alignment
message from its peer, then M2PA starts timer T2.

NEW TEXT 4.1.3

Once the association is established and M2PA has received a Start
Request from MTP3, M2PA sends the Link Status Alignment message to
its peer and moves to the NOT_ALIGNED state ([Q.703] Figure 9
Sheet 1).


----------------
OLD TEXT 4.1.3
M2PA stops timer T2 when it has received the Link Status Alignment
message from its peer.

NEW TEXT 4.1.3
M2PA stops timer T2 when it has received a Link Status Alignment
message or Link Status Proving Normal or Link Status Emergency
message from its peer. It sends a Link Status Proving message,
starts timer T3, and then moves to the ALIGNED state ([Q.703]
Figure 9 Sheet 2).


----------
OLD TEXT 4.1.3
Note: Between the time M2PA sends Link Status Alignment to its peer
and receives Link Status Alignment from its peer, M2PA may receive
Link Status Out of Service from its peer. This message is
ignored. After receiving Link Status Alignment from the peer, receipt
of a Link Status Out of Service message causes M2PA to send Out of
Service to MTP3 and return to the Out of Service state.

NEW TEXT 4.1.3
Note: During the NOT_ALIGNED state M2PA may receive Link Status Out
of Service from its peer. This message is ignored. After moving to
the ALIGNED state, receipt of a Link Status Out of Service message
causes M2PA to send Out of Service to MTP3 and return to the Out of
Service state ([Q.703] Figure 9 Sheet 4).

----------
OLD TEXT 4.1.3

When M2PA has both sent and received the Link Status Alignment
message, it has completed alignment.  M2PA starts the aligned timer
T3 and moves to the proving state.

M2PA stops timer T3 when it receives a Proving Normal or Proving
Emergency message and starts proving period timer T4.

If timer T3 expires, then M2PA reports to MTP3 that the link is out of
service.  M2PA sends Link Status Out of Service to its peer.  M2PA
SHOULD leave the association established.  M2PA waits for MTP3 to
initiate the alignment procedure again.

During the proving period (i.e., after M2PA starts the proving period
timer T4), M2PA sends Link Status Proving messages to its peer at an
interval defined by the protocol parameter Proving_Interval. M2PA
sends either the Proving Normal or Proving Emergency message,
according to the Emergency and Emergency Ceases commands from
MTP3. M2PA uses the value of T4 corresponding to the Normal or
Emergency condition. However, if M2PA receives a Link Status Proving
Emergency message from its peer, then M2PA SHALL initiate the
Emergency proving period value for T4, but it SHALL continue to send
the Proving message (Normal or Emergency) determined by its own upper
layer MTP3.

NEW TEXT 4.1.3

Upon receiving either Link Status Proving Normal or Link Status
Proving Emergency during the ALIGNED state, M2PA stops timer T3,
starts proving period timer T4, and enters the PROVING state
([Q.703] Figure 9 Sheet 3).

If timer T3 expires, then M2PA reports to MTP3 that the link is
out of service, sends Link Status Out of Service to its peer and
moves to the OUT_OF_SERVICE state ([Q.703 Figure 9 Sheet 4). M2PA
SHOULD leave the association established.

During the PROVING state M2PA periodically sends either Link
Status Proving Normal or Link Status Proving Emergency according
to the Emergency and Emergency Ceases commands from MTP3
([Q.703] Figure 9 Sheets 5 and 6). M2PA uses the value of T4
corresponding to the Normal or Emergency condition. However, if
M2PA receives a Link Status Proving Emergency message from its
peer, then M2PA SHALL initiate the Emergency proving period value
for T4, but it SHALL continue to send the Proving message
(Normal or Emergency) determined by its own upper layer MTP3
([Q.703] Figure 9 Sheet 6).

Note: If M2PA receives a Link Status Alignment message during the
PROVING state, then M2PA stops timer T4, starts timer T3, and
moves to the ALIGNED state ([Q.703] Figure 9 Sheet 5).


************************************************
NOTE: There seems to be a differences between M2PA
and Q.703  concerning reception of LSR or LSPO
during PROVING. Currently M2PA indicates these
events should be recorded. This memory is used to
move from PROVING straight to either IN_SERVICE or
PROCESSOR_OUTAGE without starting timer T1. Unless
I am mistaken, Q.703 has no such memory -- these
events are ignored during PROVING, and timer T1
is always started upon expiry of T4 (end of
proving). Changing this in M2PA will require that
the periodic sending of LSR or LSPO become SHOULD
rather than MAY. Leaving this unchanged should
require that the M2PA state machine figure change
to show transitions from PROVING to either
ALIGNED_READY or ALIGNED_NOT_READY or IN_SERVICE or
PROCESSOR_OUTAGE. I don't know how to make this
change to the figure -- no room left.

I'm keeping the memory of these events during
PROVING (status quo) in the following revision.
************************************************

----------
OLD TEXT 4.1.3

When the proving period timer T4 expires, M2PA changes to the Aligned
Ready state. M2PA SHALL start the timer T1 and send a Link Status
Ready message to its peer. These Link Status Ready message is used to
verify that both ends have completed proving. M2PA MAY send additional
Link Status Ready messages while timer T1 is running.

NEW TEXT 4.1.3

If M2PA receives either Link Status Ready or Link Status
Processor Outage during PROVING, then it records the event
for processing upon expiry of timer T4.

When timer T4 expires and a Link Status Ready message has
been received, then M2PA, depending on local processor
outage state, either sends a Link Status Processor Outage
message and moves to the PROCESSOR_OUTAGE state or sends
a Link Status Ready message and moves to the IN_SERVICE
state.

When timer T4 expires and a Link Status Processor Outage
message has been received, then M2PA notifies MTP3 of the
remote processor outage condition, and, depending on local
processor outage status, M2PA either sends a Link Status
Processor Outage message or sends a Link Status Ready
message, and then moves to the PROCESSOR_OUTAGE state.

When timer T4 expires and neither Link Status Ready nor
Link Status Processor Outage have been received, then M2PA
starts timer T1. If a Local Processor Outage Notification
had been received from MTP3, then M2PA sends a Link Status
Processor Outage message and moves to the ALIGNED_NOT_READY
state, otherwise it sends a Link Status Ready message and
moves to the ALIGNED_READY state ([Q.703] Figure 8 Sheet 5).

During the ALIGNED_READY state, M2PA MAY periodically send
Link Status Ready messages.

During the ALIGNED_NOT_READY and PROCESSOR_OUTAGE states,
M2PA MAY periodically send Link Status Processor Outage
messages.


----------
OLD TEXT 4.1.3

M2PA SHALL stop timer T1 when it receives a Link Status Ready or User
Data message from its peer. If timer T1 expires, then M2PA reports to
MTP3 that the link is out of service. M2PA sends Link Status Out of
Service to its peer. M2PA SHOULD leave the association
established. M2PA waits for MTP3 to initiate the alignment procedure
again.

Note that if M2PA has already received a Link Status Ready message
from its peer when its timer T4 expires, there is no need to start
timer T1. M2PA can just send Link Status Ready to the peer and
continue along.

NEW TEXT 4.1.3

M2PA SHALL stop timer T1 when it receives either a Link Status Ready
or User Data or Link Status Processor Outage message from its peer
([Q.703] Figure 8 Sheets 7 and 9). If either local or remote processor
outage is in effect, then M2PA moves to the PROCESSOR_OUTAGE state,
otherwise it moves to the IN_SERVICE state ([Q.703] Figure 8 Sheet 7).

If timer T1 expires, then M2PA reports to MTP3 that the link is
out of service. M2PA sends Link Status Out of Service to its peer,
and moves to the OUT_OF_SERVICE state. M2PA SHOULD leave the
association established.


----------
OLD TEXT 4.1.3

Note that if M2PA has already received a Link Status Ready message
from its peer when its timer T4 expires, there is no need to start
timer T1. M2PA can just send Link Status Ready to the peer and
continue along.


NEW TEXT 4.1.3
[deleted -- content moved to PROVING discussion]


----------
OLD TEXT 4.1.3

If there is a local processor outage condition during the alignment
procedure, M2PA sends Link Status Processor Outage to its peer. When
the local processor outage condition ends, then M2PA SHALL send Link
Status Processor Outage Ended to its peer. M2PA SHALL attempt to
complete the alignment procedure during the local processor outage
condition.

If M2PA receives a Link Status Processor Outage during alignment, and
M2PA had received a Start Request from its MTP3, M2PA SHALL report
Remote Processor Outage to MTP3. M2PA SHALL attempt to complete the
alignment procedure during the remote processor outage condition.

NEW TEXT 4.1.3

[deleted -- Behavior has been clarified above. LSPO should only
be received during PROVING or ALIGNED_READY or ALIGNED_NOT_READY states.
LSPO should only be sent during ALIGNED_NOT_READY or PROCESSOR_OUTAGE
states].

----------
OLD TEXT 4.1.3

When all of the following are true:

   (a) M2PA has received a Start Request from MTP3.

   (b) M2PA's proving period T4 has expired.

   (c) M2PA has sent a Link Status Ready to its peer.

   (d) M2PA has received a Link Status Ready OR User Data
       message from its peer.

   (e) M2PA has not received Link Status Out of Service from its peer
       since it received Link Status Alignment.

   then M2PA SHALL send Link In Service to its MTP3.

NEW TEXT 4.1.3

When all of the following are true:

   (a) M2PA has received a Start Request from MTP3.

   (b) M2PA has not received a Stop Request from MTP3
       since the Start Request.

   (c) The association has remained established.

   (d) M2PA has received EITHER (a Link Status Aligned and
       a Link Status Proving message)) OR (two Link
       Status Proving messages).

   (e) M2PA's proving period timer T4 has expired.

   (f) M2PA has received a Link Status Ready OR User Data
       OR Link Status Processor Outage Ended message from
       its peer.

   (g) M2PA has not received Link Status Out of Service from
       its peer since the ALIGNED state.

   then M2PA SHALL send Link In Service to its MTP3.



-----------------
OLD TEXT 4.1.4

When M2PA detects a local processor outage, it sends a Link Status
message to its peer with status Processor Outage. M2PA SHALL also
cease sending User Data messages to SCTP for transmission. M2PA SHALL
stop acknowledging received User Data messages.

M2PA MAY send additional Link Status Processor Outage messages as long
as there is a local processor outage and the link is in service. If
the link is out of service, M2PA SHOULD locally mark that it is in
local processor outage.

NEW TEXT 4.1.4

During PROCESSOR_OUTAGE or IN_SERVICE states, when MTP3 notifies
M2PA of local processor outage, M2PA sends a Link Status Processor
Outage message to its peer and moves to the PROCESSOR_OUTAGE state
([Q.703] Figure 8 Sheet 11). M2PA SHALL also cease sending User Data
messages to SCTP for transmission. M2PA SHALL stop acknowledging
received User Data messages.

M2PA MAY periodically send Link Status Processor Outage messages
during the PROCESSOR_OUTAGE state while local processor outage
is in effect.


-----------------
OLD TEXT 4.1.4

MTP3 may send a Flush Buffers or Continue command to M2PA as part of
its processor outage procedure (See section 4.2.4 Flush Buffers and
Continue). The use of Flush Buffers and Continue is specified in Q.703
[Q.703] and T1.111.3 [T1.111].

Alternatively, MTP3 may perform data retrieval as part of a changeover
procedure.

When the processor outage ceases, MTP3 sends a Local Processor
Recovered indication to M2PA. The local M2PA notifies its peer by
sending a Link Status message with status Processor Outage Ended. The
local M2PA SHALL resume receiving and transmitting messages.


NEW TEXT 4.1.4

MTP3 notifies M2PA of the onset of local processor outage condition
by sending a Local Processor Outage Request. Prior to the
PROCESSOR_OUTAGE state, M2PA records this event for future
processing. If this event is received during PROCESSOR_OUTAGE, then
M2PA sends a Link Status Processor Outage message to its peer and
records the event ([Q.703] Figure 8 Sheet 12). If this event is
received during IN_SERVICE, then M2PA sends a Link Status Processor
Outage message to its peer, records the local processor outage
state, and moves to the PROCESSOR_OUTAGE state ([Q.703] Figure 8
Sheet 11).

MTP3 notifies M2PA of the clearing of the local processor condition
by sending the Local Processor Outage Recovered Request. Prior to
the PROCESSOR_OUTAGE state, M2PA records this event by clearing its
record of the local processor outage condition. When this
indication is received during PROCESSOR_OUTAGE state, M2PA sends
a Link Status Processor Outage Ended message to its peer
([Q.703] Figure 8 Sheet 13).

M2PA timers T6 and T7 are stopped and remote busy status is
cleared upon transition to the PROCESSOR_OUTAGE state.

During PROCESSOR_OUTAGE if a Link Status Processor Outage Ended
message is received from the peer, M2PA notifies MTP3 that
remote processor has recovered ([Q.703] Figure 8 Sheet 12).

The use of Flush Buffers and Continue is specified in Q.703
[Q.703] and T1.111.3 [T1.111].

During PROCESSOR_OUTAGE if a Flush Buffers Request is received
from MTP3, then M2PA flushes the transmit message buffers and
records the event for future processing ([Q.703] Figure 8
Sheet 13).

During PROCESSOR_OUTAGE if a Continue Request is received
from MTP3, then M2PA records the event for future processing
([Q.703] Figure 8 Sheet 13).

If during PROCESSOR_OUTAGE M2PA determines that neither local
nor remote processor outage is in effect and either a Flush Buffers
Request or a Continue Request was received from MTP3, then it moves
to the IN_SERVICE state and begins receiving and transmitting
normally ([Q.703] Figure 8 Sheet 13).


>
>tomG.


jeffrey.craig@tekelec.com wrote:
>
> Hello Franz,
>
> >an remark to your M2PA state event machine. In the case
> >'Proving' (T4 expired) is finished and local Processor
> >Outage is occurred, I missing an arrow from state
> >'PROVING' to state 'ALIGNED_NOT_READY'.
>
> You are correct, it's an error. Here is another try....
> (It would be a lot easier if M2PA simply referred to Q.703 SDLs.)
>
> Regards,
>
> Jeff
>
>                    +-----------+
>                    |   POWER   |
>                    |    OFF    |
>                    +-----------+
>                          |
>                          | Power On
>                          |
>        Flush Buffers     |  +-----------------------------------------+
>        OR Retrieval      |  |                                         |
>               +-------+  |  |                                         |
>               |       |  V  V                                         |
>               |    +-----------+                                      |
>               +--->|  OUT OF   |                                      |
>   +--------------->|  SERVICE  |<--+                                  |
>   |                +-----------+   | Link Configured                  |
>   |                      |   |     | (Associate)                      |
>   |                      |   +-----+                                  |
>   |     Receive LS OOS   |                                            |
>   |           +-------+  | MTP3 Start                                 |
>   |           |       |  V                                            |
>   |           |    +-----------+                                      |
>   |           +--->|    NOT    |                                      |
>   +<---------------|  ALIGNED  |------------------------------------->+
>   |     MTP3 Stop  +-----------+     SCTP Comm Error                  |
>   |  OR T2 Expiry        |        OR SCTP Comm Lost                   |
>   |                      |                                            |
>   |                      |    Receive LS Align                        |
>   |                      | OR Receive LS Prov Norm                    |
>   |     Receive LS Align | OR Receive LS Prov Emer                    |
>   |           +-------+  |                                            |
>   |           |       |  V                                            |
>   |           |    +-----------+                                      |
>   |           +--->|  ALIGNED  |                                      |
>   +<---------------|           |------------------------------------->+
>   |     MTP3 Stop  +-----------+     SCTP Comm Error                  |
>   |  OR T3 Expiry      ^    |     OR SCTP Comm Lost                   |
>   |  OR Receive LS OOS |    |                                         |
>   |                    |    |    Receive LS Prov Norm                 |
>   |                    |    | OR Receive LS Prov Emer                 |
>   |                    |    |                                         |
>   |   Receive LS Align |    |                                         |
>   |                    |    |                                         |
>   |                    |    |                                         |
>   | Receive LS Prov X  |    |                                         |
>   |           +------+ |    |                                         |
>   |           |      | |    V                                         |
>   |           |    +-----------+                                      |
>   |           +--->|  PROVING  |------------------------------------->+
>   |<---------------|           |------+          SCTP Comm Error      |
>   |     MTP3 Stop  +-----------+      |         OR SCTP Comm Lost     |
>   |  OR Receive LS OOS   |            |                               |
>   |                      |  T4 Expiry |                               |
>   |                      |   (LPO)    |                               |
>   |                      |            |                               |
>   |                      |            | MTP3 Stop            +------->+
>   |                      |            | OR Receive LS OOS    |        |
>   |                      |            | OR SCTP Comm Error   |        |
>   |                      |            | OR SCTP Comm Lost    |        |
>   |                      |            | OR T1 Expiry         |        |
>   |           T4 Expiry  |            | OR M2PA Link Failure |        |
>   |            (No LPO)  |            | OR Receive LS Align  |        |
>   |                      V            |                      |        |
>   |                +-----------+ LPO  V            +-----------+      |
>   |                |  ALIGNED  |------+----------->|  ALIGNED  |      |
>   +<---------------|   READY   |<------------------| NOT READY |      |
>   |     MTP3 Stop  +-----------+       LPR         +-----------+      |
>   |  OR T1 Expiry        |   |                         | Rcv LS Ready |
>   |  OR Receive LS OOS   |   |                         | OR Rcv Data  |
>   |  OR Receive LS Align |   +------------------------>+ OR Rcv LS PO |
>   |  OR SCTP Comm Error  |   Receive LS PO             |              |
>   |  OR SCTP Comm Lost   |                             |              |
>   |                      |    Receive LS Ready         |              |
>   |                      | OR Receive User Data        |              |
>   |                      V                             V              |
>   |                +-----------+ LPO or Rcv LS PO  +-----------+      |
>   |                |    IN     |------------------>| PROCESSOR |      |
>   +<---------------|  SERVICE  |<------------------|  OUTAGE   |      |
>                    +-----------+    MTP3 Flush     +-----------+      |
>        MTP3 Stop                 OR MTP3 Continue            |        |
>     OR Receive LS OOS                                        |        |
>     OR SCTP Comm Error                     MTP3 Stop         |        |
>     OR SCTP Comm Lost                   OR Receive LS OOS    |        |
>     OR T6 Expiry                        OR SCTP Comm Error   |        |
>     OR T7 Expiry                        OR SCTP Comm Lost    |        |
>     OR M2PA Link Failure                OR M2PA Link Failure |        |
>     OR Receive LS Align                 OR Receive LS Align  +------->+

--
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 19 11:33:17 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA00739
	for <sigtran-archive@odin.ietf.org>; Wed, 19 Feb 2003 11:33:17 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1JGdRN24726
	for sigtran-archive@odin.ietf.org; Wed, 19 Feb 2003 11:39:27 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1JGd8p24713;
	Wed, 19 Feb 2003 11:39:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1JGc6p24675
	for <sigtran@optimus.ietf.org>; Wed, 19 Feb 2003 11:38:06 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA00667
	for <sigtran@ietf.org>; Wed, 19 Feb 2003 11:31:16 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:10V6Sll11ct0Pz5xhLuYlLsKukTKlvbF@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1JGZ5D21266;
	Wed, 19 Feb 2003 09:35:05 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1JGZ4c16588;
	Wed, 19 Feb 2003 09:35:04 -0700
Date: Wed, 19 Feb 2003 09:35:04 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: jeffrey.craig@tekelec.com
Cc: Tom George <Tom.George@alcatel.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft 07 comment: LPO
Message-ID: <20030219093504.A16359@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: jeffrey.craig@tekelec.com,
	Tom George <Tom.George@alcatel.com>, sigtran@ietf.org
References: <OF71910696.3AE98EF8-ON85256CD2.00448044@nc.tekelec.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF71910696.3AE98EF8-ON85256CD2.00448044@nc.tekelec.com>; from jeffrey.craig@tekelec.com on Wed, Feb 19, 2003 at 11:18:39AM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Jeffrey,

That would be fine, but ANSI T1.111.3 is different than Q.703, particularly
as regards processor outage handling and clear buffers/resume behavior.
Did you want to add SDL references for ANSI T1.111.3?  An what about JTTC
(which is rather different again)?

I agree that M2PA should not conflict with the SDLs, but referencing just
Q.703 seems inadequate.

--brian

jeffrey.craig@tekelec.com wrote:                                             (Wed, 19 Feb 2003 11:18:39)
> 
> Hello Tom,
> 
> My lengthy responses are below.
> 
> Regards,
> 
> Jeff
> 
> 
> 
> 
>                                                                                                                                   
>                       Tom George                                                                                                  
>                       <Tom.George@alca         To:                                                                                
>                       tel.com>                 cc:      sigtran@ietf.org                                                          
>                       Sent by:                 Subject: Re: AW: [Sigtran] M2PA draft 07 comment: LPO                              
>                       sigtran-admin@ie                                                                                            
>                       tf.org                                                                                                      
>                                                                                                                                   
>                                                                                                                                   
>                       02/18/03 05:59                                                                                              
>                       PM                                                                                                          
>                                                                                                                                   
>                                                                                                                                   
> 
> 
> 
> 
> >Jeff,
> >
> >One thing that confuses me about this diagram is the distinction between
> >ALIGNED and PROVING. It looks like both ends would get to ALIGNED, start
> >T3 and wait for the other end to send LS Proving. But they don't send LS
> >Proving until they go to PROVING, so it looks like both ends would just
> >sit and let their T3s expire.
> 
> Multiple-paged SDLs are such fun, aren't they? I obviously have joined
> you in confusion, considering my previous timer T1 post.
> 
> Q.703 figure 9 sheet 2 shows that either LSPx should be sent
> upon transition to ALIGNED from NOT_ALIGNED. LSPx is repeatedly sent
> during PROVING. This ensures that during successful alignment a peer
> will receive LSPx during ALIGNED, will disable T3, and move to
> PROVING (shown on figure 9 sheet 3).
> 
> The ALIGNED state is used to ensure that a LSPx is received.
> The PROVING state is used to ensure the viability of the link under
> greater load, greater message size. It may also be used for MTU
> discovery, opening SCTP windows, etc.
> 
> >
> >So I have the sending and receiving of LS Proving messages going on in
> >the PROVING state.
> >
> 
> M2PA PROVING procedure matches Q.703, but INITIAL_ALIGNMENT
> does not currently match, leading to the confusion over when
> to send/receive LSA and LSPx. That is why I suggest that the
> NOT_ALIGNED and ALIGNED states be used in place of the singular
> INITIAL_ALIGNMENT state.
> 
> >
> >BTW, I would be very happy to reference Q.703 wherever possible and am
> >open to suggestions on how to do it. Unfortunately, there are parts of
> >the SDLs that do not apply to M2PA (e.g., sending FISUs), so I have been
> >stumped on how to do this.
> 
> Yes, I understand that it is a challenge. In section "4.1.3 Link
> Alignment", I think it would be helpful to reference the Q.703 SDLs
> as a guide and list the basic equivalence between the M2PA and
> MTP2 primitives:
> 
> LSA  == SIO
> LSB  == SIB
> LSO  == SIOS
> LSPN == SIN
> LSPE == SIE
> LSPO == SIPO
> Data == MSU
> 
> For each paragraph in section 4.1.3, one could reference the applicable
> SDL figure and sheet. This may be excessive, but I think it would add
> clarity. Here follows my attempt to revise the draft based on the
> recent alignment and LPO discussions:
> 
> ----------------
> OLD TEXT 4.1.3
> After the association is established, M2PA SHALL send a Link Status
> Out of Service message to its peer.
> 
> NEW TEXT 4.1.3
> After the association is established, M2PA SHALL send a Link Status
> Out of Service message to its peer and, prior to receiving a Start
> Request from MTP3, MAY periodically send Link Status Out of
> Service messages ([Q.703] Figure 4 and Figure 8 Sheet 1). During
> the Out of Service State M2PA SHOULD ignore received Link Status
> Out of Service messages and SHOULD respond to all other received
> messages by sending a Link Status Out of Service message."
> 
> 
> ----------------
> OLD TEXT 4.1.3
> 
> Once the association is established and M2PA has received a Start
> Request from MTP3, M2PA sends the Link Status Alignment message to
> its peer. If M2PA has not already received the Link Status Alignment
> message from its peer, then M2PA starts timer T2.
> 
> NEW TEXT 4.1.3
> 
> Once the association is established and M2PA has received a Start
> Request from MTP3, M2PA sends the Link Status Alignment message to
> its peer and moves to the NOT_ALIGNED state ([Q.703] Figure 9
> Sheet 1).
> 
> 
> ----------------
> OLD TEXT 4.1.3
> M2PA stops timer T2 when it has received the Link Status Alignment
> message from its peer.
> 
> NEW TEXT 4.1.3
> M2PA stops timer T2 when it has received a Link Status Alignment
> message or Link Status Proving Normal or Link Status Emergency
> message from its peer. It sends a Link Status Proving message,
> starts timer T3, and then moves to the ALIGNED state ([Q.703]
> Figure 9 Sheet 2).
> 
> 
> ----------
> OLD TEXT 4.1.3
> Note: Between the time M2PA sends Link Status Alignment to its peer
> and receives Link Status Alignment from its peer, M2PA may receive
> Link Status Out of Service from its peer. This message is
> ignored. After receiving Link Status Alignment from the peer, receipt
> of a Link Status Out of Service message causes M2PA to send Out of
> Service to MTP3 and return to the Out of Service state.
> 
> NEW TEXT 4.1.3
> Note: During the NOT_ALIGNED state M2PA may receive Link Status Out
> of Service from its peer. This message is ignored. After moving to
> the ALIGNED state, receipt of a Link Status Out of Service message
> causes M2PA to send Out of Service to MTP3 and return to the Out of
> Service state ([Q.703] Figure 9 Sheet 4).
> 
> ----------
> OLD TEXT 4.1.3
> 
> When M2PA has both sent and received the Link Status Alignment
> message, it has completed alignment.  M2PA starts the aligned timer
> T3 and moves to the proving state.
> 
> M2PA stops timer T3 when it receives a Proving Normal or Proving
> Emergency message and starts proving period timer T4.
> 
> If timer T3 expires, then M2PA reports to MTP3 that the link is out of
> service.  M2PA sends Link Status Out of Service to its peer.  M2PA
> SHOULD leave the association established.  M2PA waits for MTP3 to
> initiate the alignment procedure again.
> 
> During the proving period (i.e., after M2PA starts the proving period
> timer T4), M2PA sends Link Status Proving messages to its peer at an
> interval defined by the protocol parameter Proving_Interval. M2PA
> sends either the Proving Normal or Proving Emergency message,
> according to the Emergency and Emergency Ceases commands from
> MTP3. M2PA uses the value of T4 corresponding to the Normal or
> Emergency condition. However, if M2PA receives a Link Status Proving
> Emergency message from its peer, then M2PA SHALL initiate the
> Emergency proving period value for T4, but it SHALL continue to send
> the Proving message (Normal or Emergency) determined by its own upper
> layer MTP3.
> 
> NEW TEXT 4.1.3
> 
> Upon receiving either Link Status Proving Normal or Link Status
> Proving Emergency during the ALIGNED state, M2PA stops timer T3,
> starts proving period timer T4, and enters the PROVING state
> ([Q.703] Figure 9 Sheet 3).
> 
> If timer T3 expires, then M2PA reports to MTP3 that the link is
> out of service, sends Link Status Out of Service to its peer and
> moves to the OUT_OF_SERVICE state ([Q.703 Figure 9 Sheet 4). M2PA
> SHOULD leave the association established.
> 
> During the PROVING state M2PA periodically sends either Link
> Status Proving Normal or Link Status Proving Emergency according
> to the Emergency and Emergency Ceases commands from MTP3
> ([Q.703] Figure 9 Sheets 5 and 6). M2PA uses the value of T4
> corresponding to the Normal or Emergency condition. However, if
> M2PA receives a Link Status Proving Emergency message from its
> peer, then M2PA SHALL initiate the Emergency proving period value
> for T4, but it SHALL continue to send the Proving message
> (Normal or Emergency) determined by its own upper layer MTP3
> ([Q.703] Figure 9 Sheet 6).
> 
> Note: If M2PA receives a Link Status Alignment message during the
> PROVING state, then M2PA stops timer T4, starts timer T3, and
> moves to the ALIGNED state ([Q.703] Figure 9 Sheet 5).
> 
> 
> ************************************************
> NOTE: There seems to be a differences between M2PA
> and Q.703  concerning reception of LSR or LSPO
> during PROVING. Currently M2PA indicates these
> events should be recorded. This memory is used to
> move from PROVING straight to either IN_SERVICE or
> PROCESSOR_OUTAGE without starting timer T1. Unless
> I am mistaken, Q.703 has no such memory -- these
> events are ignored during PROVING, and timer T1
> is always started upon expiry of T4 (end of
> proving). Changing this in M2PA will require that
> the periodic sending of LSR or LSPO become SHOULD
> rather than MAY. Leaving this unchanged should
> require that the M2PA state machine figure change
> to show transitions from PROVING to either
> ALIGNED_READY or ALIGNED_NOT_READY or IN_SERVICE or
> PROCESSOR_OUTAGE. I don't know how to make this
> change to the figure -- no room left.
> 
> I'm keeping the memory of these events during
> PROVING (status quo) in the following revision.
> ************************************************
> 
> ----------
> OLD TEXT 4.1.3
> 
> When the proving period timer T4 expires, M2PA changes to the Aligned
> Ready state. M2PA SHALL start the timer T1 and send a Link Status
> Ready message to its peer. These Link Status Ready message is used to
> verify that both ends have completed proving. M2PA MAY send additional
> Link Status Ready messages while timer T1 is running.
> 
> NEW TEXT 4.1.3
> 
> If M2PA receives either Link Status Ready or Link Status
> Processor Outage during PROVING, then it records the event
> for processing upon expiry of timer T4.
> 
> When timer T4 expires and a Link Status Ready message has
> been received, then M2PA, depending on local processor
> outage state, either sends a Link Status Processor Outage
> message and moves to the PROCESSOR_OUTAGE state or sends
> a Link Status Ready message and moves to the IN_SERVICE
> state.
> 
> When timer T4 expires and a Link Status Processor Outage
> message has been received, then M2PA notifies MTP3 of the
> remote processor outage condition, and, depending on local
> processor outage status, M2PA either sends a Link Status
> Processor Outage message or sends a Link Status Ready
> message, and then moves to the PROCESSOR_OUTAGE state.
> 
> When timer T4 expires and neither Link Status Ready nor
> Link Status Processor Outage have been received, then M2PA
> starts timer T1. If a Local Processor Outage Notification
> had been received from MTP3, then M2PA sends a Link Status
> Processor Outage message and moves to the ALIGNED_NOT_READY
> state, otherwise it sends a Link Status Ready message and
> moves to the ALIGNED_READY state ([Q.703] Figure 8 Sheet 5).
> 
> During the ALIGNED_READY state, M2PA MAY periodically send
> Link Status Ready messages.
> 
> During the ALIGNED_NOT_READY and PROCESSOR_OUTAGE states,
> M2PA MAY periodically send Link Status Processor Outage
> messages.
> 
> 
> ----------
> OLD TEXT 4.1.3
> 
> M2PA SHALL stop timer T1 when it receives a Link Status Ready or User
> Data message from its peer. If timer T1 expires, then M2PA reports to
> MTP3 that the link is out of service. M2PA sends Link Status Out of
> Service to its peer. M2PA SHOULD leave the association
> established. M2PA waits for MTP3 to initiate the alignment procedure
> again.
> 
> Note that if M2PA has already received a Link Status Ready message
> from its peer when its timer T4 expires, there is no need to start
> timer T1. M2PA can just send Link Status Ready to the peer and
> continue along.
> 
> NEW TEXT 4.1.3
> 
> M2PA SHALL stop timer T1 when it receives either a Link Status Ready
> or User Data or Link Status Processor Outage message from its peer
> ([Q.703] Figure 8 Sheets 7 and 9). If either local or remote processor
> outage is in effect, then M2PA moves to the PROCESSOR_OUTAGE state,
> otherwise it moves to the IN_SERVICE state ([Q.703] Figure 8 Sheet 7).
> 
> If timer T1 expires, then M2PA reports to MTP3 that the link is
> out of service. M2PA sends Link Status Out of Service to its peer,
> and moves to the OUT_OF_SERVICE state. M2PA SHOULD leave the
> association established.
> 
> 
> ----------
> OLD TEXT 4.1.3
> 
> Note that if M2PA has already received a Link Status Ready message
> from its peer when its timer T4 expires, there is no need to start
> timer T1. M2PA can just send Link Status Ready to the peer and
> continue along.
> 
> 
> NEW TEXT 4.1.3
> [deleted -- content moved to PROVING discussion]
> 
> 
> ----------
> OLD TEXT 4.1.3
> 
> If there is a local processor outage condition during the alignment
> procedure, M2PA sends Link Status Processor Outage to its peer. When
> the local processor outage condition ends, then M2PA SHALL send Link
> Status Processor Outage Ended to its peer. M2PA SHALL attempt to
> complete the alignment procedure during the local processor outage
> condition.
> 
> If M2PA receives a Link Status Processor Outage during alignment, and
> M2PA had received a Start Request from its MTP3, M2PA SHALL report
> Remote Processor Outage to MTP3. M2PA SHALL attempt to complete the
> alignment procedure during the remote processor outage condition.
> 
> NEW TEXT 4.1.3
> 
> [deleted -- Behavior has been clarified above. LSPO should only
> be received during PROVING or ALIGNED_READY or ALIGNED_NOT_READY states.
> LSPO should only be sent during ALIGNED_NOT_READY or PROCESSOR_OUTAGE
> states].
> 
> ----------
> OLD TEXT 4.1.3
> 
> When all of the following are true:
> 
>    (a) M2PA has received a Start Request from MTP3.
> 
>    (b) M2PA's proving period T4 has expired.
> 
>    (c) M2PA has sent a Link Status Ready to its peer.
> 
>    (d) M2PA has received a Link Status Ready OR User Data
>        message from its peer.
> 
>    (e) M2PA has not received Link Status Out of Service from its peer
>        since it received Link Status Alignment.
> 
>    then M2PA SHALL send Link In Service to its MTP3.
> 
> NEW TEXT 4.1.3
> 
> When all of the following are true:
> 
>    (a) M2PA has received a Start Request from MTP3.
> 
>    (b) M2PA has not received a Stop Request from MTP3
>        since the Start Request.
> 
>    (c) The association has remained established.
> 
>    (d) M2PA has received EITHER (a Link Status Aligned and
>        a Link Status Proving message)) OR (two Link
>        Status Proving messages).
> 
>    (e) M2PA's proving period timer T4 has expired.
> 
>    (f) M2PA has received a Link Status Ready OR User Data
>        OR Link Status Processor Outage Ended message from
>        its peer.
> 
>    (g) M2PA has not received Link Status Out of Service from
>        its peer since the ALIGNED state.
> 
>    then M2PA SHALL send Link In Service to its MTP3.
> 
> 
> 
> -----------------
> OLD TEXT 4.1.4
> 
> When M2PA detects a local processor outage, it sends a Link Status
> message to its peer with status Processor Outage. M2PA SHALL also
> cease sending User Data messages to SCTP for transmission. M2PA SHALL
> stop acknowledging received User Data messages.
> 
> M2PA MAY send additional Link Status Processor Outage messages as long
> as there is a local processor outage and the link is in service. If
> the link is out of service, M2PA SHOULD locally mark that it is in
> local processor outage.
> 
> NEW TEXT 4.1.4
> 
> During PROCESSOR_OUTAGE or IN_SERVICE states, when MTP3 notifies
> M2PA of local processor outage, M2PA sends a Link Status Processor
> Outage message to its peer and moves to the PROCESSOR_OUTAGE state
> ([Q.703] Figure 8 Sheet 11). M2PA SHALL also cease sending User Data
> messages to SCTP for transmission. M2PA SHALL stop acknowledging
> received User Data messages.
> 
> M2PA MAY periodically send Link Status Processor Outage messages
> during the PROCESSOR_OUTAGE state while local processor outage
> is in effect.
> 
> 
> -----------------
> OLD TEXT 4.1.4
> 
> MTP3 may send a Flush Buffers or Continue command to M2PA as part of
> its processor outage procedure (See section 4.2.4 Flush Buffers and
> Continue). The use of Flush Buffers and Continue is specified in Q.703
> [Q.703] and T1.111.3 [T1.111].
> 
> Alternatively, MTP3 may perform data retrieval as part of a changeover
> procedure.
> 
> When the processor outage ceases, MTP3 sends a Local Processor
> Recovered indication to M2PA. The local M2PA notifies its peer by
> sending a Link Status message with status Processor Outage Ended. The
> local M2PA SHALL resume receiving and transmitting messages.
> 
> 
> NEW TEXT 4.1.4
> 
> MTP3 notifies M2PA of the onset of local processor outage condition
> by sending a Local Processor Outage Request. Prior to the
> PROCESSOR_OUTAGE state, M2PA records this event for future
> processing. If this event is received during PROCESSOR_OUTAGE, then
> M2PA sends a Link Status Processor Outage message to its peer and
> records the event ([Q.703] Figure 8 Sheet 12). If this event is
> received during IN_SERVICE, then M2PA sends a Link Status Processor
> Outage message to its peer, records the local processor outage
> state, and moves to the PROCESSOR_OUTAGE state ([Q.703] Figure 8
> Sheet 11).
> 
> MTP3 notifies M2PA of the clearing of the local processor condition
> by sending the Local Processor Outage Recovered Request. Prior to
> the PROCESSOR_OUTAGE state, M2PA records this event by clearing its
> record of the local processor outage condition. When this
> indication is received during PROCESSOR_OUTAGE state, M2PA sends
> a Link Status Processor Outage Ended message to its peer
> ([Q.703] Figure 8 Sheet 13).
> 
> M2PA timers T6 and T7 are stopped and remote busy status is
> cleared upon transition to the PROCESSOR_OUTAGE state.
> 
> During PROCESSOR_OUTAGE if a Link Status Processor Outage Ended
> message is received from the peer, M2PA notifies MTP3 that
> remote processor has recovered ([Q.703] Figure 8 Sheet 12).
> 
> The use of Flush Buffers and Continue is specified in Q.703
> [Q.703] and T1.111.3 [T1.111].
> 
> During PROCESSOR_OUTAGE if a Flush Buffers Request is received
> from MTP3, then M2PA flushes the transmit message buffers and
> records the event for future processing ([Q.703] Figure 8
> Sheet 13).
> 
> During PROCESSOR_OUTAGE if a Continue Request is received
> from MTP3, then M2PA records the event for future processing
> ([Q.703] Figure 8 Sheet 13).
> 
> If during PROCESSOR_OUTAGE M2PA determines that neither local
> nor remote processor outage is in effect and either a Flush Buffers
> Request or a Continue Request was received from MTP3, then it moves
> to the IN_SERVICE state and begins receiving and transmitting
> normally ([Q.703] Figure 8 Sheet 13).
> 
> 
> >
> >tomG.
> 
> 
> jeffrey.craig@tekelec.com wrote:
> >
> > Hello Franz,
> >
> > >an remark to your M2PA state event machine. In the case
> > >'Proving' (T4 expired) is finished and local Processor
> > >Outage is occurred, I missing an arrow from state
> > >'PROVING' to state 'ALIGNED_NOT_READY'.
> >
> > You are correct, it's an error. Here is another try....
> > (It would be a lot easier if M2PA simply referred to Q.703 SDLs.)
> >
> > Regards,
> >
> > Jeff
> >
> >                    +-----------+
> >                    |   POWER   |
> >                    |    OFF    |
> >                    +-----------+
> >                          |
> >                          | Power On
> >                          |
> >        Flush Buffers     |  +-----------------------------------------+
> >        OR Retrieval      |  |                                         |
> >               +-------+  |  |                                         |
> >               |       |  V  V                                         |
> >               |    +-----------+                                      |
> >               +--->|  OUT OF   |                                      |
> >   +--------------->|  SERVICE  |<--+                                  |
> >   |                +-----------+   | Link Configured                  |
> >   |                      |   |     | (Associate)                      |
> >   |                      |   +-----+                                  |
> >   |     Receive LS OOS   |                                            |
> >   |           +-------+  | MTP3 Start                                 |
> >   |           |       |  V                                            |
> >   |           |    +-----------+                                      |
> >   |           +--->|    NOT    |                                      |
> >   +<---------------|  ALIGNED  |------------------------------------->+
> >   |     MTP3 Stop  +-----------+     SCTP Comm Error                  |
> >   |  OR T2 Expiry        |        OR SCTP Comm Lost                   |
> >   |                      |                                            |
> >   |                      |    Receive LS Align                        |
> >   |                      | OR Receive LS Prov Norm                    |
> >   |     Receive LS Align | OR Receive LS Prov Emer                    |
> >   |           +-------+  |                                            |
> >   |           |       |  V                                            |
> >   |           |    +-----------+                                      |
> >   |           +--->|  ALIGNED  |                                      |
> >   +<---------------|           |------------------------------------->+
> >   |     MTP3 Stop  +-----------+     SCTP Comm Error                  |
> >   |  OR T3 Expiry      ^    |     OR SCTP Comm Lost                   |
> >   |  OR Receive LS OOS |    |                                         |
> >   |                    |    |    Receive LS Prov Norm                 |
> >   |                    |    | OR Receive LS Prov Emer                 |
> >   |                    |    |                                         |
> >   |   Receive LS Align |    |                                         |
> >   |                    |    |                                         |
> >   |                    |    |                                         |
> >   | Receive LS Prov X  |    |                                         |
> >   |           +------+ |    |                                         |
> >   |           |      | |    V                                         |
> >   |           |    +-----------+                                      |
> >   |           +--->|  PROVING  |------------------------------------->+
> >   |<---------------|           |------+          SCTP Comm Error      |
> >   |     MTP3 Stop  +-----------+      |         OR SCTP Comm Lost     |
> >   |  OR Receive LS OOS   |            |                               |
> >   |                      |  T4 Expiry |                               |
> >   |                      |   (LPO)    |                               |
> >   |                      |            |                               |
> >   |                      |            | MTP3 Stop            +------->+
> >   |                      |            | OR Receive LS OOS    |        |
> >   |                      |            | OR SCTP Comm Error   |        |
> >   |                      |            | OR SCTP Comm Lost    |        |
> >   |                      |            | OR T1 Expiry         |        |
> >   |           T4 Expiry  |            | OR M2PA Link Failure |        |
> >   |            (No LPO)  |            | OR Receive LS Align  |        |
> >   |                      V            |                      |        |
> >   |                +-----------+ LPO  V            +-----------+      |
> >   |                |  ALIGNED  |------+----------->|  ALIGNED  |      |
> >   +<---------------|   READY   |<------------------| NOT READY |      |
> >   |     MTP3 Stop  +-----------+       LPR         +-----------+      |
> >   |  OR T1 Expiry        |   |                         | Rcv LS Ready |
> >   |  OR Receive LS OOS   |   |                         | OR Rcv Data  |
> >   |  OR Receive LS Align |   +------------------------>+ OR Rcv LS PO |
> >   |  OR SCTP Comm Error  |   Receive LS PO             |              |
> >   |  OR SCTP Comm Lost   |                             |              |
> >   |                      |    Receive LS Ready         |              |
> >   |                      | OR Receive User Data        |              |
> >   |                      V                             V              |
> >   |                +-----------+ LPO or Rcv LS PO  +-----------+      |
> >   |                |    IN     |------------------>| PROCESSOR |      |
> >   +<---------------|  SERVICE  |<------------------|  OUTAGE   |      |
> >                    +-----------+    MTP3 Flush     +-----------+      |
> >        MTP3 Stop                 OR MTP3 Continue            |        |
> >     OR Receive LS OOS                                        |        |
> >     OR SCTP Comm Error                     MTP3 Stop         |        |
> >     OR SCTP Comm Lost                   OR Receive LS OOS    |        |
> >     OR T6 Expiry                        OR SCTP Comm Error   |        |
> >     OR T7 Expiry                        OR SCTP Comm Lost    |        |
> >     OR M2PA Link Failure                OR M2PA Link Failure |        |
> >     OR Receive LS Align                 OR Receive LS Align  +------->+
> 
> --
> * Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
> * 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
> * Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
> * I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
> * to advertise or represent its position,  operations, or activities. *
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 
> 
> 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 19 13:16:35 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA03839
	for <sigtran-archive@odin.ietf.org>; Wed, 19 Feb 2003 13:16:35 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1JIMkg31796
	for sigtran-archive@odin.ietf.org; Wed, 19 Feb 2003 13:22:46 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1JIMBp31786;
	Wed, 19 Feb 2003 13:22:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1JIKBp31732
	for <sigtran@optimus.ietf.org>; Wed, 19 Feb 2003 13:20:11 -0500
Received: from w2ksjexg01.ciena.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA03745
	for <sigtran@ietf.org>; Wed, 19 Feb 2003 13:13:29 -0500 (EST)
Received: from wntcsdexg01.csd.ciena.com ([10.34.31.31]) by w2ksjexg01.ciena.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2653.13)
	id FG7XGAZY; Wed, 19 Feb 2003 10:17:02 -0800
Received: by webdev-llnt.oni.com with Internet Mail Service (5.5.2653.19)
	id <Y5W9M7CZ>; Wed, 19 Feb 2003 10:17:17 -0800
Message-ID: <2135200C183FD5119588009027DE572302836CD0@webdev-owa.oni.com>
From: "Ong, Lyndon" <LyOng@ciena.com>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Date: Wed, 19 Feb 2003 10:17:08 -0800
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Sigtran] status of security framework
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Hi Folks,

FYI, I have requested that the security framework
(draft-ietf-sigtran-security-01.txt)
be considered for Proposed Standard by our A-Ds and the rest of the IESG.
Thanks
very much to Michael Tuexen, John Loughney and Javier Pastor for their work
on this.

Cheers,

L. Ong
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 19 14:55:34 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA06944
	for <sigtran-archive@odin.ietf.org>; Wed, 19 Feb 2003 14:55:34 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1JK1mI06034
	for sigtran-archive@odin.ietf.org; Wed, 19 Feb 2003 15:01:48 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1JK15p06003;
	Wed, 19 Feb 2003 15:01:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1JJxTp05915
	for <sigtran@optimus.ietf.org>; Wed, 19 Feb 2003 14:59:29 -0500
Received: from rainier.illuminet.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA06806
	for <sigtran@ietf.org>; Wed, 19 Feb 2003 14:52:44 -0500 (EST)
Received: from olwinexsmtp01.corp.illuminet.com (olwinexsmtp01.corp.illuminet.com [10.55.13.9]) by rainier.illuminet.com (8.8.8/8.8.8) with ESMTP id LAA04560; Wed, 19 Feb 2003 11:55:19 -0800 (PST)
Received: by olwinexsmtp01.corp.illuminet.com with Internet Mail Service (5.5.2656.59)
	id <192QY9CP>; Wed, 19 Feb 2003 11:55:19 -0800
Message-ID: <90B62898D7B43B448D1A7297980A8C5F6CE5A2@opwinex01.corp.illuminet.com>
From: Kevin McCandless <KMcCandless@verisign.com>
To: "'Ong, Lyndon'" <LyOng@ciena.com>,
        "'sigtran@ietf.org'"
	 <sigtran@ietf.org>
Subject: RE: [Sigtran] Last Call on signaling applicability statement
Date: Wed, 19 Feb 2003 11:55:18 -0800
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2656.59)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Lyndon:

It appears the list is still working out issues with M2PA.  Should there be
a version 8 of M2PA to address current comments and another short period of
review?

Thanks,

kevin....

> -----Original Message-----
> From: Ong, Lyndon [mailto:LyOng@ciena.com]
> Sent: Tuesday, February 18, 2003 4:51 PM
> To: 'sigtran@ietf.org'
> Subject: [Sigtran] Last Call on signaling applicability statement
> 
> 
> Hi Folks,
> 
> The Last Call on the security framework has now closed.  Last Call on
> M2PA actually closed last week, although I saw a few more comments
> dribble in.
> 
> I've had a request now to start Last Call on the Signaling 
> Applicability
> Statement,
> draft-ietf-sigtran-signalling-over-sctp-applic-07.txt
> 
> Please send in your comments starting today and ending Last 
> Call on Tuesday,
> March 4th.
> 
> Thanks,
> 
> L. Ong
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 19 15:58:54 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA08660
	for <sigtran-archive@odin.ietf.org>; Wed, 19 Feb 2003 15:58:54 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1JL59M09620
	for sigtran-archive@odin.ietf.org; Wed, 19 Feb 2003 16:05:09 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1JL4mp09566;
	Wed, 19 Feb 2003 16:04:49 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1JL34p09531
	for <sigtran@optimus.ietf.org>; Wed, 19 Feb 2003 16:03:04 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA08595
	for <sigtran@ietf.org>; Wed, 19 Feb 2003 15:56:18 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1JL06eW011467;
	Wed, 19 Feb 2003 15:00:06 -0600 (CST)
Message-ID: <3E53F056.B1C25EB9@alcatel.com>
Date: Wed, 19 Feb 2003 15:00:06 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: M2PA version 8? (was: Re: [Sigtran] Last Call on signaling applicability 
 statement)
References: <90B62898D7B43B448D1A7297980A8C5F6CE5A2@opwinex01.corp.illuminet.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

This looks like the way to do it. I could release version 8 when I have
the current batch of comments resolved.

tomG.

Kevin McCandless wrote:
> 
> Lyndon:
> 
> It appears the list is still working out issues with M2PA.  Should there be
> a version 8 of M2PA to address current comments and another short period of
> review?
> 
> Thanks,
> 
> kevin....
> 
> > -----Original Message-----
> > From: Ong, Lyndon [mailto:LyOng@ciena.com]
> > Sent: Tuesday, February 18, 2003 4:51 PM
> > To: 'sigtran@ietf.org'
> > Subject: [Sigtran] Last Call on signaling applicability statement
> >
> >
> > Hi Folks,
> >
> > The Last Call on the security framework has now closed.  Last Call on
> > M2PA actually closed last week, although I saw a few more comments
> > dribble in.
> >
> > I've had a request now to start Last Call on the Signaling
> > Applicability
> > Statement,
> > draft-ietf-sigtran-signalling-over-sctp-applic-07.txt
> >
> > Please send in your comments starting today and ending Last
> > Call on Tuesday,
> > March 4th.
> >
> > Thanks,
> >
> > L. Ong
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 19 16:04:08 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA08808
	for <sigtran-archive@odin.ietf.org>; Wed, 19 Feb 2003 16:04:08 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1JLANd10496
	for sigtran-archive@odin.ietf.org; Wed, 19 Feb 2003 16:10:23 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1JLA7p10483;
	Wed, 19 Feb 2003 16:10:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1JL9Kp10453
	for <sigtran@optimus.ietf.org>; Wed, 19 Feb 2003 16:09:20 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA08755
	for <sigtran@ietf.org>; Wed, 19 Feb 2003 16:02:33 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1JL6MeW013015;
	Wed, 19 Feb 2003 15:06:23 -0600 (CST)
Message-ID: <3E53F1CE.D8849AA8@alcatel.com>
Date: Wed, 19 Feb 2003 15:06:22 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft 07 comment: LPO
References: <OF71910696.3AE98EF8-ON85256CD2.00448044@nc.tekelec.com> <20030219093504.A16359@openss7.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

So where do we go with this? IMHO referencing 2 or even 3 documents in
each paragraph would be unwieldy. 

tomG.

"Brian F. G. Bidulock" wrote:
> 
> Jeffrey,
> 
> That would be fine, but ANSI T1.111.3 is different than Q.703, particularly
> as regards processor outage handling and clear buffers/resume behavior.
> Did you want to add SDL references for ANSI T1.111.3?  An what about JTTC
> (which is rather different again)?
> 
> I agree that M2PA should not conflict with the SDLs, but referencing just
> Q.703 seems inadequate.
> 
> --brian
  [...]

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 19 16:09:17 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA08928
	for <sigtran-archive@odin.ietf.org>; Wed, 19 Feb 2003 16:09:17 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1JLFWC10653
	for sigtran-archive@odin.ietf.org; Wed, 19 Feb 2003 16:15:32 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1JLFCp10640;
	Wed, 19 Feb 2003 16:15:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1JLERp10582
	for <sigtran@optimus.ietf.org>; Wed, 19 Feb 2003 16:14:27 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA08858
	for <sigtran@ietf.org>; Wed, 19 Feb 2003 16:07:40 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1JLBTeW014046;
	Wed, 19 Feb 2003 15:11:29 -0600 (CST)
Message-ID: <3E53F301.1D46F297@alcatel.com>
Date: Wed, 19 Feb 2003 15:11:29 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
References: <3E52F4FC.1030303@pt.com> <20030218202449.A30468@openss7.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Here is my new improved wording in 4.2.6:

For data retrieval, MTP3 requests the Backward Sequence Number to be
Transmitted (BSNT) from M2PA through the Retrieve BSNT request. M2PA
determines the Forward Sequence Number of the last User Data message
received from the peer and delivered to the upper layer MTP3. This
value is the BSNT. 

tomG.

"Brian F. G. Bidulock" wrote:
> 
> Andrew,
> 
> The draft is incorrect.
> 
> The BSNT is the last acknowledged received DATA (data should not be
> acknowledged unless it is locally "deliverable" to MTP3).
> 
> Otherwise there is a slim chance that message loss will occur on changeover
> resulting from processor outage or busy conditions.
> 
> The statement in section 4.2.6 should be changed to reflect this.
> 
> --brian
> 
> Andrew Booth wrote:                                                                 (Tue, 18 Feb 2003 22:07:40)
> > Hi,
> >
> > I posted these a while ago and haven't seen any responses.  Question (2)
> > is really the only relevant one.
> >
> > Thanks for any info,
> > Andrew
> >
> > --
> > Andrew Booth, Software Developer
> > Signaling Systems Group, Performance Technologies
> > 613-237-4344
> >
> > Visit www.pt.com/mailing.html to subscribe to our Packets & Signals eLetter.
> >
> 
> > From: Andrew Booth <abooth@pt.com>
> > To: sigtran@ietf.org
> > Date: Thu, 06 Feb 2003 12:31:41 -0500
> > Subject: [Sigtran] More M2PA 7 questions
> >
> > Hi All,
> >
> > A few more questions on M2PA draft 7.
> >
> > 1) Does every User Data received from the peer increases the expected FSN
> > for the subsequent message, even in cases where the received message is
> > later
> > discarded as part of a Flush Buffers or Changeover procedure?
> >
> > 2) Suppose that an M2PA implementation has received messages from a peer
> > and that some of the messages are buffered at the M2PA layer (perhaps it is
> > in processor outage).  Suppose now the link transitions to OOS and the
> > MTP3 sends a Retrieve BSNT to M2PA.
> >
> > According to 4.2.6,
> >
> >    "M2PA determines the Forward Sequence Number of the last User Data
> >    message received from the peer.  This is the BSNT."
> >
> > In the M2PA situation it seems more appropriate to respond with the FSN
> > of the last message which was delivered to MTP3.
> > Am I missing something?
> >
> > 3) How should T7 be handled when remote processor outage abates?  Should
> > we immediately check for excessive ack delay, or reset T7 somehow?
> >
> > 4) In section 5.1 we transition to proving without starting T3 and to
> > aligned ready without receiving LS Proving.  Are these omissions
> > intentional?
> >
> > 5) In section 5.8 there is some introductory text :
> >
> > "Note that in this example, the sequence numbers and messages requested
> > by MTP3 are sent from SCTP to M2PA in the Communication Lost
> > primitive. In general, the retrieval of sequence numbers and messages
> > is implementation dependent."
> >
> > Why are messages retrieved from SCTP?  Is this text indicating that the
> > M2PA retransmit buffer is implemented in the SCTP?  Just a little confused.
> >
> > 6) In section 4.2.6, is the following correct or should it be rephrased
> > in the context of a circular buffer?
> > "any transmitted User Data messages beginning with the first
> >        unacknowledged message with FSN greater than FSNC."
> >
> >
> > Thanks for any info,
> > Andrew
> >
> > --
> > Andrew Booth, Software Developer
> > Signaling Systems Group, Performance Technologies
> > 613-237-4344
> >
> > Visit www.pt.com/mailing.html to subscribe to our Packets & Signals eLetter.
> >
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> 
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 19 17:00:26 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA10274
	for <sigtran-archive@odin.ietf.org>; Wed, 19 Feb 2003 17:00:26 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1JM6gU13474
	for sigtran-archive@odin.ietf.org; Wed, 19 Feb 2003 17:06:42 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1JM6Rp13463;
	Wed, 19 Feb 2003 17:06:27 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1JM5wp13442
	for <sigtran@optimus.ietf.org>; Wed, 19 Feb 2003 17:05:58 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA10234
	for <sigtran@ietf.org>; Wed, 19 Feb 2003 16:59:10 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1JM2xJT022944;
	Wed, 19 Feb 2003 16:02:59 -0600 (CST)
Message-ID: <3E53FF13.8A566E9A@alcatel.com>
Date: Wed, 19 Feb 2003 16:02:59 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
References: <3E52F4FC.1030303@pt.com> <20030218202449.A30468@openss7.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Andrew and others,

Here are some comments on the other items....

tomG.

> Hi All,
> 
> A few more questions on M2PA draft 7.
> 
> 1) Does every User Data received from the peer increases the expected FSN
> for the subsequent message, even in cases where the received message is 
> later
> discarded as part of a Flush Buffers or Changeover procedure?

I am not sure what you are asking. Are you asking if the FSNs are ever
reset? 

> 
> 2) Suppose that an M2PA implementation has received messages from a peer
> and that some of the messages are buffered at the M2PA layer (perhaps it is
> in processor outage).  Suppose now the link transitions to OOS and the
> MTP3 sends a Retrieve BSNT to M2PA.
> 
> According to 4.2.6,
> 
>    "M2PA determines the Forward Sequence Number of the last User Data
>    message received from the peer.  This is the BSNT."
> 
> In the M2PA situation it seems more appropriate to respond with the FSN
> of the last message which was delivered to MTP3.
> Am I missing something?

This one is discussed in another message.

> 
> 3) How should T7 be handled when remote processor outage abates?  Should 
> we immediately check for excessive ack delay, or reset T7 somehow?

From our earlier discussion of T7 during Busy, I think the answer is
that T7 is not running during PO. Any messages would be flushed, so T7
would not start again until a new message was sent. 
 

> 
> 4) In section 5.1 we transition to proving without starting T3 and to 
> aligned ready without receiving LS Proving.  Are these omissions 
> intentional?

These were unintentionally omitted. I added those events to the diagram.


> 
> 5) In section 5.8 there is some introductory text :
> 
> "Note that in this example, the sequence numbers and messages requested
> by MTP3 are sent from SCTP to M2PA in the Communication Lost
> primitive. In general, the retrieval of sequence numbers and messages
> is implementation dependent."
> 
> Why are messages retrieved from SCTP?  Is this text indicating that the 
> M2PA retransmit buffer is implemented in the SCTP?  Just a little confused.

This is left over from an earlier version. I removed that paragraph.


> 
> 6) In section 4.2.6, is the following correct or should it be rephrased 
> in the context of a circular buffer?
> "any transmitted User Data messages beginning with the first
>        unacknowledged message with FSN greater than FSNC." 

There is a general warning about this in 4.2.1.


> 
> 
> Thanks for any info,
> Andrew
> 
> -- 
> Andrew Booth, Software Developer
> Signaling Systems Group, Performance Technologies
> 613-237-4344



-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 19 17:12:08 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA10876
	for <sigtran-archive@odin.ietf.org>; Wed, 19 Feb 2003 17:12:08 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1JMIPE14888
	for sigtran-archive@odin.ietf.org; Wed, 19 Feb 2003 17:18:25 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1JMIBp14881;
	Wed, 19 Feb 2003 17:18:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1JMHtp14836
	for <sigtran@optimus.ietf.org>; Wed, 19 Feb 2003 17:17:55 -0500
Received: from w2ksjexg01.ciena.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA10845
	for <sigtran@ietf.org>; Wed, 19 Feb 2003 17:11:07 -0500 (EST)
Received: from wntcsdexg01.csd.ciena.com ([10.34.31.31]) by w2ksjexg01.ciena.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2653.13)
	id F2XT8WR3; Wed, 19 Feb 2003 14:14:41 -0800
Received: by webdev-llnt.oni.com with Internet Mail Service (5.5.2653.19)
	id <Y5W9M7Q1>; Wed, 19 Feb 2003 14:14:56 -0800
Message-ID: <2135200C183FD5119588009027DE572302836CDA@webdev-owa.oni.com>
From: "Ong, Lyndon" <LyOng@ciena.com>
To: "'Kevin McCandless'" <KMcCandless@verisign.com>,
        "Ong, Lyndon"
	 <LyOng@ciena.com>,
        "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: RE: [Sigtran] Last Call on signaling applicability statement
Date: Wed, 19 Feb 2003 14:14:50 -0800
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Hi Kevin,

I believe that Tom is working on an updated version for
next week sometime.  Hopefully that version can be sent
to IESG for consideration.

Cheers,

Lyndon

-----Original Message-----
From: Kevin McCandless [mailto:KMcCandless@verisign.com]
Sent: Wednesday, February 19, 2003 11:55 AM
To: 'Ong, Lyndon'; 'sigtran@ietf.org'
Subject: RE: [Sigtran] Last Call on signaling applicability statement


Lyndon:

It appears the list is still working out issues with M2PA.  Should there be
a version 8 of M2PA to address current comments and another short period of
review?

Thanks,

kevin....

> -----Original Message-----
> From: Ong, Lyndon [mailto:LyOng@ciena.com]
> Sent: Tuesday, February 18, 2003 4:51 PM
> To: 'sigtran@ietf.org'
> Subject: [Sigtran] Last Call on signaling applicability statement
> 
> 
> Hi Folks,
> 
> The Last Call on the security framework has now closed.  Last Call on
> M2PA actually closed last week, although I saw a few more comments
> dribble in.
> 
> I've had a request now to start Last Call on the Signaling 
> Applicability
> Statement,
> draft-ietf-sigtran-signalling-over-sctp-applic-07.txt
> 
> Please send in your comments starting today and ending Last 
> Call on Tuesday,
> March 4th.
> 
> Thanks,
> 
> L. Ong
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 20 00:52:58 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA20225
	for <sigtran-archive@odin.ietf.org>; Thu, 20 Feb 2003 00:52:58 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1K5xNt06251
	for sigtran-archive@odin.ietf.org; Thu, 20 Feb 2003 00:59:23 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1K5x9p06242;
	Thu, 20 Feb 2003 00:59:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1K5vrp06205
	for <sigtran@optimus.ietf.org>; Thu, 20 Feb 2003 00:57:53 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA20194
	for <sigtran@ietf.org>; Thu, 20 Feb 2003 00:50:56 -0500 (EST)
Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF04065868.DEB287AF-ONC2256CD3.001E4B2A@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Thu, 20 Feb 2003 07:51:16 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/20/2003
 07:51:20 AM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Tom,
> > 3) How should T7 be handled when remote processor
> outage abates?  Should
> > we immediately check for excessive ack delay, or
> reset T7 somehow?
>
> From our earlier discussion of T7 during Busy, I
> think the answer is
> that T7 is not running during PO. Any messages would
> be flushed, so T7
> would not start again until a new message was sent.
[TOLGA]Will this flushing occur always? IMO, only if the processor outage
continues for long term, which is decided by MTP3.
Will the 08 draft follow your interpretation of how messages should be
handled during PO or not, i.e. they should be considered as sent either
before or after processor outage? If yes, should T7 be stopped for any
message, which was sent before receiving LS(PO)?

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 20 01:43:08 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA21398
	for <sigtran-archive@odin.ietf.org>; Thu, 20 Feb 2003 01:43:08 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1K6nYk09512
	for sigtran-archive@odin.ietf.org; Thu, 20 Feb 2003 01:49:34 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1K6nCp09498;
	Thu, 20 Feb 2003 01:49:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1K6mCp09464
	for <sigtran@optimus.ietf.org>; Thu, 20 Feb 2003 01:48:12 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA21368
	for <sigtran@ietf.org>; Thu, 20 Feb 2003 01:41:15 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:EGa5sAOy5gma7s+HZIpMU7kRhGrAxXcM@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1K6j4D01280;
	Wed, 19 Feb 2003 23:45:04 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1K6j2X06404;
	Wed, 19 Feb 2003 23:45:02 -0700
Date: Wed, 19 Feb 2003 23:45:02 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tom George <Tom.George@alcatel.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
Message-ID: <20030219234502.A31291@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tom George <Tom.George@alcatel.com>, sigtran@ietf.org
References: <3E52F4FC.1030303@pt.com> <20030218202449.A30468@openss7.org> <3E53F301.1D46F297@alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E53F301.1D46F297@alcatel.com>; from Tom.George@alcatel.com on Wed, Feb 19, 2003 at 03:11:29PM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Tom,

Well....

MTP2 is allowed to acknowledge messages which have not actually been
delivered to MTP3.  It would be better to say that "the BSNT is the
FSN of the last User Data message received for which an acknowledgement
was sent."  The timing of actual deliveries to MTP3 is implementation
dependent.

--brian

Tom George wrote:                                                                      (Wed, 19 Feb 2003 15:11:29)
> Here is my new improved wording in 4.2.6:
> 
> For data retrieval, MTP3 requests the Backward Sequence Number to be
> Transmitted (BSNT) from M2PA through the Retrieve BSNT request. M2PA
> determines the Forward Sequence Number of the last User Data message
> received from the peer and delivered to the upper layer MTP3. This
> value is the BSNT. 
> 
> tomG.
> 
> "Brian F. G. Bidulock" wrote:
> > 
> > Andrew,
> > 
> > The draft is incorrect.
> > 
> > The BSNT is the last acknowledged received DATA (data should not be
> > acknowledged unless it is locally "deliverable" to MTP3).
> > 
> > Otherwise there is a slim chance that message loss will occur on changeover
> > resulting from processor outage or busy conditions.
> > 
> > The statement in section 4.2.6 should be changed to reflect this.
> > 
> > --brian
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 20 01:53:39 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA21814
	for <sigtran-archive@odin.ietf.org>; Thu, 20 Feb 2003 01:53:39 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1K705110214
	for sigtran-archive@odin.ietf.org; Thu, 20 Feb 2003 02:00:05 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1K6xop10166;
	Thu, 20 Feb 2003 01:59:51 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1K6wwp10139
	for <sigtran@optimus.ietf.org>; Thu, 20 Feb 2003 01:58:58 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA21800
	for <sigtran@ietf.org>; Thu, 20 Feb 2003 01:52:00 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:lkqdwW1o22jVoEhvEp9nKTf3lhF1khxw@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1K6tnD01392;
	Wed, 19 Feb 2003 23:55:50 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1K6tnL06552;
	Wed, 19 Feb 2003 23:55:49 -0700
Date: Wed, 19 Feb 2003 23:55:49 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
Message-ID: <20030219235549.B31291@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OF04065868.DEB287AF-ONC2256CD3.001E4B2A@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF04065868.DEB287AF-ONC2256CD3.001E4B2A@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Thu, Feb 20, 2003 at 07:51:16AM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

TOLGA,

Normal Q.703 procedures should prevail.

When SIPO is sent T7 is stopped (see
Figure 13/Q.703 sheet 2 of 7).

When the link resumes from processor outage,
T7 is started if the retransmission buffer
contains unacknowledged messages (see
Figure 13/Q.703 sheet 5 of 7 under "Send MSU").

When LSPO is sent, T7 must be stopped for all
messages.

When LSPR is sent, T7 must be started for any
messages still present in the retransmission buffer.

It's that simple.

Flush buffers is only performed after a Stop, in
which case no timers are running.

--brian

TOLGA ASVEREN wrote:                                                                 (Thu, 20 Feb 2003 07:51:16)
> 
> Tom,
> > > 3) How should T7 be handled when remote processor
> > outage abates?  Should
> > > we immediately check for excessive ack delay, or
> > reset T7 somehow?
> >
> > From our earlier discussion of T7 during Busy, I
> > think the answer is
> > that T7 is not running during PO. Any messages would
> > be flushed, so T7
> > would not start again until a new message was sent.
> [TOLGA]Will this flushing occur always? IMO, only if the processor outage
> continues for long term, which is decided by MTP3.
> Will the 08 draft follow your interpretation of how messages should be
> handled during PO or not, i.e. they should be considered as sent either
> before or after processor outage? If yes, should T7 be stopped for any
> message, which was sent before receiving LS(PO)?
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 20 04:48:11 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA05667
	for <sigtran-archive@odin.ietf.org>; Thu, 20 Feb 2003 04:48:10 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1K9sf131313
	for sigtran-archive@odin.ietf.org; Thu, 20 Feb 2003 04:54:41 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1K9sRp31305;
	Thu, 20 Feb 2003 04:54:27 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1K9rLp31278
	for <sigtran@optimus.ietf.org>; Thu, 20 Feb 2003 04:53:21 -0500
Received: from mailrelay03.wind.it (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA05629
	for <sigtran@ietf.org>; Thu, 20 Feb 2003 04:46:19 -0500 (EST)
Received: from mailrelay03.wind.it ([212.141.48.56]) by mailrelay03.wind.it with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2654.52)
	id 1VD48379; Thu, 20 Feb 2003 10:47:13 +0100
Received: by wexrm3co.wind with Internet Mail Service (5.5.2653.19)
	id <1VDG3Q70>; Thu, 20 Feb 2003 10:50:06 +0100
Message-ID: <AEA4C681309E384EB558453487484F1020CC4E@wexrm3a.wind>
From: Troiano Marco <Marco.Troiano@mail.wind.it>
To: "Sigtran (Posta elettronica)" <sigtran@ietf.org>
Date: Thu, 20 Feb 2003 10:50:05 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: multipart/mixed;
	boundary="------------InterScan_NT_MIME_Boundary"
Subject: [Sigtran] SCTP
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

--------------InterScan_NT_MIME_Boundary
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C2D8C5.1E52529A"

------_=_NextPart_001_01C2D8C5.1E52529A
Content-Type: text/plain;
	charset="iso-8859-1"

Hi ALL,
I am working on SCTP and I would like to ask you what are possible values
for "cookie information" and "heartbeat information" in an SCTP control
packet?What is the typical length of these fields?

Thanks in advance
Marco







------_=_NextPart_001_01C2D8C5.1E52529A
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Diso-8859-1">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
5.5.2655.35">
<TITLE>SCTP</TITLE>
</HEAD>
<BODY>

<P><FONT SIZE=3D2 FACE=3D"Courier New">Hi ALL,</FONT>
<BR><FONT SIZE=3D2 FACE=3D"Courier New">I am working on SCTP and I =
would like to ask you what are possible values for &quot;cookie =
information&quot; and &quot;heartbeat information&quot; in an SCTP =
control packet?What is the typical length of these fields?</FONT></P>

<P><FONT SIZE=3D2 FACE=3D"Courier New">Thanks in advance</FONT>
<BR><FONT SIZE=3D2 FACE=3D"Courier New">Marco</FONT>
</P>
<BR>
<BR>
<BR>
<BR>
<BR>

</BODY>
</HTML>
------_=_NextPart_001_01C2D8C5.1E52529A--

--------------InterScan_NT_MIME_Boundary--

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 20 05:06:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA06071
	for <sigtran-archive@odin.ietf.org>; Thu, 20 Feb 2003 05:06:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1KADSN32621
	for sigtran-archive@odin.ietf.org; Thu, 20 Feb 2003 05:13:28 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KADBp32598;
	Thu, 20 Feb 2003 05:13:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KAC0p32553
	for <sigtran@optimus.ietf.org>; Thu, 20 Feb 2003 05:12:00 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA06027
	for <sigtran@ietf.org>; Thu, 20 Feb 2003 05:04:58 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:4taLlRnLxzUdLzKUbo8yKoAzaR7/A8LX@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1KA8mD04826;
	Thu, 20 Feb 2003 03:08:48 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1KA8mn09403;
	Thu, 20 Feb 2003 03:08:48 -0700
Date: Thu, 20 Feb 2003 03:08:47 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Troiano Marco <Marco.Troiano@mail.wind.it>
Cc: "Sigtran (Posta elettronica)" <sigtran@ietf.org>
Subject: Re: [Sigtran] SCTP
Message-ID: <20030220030847.A9344@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Troiano Marco <Marco.Troiano@mail.wind.it>,
	"Sigtran (Posta elettronica)" <sigtran@ietf.org>
References: <AEA4C681309E384EB558453487484F1020CC4E@wexrm3a.wind>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <AEA4C681309E384EB558453487484F1020CC4E@wexrm3a.wind>; from Marco.Troiano@mail.wind.it on Thu, Feb 20, 2003 at 10:50:05AM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Troiano,

Please direct SCTP implementation questions and issues to
sctp-impl@external.cisco.com

--brian

Troiano Marco wrote:                                    (Thu, 20 Feb 2003 10:50:05)
> 
>    Hi ALL,
>    I  am  working  on  SCTP and I would like to ask you what are possible
>    values for "cookie information" and "heartbeat information" in an SCTP
>    control packet?What is the typical length of these fields?
> 
>    Thanks in advance
>    Marco


-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 20 05:13:48 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA06254
	for <sigtran-archive@odin.ietf.org>; Thu, 20 Feb 2003 05:13:48 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1KAKJw00401
	for sigtran-archive@odin.ietf.org; Thu, 20 Feb 2003 05:20:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KAK7p00391;
	Thu, 20 Feb 2003 05:20:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KAJDp00331
	for <sigtran@optimus.ietf.org>; Thu, 20 Feb 2003 05:19:13 -0500
Received: from hindon.hss.co.in (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA06193
	for <sigtran@ietf.org>; Thu, 20 Feb 2003 05:12:07 -0500 (EST)
From: mrajpal@hss.hns.com
Received: from hindon.hss.co.in (localhost [127.0.0.1])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h1KAEJs17560
	for <sigtran@ietf.org>; Thu, 20 Feb 2003 15:44:19 +0530 (IST)
Received: from ultra.hss.co.in (ultra.hss.hns.com [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h1KAEG217538;
	Thu, 20 Feb 2003 15:44:17 +0530 (IST)
Received: from sandesh.hss.hns.com (localhost [127.0.0.1])
	by ultra.hss.co.in (8.10.0/8.10.0) with SMTP id h1KAGle03682;
	Thu, 20 Feb 2003 15:46:48 +0530 (IST)
Received: by sandesh.hss.hns.com(Lotus SMTP MTA v4.6.3  (733.2 10-16-1998))  id 65256CD3.0037AA1A ; Thu, 20 Feb 2003 15:38:00 +0530
X-Lotus-FromDomain: HSS
To: Troiano Marco <Marco.Troiano@mail.wind.it>
cc: "Sigtran (Posta elettronica)" <sigtran@ietf.org>
Message-ID: <65256CD3.0037A7FF.00@sandesh.hss.hns.com>
Date: Thu, 20 Feb 2003 15:42:57 +0530
Subject: Re: [Sigtran] SCTP
Mime-Version: 1.0
Content-type: multipart/mixed; 
	Boundary="0__=ksaZgdOyu0PMAaA6BuiwyjesNb6cs8gL5izclBqXwQyjEP8VFdkHwAvL"
Content-Disposition: inline
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>



--0__=ksaZgdOyu0PMAaA6BuiwyjesNb6cs8gL5izclBqXwQyjEP8VFdkHwAvL
Content-type: text/plain; charset=us-ascii
Content-Disposition: inline




hi marco,

This is the sigtran mailing list and you should address SCTP related queries to
tsvwg (tsvwg@ietf.org) or  sctp implementors list (sctp-impl@external.cisco.com)

btw, the cookie information is of relevance only to the sender of the init-ack.
It can contain anything and the receiver just needs to echo it back to the
sender. Typically a sender stores the state information received from the init
chunk in the cookie and encrypts it before sending it to the peer. On receiving
it back in the cookie echo, it decrypts it back and uses this information to
create the TCB. It also stores the time stamp in the cookie to make sure the
cookie echoed back is not a stale one.

regarding heartbeat infromation, the peer is expected to echo the information
recieved as it is and is again of relevance only to the sender.


regards
manish

www.hssworld.com





Troiano Marco <Marco.Troiano@mail.wind.it> on 02/20/2003 03:20:05 PM

To:   "Sigtran (Posta elettronica)" <sigtran@ietf.org>
cc:    (bcc: Manish Rajpal/HSS)

Subject:  [Sigtran] SCTP




Hi ALL,
I am working on SCTP and I would like to ask you what are possible values
for "cookie information" and "heartbeat information" in an SCTP control
packet?What is the typical length of these fields?

Thanks in advance
Marco








--0__=ksaZgdOyu0PMAaA6BuiwyjesNb6cs8gL5izclBqXwQyjEP8VFdkHwAvL
Content-type: text/html; 
	name="att1.htm"
Content-Disposition: attachment; filename="att1.htm"
Content-Description: Internet HTML
Content-Transfer-Encoding: base64
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--0__=ksaZgdOyu0PMAaA6BuiwyjesNb6cs8gL5izclBqXwQyjEP8VFdkHwAvL--

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 20 06:33:14 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA07613
	for <sigtran-archive@odin.ietf.org>; Thu, 20 Feb 2003 06:33:14 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1KBdlx05389
	for sigtran-archive@odin.ietf.org; Thu, 20 Feb 2003 06:39:47 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KBdZp05380;
	Thu, 20 Feb 2003 06:39:35 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KBcwp05354
	for <sigtran@optimus.ietf.org>; Thu, 20 Feb 2003 06:38:58 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA07597
	for <sigtran@ietf.org>; Thu, 20 Feb 2003 06:31:52 -0500 (EST)
Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OFB31DC12C.36033690-ONC2256CD3.003EA381@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Thu, 20 Feb 2003 13:32:13 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/20/2003
 01:32:18 PM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,
--- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
> TOLGA,
>
> Normal Q.703 procedures should prevail.
>
> When SIPO is sent T7 is stopped (see
> Figure 13/Q.703 sheet 2 of 7).
>
> When the link resumes from processor outage,
> T7 is started if the retransmission buffer
> contains unacknowledged messages (see
> Figure 13/Q.703 sheet 5 of 7 under "Send MSU").
>
> When LSPO is sent, T7 must be stopped for all
> messages.
[TOLGA]Even for the case, where T7 is running for a message, which is sent
before LSPO? So, won't the peer which receives LSPO assume, any incoming
MSU after LSP is in fact sent not during remote processor outage and
acknowledge it?
>
> When LSPR is sent, T7 must be started for any
> messages still present in the retransmission buffer.
>
> It's that simple.
>
> Flush buffers is only performed after a Stop, in
> which case no timers are running.
>
> --brian
>
> TOLGA ASVEREN wrote:
>                                 (Thu, 20 Feb 2003
> 07:51:16)
> >
> > Tom,
> > > > 3) How should T7 be handled when remote
> processor
> > > outage abates?  Should
> > > > we immediately check for excessive ack delay,
> or
> > > reset T7 somehow?
> > >
> > > From our earlier discussion of T7 during Busy, I
> > > think the answer is
> > > that T7 is not running during PO. Any messages
> would
> > > be flushed, so T7
> > > would not start again until a new message was
> sent.
> > [TOLGA]Will this flushing occur always? IMO, only
> if the processor outage
> > continues for long term, which is decided by MTP3.
> > Will the 08 draft follow your interpretation of
> how messages should be
> > handled during PO or not, i.e. they should be
> considered as sent either
> > before or after processor outage? If yes, should
> T7 be stopped for any
> > message, which was sent before receiving LS(PO)?

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 20 08:03:17 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA11347
	for <sigtran-archive@odin.ietf.org>; Thu, 20 Feb 2003 08:03:17 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1KD9oE11079
	for sigtran-archive@odin.ietf.org; Thu, 20 Feb 2003 08:09:50 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KD9Lp11070;
	Thu, 20 Feb 2003 08:09:22 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KD7gp11029
	for <sigtran@optimus.ietf.org>; Thu, 20 Feb 2003 08:07:42 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA11179
	for <sigtran@ietf.org>; Thu, 20 Feb 2003 08:00:32 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:B4XT329d5rQtQ0QLRicoWesJsU9YjObQ@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1KD4KD07717;
	Thu, 20 Feb 2003 06:04:20 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1KD4KZ15685;
	Thu, 20 Feb 2003 06:04:20 -0700
Date: Thu, 20 Feb 2003 06:04:19 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
Message-ID: <20030220060419.A13789@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OFB31DC12C.36033690-ONC2256CD3.003EA381@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OFB31DC12C.36033690-ONC2256CD3.003EA381@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Thu, Feb 20, 2003 at 01:32:13PM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

TOLGA,

TOLGA ASVEREN wrote:                                                                 (Thu, 20 Feb 2003 13:32:13)
> 
> Brian,
> --- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
> > TOLGA,
> >
> > Normal Q.703 procedures should prevail.
> >
> > When SIPO is sent T7 is stopped (see
> > Figure 13/Q.703 sheet 2 of 7).
> >
> > When the link resumes from processor outage,
> > T7 is started if the retransmission buffer
> > contains unacknowledged messages (see
> > Figure 13/Q.703 sheet 5 of 7 under "Send MSU").
> >
> > When LSPO is sent, T7 must be stopped for all
> > messages.
> [TOLGA]Even for the case, where T7 is running for a message, which is sent
> before LSPO? So, won't the peer which receives LSPO assume, any incoming
> MSU after LSP is in fact sent not during remote processor outage and
> acknowledge it?

So what? The peer to which LSPO is sent is welcome
to acknowledge messages.   It is just not required
to do so within T7 until after LSPR is sent.

Read the SDLs (RC/TXC).  This is the same for Q.703.

--brian

> >
> > When LSPR is sent, T7 must be started for any
> > messages still present in the retransmission buffer.
> >
> > It's that simple.
> >
> > Flush buffers is only performed after a Stop, in
> > which case no timers are running.
> >
> > --brian
> >
> > TOLGA ASVEREN wrote:
> >                                 (Thu, 20 Feb 2003
> > 07:51:16)
> > >
> > > Tom,
> > > > > 3) How should T7 be handled when remote
> > processor
> > > > outage abates?  Should
> > > > > we immediately check for excessive ack delay,
> > or
> > > > reset T7 somehow?
> > > >
> > > > From our earlier discussion of T7 during Busy, I
> > > > think the answer is
> > > > that T7 is not running during PO. Any messages
> > would
> > > > be flushed, so T7
> > > > would not start again until a new message was
> > sent.
> > > [TOLGA]Will this flushing occur always? IMO, only
> > if the processor outage
> > > continues for long term, which is decided by MTP3.
> > > Will the 08 draft follow your interpretation of
> > how messages should be
> > > handled during PO or not, i.e. they should be
> > considered as sent either
> > > before or after processor outage? If yes, should
> > T7 be stopped for any
> > > message, which was sent before receiving LS(PO)?
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 20 09:58:07 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA16476
	for <sigtran-archive@odin.ietf.org>; Thu, 20 Feb 2003 09:58:07 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1KF4ii18526
	for sigtran-archive@odin.ietf.org; Thu, 20 Feb 2003 10:04:44 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KF4Wp18499;
	Thu, 20 Feb 2003 10:04:32 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KF3Qp18429
	for <sigtran@optimus.ietf.org>; Thu, 20 Feb 2003 10:03:26 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA16422
	for <sigtran@ietf.org>; Thu, 20 Feb 2003 09:56:17 -0500 (EST)
Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OFC1B7DB11.F0C903BE-ONC2256CD3.0051B687@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Thu, 20 Feb 2003 16:56:36 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/20/2003
 04:56:41 PM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

I just think because in M2PA it is the normal behavior -as far as I
understand- that received messages during remote processor outage are
acknowledged, I would think there is no need to stop T7, which is started
while sending a message, sent before LSPO.

    Tolga



TOLGA,

TOLGA ASVEREN wrote:
(Thu, 20 Feb 2003 13:32:13)
>
> Brian,
> --- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
> > TOLGA,
> >
> > Normal Q.703 procedures should prevail.
> >
> > When SIPO is sent T7 is stopped (see
> > Figure 13/Q.703 sheet 2 of 7).
> >
> > When the link resumes from processor outage,
> > T7 is started if the retransmission buffer
> > contains unacknowledged messages (see
> > Figure 13/Q.703 sheet 5 of 7 under "Send MSU").
> >
> > When LSPO is sent, T7 must be stopped for all
> > messages.
> [TOLGA]Even for the case, where T7 is running for a message, which is
sent
> before LSPO? So, won't the peer which receives LSPO assume, any incoming
> MSU after LSP is in fact sent not during remote processor outage and
> acknowledge it?

So what? The peer to which LSPO is sent is welcome
to acknowledge messages.   It is just not required
to do so within T7 until after LSPR is sent.

Read the SDLs (RC/TXC).  This is the same for Q.703.






_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 20 11:47:04 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA20832
	for <sigtran-archive@odin.ietf.org>; Thu, 20 Feb 2003 11:47:04 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1KGrhH27911
	for sigtran-archive@odin.ietf.org; Thu, 20 Feb 2003 11:53:43 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KGrPp27852;
	Thu, 20 Feb 2003 11:53:25 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KGmap27468
	for <sigtran@optimus.ietf.org>; Thu, 20 Feb 2003 11:48:36 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA20595
	for <sigtran@ietf.org>; Thu, 20 Feb 2003 11:41:26 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1KGjFqe023821;
	Thu, 20 Feb 2003 10:45:16 -0600 (CST)
Message-ID: <3E55061B.904CFEC6@alcatel.com>
Date: Thu, 20 Feb 2003 10:45:15 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
References: <3E52F4FC.1030303@pt.com> <20030218202449.A30468@openss7.org> <3E53F301.1D46F297@alcatel.com> <20030219234502.A31291@openss7.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Brian,

Good point. Now it says:

M2PA
determines the Forward Sequence Number of the last User Data message
received from the peer for which an acknowledgement was sent. 

tomG.


"Brian F. G. Bidulock" wrote:
> 
> Tom,
> 
> Well....
> 
> MTP2 is allowed to acknowledge messages which have not actually been
> delivered to MTP3.  It would be better to say that "the BSNT is the
> FSN of the last User Data message received for which an acknowledgement
> was sent."  The timing of actual deliveries to MTP3 is implementation
> dependent.
> 
> --brian
  [...]

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 20 16:34:22 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA00467
	for <sigtran-archive@odin.ietf.org>; Thu, 20 Feb 2003 16:34:21 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1KLf6H18298
	for sigtran-archive@odin.ietf.org; Thu, 20 Feb 2003 16:41:06 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KLenp18270;
	Thu, 20 Feb 2003 16:40:50 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KLaap17295
	for <sigtran@optimus.ietf.org>; Thu, 20 Feb 2003 16:36:36 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA00215
	for <sigtran@ietf.org>; Thu, 20 Feb 2003 16:29:20 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1KLXAJT005556;
	Thu, 20 Feb 2003 15:33:10 -0600 (CST)
Message-ID: <3E554996.2713B9@alcatel.com>
Date: Thu, 20 Feb 2003 15:33:10 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft comments and questions.
References: <OF2DF310C4.A6055B45-ON85256CCB.00467B35@nc.tekelec.com> <20030212082103.B21356@openss7.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Regarding item 5 in John's list:

IMHO the newer version of M2PA SHOULD send a LS OOS in reply to the
unsupported LSA, instead of just discarding it. That would be more
effective in stopping the alignment procedure. 

tomG.


"Brian F. G. Bidulock" wrote:
> 
> Jeff,
> 
> jeffrey.craig@tekelec.com wrote:                                             (Wed, 12 Feb 2003 08:33:17)
> > >
> > >Management procedures are implementation dependent.
> >
> > (Jeff)
> > I agree that requesting that the upper layer stop trying to align the link
> > is
> > outside the scope of the M2PA draft. There is no MTP3 primitive for this.
> > There is nothing preventing the implementor from maintaining and
> > reporting 'M2PA version error' measurements.
> >
> > The initial question was: "What is the definition of 'SHALL NOT complete
> > the alignment procedure'".
> >
> > The draft currently implies that the unsupported LSA is simply ignored.
> > This means that alignment will ultimately fail due to T2 expiry by the
> > receiver of the unsupported LSA. This also means that the sender is
> > likely to advance to PROVING. I see no problem with this.
> 
> The sender should not necessarily advance to proving, unless it likes
> the verion of a received ALIGNMENT or PROVING message.
> 
> >
> > So the text at 4.1.8 should be clarified:
> >
> > OLD TEXT
> > In case a newer version of M2PA is incompatible with an older version,
> > the newer version SHOULD recognize this and prevent the alignment of
> > the link. If a Link Status Alignment message with an unsupported
> > version is received by the newer version, the receiving end's M2PA
> > SHALL NOT complete the alignment procedure.
> >
> > NEW TEXT
> > In case a newer version of M2PA is incompatible with an older version,
> > the newer version SHOULD recognize this and prevent the alignment of
> > the link. If a Link Status Alignment message with an unsupported
> > version is received by the newer version, the receiving end's M2PA
> > MUST discard the message.
> 
> I would say that if it receives any message with an unsupported version
> it SHOULD discard the message and not act upon it.
> 
> (SHOULD not MUST, because "unsupported" is not clearly defined.)
> 
> --brian
> 
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 20 16:43:35 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA00782
	for <sigtran-archive@odin.ietf.org>; Thu, 20 Feb 2003 16:43:35 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1KLoK318618
	for sigtran-archive@odin.ietf.org; Thu, 20 Feb 2003 16:50:20 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KLo1p18585;
	Thu, 20 Feb 2003 16:50:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KLm9p18520
	for <sigtran@optimus.ietf.org>; Thu, 20 Feb 2003 16:48:09 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA00737
	for <sigtran@ietf.org>; Thu, 20 Feb 2003 16:40:53 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:gr0/kCS0yrE/5thqHxZaai/IfBWDdJHJ@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1KLihD15401;
	Thu, 20 Feb 2003 14:44:43 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1KLihQ10077;
	Thu, 20 Feb 2003 14:44:43 -0700
Date: Thu, 20 Feb 2003 14:44:43 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
Message-ID: <20030220144443.A9189@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OFC1B7DB11.F0C903BE-ONC2256CD3.0051B687@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OFC1B7DB11.F0C903BE-ONC2256CD3.0051B687@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Thu, Feb 20, 2003 at 04:56:36PM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

TOLGA,

You need to stop T7 because the other side is not required
(and should not) acknowledge messages after its receives LSPO.
So, if you don not stop T7 your link will fail prematurely if
T7 is less than MTP3 timer T1.

Also, MTP3 needs to know that no messages are sent from the
moment that processor outage is reported to L3 for time-controlled
changeover procedures to be properly refererenced to the last
message sent.

On this thread we already established that T7 is to be stopped
when LSPO is sent, and User Data is no longer sent.  The question
was when to restart T7.  It is restarted after LSPR is sent for
any messages in the retranmission buffer at the time that LSPR
is sent.

--brian

TOLGA ASVEREN wrote:                                                                 (Thu, 20 Feb 2003 16:56:36)
> 
> Brian,
> 
> I just think because in M2PA it is the normal behavior -as far as I
> understand- that received messages during remote processor outage are
> acknowledged, I would think there is no need to stop T7, which is started
> while sending a message, sent before LSPO.
> 
>     Tolga
> 
> 
> 
> TOLGA,
> 
> TOLGA ASVEREN wrote:
> (Thu, 20 Feb 2003 13:32:13)
> >
> > Brian,
> > --- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
> > > TOLGA,
> > >
> > > Normal Q.703 procedures should prevail.
> > >
> > > When SIPO is sent T7 is stopped (see
> > > Figure 13/Q.703 sheet 2 of 7).
> > >
> > > When the link resumes from processor outage,
> > > T7 is started if the retransmission buffer
> > > contains unacknowledged messages (see
> > > Figure 13/Q.703 sheet 5 of 7 under "Send MSU").
> > >
> > > When LSPO is sent, T7 must be stopped for all
> > > messages.
> > [TOLGA]Even for the case, where T7 is running for a message, which is
> sent
> > before LSPO? So, won't the peer which receives LSPO assume, any incoming
> > MSU after LSP is in fact sent not during remote processor outage and
> > acknowledge it?
> 
> So what? The peer to which LSPO is sent is welcome
> to acknowledge messages.   It is just not required
> to do so within T7 until after LSPR is sent.
> 
> Read the SDLs (RC/TXC).  This is the same for Q.703.
> 
> 
> 
> 
> 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 20 16:44:31 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA00821
	for <sigtran-archive@odin.ietf.org>; Thu, 20 Feb 2003 16:44:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1KLpG418668
	for sigtran-archive@odin.ietf.org; Thu, 20 Feb 2003 16:51:16 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KLp5p18656;
	Thu, 20 Feb 2003 16:51:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KLnup18560
	for <sigtran@optimus.ietf.org>; Thu, 20 Feb 2003 16:49:56 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA00767
	for <sigtran@ietf.org>; Thu, 20 Feb 2003 16:42:39 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:vzAWLRvO4Z0nIAt1zXt8dfjfl5HKYVqq@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1KLkUD15426;
	Thu, 20 Feb 2003 14:46:30 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1KLkTx10094;
	Thu, 20 Feb 2003 14:46:29 -0700
Date: Thu, 20 Feb 2003 14:46:29 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tom George <Tom.George@alcatel.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft comments and questions.
Message-ID: <20030220144629.B9189@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tom George <Tom.George@alcatel.com>, sigtran@ietf.org
References: <OF2DF310C4.A6055B45-ON85256CCB.00467B35@nc.tekelec.com> <20030212082103.B21356@openss7.org> <3E554996.2713B9@alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E554996.2713B9@alcatel.com>; from Tom.George@alcatel.com on Thu, Feb 20, 2003 at 03:33:10PM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Tom,

How can an implementation not support LSA?

--brian

Tom George wrote:                                                                      (Thu, 20 Feb 2003 15:33:10)
> Regarding item 5 in John's list:
> 
> IMHO the newer version of M2PA SHOULD send a LS OOS in reply to the
> unsupported LSA, instead of just discarding it. That would be more
> effective in stopping the alignment procedure. 
> 
> tomG.
> 
> 
> "Brian F. G. Bidulock" wrote:
> > 
> > Jeff,
> > 
> > jeffrey.craig@tekelec.com wrote:                                             (Wed, 12 Feb 2003 08:33:17)
> > > >
> > > >Management procedures are implementation dependent.
> > >
> > > (Jeff)
> > > I agree that requesting that the upper layer stop trying to align the link
> > > is
> > > outside the scope of the M2PA draft. There is no MTP3 primitive for this.
> > > There is nothing preventing the implementor from maintaining and
> > > reporting 'M2PA version error' measurements.
> > >
> > > The initial question was: "What is the definition of 'SHALL NOT complete
> > > the alignment procedure'".
> > >
> > > The draft currently implies that the unsupported LSA is simply ignored.
> > > This means that alignment will ultimately fail due to T2 expiry by the
> > > receiver of the unsupported LSA. This also means that the sender is
> > > likely to advance to PROVING. I see no problem with this.
> > 
> > The sender should not necessarily advance to proving, unless it likes
> > the verion of a received ALIGNMENT or PROVING message.
> > 
> > >
> > > So the text at 4.1.8 should be clarified:
> > >
> > > OLD TEXT
> > > In case a newer version of M2PA is incompatible with an older version,
> > > the newer version SHOULD recognize this and prevent the alignment of
> > > the link. If a Link Status Alignment message with an unsupported
> > > version is received by the newer version, the receiving end's M2PA
> > > SHALL NOT complete the alignment procedure.
> > >
> > > NEW TEXT
> > > In case a newer version of M2PA is incompatible with an older version,
> > > the newer version SHOULD recognize this and prevent the alignment of
> > > the link. If a Link Status Alignment message with an unsupported
> > > version is received by the newer version, the receiving end's M2PA
> > > MUST discard the message.
> > 
> > I would say that if it receives any message with an unsupported version
> > it SHOULD discard the message and not act upon it.
> > 
> > (SHOULD not MUST, because "unsupported" is not clearly defined.)
> > 
> > --brian
> > 
> > --
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> -- 
> * Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
> * 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
> * Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
> * I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
> * to advertise or represent its position,  operations, or activities. *
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 20 16:49:34 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA00980
	for <sigtran-archive@odin.ietf.org>; Thu, 20 Feb 2003 16:49:34 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1KLuJL18858
	for sigtran-archive@odin.ietf.org; Thu, 20 Feb 2003 16:56:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KLu5p18850;
	Thu, 20 Feb 2003 16:56:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KLtkp18831
	for <sigtran@optimus.ietf.org>; Thu, 20 Feb 2003 16:55:46 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA00961
	for <sigtran@ietf.org>; Thu, 20 Feb 2003 16:48:30 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1KLqJqe021769;
	Thu, 20 Feb 2003 15:52:20 -0600 (CST)
Message-ID: <3E554E13.7EA54FB0@alcatel.com>
Date: Thu, 20 Feb 2003 15:52:19 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft comments and questions.
References: <20030211150947.10241.qmail@mail.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Summing up the discussion, here is what I plan to do for each of these
items....

John Li wrote:
> 
> All,
> I sent the following comments and questions to the group a while ago, and have not got responses yet. here goes again.
> thanks.
> John
> 
> 1.      In section 1.5: the following statement is somewhat confusing.
> ?The SG is an IPSP equipped with both traditional SS7 and IP
> network connections. In effect, the Signaling Gateway acts as a
> Signal Transfer Point (STP)?
> 
> Perhaps we can say "The SG interconnects SS7 and IP networks by
> converting signaling traffic" or just refer to the definition in
> section 1.2

I reworded this paragraph slightly. 
          
	  Figure 2 shows an example of M2PA used in a Signaling
	  Gateway (SG). The SG is an IPSP equipped with both
	  traditional SS7 and IP network connections.  Any of the
	  nodes in the diagram could have SCCP or other SS7
	  layers. The Signaling Gateway acts as a Signal Transfer
	  Point (STP).  Other STPs MAY or MAY NOT be present in the
	  SS7 path between the SEP and the SG.


> 
> 2.      Section 2.1.5 Message length
> Should we specify the valid value range of the length?

No. This is determined by MTP3. 

No change to the draft is necessary.


> 
> 3.      Section 2.3.1 User Data
> 
> statement "M2PA does not add padding to the MTP3 message.".
> I think we should make it a requirement: ?M2PA should not add padding
> to the MTP3 message.?

I changed this to "M2PA SHALL NOT add padding to the MTP3
message." It has to be a firm requirement.


> 
> 4.      In section 4.1.3 Link alignment
> 
> perhaps we can specify the recommended practice for choosing
> emergency proving vs. normal proving during alignment procedure.

No. This is determined by MTP3. 

No change to the draft is necessary.


> 
> 5.      Section 4.1.8:
> "If a Link Status Alignment message with an unsupported
> received by the newer version, the receiving end's M2PA
> SHALL NOT complete the alignment procedure."
> 
> can we clearly define the behavior of not completing the alignment
> procedure?

I changed this sentence to: "If a Link Status Alignment
message with an unsupported version is received by the newer
version, the receiving end's M2PA SHOULD reply with a Link
Status Out of Service message and not complete the alignment
procedure."


> 
> 6.      section 4.2.1 sending and receiving messages
> When M2PA receives a User Data message with Backward Sequence Number
> equal to 'n', it may remove all messages with Forward Sequence Number
> <= n from its queue.
> 
> This is incorrect. Consider at the time SN is cycled to 0 from 127
> (assume 128 buffers), there are 127, 126, etc are waiting for
> acknowledgement. We should replace it with ?The message n and all
> outstanding (not yet acknowledged) messages sent before the message n
> may be removed from its queue.

There is a general warning about this in this section
(4.2.1).


> 
> 7.      section 4.2.1
> If M2PA receives an empty User Data message, it SHALL NOT send an
> acknowledgement of that message.
> 
> What is empty User Data message? FISUs?

It's an Ack (sections 2.3.1, 4.2.1). 

No change to the draft is necessary.


-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 20 16:54:15 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA01210
	for <sigtran-archive@odin.ietf.org>; Thu, 20 Feb 2003 16:54:15 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1KM10119486
	for sigtran-archive@odin.ietf.org; Thu, 20 Feb 2003 17:01:00 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KM0ip19478;
	Thu, 20 Feb 2003 17:00:44 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KLx7p19072
	for <sigtran@optimus.ietf.org>; Thu, 20 Feb 2003 16:59:07 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA01072
	for <sigtran@ietf.org>; Thu, 20 Feb 2003 16:51:51 -0500 (EST)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1KLtfqe022472;
	Thu, 20 Feb 2003 15:55:41 -0600 (CST)
Message-ID: <3E554EDD.1A172C4E@alcatel.com>
Date: Thu, 20 Feb 2003 15:55:41 -0600
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA draft comments and questions.
References: <OF2DF310C4.A6055B45-ON85256CCB.00467B35@nc.tekelec.com> <20030212082103.B21356@openss7.org> <3E554996.2713B9@alcatel.com> <20030220144629.B9189@openss7.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

That would be tragic. 8-) I meant an LSA with an unsupported version
number. 

I put my updated wording in the other message I just sent out.

tomG.

"Brian F. G. Bidulock" wrote:
> 
> Tom,
> 
> How can an implementation not support LSA?
> 
> --brian
> 
> Tom George wrote:                                                                      (Thu, 20 Feb 2003 15:33:10)
> > Regarding item 5 in John's list:
> >
> > IMHO the newer version of M2PA SHOULD send a LS OOS in reply to the
> > unsupported LSA, instead of just discarding it. That would be more
> > effective in stopping the alignment procedure.
> >
> > tomG.
> >
> >
> > "Brian F. G. Bidulock" wrote:
> > >
> > > Jeff,
> > >
> > > jeffrey.craig@tekelec.com wrote:                                             (Wed, 12 Feb 2003 08:33:17)
> > > > >
> > > > >Management procedures are implementation dependent.
> > > >
> > > > (Jeff)
> > > > I agree that requesting that the upper layer stop trying to align the link
> > > > is
> > > > outside the scope of the M2PA draft. There is no MTP3 primitive for this.
> > > > There is nothing preventing the implementor from maintaining and
> > > > reporting 'M2PA version error' measurements.
> > > >
> > > > The initial question was: "What is the definition of 'SHALL NOT complete
> > > > the alignment procedure'".
> > > >
> > > > The draft currently implies that the unsupported LSA is simply ignored.
> > > > This means that alignment will ultimately fail due to T2 expiry by the
> > > > receiver of the unsupported LSA. This also means that the sender is
> > > > likely to advance to PROVING. I see no problem with this.
> > >
> > > The sender should not necessarily advance to proving, unless it likes
> > > the verion of a received ALIGNMENT or PROVING message.
> > >
> > > >
> > > > So the text at 4.1.8 should be clarified:
> > > >
> > > > OLD TEXT
> > > > In case a newer version of M2PA is incompatible with an older version,
> > > > the newer version SHOULD recognize this and prevent the alignment of
> > > > the link. If a Link Status Alignment message with an unsupported
> > > > version is received by the newer version, the receiving end's M2PA
> > > > SHALL NOT complete the alignment procedure.
> > > >
> > > > NEW TEXT
> > > > In case a newer version of M2PA is incompatible with an older version,
> > > > the newer version SHOULD recognize this and prevent the alignment of
> > > > the link. If a Link Status Alignment message with an unsupported
> > > > version is received by the newer version, the receiving end's M2PA
> > > > MUST discard the message.
> > >
> > > I would say that if it receives any message with an unsupported version
> > > it SHOULD discard the message and not act upon it.
> > >
> > > (SHOULD not MUST, because "unsupported" is not clearly defined.)
> > >
> > > --brian
> > >
> > > --
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> > --
> > * Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
> > * 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
> > * Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
> > * I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
> > * to advertise or represent its position,  operations, or activities. *
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/

-- 
* Tom George       PB7-24001 (near B-21)      Tom.George@alcatel.com  *
* 972-519-3168   Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
* Alcatel;  3400 West Plano Pkwy,  MS PB7-SWPRO;  Plano, Texas 75075  *
* I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
* to advertise or represent its position,  operations, or activities. *
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 20 18:27:10 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA03308
	for <sigtran-archive@odin.ietf.org>; Thu, 20 Feb 2003 18:27:10 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1KNXvR25373
	for sigtran-archive@odin.ietf.org; Thu, 20 Feb 2003 18:33:57 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KNXbp25345;
	Thu, 20 Feb 2003 18:33:37 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1KNW1p25309
	for <sigtran@optimus.ietf.org>; Thu, 20 Feb 2003 18:32:01 -0500
Received: from pt.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA03249
	for <sigtran@ietf.org>; Thu, 20 Feb 2003 18:24:43 -0500 (EST)
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id SAA00631;
	Thu, 20 Feb 2003 18:23:18 -0500
Message-ID: <3E55648F.7010505@pt.com>
Date: Thu, 20 Feb 2003 18:28:15 -0500
From: Andrew Booth <abooth@pt.com>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.1) Gecko/20020826
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Tom George <Tom.George@alcatel.com>
CC: sigtran@ietf.org
Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
References: <3E52F4FC.1030303@pt.com> <20030218202449.A30468@openss7.org> <3E53F301.1D46F297@alcatel.com> <20030219234502.A31291@openss7.org> <3E55061B.904CFEC6@alcatel.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Tom, Brian,

Is M2PA also allowed to withhold acknowledgements briefly? i.e. can't I 
have delivered a message to MTP3 but not yet sent the acknowledgement? 
 Perhaps

   M2PA determines the Forward Sequence Number of the last User Data message
   received from the peer which has been marked for acknowledgement i.e. which
   is guaranteed to be delivered to MTP3.

Andrew

Tom George wrote:

>Brian,
>
>Good point. Now it says:
>
>M2PA
>determines the Forward Sequence Number of the last User Data message
>received from the peer for which an acknowledgement was sent. 
>
>tomG.
>
>
>"Brian F. G. Bidulock" wrote:
>  
>
>>Tom,
>>
>>Well....
>>
>>MTP2 is allowed to acknowledge messages which have not actually been
>>delivered to MTP3.  It would be better to say that "the BSNT is the
>>FSN of the last User Data message received for which an acknowledgement
>>was sent."  The timing of actual deliveries to MTP3 is implementation
>>dependent.
>>
>>--brian
>>    
>>
>  [...]
>
>  
>

-- 
Andrew Booth, Software Developer
Signaling Systems Group, Performance Technologies
613-237-4344

Visit www.pt.com/mailing.html to subscribe to our Packets & Signals eLetter.


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 20 19:30:34 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA04741
	for <sigtran-archive@odin.ietf.org>; Thu, 20 Feb 2003 19:30:34 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1L0bN330063
	for sigtran-archive@odin.ietf.org; Thu, 20 Feb 2003 19:37:23 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1L0b5p29533;
	Thu, 20 Feb 2003 19:37:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1L0Ypp29355
	for <sigtran@optimus.ietf.org>; Thu, 20 Feb 2003 19:34:51 -0500
Received: from pt.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA04722
	for <sigtran@ietf.org>; Thu, 20 Feb 2003 19:27:31 -0500 (EST)
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id TAA01020;
	Thu, 20 Feb 2003 19:26:05 -0500
Message-ID: <3E557347.9080202@pt.com>
Date: Thu, 20 Feb 2003 19:31:03 -0500
From: Andrew Booth <abooth@pt.com>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.1) Gecko/20020826
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Tom George <Tom.George@alcatel.com>
CC: sigtran@ietf.org
Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
References: <3E52F4FC.1030303@pt.com> <20030218202449.A30468@openss7.org> <3E53FF13.8A566E9A@alcatel.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Tom,

Thanks for the info.  One comment below...

Tom George wrote:

>Andrew and others,
>
>Here are some comments on the other items....
>
>tomG.
>
>  
>
>>Hi All,
>>
>>A few more questions on M2PA draft 7.
>>
>>1) Does every User Data received from the peer increases the expected FSN
>>for the subsequent message, even in cases where the received message is 
>>later
>>discarded as part of a Flush Buffers or Changeover procedure?
>>    
>>
>
>I am not sure what you are asking. Are you asking if the FSNs are ever
>reset? 
>
In "4.1.4 Processor Outage" as described in M2PA draft 7:

   MTP3 may send a Flush Buffers or Continue command to M2PA as part of
   its processor outage procedure (See section 4.2.4 Flush Buffers and
   Continue). The use of Flush Buffers and Continue is specified in Q.703
   [Q.703] and T1.111.3 [T1.111].

In q.703 there is a reference to "synchronization of the level 2 
sequence numbers" after flushing buffers.  I'm confused about the 
concept of synchronizing sequence numbers in M2PA, and was wondering if 
this reference could be clarified.  Do we need to do anything special 
with sequence numbers after a Flush Buffers?

Andrew

>
>  
>
>>2) Suppose that an M2PA implementation has received messages from a peer
>>and that some of the messages are buffered at the M2PA layer (perhaps it is
>>in processor outage).  Suppose now the link transitions to OOS and the
>>MTP3 sends a Retrieve BSNT to M2PA.
>>
>>According to 4.2.6,
>>
>>   "M2PA determines the Forward Sequence Number of the last User Data
>>   message received from the peer.  This is the BSNT."
>>
>>In the M2PA situation it seems more appropriate to respond with the FSN
>>of the last message which was delivered to MTP3.
>>Am I missing something?
>>    
>>
>
>This one is discussed in another message.
>
>  
>
>>3) How should T7 be handled when remote processor outage abates?  Should 
>>we immediately check for excessive ack delay, or reset T7 somehow?
>>    
>>
>
>>From our earlier discussion of T7 during Busy, I think the answer is
>that T7 is not running during PO. Any messages would be flushed, so T7
>would not start again until a new message was sent. 
> 
>
>  
>
>>4) In section 5.1 we transition to proving without starting T3 and to 
>>aligned ready without receiving LS Proving.  Are these omissions 
>>intentional?
>>    
>>
>
>These were unintentionally omitted. I added those events to the diagram.
>
>
>  
>
>>5) In section 5.8 there is some introductory text :
>>
>>"Note that in this example, the sequence numbers and messages requested
>>by MTP3 are sent from SCTP to M2PA in the Communication Lost
>>primitive. In general, the retrieval of sequence numbers and messages
>>is implementation dependent."
>>
>>Why are messages retrieved from SCTP?  Is this text indicating that the 
>>M2PA retransmit buffer is implemented in the SCTP?  Just a little confused.
>>    
>>
>
>This is left over from an earlier version. I removed that paragraph.
>
>
>  
>
>>6) In section 4.2.6, is the following correct or should it be rephrased 
>>in the context of a circular buffer?
>>"any transmitted User Data messages beginning with the first
>>       unacknowledged message with FSN greater than FSNC." 
>>    
>>
>
>There is a general warning about this in 4.2.1.
>
>
>  
>
>>Thanks for any info,
>>Andrew
>>
>>-- 
>>Andrew Booth, Software Developer
>>Signaling Systems Group, Performance Technologies
>>613-237-4344
>>    
>>
>
>
>
>  
>

-- 
Andrew Booth, Software Developer
Signaling Systems Group, Performance Technologies
613-237-4344

Visit www.pt.com/mailing.html to subscribe to our Packets & Signals eLetter.


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 21 00:32:49 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA10063
	for <sigtran-archive@odin.ietf.org>; Fri, 21 Feb 2003 00:32:49 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1L5diX14845
	for sigtran-archive@odin.ietf.org; Fri, 21 Feb 2003 00:39:44 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1L5dVp14836;
	Fri, 21 Feb 2003 00:39:32 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1L5ccp14811
	for <sigtran@optimus.ietf.org>; Fri, 21 Feb 2003 00:38:38 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA10053
	for <sigtran@ietf.org>; Fri, 21 Feb 2003 00:31:11 -0500 (EST)
Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF259CC347.7CEE3FD2-ONC2256CD4.001DDD3A@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Fri, 21 Feb 2003 07:31:32 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/21/2003
 07:31:36 AM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


So, it seems something different than Tom's previos interpretation
regarding handling User Data, which is sent before LSPO but arrives after
it, will be put to the draft(probably somewhere deep in the messages in one
of the long threads the proposed mechanism is described(?))
I am looking forward for the new version, to see what we are going to have
to address this issue(and to see whether stopping T7 started while sending
such a message is really reasonable or not).

     Tolga
--- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
> TOLGA,
>
> You need to stop T7 because the other side is not
> required
> (and should not) acknowledge messages after its
> receives LSPO.
> So, if you don not stop T7 your link will fail
> prematurely if
> T7 is less than MTP3 timer T1.
>
> Also, MTP3 needs to know that no messages are sent
> from the
> moment that processor outage is reported to L3 for
> time-controlled
> changeover procedures to be properly refererenced to
> the last
> message sent.
>
> On this thread we already established that T7 is to
> be stopped
> when LSPO is sent, and User Data is no longer sent.
> The question
> was when to restart T7.  It is restarted after LSPR
> is sent for
> any messages in the retranmission buffer at the time
> that LSPR
> is sent.
>
> --brian
>
> TOLGA ASVEREN wrote:
>                                 (Thu, 20 Feb 2003
> 16:56:36)
> >
> > Brian,
> >
> > I just think because in M2PA it is the normal
> behavior -as far as I
> > understand- that received messages during remote
> processor outage are
> > acknowledged, I would think there is no need to
> stop T7, which is started
> > while sending a message, sent before LSPO.
> >
> >     Tolga
> >
> >
> >
> > TOLGA,
> >
> > TOLGA ASVEREN wrote:
> > (Thu, 20 Feb 2003 13:32:13)
> > >
> > > Brian,
> > > --- "Brian F. G. Bidulock"
> <bidulock@openss7.org> wrote:
> > > > TOLGA,
> > > >
> > > > Normal Q.703 procedures should prevail.
> > > >
> > > > When SIPO is sent T7 is stopped (see
> > > > Figure 13/Q.703 sheet 2 of 7).
> > > >
> > > > When the link resumes from processor outage,
> > > > T7 is started if the retransmission buffer
> > > > contains unacknowledged messages (see
> > > > Figure 13/Q.703 sheet 5 of 7 under "Send
> MSU").
> > > >
> > > > When LSPO is sent, T7 must be stopped for all
> > > > messages.
> > > [TOLGA]Even for the case, where T7 is running
> for a message, which is
> > sent
> > > before LSPO? So, won't the peer which receives
> LSPO assume, any incoming
> > > MSU after LSP is in fact sent not during remote
> processor outage and
> > > acknowledge it?
> >
> > So what? The peer to which LSPO is sent is welcome
> > to acknowledge messages.   It is just not required
> > to do so within T7 until after LSPR is sent.
> >
> > Read the SDLs (RC/TXC).  This is the same for
> Q.703.

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 21 01:57:48 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA11173
	for <sigtran-archive@odin.ietf.org>; Fri, 21 Feb 2003 01:57:48 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1L74ha20223
	for sigtran-archive@odin.ietf.org; Fri, 21 Feb 2003 02:04:43 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1L74Pp19081;
	Fri, 21 Feb 2003 02:04:26 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1L73mp19044
	for <sigtran@optimus.ietf.org>; Fri, 21 Feb 2003 02:03:48 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA11159
	for <sigtran@ietf.org>; Fri, 21 Feb 2003 01:56:21 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:sAN90hAZ6yoA1Vu2fsw7Wuuw8NYR0+ZK@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1L70AD22726;
	Fri, 21 Feb 2003 00:00:10 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1L708l21103;
	Fri, 21 Feb 2003 00:00:08 -0700
Date: Fri, 21 Feb 2003 00:00:08 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
Message-ID: <20030221000008.A20827@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OF259CC347.7CEE3FD2-ONC2256CD4.001DDD3A@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF259CC347.7CEE3FD2-ONC2256CD4.001DDD3A@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Fri, Feb 21, 2003 at 07:31:32AM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

TOLGA,

No it is not different.  When the sender sends LSPO it must stop sending
data (for MTP3 compatibility).  If data arrives after LSPO it may be
acknowledged (but should not for MTP3 compatibility at the receiver).
In either case, T7 is stopped so the sender cannot expect acknowledgement
but will accept acknowledgement ala Q.703 if one is given.

This was the conclusion and concensus from the "M2PA Flush Buffers" thread
in which you were involved ending February 5.

--brian

TOLGA ASVEREN wrote:                                                                 (Fri, 21 Feb 2003 07:31:32)
> 
> So, it seems something different than Tom's previos interpretation
> regarding handling User Data, which is sent before LSPO but arrives after
> it, will be put to the draft(probably somewhere deep in the messages in one
> of the long threads the proposed mechanism is described(?))
> I am looking forward for the new version, to see what we are going to have
> to address this issue(and to see whether stopping T7 started while sending
> such a message is really reasonable or not).
> 
>      Tolga
> --- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
> > TOLGA,
> >
> > You need to stop T7 because the other side is not
> > required
> > (and should not) acknowledge messages after its
> > receives LSPO.
> > So, if you don not stop T7 your link will fail
> > prematurely if
> > T7 is less than MTP3 timer T1.
> >
> > Also, MTP3 needs to know that no messages are sent
> > from the
> > moment that processor outage is reported to L3 for
> > time-controlled
> > changeover procedures to be properly refererenced to
> > the last
> > message sent.
> >
> > On this thread we already established that T7 is to
> > be stopped
> > when LSPO is sent, and User Data is no longer sent.
> > The question
> > was when to restart T7.  It is restarted after LSPR
> > is sent for
> > any messages in the retranmission buffer at the time
> > that LSPR
> > is sent.
> >
> > --brian
> >
> > TOLGA ASVEREN wrote:
> >                                 (Thu, 20 Feb 2003
> > 16:56:36)
> > >
> > > Brian,
> > >
> > > I just think because in M2PA it is the normal
> > behavior -as far as I
> > > understand- that received messages during remote
> > processor outage are
> > > acknowledged, I would think there is no need to
> > stop T7, which is started
> > > while sending a message, sent before LSPO.
> > >
> > >     Tolga
> > >
> > >
> > >
> > > TOLGA,
> > >
> > > TOLGA ASVEREN wrote:
> > > (Thu, 20 Feb 2003 13:32:13)
> > > >
> > > > Brian,
> > > > --- "Brian F. G. Bidulock"
> > <bidulock@openss7.org> wrote:
> > > > > TOLGA,
> > > > >
> > > > > Normal Q.703 procedures should prevail.
> > > > >
> > > > > When SIPO is sent T7 is stopped (see
> > > > > Figure 13/Q.703 sheet 2 of 7).
> > > > >
> > > > > When the link resumes from processor outage,
> > > > > T7 is started if the retransmission buffer
> > > > > contains unacknowledged messages (see
> > > > > Figure 13/Q.703 sheet 5 of 7 under "Send
> > MSU").
> > > > >
> > > > > When LSPO is sent, T7 must be stopped for all
> > > > > messages.
> > > > [TOLGA]Even for the case, where T7 is running
> > for a message, which is
> > > sent
> > > > before LSPO? So, won't the peer which receives
> > LSPO assume, any incoming
> > > > MSU after LSP is in fact sent not during remote
> > processor outage and
> > > > acknowledge it?
> > >
> > > So what? The peer to which LSPO is sent is welcome
> > > to acknowledge messages.   It is just not required
> > > to do so within T7 until after LSPR is sent.
> > >
> > > Read the SDLs (RC/TXC).  This is the same for
> > Q.703.
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 21 02:44:31 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA22251
	for <sigtran-archive@odin.ietf.org>; Fri, 21 Feb 2003 02:44:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1L7pRZ00529
	for sigtran-archive@odin.ietf.org; Fri, 21 Feb 2003 02:51:27 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1L7pAp00519;
	Fri, 21 Feb 2003 02:51:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1L7oVp00491
	for <sigtran@optimus.ietf.org>; Fri, 21 Feb 2003 02:50:31 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA22219
	for <sigtran@ietf.org>; Fri, 21 Feb 2003 02:43:02 -0500 (EST)
Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF4391B5EF.3CA82745-ONC2256CD4.002928FF@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Fri, 21 Feb 2003 09:43:22 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/21/2003
 09:43:25 AM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Thanks , now it is clear, so that thread was concluded(sometimes hard to
decide for this) and we will have a slightly modified version of Tom's
initial interpretation, where data is not discarded but not acknowledged
either. Then I also would think stopping T7 is fine.

One more question:
During changeover, will data received during RPO but not acknowledged,
considered as received successfully or not, i.e. in case of changeover,
will they be delivered to MTP3 or not?

    Tolga



TOLGA,

No it is not different.  When the sender sends LSPO it must stop sending
data (for MTP3 compatibility).  If data arrives after LSPO it may be
acknowledged (but should not for MTP3 compatibility at the receiver).
In either case, T7 is stopped so the sender cannot expect acknowledgement
but will accept acknowledgement ala Q.703 if one is given.

This was the conclusion and concensus from the "M2PA Flush Buffers" thread
in which you were involved ending February 5.

--brian

TOLGA ASVEREN wrote:
(Fri, 21 Feb 2003 07:31:32)
>
> So, it seems something different than Tom's previos interpretation
> regarding handling User Data, which is sent before LSPO but arrives after
> it, will be put to the draft(probably somewhere deep in the messages in
one
> of the long threads the proposed mechanism is described(?))
> I am looking forward for the new version, to see what we are going to
have
> to address this issue(and to see whether stopping T7 started while
sending
> such a message is really reasonable or not).
>
>      Tolga
> --- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
> > TOLGA,
> >
> > You need to stop T7 because the other side is not
> > required
> > (and should not) acknowledge messages after its
> > receives LSPO.
> > So, if you don not stop T7 your link will fail
> > prematurely if
> > T7 is less than MTP3 timer T1.
> >
> > Also, MTP3 needs to know that no messages are sent
> > from the
> > moment that processor outage is reported to L3 for
> > time-controlled
> > changeover procedures to be properly refererenced to
> > the last
> > message sent.
> >
> > On this thread we already established that T7 is to
> > be stopped
> > when LSPO is sent, and User Data is no longer sent.
> > The question
> > was when to restart T7.  It is restarted after LSPR
> > is sent for
> > any messages in the retranmission buffer at the time
> > that LSPR
> > is sent.
> >
> > --brian
> >
> > TOLGA ASVEREN wrote:
> >                                 (Thu, 20 Feb 2003
> > 16:56:36)
> > >
> > > Brian,
> > >
> > > I just think because in M2PA it is the normal
> > behavior -as far as I
> > > understand- that received messages during remote
> > processor outage are
> > > acknowledged, I would think there is no need to
> > stop T7, which is started
> > > while sending a message, sent before LSPO.
> > >
> > >     Tolga
> > >
> > >
> > >
> > > TOLGA,
> > >
> > > TOLGA ASVEREN wrote:
> > > (Thu, 20 Feb 2003 13:32:13)
> > > >
> > > > Brian,
> > > > --- "Brian F. G. Bidulock"
> > <bidulock@openss7.org> wrote:
> > > > > TOLGA,
> > > > >
> > > > > Normal Q.703 procedures should prevail.
> > > > >
> > > > > When SIPO is sent T7 is stopped (see
> > > > > Figure 13/Q.703 sheet 2 of 7).
> > > > >
> > > > > When the link resumes from processor outage,
> > > > > T7 is started if the retransmission buffer
> > > > > contains unacknowledged messages (see
> > > > > Figure 13/Q.703 sheet 5 of 7 under "Send
> > MSU").
> > > > >
> > > > > When LSPO is sent, T7 must be stopped for all
> > > > > messages.
> > > > [TOLGA]Even for the case, where T7 is running
> > for a message, which is
> > > sent
> > > > before LSPO? So, won't the peer which receives
> > LSPO assume, any incoming
> > > > MSU after LSP is in fact sent not during remote
> > processor outage and
> > > > acknowledge it?
> > >
> > > So what? The peer to which LSPO is sent is welcome
> > > to acknowledge messages.   It is just not required
> > > to do so within T7 until after LSPR is sent.
> > >
> > > Read the SDLs (RC/TXC).  This is the same for
> > Q.703.
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 21 04:12:51 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA23870
	for <sigtran-archive@odin.ietf.org>; Fri, 21 Feb 2003 04:12:51 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1L9Jog06401
	for sigtran-archive@odin.ietf.org; Fri, 21 Feb 2003 04:19:50 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1L9JZp06386;
	Fri, 21 Feb 2003 04:19:35 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1L9Ilp06364
	for <sigtran@optimus.ietf.org>; Fri, 21 Feb 2003 04:18:47 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA23850
	for <sigtran@ietf.org>; Fri, 21 Feb 2003 04:11:16 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:/ZC3f7rgkDr/LpLke8uTGyTJ+tyo0exe@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1L9F6D24413;
	Fri, 21 Feb 2003 02:15:06 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1L9F5d22946;
	Fri, 21 Feb 2003 02:15:05 -0700
Date: Fri, 21 Feb 2003 02:15:05 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
Message-ID: <20030221021505.A22911@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OF4391B5EF.3CA82745-ONC2256CD4.002928FF@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF4391B5EF.3CA82745-ONC2256CD4.002928FF@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Fri, Feb 21, 2003 at 09:43:22AM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

TOLGA,

TOLGA ASVEREN wrote:                                                                 (Fri, 21 Feb 2003 09:43:22)
> 
> Thanks , now it is clear, so that thread was concluded(sometimes hard to
> decide for this) and we will have a slightly modified version of Tom's
> initial interpretation, where data is not discarded but not acknowledged
> either. Then I also would think stopping T7 is fine.
> 
> One more question:
> During changeover, will data received during RPO but not acknowledged,
> considered as received successfully or not, i.e. in case of changeover,
> will they be delivered to MTP3 or not?

Unacknowledged messages must never be delivered to MTP3 or duplication
will result on changeover.  This is true for Q.703 as well as M2PA.

--brian


> 
>     Tolga
> 
> 
> 
> TOLGA,
> 
> No it is not different.  When the sender sends LSPO it must stop sending
> data (for MTP3 compatibility).  If data arrives after LSPO it may be
> acknowledged (but should not for MTP3 compatibility at the receiver).
> In either case, T7 is stopped so the sender cannot expect acknowledgement
> but will accept acknowledgement ala Q.703 if one is given.
> 
> This was the conclusion and concensus from the "M2PA Flush Buffers" thread
> in which you were involved ending February 5.
> 
> --brian
> 
> TOLGA ASVEREN wrote:
> (Fri, 21 Feb 2003 07:31:32)
> >
> > So, it seems something different than Tom's previos interpretation
> > regarding handling User Data, which is sent before LSPO but arrives after
> > it, will be put to the draft(probably somewhere deep in the messages in
> one
> > of the long threads the proposed mechanism is described(?))
> > I am looking forward for the new version, to see what we are going to
> have
> > to address this issue(and to see whether stopping T7 started while
> sending
> > such a message is really reasonable or not).
> >
> >      Tolga
> > --- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
> > > TOLGA,
> > >
> > > You need to stop T7 because the other side is not
> > > required
> > > (and should not) acknowledge messages after its
> > > receives LSPO.
> > > So, if you don not stop T7 your link will fail
> > > prematurely if
> > > T7 is less than MTP3 timer T1.
> > >
> > > Also, MTP3 needs to know that no messages are sent
> > > from the
> > > moment that processor outage is reported to L3 for
> > > time-controlled
> > > changeover procedures to be properly refererenced to
> > > the last
> > > message sent.
> > >
> > > On this thread we already established that T7 is to
> > > be stopped
> > > when LSPO is sent, and User Data is no longer sent.
> > > The question
> > > was when to restart T7.  It is restarted after LSPR
> > > is sent for
> > > any messages in the retranmission buffer at the time
> > > that LSPR
> > > is sent.
> > >
> > > --brian
> > >
> > > TOLGA ASVEREN wrote:
> > >                                 (Thu, 20 Feb 2003
> > > 16:56:36)
> > > >
> > > > Brian,
> > > >
> > > > I just think because in M2PA it is the normal
> > > behavior -as far as I
> > > > understand- that received messages during remote
> > > processor outage are
> > > > acknowledged, I would think there is no need to
> > > stop T7, which is started
> > > > while sending a message, sent before LSPO.
> > > >
> > > >     Tolga
> > > >
> > > >
> > > >
> > > > TOLGA,
> > > >
> > > > TOLGA ASVEREN wrote:
> > > > (Thu, 20 Feb 2003 13:32:13)
> > > > >
> > > > > Brian,
> > > > > --- "Brian F. G. Bidulock"
> > > <bidulock@openss7.org> wrote:
> > > > > > TOLGA,
> > > > > >
> > > > > > Normal Q.703 procedures should prevail.
> > > > > >
> > > > > > When SIPO is sent T7 is stopped (see
> > > > > > Figure 13/Q.703 sheet 2 of 7).
> > > > > >
> > > > > > When the link resumes from processor outage,
> > > > > > T7 is started if the retransmission buffer
> > > > > > contains unacknowledged messages (see
> > > > > > Figure 13/Q.703 sheet 5 of 7 under "Send
> > > MSU").
> > > > > >
> > > > > > When LSPO is sent, T7 must be stopped for all
> > > > > > messages.
> > > > > [TOLGA]Even for the case, where T7 is running
> > > for a message, which is
> > > > sent
> > > > > before LSPO? So, won't the peer which receives
> > > LSPO assume, any incoming
> > > > > MSU after LSP is in fact sent not during remote
> > > processor outage and
> > > > > acknowledge it?
> > > >
> > > > So what? The peer to which LSPO is sent is welcome
> > > > to acknowledge messages.   It is just not required
> > > > to do so within T7 until after LSPR is sent.
> > > >
> > > > Read the SDLs (RC/TXC).  This is the same for
> > > Q.703.
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 
> 
> 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 21 05:46:03 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA25500
	for <sigtran-archive@odin.ietf.org>; Fri, 21 Feb 2003 05:46:03 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1LAr4Q12091
	for sigtran-archive@odin.ietf.org; Fri, 21 Feb 2003 05:53:04 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1LAqgp12063;
	Fri, 21 Feb 2003 05:52:43 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1LApkp12038
	for <sigtran@optimus.ietf.org>; Fri, 21 Feb 2003 05:51:46 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA25478
	for <sigtran@ietf.org>; Fri, 21 Feb 2003 05:44:12 -0500 (EST)
Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF24FAFD42.687DD8CF-ONC2256CD4.003A2896@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Fri, 21 Feb 2003 12:44:30 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/21/2003
 12:44:38 PM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,
--- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
> TOLGA,
>
> TOLGA ASVEREN wrote:
>                                 (Fri, 21 Feb 2003
> 09:43:22)
> >
> > Thanks , now it is clear, so that thread was
> concluded(sometimes hard to
> > decide for this) and we will have a slightly
> modified version of Tom's
> > initial interpretation, where data is not
> discarded but not acknowledged
> > either. Then I also would think stopping T7 is
> fine.
> >
> > One more question:
> > During changeover, will data received during RPO
> but not acknowledged,
> > considered as received successfully or not, i.e.
> in case of changeover,
> > will they be delivered to MTP3 or not?
>
> Unacknowledged messages must never be delivered to
> MTP3 or duplication
> will result on changeover.  This is true for Q.703
> as well as M2PA.
[TOLGA]Why so? I would think, when Retrieve BSTN is received, it might be
good to deliver data received successfully but not delivered to MTP3
because of RPO, and then reply to Retrieve BSTN request. This migth save us
to retranmit them on another link (I am not sure though, whether this will
create a problem for MTP3, Q.704 Figure30 Sheet 2 of 6 does not show what
to do in such a case, but I would think this should be fine, I assume it is
possible that MTP2 is delivering a message to MTP3, while MTP3 is sending
Retrieve BSTN. So, MTP3 should be prepared to get something different than
BSTN after sending Retrive BSTN (?) )

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 21 06:25:32 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA26402
	for <sigtran-archive@odin.ietf.org>; Fri, 21 Feb 2003 06:25:32 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1LBWYB14710
	for sigtran-archive@odin.ietf.org; Fri, 21 Feb 2003 06:32:34 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1LBWAp14671;
	Fri, 21 Feb 2003 06:32:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1LBVcp14647
	for <sigtran@optimus.ietf.org>; Fri, 21 Feb 2003 06:31:38 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA26372
	for <sigtran@ietf.org>; Fri, 21 Feb 2003 06:24:04 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:9evWPeWTdVcdJTZDNg4/X6jSAWAQRX8a@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1LBRtD26562;
	Fri, 21 Feb 2003 04:27:55 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1LBRtP25773;
	Fri, 21 Feb 2003 04:27:55 -0700
Date: Fri, 21 Feb 2003 04:27:55 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
Message-ID: <20030221042755.A25531@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OF24FAFD42.687DD8CF-ONC2256CD4.003A2896@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF24FAFD42.687DD8CF-ONC2256CD4.003A2896@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Fri, Feb 21, 2003 at 12:44:30PM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

TOLGA,

TOLGA ASVEREN wrote:                                                                 (Fri, 21 Feb 2003 12:44:30)
> 
> Brian,
> --- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
> > TOLGA,
> >
> > TOLGA ASVEREN wrote:
> >                                 (Fri, 21 Feb 2003
> > 09:43:22)
> > >
> > > Thanks , now it is clear, so that thread was
> > concluded(sometimes hard to
> > > decide for this) and we will have a slightly
> > modified version of Tom's
> > > initial interpretation, where data is not
> > discarded but not acknowledged
> > > either. Then I also would think stopping T7 is
> > fine.
> > >
> > > One more question:
> > > During changeover, will data received during RPO
> > but not acknowledged,
> > > considered as received successfully or not, i.e.
> > in case of changeover,
> > > will they be delivered to MTP3 or not?
> >
> > Unacknowledged messages must never be delivered to
> > MTP3 or duplication
> > will result on changeover.  This is true for Q.703
> > as well as M2PA.
> [TOLGA]Why so? I would think, when Retrieve BSTN is received, it might be
> good to deliver data received successfully but not delivered to MTP3
> because of RPO, and then reply to Retrieve BSTN request. This migth save us
> to retranmit them on another link (I am not sure though, whether this will
> create a problem for MTP3, Q.704 Figure30 Sheet 2 of 6 does not show what
> to do in such a case, but I would think this should be fine, I assume it is
> possible that MTP2 is delivering a message to MTP3, while MTP3 is sending
> Retrieve BSTN. So, MTP3 should be prepared to get something different than
> BSTN after sending Retrive BSTN (?) )

Bad idea.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 21 11:12:07 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA04935
	for <sigtran-archive@odin.ietf.org>; Fri, 21 Feb 2003 11:12:07 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1LGJEx01761
	for sigtran-archive@odin.ietf.org; Fri, 21 Feb 2003 11:19:14 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1LGIsp01752;
	Fri, 21 Feb 2003 11:18:54 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1LGHPp01712
	for <sigtran@optimus.ietf.org>; Fri, 21 Feb 2003 11:17:25 -0500
Received: from CNRI.Reston.VA.US (localhost [127.0.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA04885;
	Fri, 21 Feb 2003 11:09:47 -0500 (EST)
Message-Id: <200302211609.LAA04885@ietf.org>
To: IETF-Announce: ;
Cc: sigtran@ietf.org
From: The IESG <iesg-secretary@ietf.org>
Reply-to: iesg@ietf.org
Date: Fri, 21 Feb 2003 11:09:46 -0500
Subject: [Sigtran] Last Call: Security Considerations for SIGTRAN Protocols to
 Proposed Standard
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


The IESG has received a request from the Signaling Transport Working
Group to consider Security Considerations for SIGTRAN Protocols
<draft-ietf-sigtran-security-01.txt> as a Proposed Standard.

The IESG plans to make a decision in the next few weeks, and solicits
final comments on this action.  Please send any comments to the
iesg@ietf.org or ietf@ietf.org mailing lists by 2003-3-7.

Files can be obtained via
http://www.ietf.org/internet-drafts/draft-ietf-sigtran-security-01.txt



_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 24 06:35:02 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA08246
	for <sigtran-archive@odin.ietf.org>; Mon, 24 Feb 2003 06:35:02 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1OBhVD18368
	for sigtran-archive@odin.ietf.org; Mon, 24 Feb 2003 06:43:31 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1OBhCp18352;
	Mon, 24 Feb 2003 06:43:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1OBf9p18313
	for <sigtran@optimus.ietf.org>; Mon, 24 Feb 2003 06:41:09 -0500
Received: from albatross.wise.edt.ericsson.se (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA08201
	for <sigtran@ietf.org>; Mon, 24 Feb 2003 06:32:06 -0500 (EST)
Received: from esealnt612.al.sw.ericsson.se (esealnt612.al.sw.ericsson.se [153.88.254.71])
	by albatross.wise.edt.ericsson.se (8.12.1/8.12.1/WIREfire-1.4) with ESMTP id h1OBZwKV002629;
	Mon, 24 Feb 2003 12:35:59 +0100 (MET)
Received: from ESEALNT745.al.sw.ericsson.se ([153.88.251.5]) by esealnt612.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55)
	id FDV2A2N9; Mon, 24 Feb 2003 12:35:58 +0100
Received: by ESEALNT745.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <DW9DFF5M>; Mon, 24 Feb 2003 12:35:58 +0100
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B659B2@EESTQNT105.es.eu.ericsson.se>
X-Sybari-Trust: 6449b1e4 9ffcebbb 7a95d2f4 00000138
From: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>,
        Coene Lode
	 <Lode.Coene@siemens.atea.be>
Cc: sigtran@ietf.org
Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-signalling-over-sctp-
	 applic-04.txt
Date: Mon, 24 Feb 2003 12:35:15 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h1OBfAp18315
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Hi Brian, 

All the points you refer here were taken into account during the session edition in Minneapolis. And the result was version 5. If there is any particular point that is not there now, and you want to discuss it, just point it, and we'll show the reasons why we included or excluded it.

Thanks // Javier.


-----Original Message-----
From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
Sent: miércoles, 19 de febrero de 2003 10:57
To: Coene Lode
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] I-D
ACTION:draft-ietf-sigtran-signalling-over-sctp- applic-04.txt


Coene,

Although many of these comments were handled between draft-04 and draft-07,
there are a number that have not been addressed.

Please consider the unresolved comments in this list as comments for LC.

--brian

Coene Lode wrote:                                                                    (Thu, 07 Mar 2002 09:10:05)
> Brian,
> 
> Thank you for the detailed list.
> I will not answer them point for point as the document itself was/is not
> fully baked yet. 
> With your input It looks like we'll get much nearer that goal.
> We will take into account all the points you mentioned during our editing in
> Minneapolis.
> I hope that after a new release all your points will have taken care of.
> 
> yours sincerly,
> Lode
> 
> Siemens atea
> atealaan 34          2200 Herentals, Belgium
> E-mail: lode.coene@siemens.atea.be
> Tel: +32-14-252081
> Fax: +32-14-253212
> 
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: Wednesday, March 06, 2002 11:47 AM
> > To: Coene Lode
> > Cc: sigtran@ietf.org
> > Subject: Re: [Sigtran] I-D
> > ACTION:draft-ietf-sigtran-signalling-over-sctp-applic-04.txt
> > 
> > 
> > Lode,
> > 
> > A few comments on this draft:
> > 
> > Section 2:
> > 
> > >   ...            Message boundaries are preserved during data
> > >   transport and so no message delineation is needed. ...
> > 
> > Yet, each UA message has a common header which contains a label
> > and a length which indicates the length of the message.  This is
> > a message delimitation.  Why is it present in each UA if it is
> > not required?
> > 
> > >                                                  ... The 
> > user data can
> > >   be delivered by the order of transmission within a stream(in
> > >   sequence delivery) or the order of arrival.
> > 
> > Unordered delivery in SCTP and "order of arrival" delivery 
> > are different
> > things.  SCTP is permitted to place a priority on out of order packets
> > from various streams.  Preference for delivery of stream "0" unordered
> > packets ahead of other stream unordered packets is of 
> > significant advantage
> > to most UAs (regarless of order of arrival).  Please use the term
> > "unordered delivery" instead of "order of arrival".
> > 
> > >   SCTP can be used to provide redundancy and fault tolerance at the
> > >   transport layer and below. Telephony applications needing 
> > this level
> > >   of fault tolerance can make use of SCTP's multi-homing support.
> > 
> > "Fault tolerance" usually has a different connotation when used in
> > telephony related computing.  It indicates a level of reliability
> > (greater than 5 9's).  Please reword as follows:
> > 
> > <   SCTP can be used to provide redundancy at the transport layer and
> > <   below. Telephony applications needing this level of redundancy can
> > <   make use of SCTP's multi-homing support.
> > 
> > Section 3.1:
> > 
> > >                                               ...  Additionally,
> > >   telephony applications may use SCTP stream based flow control
> > >   [SCTPFLOW].
> > 
> > I believe that stream based flow controls were dropped (in favor of
> > application level flow controls).  Please remove this sentence.
> > 
> > Section 3.2:
> > 
> > >   Telephony systems often must achieve high availability in 
> > operation.
> > 
> > Again, "high availability" has the connotation of a specific level
> > of reliability in telephony related computing (in this case less than
> > 5 9's).  Please reword as:
> > 
> > <   Telephony systems often must achieve redundancy in operation.
> > 
> > or
> > 
> > <   Telephony systems often must have no single point of failure in
> > <   operation.
> > 
> > >   transport layer and below must be detected very fast so that the
> > 
> > grammar, reword as:
> > 
> > <   transport layer and below must be detected quickly so that the
> > 
> > >   For example, they are often required to be able to preserve stable
> > >   calls during a component failure. Therefore error 
> > situations at the
> > >   transport layer and below must be detected very fast so that the
> > >   application can take approriate steps to recover and preserve the
> > >   stable calls. ...
> > 
> > This example is a poor one and shows an utter lack of understanding
> > for telephony.  Stable calls have *no* signalling associated 
> > with them.
> > It is only a call in transition that has signalling.  Signalling
> > transport need not worry about affecting stable calls.  Only redundant
> > systems maintaining call state need worry about this.  While this is
> > a system requirement for an MG or MGC, it has nothing to do with
> > signalling components.  Also, sharing of call state between redundant
> > systems is explicitly excluded from the scope of the UAs.  Please
> > remove this example: it makes the applicability statement 
> > sound foolish.
> > 
> > On the other hand, telephony applications pose requirements on the
> > availability of signalling paths to meet service availability and
> > performance requirements.  SS7 has such signalling relation 
> > availability
> > requirements.  These are the telephony requirements which impose on
> > SCTP's ability to detect failures.  (Again, nothing to do with call
> > state.)  Please replace this example in section 3.2 with words to this
> > effect.
> > 
> > Section 3.2.2:
> > 
> > >   For faster detection of (un)availability of idle paths, the
> > >   telephony application may consider lowering the SCTP parameter
> > >   HB.interval. It should be noted this will result in a 
> > higher traffic
> > >   load.
> > 
> > It is not always true that lowered HB.interval will incresae traffic
> > load.  Please change "will result" to "might result".
> > 
> > Section 3.2.3:
> > 
> > >   Setting Path.Max.Retrans and Association.Max.Retrans SCTP 
> > parameters
> > >   to lower values will speed up both destination address and peer
> > >   failure detection. However, if these values are set too low, the
> > >   probability of false detections will increase.
> > 
> > What false detection?  A retransmission is a retransmission.  There is
> > no "false" retransmission or "false detection" mentioned anywhere.
> > In tests we have shown that Path.Max.Retrans of zero (0) (changing
> > destination address on every retransmission) works quite 
> > fine.  I don't
> > understand what you are trying to warn against here...
> > 
> > Section 3.3:
> > 
> > >   Telephony applications often require short end-to-end message
> > >   delays.  The methods described in section 3.2.1 on 
> > lowering RTO and
> > >   not using delayed SACK may be considered.
> > 
> > Not using delayed SACK does nothing to shorten end-to-end message
> > delay.  Please remove it.
> > 
> > Section 4:
> > 
> > >   Users Adaptation Layers are defined to substitute the telephony
> > >   signaling protocol that is below of the telephony 
> > signaling protocol
> > >   to be relayed.
> > 
> > "that is below of the"????  This sentence is so malformed I don't know
> > what it means.  Please correct.
> > 
> > Section 4:
> > 
> > >      - Support management of SCTP transport associations and traffic
> > >         between SGs and ISEPs or two ISEPs
> > 
> > It is a sore point that several UAs do *not* adequately support
> > management between "two ISEPs."
> > 
> > Section 4:
> > 
> > >                 SP:   Signaling Protocol
> > 
> > What is shown is not a signalling protocol, but a signalling protocol
> > user.  Please change to SPU (Signalling Protocol User).
> > 
> > >                 TTST: Traditional Telephony Signaling Transport
> > 
> > The only traditional SS7 signalling that qualifies as a transport
> > is TCAP.  I suggest you use "Protocol" instead of "Transport".
> > Please replace TTST with TTSP (Traditional Telephony Signalling
> > Protocol).
> > 
> > >   It is also referred as SG to AS communication. AS I the name that
> > 
> > Typo: change "AS I the name" to "AS is the name"
> > 
> > >   It is also referred as IPSP communication. IPSP is the name of the
> > >   role that an IP-based node plays UAL usually gives to the ISEP
> > >   nodes. It stands for IP Signaling Point.
> > 
> > English usage.  Please reword this entire paragraph (especially the
> > middle line "IP-based node plays UAL usually").
> > 
> > Section 4.1:
> > 
> > >                                             ... The default 
> > orientation
> > >   would be for the SG to take on the role of server while 
> > the ISEP is
> > >   the client. ...
> > 
> > Grammar: change "would be for" to "is for".
> > 
> > >   Examples of the upper layers to be transported would be Q.931 and
> > >   QSIG.
> > 
> > Grammar: change "would be" to "are".
> > 
> > >   communication where the ISEP role is typically played by a node
> > >   called MGC defined in [RFC2719].
> > 
> > Grammar: change "called MGC defined" to "called an MGC, as defined".
> > 
> > >   The value assigned by IANA for the Payload Protocol 
> > Identifier in the
> > >   SCTP Payload Data chunk is ô1ö
> > 
> > Non-ascii characters at the end of this sentence.
> > 
> > Section 4.2:
> > 
> > >                                                   ... It is 
> > basically
> > >   implemented in an interworking scenario with SG.
> > 
> > Usage: change "basically" to "typically".
> > 
> > >           AN    û Access Network
> > >           NIF   û Nodal Interworking Function
> > >           LAPV5 û Link Access Protocol for the V5 channel
> > 
> > Non-ascii characters in text.
> > 
> > Section 4.3:
> > 
> > >   The DUA is built on top of IUA defining the necessary 
> > extensions to
> > >   IUA for a DPNSS/DASS2 transport. ...
> > 
> > grammar: change "top of IAU defining" to "top of IUA and defines"
> > 
> > >                                ... DPNSS stands for Digital Private
> > >   Network Signaling System and DASS2 for Digital Access Signaling
> > >   System No 2
> > 
> > Missing period at end of sentence.
> > 
> > >   The value assigned by IANA for the Payload Protocol 
> > Identifier in the
> > >   SCTP Payload Data chunk is ôTBDö.
> > 
> > Non-ascii text at end of sentence.
> > 
> > Does not DUA have a PPI assigned yet?
> > 
> > Section 4.4:
> > 
> > >   This protocol would be mainly used between a Signaling 
> > Gateway (SG)
> > >   and Media Gateway Controler (MGC). ...
> > 
> > Grammar: change "would be" to "is"; change "mainly" to "typically" or
> > "largely" or "usually".
> > 
> > >   The only SS7 MTP2 User is MTP3 that is the protocol transported by
> > >   this UAL.
> > 
> > Really bad sentence formation:  change to:
> > 
> > <   MTP3 is the only MTP2-User that is transport by this UAL.
> > 
> > Besides, the statement is incorrect: MTP3b could also be transported.
> > 
> > >   The SG provides a interworking of transport functions with the IP
> > >   transport, in order to transfer the MTP2-User signaling 
> > messages to
> > >   and from an Application Server (e.g. MGC) where the peer MTP2-
> > >   User protocol layer exists.
> > 
> > This is a relly awkward paragraph.  I think you mean "interworking of
> > SS7 transport functions with IP transport functions".  Change 
> > "in order
> > to" to "to", change "the MTP2-User" to "MTP2-User"; change 
> > "to and from"
> > to "with"; change "Application Server (e.g., MGC) where the peer
> > MTP2-User protocol layer exists" to "MTP2-User at an 
> > Application Server
> > (e.g., MGC)."
> > 
> > >   The value assigned by IANA for the Payload Protocol 
> > Identifier in the
> > >   SCTP Payload Data chunk is ô2ö
> > 
> > Non-ascii text at end of sentence.
> > 
> > Section 4.5:
> > 
> > >   This protocol is used between SS7 Signaling Points 
> > employing the MTP
> > >   Level 3 protocol. The SS7 Signaling Points may also 
> > employ standard
> > >   SS7 links using the SS7 MTP Layer 2 to provide transport 
> > of MTP Layer
> > >   3 signaling messages.
> > 
> > Word usage: change "employing" to "using" and "employ" to "use".
> > 
> > There is no such thing as "MTP Layer 2" or "MTP Layer 3"; in SS7 it is
> > referred to as "MTP Level 2" and "MTP Level 3".
> > 
> > >   Both configurations: intercommunication of SS7 and IP with SG and
> > >   communication between ISEPs are possible.
> > 
> > There is no such word as "intercommunication".  "Communication" is
> > already between two things ("inter").
> > 
> > Besides, M2PA only provides for communication between SPs.  There is
> > no need for the two diagrams for M2PA.  There is no 
> > distinction between
> > the first diagram and the right-hand side of the second diagram for
> > M2PA.
> > 
> > >   These figures are only an example. Other configurations 
> > are possible.
> > >   For example, IPSPs without traditional SS7 links could use the
> > >   protocol layers MTP3/M2PA/SCTP/IP to route SS7 messages 
> > in a network
> > >   with all IP links.
> > >
> > >   Another example is that two SGs could be connected over 
> > an IP network
> > >   to form an SG mated pair similar to the way STPs are 
> > provisioned in
> > >   traditional SS7 networks.
> > 
> > These statements sound foolish.  M2PA is really just a different sort
> > of SS7 link.  Plesae remove.
> > 
> > >   The value assigned by IANA for the Payload Protocol 
> > Identifier in the
> > >   SCTP Payload Data chunk is TBD
> > 
> > Does not M2PA have a PPI yet?
> > 
> > >      b. M2PA: SG-IPSP connection is an SS7 link.
> > >         M2UA: SG-MGC connection is not an SS7 link. It is an
> > >               extension of MTP to a remote entity.
> > 
> > There is nothing stopping M2UA from being used in a peer-to-peer
> > fashion.
> > 
> > >      c. M2PA: SG is an SS7 node with a point code.
> > >         M2UA: SG is not an SS7 node and has no point code.
> > 
> > This is not necessarily a correct statement.  There is nothing
> > restricting an M2UA SG from having a point code or being an
> > SS7 node in its own right.
> > 
> > >      d. M2PA: SG can have upper SS7 layers, e.g., SCCP.
> > >         M2UA: SG does not have upper SS7 layers since it 
> > has no MTP3.
> > 
> > Same is true here.  The above statement is false.
> > 
> > >      e. M2PA: relies on MTP3 for management procedures.
> > >         M2UA: uses M2UA management procedures.
> > 
> > If need explanation of the absurdity of this statement, consider
> > this SG:
> >                      ____
> >                     |
> >                     |
> >                 +------+
> >                 | M2UA |
> >                 +------+
> >                 | M2PA |
> >                 +------+
> >                     |
> >              _______|
> > 
> > Which is referred to in the M2UA document.
> > 
> > This SG is an M2PA SG, but it has no point code, has no local MTP3,
> > has no local upper layers (SCCP), and relies on M2UA for management
> > procedures.
> > 
> > Section 4.6:
> > 
> > >      - Interconnection of SS7 and IP nodes
> > 
> > Grammar: replace "Interconection" with "Connection".
> > 
> > 
> > >   It works using the client-server philosophy. ...
> > 
> > This might be ok in German, but is awkward in English.  Reword as:
> > 
> > <   M3UA uses a client-server architecture.
> > 
> > >                                            ... ISEP is 
> > recommended to
> > >   be client when talking with a SG.
> > 
> > <   It is recommended that the ISEP act as the client and initiate
> > <   SCTP associations with the SG.
> > 
> > >                                 ... The reserved port by 
> > IANA is 2905
> > >   to listen to possible client connections.
> > 
> > <   The port reserved by IANA is port number 2905.  This is the port
> > <   upon which the SG should listen for client connections.
> > 
> > >   The assigned payload protocol identifier for the SCTP 
> > DATA chunks is
> > >   ô3ö.
> > 
> > Non-ascii text at end of sentence.
> > 
> > 
> > Section 4.7:
> > 
> > >                 ... SUA can support only non-call related signaling.
> > 
> > Almost all SUA signalling is call related.  I think you mean
> > non-call-associated signalling; but, then, *all* SS7 signalling is
> > non-call-associated signalling.  Please strike this sentence.
> > 
> > >   SUA does not pose stringent timing constraints on SCTP due to the
> > >   fact that SUA applications have broad timing requirement 
> > (from 10 of
> > >   seconds to hours) which the applications guard themselves and the
> > >   timing supervision of the application is end-to-end, not hop-by-
> > >   hop(as with ISUP).
> > 
> > This statement is false and misleading.  SCCP, particularly for the
> > cited applications: MAP, INAP, SMS, BSSAP, RANAP, has very stringent
> > timing requirements, often more stringent than ISUP due to the fact
> > that these represent transaction which are made during ISUP 
> > call setup.
> > 
> > Tens of seconds to hours is false; and atypical.  Transaction timers
> > an performance requirements for INAP, for example, are 1 
> > second maximum
> > with 100 ms nominal query/response times.  Please remove this 
> > paragraph.
> > 
> > >      - Interconnection of SS7 and IP
> > 
> > Change "Interconnection" to "Connection".
> > 
> > >   connections. The payload protocol identifier for the SCTP 
> > DATA chunks
> > >   is ô4ö.
> > 
> > Non-ascii text.
> > 
> > 
> > General:
> > 
> > There is a mixture of "signalling" (two l's British/Canadian spelling)
> > and "signaling" (American spelling).  As the title contains signalling
> > with two l's, please use signalling with two l's throughout.
> > 
> > --brian
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 24 06:37:50 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA08361
	for <sigtran-archive@odin.ietf.org>; Mon, 24 Feb 2003 06:37:50 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1OBkJX18467
	for sigtran-archive@odin.ietf.org; Mon, 24 Feb 2003 06:46:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1OBk4p18457;
	Mon, 24 Feb 2003 06:46:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1OBjFp18435
	for <sigtran@optimus.ietf.org>; Mon, 24 Feb 2003 06:45:15 -0500
Received: from albatross.wise.edt.ericsson.se (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA08302
	for <sigtran@ietf.org>; Mon, 24 Feb 2003 06:36:15 -0500 (EST)
Received: from esealnt612.al.sw.ericsson.se (esealnt612.al.sw.ericsson.se [153.88.254.71])
	by albatross.wise.edt.ericsson.se (8.12.1/8.12.1/WIREfire-1.4) with ESMTP id h1OBe7KW003906;
	Mon, 24 Feb 2003 12:40:07 +0100 (MET)
Received: from ESEALNT744.al.sw.ericsson.se ([153.88.251.4]) by esealnt612.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55)
	id FDV2AJML; Mon, 24 Feb 2003 12:40:06 +0100
Received: by ESEALNT744.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <1ZX17AC6>; Mon, 24 Feb 2003 12:40:06 +0100
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B659B3@EESTQNT105.es.eu.ericsson.se>
X-Sybari-Trust: 9d48c6a4 9ffcebbb 7a95d2f4 00000138
From: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>,
        "'Coene Lode'"
	 <Lode.Coene@siemens.atea.be>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Recall: [Sigtran] I-D ACTION:draft-ietf-sigtran-signalling-over-s
	ctp- applic-04.txt
Date: Mon, 24 Feb 2003 12:39:27 +0100
Expiry-Date: Wed, 26 Feb 2003 12:40:00 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Javier Pastor-Balbas (ECE) would like to recall the message, "[Sigtran] I-D ACTION:draft-ietf-sigtran-signalling-over-sctp- applic-04.txt".
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 24 07:23:06 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA11031
	for <sigtran-archive@odin.ietf.org>; Mon, 24 Feb 2003 07:23:06 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1OCVaa21003
	for sigtran-archive@odin.ietf.org; Mon, 24 Feb 2003 07:31:36 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1OCV8p20978;
	Mon, 24 Feb 2003 07:31:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1OCULp20916
	for <sigtran@optimus.ietf.org>; Mon, 24 Feb 2003 07:30:21 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA10897
	for <sigtran@ietf.org>; Mon, 24 Feb 2003 07:21:17 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:LaaFBlXjSJUTJlr1gP2XOcRfvFMkZh4x@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1OCP9D30259;
	Mon, 24 Feb 2003 05:25:10 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1OCP9A16574;
	Mon, 24 Feb 2003 05:25:09 -0700
Date: Mon, 24 Feb 2003 05:25:09 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>
Cc: Coene Lode <Lode.Coene@siemens.atea.be>, sigtran@ietf.org
Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-signalling-over-sctp- applic-04.txt
Message-ID: <20030224052509.B16375@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>,
	Coene Lode <Lode.Coene@siemens.atea.be>, sigtran@ietf.org
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B659B2@EESTQNT105.es.eu.ericsson.se>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <1AB3D30B989BF141BBD5C70057B2EF7C01B659B2@EESTQNT105.es.eu.ericsson.se>; from javier.pastor-balbas@ece.ericsson.se on Mon, Feb 24, 2003 at 12:35:15PM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: 8bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Javier,

Please consider all the points not included below as comments/issues for Last Call.

--brian

Javier Pastor-Balbas (ECE) wrote:                                                   (Mon, 24 Feb 2003 12:35:15)
> Hi Brian, 
> 
> All the points you refer here were taken into account during the session
> edition in Minneapolis. And the result was version 5. If there is any
> particular point that is not there now, and you want to discuss it, just
> point it, and we'll show the reasons why we included or excluded it.
> 
> Thanks // Javier.
> 
> 
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: miércoles, 19 de febrero de 2003 10:57
> To: Coene Lode
> Cc: sigtran@ietf.org
> Subject: Re: [Sigtran] I-D
> ACTION:draft-ietf-sigtran-signalling-over-sctp- applic-04.txt
> 
> 
> Coene,
> 
> Although many of these comments were handled between draft-04 and draft-07,
> there are a number that have not been addressed.
> 
> Please consider the unresolved comments in this list as comments for LC.
> 
> --brian
> 
> Coene Lode wrote:                                                                    (Thu, 07 Mar 2002 09:10:05)
> > Brian,
> > 
> > Thank you for the detailed list.
> > I will not answer them point for point as the document itself was/is not
> > fully baked yet. 
> > With your input It looks like we'll get much nearer that goal.
> > We will take into account all the points you mentioned during our editing in
> > Minneapolis.
> > I hope that after a new release all your points will have taken care of.
> > 
> > yours sincerly,
> > Lode
> > 
> > Siemens atea
> > atealaan 34          2200 Herentals, Belgium
> > E-mail: lode.coene@siemens.atea.be
> > Tel: +32-14-252081
> > Fax: +32-14-253212
> > 
> > 
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: Wednesday, March 06, 2002 11:47 AM
> > > To: Coene Lode
> > > Cc: sigtran@ietf.org
> > > Subject: Re: [Sigtran] I-D
> > > ACTION:draft-ietf-sigtran-signalling-over-sctp-applic-04.txt
> > > 
> > > 
> > > Lode,
> > > 
> > > A few comments on this draft:
> > > 
> > > Section 2:
> > > 
> > > >   ...            Message boundaries are preserved during data
> > > >   transport and so no message delineation is needed. ...
> > > 
> > > Yet, each UA message has a common header which contains a label
> > > and a length which indicates the length of the message.  This is
> > > a message delimitation.  Why is it present in each UA if it is
> > > not required?
> > > 
> > > >                                                  ... The 
> > > user data can
> > > >   be delivered by the order of transmission within a stream(in
> > > >   sequence delivery) or the order of arrival.
> > > 
> > > Unordered delivery in SCTP and "order of arrival" delivery 
> > > are different
> > > things.  SCTP is permitted to place a priority on out of order packets
> > > from various streams.  Preference for delivery of stream "0" unordered
> > > packets ahead of other stream unordered packets is of 
> > > significant advantage
> > > to most UAs (regarless of order of arrival).  Please use the term
> > > "unordered delivery" instead of "order of arrival".
> > > 
> > > >   SCTP can be used to provide redundancy and fault tolerance at the
> > > >   transport layer and below. Telephony applications needing 
> > > this level
> > > >   of fault tolerance can make use of SCTP's multi-homing support.
> > > 
> > > "Fault tolerance" usually has a different connotation when used in
> > > telephony related computing.  It indicates a level of reliability
> > > (greater than 5 9's).  Please reword as follows:
> > > 
> > > <   SCTP can be used to provide redundancy at the transport layer and
> > > <   below. Telephony applications needing this level of redundancy can
> > > <   make use of SCTP's multi-homing support.
> > > 
> > > Section 3.1:
> > > 
> > > >                                               ...  Additionally,
> > > >   telephony applications may use SCTP stream based flow control
> > > >   [SCTPFLOW].
> > > 
> > > I believe that stream based flow controls were dropped (in favor of
> > > application level flow controls).  Please remove this sentence.
> > > 
> > > Section 3.2:
> > > 
> > > >   Telephony systems often must achieve high availability in 
> > > operation.
> > > 
> > > Again, "high availability" has the connotation of a specific level
> > > of reliability in telephony related computing (in this case less than
> > > 5 9's).  Please reword as:
> > > 
> > > <   Telephony systems often must achieve redundancy in operation.
> > > 
> > > or
> > > 
> > > <   Telephony systems often must have no single point of failure in
> > > <   operation.
> > > 
> > > >   transport layer and below must be detected very fast so that the
> > > 
> > > grammar, reword as:
> > > 
> > > <   transport layer and below must be detected quickly so that the
> > > 
> > > >   For example, they are often required to be able to preserve stable
> > > >   calls during a component failure. Therefore error 
> > > situations at the
> > > >   transport layer and below must be detected very fast so that the
> > > >   application can take approriate steps to recover and preserve the
> > > >   stable calls. ...
> > > 
> > > This example is a poor one and shows an utter lack of understanding
> > > for telephony.  Stable calls have *no* signalling associated 
> > > with them.
> > > It is only a call in transition that has signalling.  Signalling
> > > transport need not worry about affecting stable calls.  Only redundant
> > > systems maintaining call state need worry about this.  While this is
> > > a system requirement for an MG or MGC, it has nothing to do with
> > > signalling components.  Also, sharing of call state between redundant
> > > systems is explicitly excluded from the scope of the UAs.  Please
> > > remove this example: it makes the applicability statement 
> > > sound foolish.
> > > 
> > > On the other hand, telephony applications pose requirements on the
> > > availability of signalling paths to meet service availability and
> > > performance requirements.  SS7 has such signalling relation 
> > > availability
> > > requirements.  These are the telephony requirements which impose on
> > > SCTP's ability to detect failures.  (Again, nothing to do with call
> > > state.)  Please replace this example in section 3.2 with words to this
> > > effect.
> > > 
> > > Section 3.2.2:
> > > 
> > > >   For faster detection of (un)availability of idle paths, the
> > > >   telephony application may consider lowering the SCTP parameter
> > > >   HB.interval. It should be noted this will result in a 
> > > higher traffic
> > > >   load.
> > > 
> > > It is not always true that lowered HB.interval will incresae traffic
> > > load.  Please change "will result" to "might result".
> > > 
> > > Section 3.2.3:
> > > 
> > > >   Setting Path.Max.Retrans and Association.Max.Retrans SCTP 
> > > parameters
> > > >   to lower values will speed up both destination address and peer
> > > >   failure detection. However, if these values are set too low, the
> > > >   probability of false detections will increase.
> > > 
> > > What false detection?  A retransmission is a retransmission.  There is
> > > no "false" retransmission or "false detection" mentioned anywhere.
> > > In tests we have shown that Path.Max.Retrans of zero (0) (changing
> > > destination address on every retransmission) works quite 
> > > fine.  I don't
> > > understand what you are trying to warn against here...
> > > 
> > > Section 3.3:
> > > 
> > > >   Telephony applications often require short end-to-end message
> > > >   delays.  The methods described in section 3.2.1 on 
> > > lowering RTO and
> > > >   not using delayed SACK may be considered.
> > > 
> > > Not using delayed SACK does nothing to shorten end-to-end message
> > > delay.  Please remove it.
> > > 
> > > Section 4:
> > > 
> > > >   Users Adaptation Layers are defined to substitute the telephony
> > > >   signaling protocol that is below of the telephony 
> > > signaling protocol
> > > >   to be relayed.
> > > 
> > > "that is below of the"????  This sentence is so malformed I don't know
> > > what it means.  Please correct.
> > > 
> > > Section 4:
> > > 
> > > >      - Support management of SCTP transport associations and traffic
> > > >         between SGs and ISEPs or two ISEPs
> > > 
> > > It is a sore point that several UAs do *not* adequately support
> > > management between "two ISEPs."
> > > 
> > > Section 4:
> > > 
> > > >                 SP:   Signaling Protocol
> > > 
> > > What is shown is not a signalling protocol, but a signalling protocol
> > > user.  Please change to SPU (Signalling Protocol User).
> > > 
> > > >                 TTST: Traditional Telephony Signaling Transport
> > > 
> > > The only traditional SS7 signalling that qualifies as a transport
> > > is TCAP.  I suggest you use "Protocol" instead of "Transport".
> > > Please replace TTST with TTSP (Traditional Telephony Signalling
> > > Protocol).
> > > 
> > > >   It is also referred as SG to AS communication. AS I the name that
> > > 
> > > Typo: change "AS I the name" to "AS is the name"
> > > 
> > > >   It is also referred as IPSP communication. IPSP is the name of the
> > > >   role that an IP-based node plays UAL usually gives to the ISEP
> > > >   nodes. It stands for IP Signaling Point.
> > > 
> > > English usage.  Please reword this entire paragraph (especially the
> > > middle line "IP-based node plays UAL usually").
> > > 
> > > Section 4.1:
> > > 
> > > >                                             ... The default 
> > > orientation
> > > >   would be for the SG to take on the role of server while 
> > > the ISEP is
> > > >   the client. ...
> > > 
> > > Grammar: change "would be for" to "is for".
> > > 
> > > >   Examples of the upper layers to be transported would be Q.931 and
> > > >   QSIG.
> > > 
> > > Grammar: change "would be" to "are".
> > > 
> > > >   communication where the ISEP role is typically played by a node
> > > >   called MGC defined in [RFC2719].
> > > 
> > > Grammar: change "called MGC defined" to "called an MGC, as defined".
> > > 
> > > >   The value assigned by IANA for the Payload Protocol 
> > > Identifier in the
> > > >   SCTP Payload Data chunk is ô1ö
> > > 
> > > Non-ascii characters at the end of this sentence.
> > > 
> > > Section 4.2:
> > > 
> > > >                                                   ... It is 
> > > basically
> > > >   implemented in an interworking scenario with SG.
> > > 
> > > Usage: change "basically" to "typically".
> > > 
> > > >           AN    û Access Network
> > > >           NIF   û Nodal Interworking Function
> > > >           LAPV5 û Link Access Protocol for the V5 channel
> > > 
> > > Non-ascii characters in text.
> > > 
> > > Section 4.3:
> > > 
> > > >   The DUA is built on top of IUA defining the necessary 
> > > extensions to
> > > >   IUA for a DPNSS/DASS2 transport. ...
> > > 
> > > grammar: change "top of IAU defining" to "top of IUA and defines"
> > > 
> > > >                                ... DPNSS stands for Digital Private
> > > >   Network Signaling System and DASS2 for Digital Access Signaling
> > > >   System No 2
> > > 
> > > Missing period at end of sentence.
> > > 
> > > >   The value assigned by IANA for the Payload Protocol 
> > > Identifier in the
> > > >   SCTP Payload Data chunk is ôTBDö.
> > > 
> > > Non-ascii text at end of sentence.
> > > 
> > > Does not DUA have a PPI assigned yet?
> > > 
> > > Section 4.4:
> > > 
> > > >   This protocol would be mainly used between a Signaling 
> > > Gateway (SG)
> > > >   and Media Gateway Controler (MGC). ...
> > > 
> > > Grammar: change "would be" to "is"; change "mainly" to "typically" or
> > > "largely" or "usually".
> > > 
> > > >   The only SS7 MTP2 User is MTP3 that is the protocol transported by
> > > >   this UAL.
> > > 
> > > Really bad sentence formation:  change to:
> > > 
> > > <   MTP3 is the only MTP2-User that is transport by this UAL.
> > > 
> > > Besides, the statement is incorrect: MTP3b could also be transported.
> > > 
> > > >   The SG provides a interworking of transport functions with the IP
> > > >   transport, in order to transfer the MTP2-User signaling 
> > > messages to
> > > >   and from an Application Server (e.g. MGC) where the peer MTP2-
> > > >   User protocol layer exists.
> > > 
> > > This is a relly awkward paragraph.  I think you mean "interworking of
> > > SS7 transport functions with IP transport functions".  Change 
> > > "in order
> > > to" to "to", change "the MTP2-User" to "MTP2-User"; change 
> > > "to and from"
> > > to "with"; change "Application Server (e.g., MGC) where the peer
> > > MTP2-User protocol layer exists" to "MTP2-User at an 
> > > Application Server
> > > (e.g., MGC)."
> > > 
> > > >   The value assigned by IANA for the Payload Protocol 
> > > Identifier in the
> > > >   SCTP Payload Data chunk is ô2ö
> > > 
> > > Non-ascii text at end of sentence.
> > > 
> > > Section 4.5:
> > > 
> > > >   This protocol is used between SS7 Signaling Points 
> > > employing the MTP
> > > >   Level 3 protocol. The SS7 Signaling Points may also 
> > > employ standard
> > > >   SS7 links using the SS7 MTP Layer 2 to provide transport 
> > > of MTP Layer
> > > >   3 signaling messages.
> > > 
> > > Word usage: change "employing" to "using" and "employ" to "use".
> > > 
> > > There is no such thing as "MTP Layer 2" or "MTP Layer 3"; in SS7 it is
> > > referred to as "MTP Level 2" and "MTP Level 3".
> > > 
> > > >   Both configurations: intercommunication of SS7 and IP with SG and
> > > >   communication between ISEPs are possible.
> > > 
> > > There is no such word as "intercommunication".  "Communication" is
> > > already between two things ("inter").
> > > 
> > > Besides, M2PA only provides for communication between SPs.  There is
> > > no need for the two diagrams for M2PA.  There is no 
> > > distinction between
> > > the first diagram and the right-hand side of the second diagram for
> > > M2PA.
> > > 
> > > >   These figures are only an example. Other configurations 
> > > are possible.
> > > >   For example, IPSPs without traditional SS7 links could use the
> > > >   protocol layers MTP3/M2PA/SCTP/IP to route SS7 messages 
> > > in a network
> > > >   with all IP links.
> > > >
> > > >   Another example is that two SGs could be connected over 
> > > an IP network
> > > >   to form an SG mated pair similar to the way STPs are 
> > > provisioned in
> > > >   traditional SS7 networks.
> > > 
> > > These statements sound foolish.  M2PA is really just a different sort
> > > of SS7 link.  Plesae remove.
> > > 
> > > >   The value assigned by IANA for the Payload Protocol 
> > > Identifier in the
> > > >   SCTP Payload Data chunk is TBD
> > > 
> > > Does not M2PA have a PPI yet?
> > > 
> > > >      b. M2PA: SG-IPSP connection is an SS7 link.
> > > >         M2UA: SG-MGC connection is not an SS7 link. It is an
> > > >               extension of MTP to a remote entity.
> > > 
> > > There is nothing stopping M2UA from being used in a peer-to-peer
> > > fashion.
> > > 
> > > >      c. M2PA: SG is an SS7 node with a point code.
> > > >         M2UA: SG is not an SS7 node and has no point code.
> > > 
> > > This is not necessarily a correct statement.  There is nothing
> > > restricting an M2UA SG from having a point code or being an
> > > SS7 node in its own right.
> > > 
> > > >      d. M2PA: SG can have upper SS7 layers, e.g., SCCP.
> > > >         M2UA: SG does not have upper SS7 layers since it 
> > > has no MTP3.
> > > 
> > > Same is true here.  The above statement is false.
> > > 
> > > >      e. M2PA: relies on MTP3 for management procedures.
> > > >         M2UA: uses M2UA management procedures.
> > > 
> > > If need explanation of the absurdity of this statement, consider
> > > this SG:
> > >                      ____
> > >                     |
> > >                     |
> > >                 +------+
> > >                 | M2UA |
> > >                 +------+
> > >                 | M2PA |
> > >                 +------+
> > >                     |
> > >              _______|
> > > 
> > > Which is referred to in the M2UA document.
> > > 
> > > This SG is an M2PA SG, but it has no point code, has no local MTP3,
> > > has no local upper layers (SCCP), and relies on M2UA for management
> > > procedures.
> > > 
> > > Section 4.6:
> > > 
> > > >      - Interconnection of SS7 and IP nodes
> > > 
> > > Grammar: replace "Interconection" with "Connection".
> > > 
> > > 
> > > >   It works using the client-server philosophy. ...
> > > 
> > > This might be ok in German, but is awkward in English.  Reword as:
> > > 
> > > <   M3UA uses a client-server architecture.
> > > 
> > > >                                            ... ISEP is 
> > > recommended to
> > > >   be client when talking with a SG.
> > > 
> > > <   It is recommended that the ISEP act as the client and initiate
> > > <   SCTP associations with the SG.
> > > 
> > > >                                 ... The reserved port by 
> > > IANA is 2905
> > > >   to listen to possible client connections.
> > > 
> > > <   The port reserved by IANA is port number 2905.  This is the port
> > > <   upon which the SG should listen for client connections.
> > > 
> > > >   The assigned payload protocol identifier for the SCTP 
> > > DATA chunks is
> > > >   ô3ö.
> > > 
> > > Non-ascii text at end of sentence.
> > > 
> > > 
> > > Section 4.7:
> > > 
> > > >                 ... SUA can support only non-call related signaling.
> > > 
> > > Almost all SUA signalling is call related.  I think you mean
> > > non-call-associated signalling; but, then, *all* SS7 signalling is
> > > non-call-associated signalling.  Please strike this sentence.
> > > 
> > > >   SUA does not pose stringent timing constraints on SCTP due to the
> > > >   fact that SUA applications have broad timing requirement 
> > > (from 10 of
> > > >   seconds to hours) which the applications guard themselves and the
> > > >   timing supervision of the application is end-to-end, not hop-by-
> > > >   hop(as with ISUP).
> > > 
> > > This statement is false and misleading.  SCCP, particularly for the
> > > cited applications: MAP, INAP, SMS, BSSAP, RANAP, has very stringent
> > > timing requirements, often more stringent than ISUP due to the fact
> > > that these represent transaction which are made during ISUP 
> > > call setup.
> > > 
> > > Tens of seconds to hours is false; and atypical.  Transaction timers
> > > an performance requirements for INAP, for example, are 1 
> > > second maximum
> > > with 100 ms nominal query/response times.  Please remove this 
> > > paragraph.
> > > 
> > > >      - Interconnection of SS7 and IP
> > > 
> > > Change "Interconnection" to "Connection".
> > > 
> > > >   connections. The payload protocol identifier for the SCTP 
> > > DATA chunks
> > > >   is ô4ö.
> > > 
> > > Non-ascii text.
> > > 
> > > 
> > > General:
> > > 
> > > There is a mixture of "signalling" (two l's British/Canadian spelling)
> > > and "signaling" (American spelling).  As the title contains signalling
> > > with two l's, please use signalling with two l's throughout.
> > > 
> > > --brian
> > > 
> > > -- 
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > 
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Mon Feb 24 18:31:26 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA00200
	for <sigtran-archive@odin.ietf.org>; Mon, 24 Feb 2003 18:31:26 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1ONe9k02080
	for sigtran-archive@odin.ietf.org; Mon, 24 Feb 2003 18:40:09 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1ONdop02027;
	Mon, 24 Feb 2003 18:39:51 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1ONcPp02002
	for <sigtran@optimus.ietf.org>; Mon, 24 Feb 2003 18:38:25 -0500
Received: from w2ksjexg01.ciena.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA00141
	for <sigtran@ietf.org>; Mon, 24 Feb 2003 18:29:10 -0500 (EST)
Received: from wntcsdexg01.csd.ciena.com ([10.34.31.31]) by w2ksjexg01.ciena.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2653.13)
	id FN36YAJ5; Mon, 24 Feb 2003 15:32:46 -0800
Received: by webdev-llnt.oni.com with Internet Mail Service (5.5.2653.19)
	id <Y5W9NAWY>; Mon, 24 Feb 2003 15:33:02 -0800
Message-ID: <2135200C183FD5119588009027DE572302836D1C@webdev-owa.oni.com>
From: "Ong, Lyndon" <LyOng@ciena.com>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Date: Mon, 24 Feb 2003 15:32:54 -0800
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Sigtran] meeting time moved
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Hi Folks,

The Sigtran WG Meeting time has been moved to Friday,
please be advised.

Cheers,

L. Ong
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 00:49:34 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA07338
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 00:49:34 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1P5wQg20740
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 00:58:26 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1P5w5p20704;
	Tue, 25 Feb 2003 00:58:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1P5u1p20662
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 00:56:01 -0500
Received: from india.mercurykr.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA07321
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 00:46:32 -0500 (EST)
Received: from srivatsp (ptil-33-162-ind.primus-india.net [203.196.162.33] (may be forged))
	by india.mercurykr.com (8.11.2/8.11.2) with SMTP id h1PAtwp04389
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 16:25:59 +0530
Message-Id: <200302251055.h1PAtwp04389@india.mercurykr.com>
Date: Tue, 25 Feb 2003 11:18:36 +0500
From: "Srivats P." <srivats@india.mercurykr.com>
Reply-To: srivatsp@email.com
To: sigtran <sigtran@ietf.org>
Subject: Re: RE: [Sigtran] Re: M3UA Notify message
Organization: Mercury Corporation, India
X-mailer: FoxMail 2.1 [en]
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

All,

Where do we go from here on this one ? Should we put it to a vote or
something ?

IMO, not stopping traffic when #ASPs < n makes more sense to me ...
however since different people are reading it in different ways ...
the IG should clarify this point.

Regards,
Srivats

At 19/02/2003 3:30:00 AM, you wrote:
>
>Hi,
>
>I beleive 'n' is minimum number of ASPs to handle the 'best traffic T'.
>To handle the 'worst traffic T+1', SG may require 'n+1' ASPs.
>
>'n'is the derived value based on the traffic report given by Traffic
>Metering and Measurement.
>
>So 'n' is fed during configuration of stack.When traffic increases from
>'T' to 'T+1' or # APSs < n - this case SG is required to send NOTIFY to
>'k' ASP.
>
>And one more thing when # APSs < n, SG still can handle the traffic with
>single ASP.
>
>When 'n' ASPs are active
>------------------------
>
>SLS-X  ---->ASP1
>SLS-Y  ---->ASP2
>
>When ASP1 goes down
>-------------------
>
>SLS-(Y,X)  ----->ASP2
>
>Instead of banning the complete traffic ,it can be handled (not 100%) by
>ASP2.
>
>
>--Samuel
>
>  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>  1000 Coit Road, Plano, Texas 75075
>  ******* The opinions expressed are not those of Alcatel USA, Inc. *******
>
>On Tue, 18 Feb 2003, Javier Pastor-Balbas (ECE) wrote:
>
>> Brian, 
>> 
>> I agree 
>> 
>> RFC simply says: 
>> 
>> 	"n ASPs is the minimum number of redundant ASPs required to handle traffic"
>> 
>> As this is the minimum number, it can be inferred that for #ASPs < n no traffic can be handled. Nothing is mentioned in the RFC about 
engineered load although it may be thought that n value calculation is based on that.  
>> 
>> I would say that the current model does not support the use of extra ASPs for loads bigger than the engineered nor the use of less than n 
for smaller loads... The message is clear saying that #ASPs<n implies no traffic, i.e. whenever you can not support
> the engineered load, stop the system. This last sentence is not in the RFC but it can be deduced as the engineered load for n can... 
>> 
>> Cheers // Javier. 
>> 
>> > -----Original Message-----
>> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
>> > Sent: Tuesday, February 18, 2003 12:49 PM
>> > To: Javier Pastor-Balbas (ECE)
>> > Cc: sigtran@ietf.org
>> > Subject: Re: [Sigtran] Re: M3UA Notify message
>> > 
>> > 
>> > Javier,
>> > 
>> > Javier Pastor-Balbas (ECE) wrote:             (Tue, 18 Feb 
>> > 2003 12:17:54)
>> > > > Javier Pastor-Balbas (ECE) wrote:             (Tue, 18 
>> > Feb 2003 07:57:23)
>> > > > > 
>> > > > > Hi Brian, 
>> > > > > 
>> > > > > According to the RFC "n" is required to handle traffic 
>> > with minimum
>> > > > > losses, what is expected from the SS7 point of view in 
>> > an interworking
>> > > > > scenario.
>> > 
>> > Please cite this passage, because I cannot find it.  I find 
>> > no mention of
>> > minimum number of ASPs and "message loss".  The only 
>> > reference to message loss
>> > that I know of is unrelated.
>> > 
>> 
>> 
>> 
>> > > > 
>> > > > That would be the numbe required to handle maximum 
>> > engineered load, which
>> > > > is the same thing.
>> > > > 
>> > > 
>> > > If you refer to the compromise load engineered to be 
>> > available 99% (aprox.)
>> > > of the time, then I agree. But this is a statically value fed at
>> > > configuration time. And the system won't work till n number 
>> > is reach since
>> > > the system will likely have quality contrains.
>> > 
>> > The maximum engineered load is a figure established at 
>> > configuration time.
>> > 
>> > Carrier-grade systems must always work with 1 processor.  The 
>> > system will just
>> > not necessarily be able to handle the offered load.  (But it might:
>> > particularly during quiet periods, off-peak, normal-day, etc.)
>> > 
>> > The offered load cannot be established at configuration time 
>> > and varies widely
>> > for telephony systems.
>> > 
>> > You must also realized that the system might, at any time,  
>> > be offered a load
>> > greater than the maximum engineered load (mass calling, busy 
>> > periods, etc.):
>> > so for carrier-grade systems even "n" processors must 
>> > implement overload
>> > controls (and they could not claim SS7 compliance if they didn't).
>> > 
>> > Look back in this thread and you will see the factors 
>> > contributing to the
>> > anticipated offered load for the start, restart and stop decision.
>> > 
>> > The decisions are not based on less than "n" processors not 
>> > allowing the
>> > system to work (where n > 1).  They are based on the negative 
>> > effects of
>> > delivering pent up traffic to processors that will likely 
>> > just congest.
>> > 
>> > --brian
>> > 
>> > -- 
>> > Brian F. G. Bidulock
>> > bidulock@openss7.org
>> > http://www.openss7.org/
>> > 
>> _______________________________________________
>> Sigtran mailing list
>> Sigtran@ietf.org
>> https://www1.ietf.org/mailman/listinfo/sigtran
>> 
>
>_______________________________________________
>Sigtran mailing list
>Sigtran@ietf.org
>https://www1.ietf.org/mailman/listinfo/sigtran
>

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 01:40:33 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA08222
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 01:40:33 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1P6nPm23786
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 01:49:25 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1P6nAp23778;
	Tue, 25 Feb 2003 01:49:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1P6mCp23744
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 01:48:12 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA08157
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 01:38:47 -0500 (EST)
Subject: Re: RE: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OFAAC2F093.20202130-ONC2256CD8.00226C1A@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Tue, 25 Feb 2003 08:39:11 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/25/2003
 08:39:15 AM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Not necessarily because I disagree with some change, but just to clarify
where the changes might take us:
NTFY was used till now only as an advisory message, this will be changed.
Starting/stopping(till now we focused more on starting but why not stopping
also? So SG can decide based on load/other dynamic conditions when to stop
traffic to/from AS) traffic will be controlled by SGP with NTFY.
IMO, this requires still a change to the original AS state machine for
transitions from INACTIVE to ACTIVE and possibly from ACTIVE to PENDING.
In general all this will change the meaning of ASPAC/ASPAC-ACK exchange, it
will mean that ASP is ready to be triggered to send/receive traffic for a
specific AS (something to be determined just by ASP, when the triggering
will happen will be decided by SGP)rather than ASP can start to
send/receive traffic for a specific AS(something to be determined by both
ASP and SGP). I believe this migth require some other changes throughout
the document. For example instead of unsolicited ASPIA-ACK, I would think
SGP should send NTFY(INACTIVE). My opinion is, if we are changing
something, let's do it clean and correct.

IMO, overall this is more than just a clarification, let's decide carefully
to prevent rollbacks (as much as possible).

    Tolga

--- "Srivats P." <srivats@india.mercurykr.com> wrote:
> All,
>
> Where do we go from here on this one ? Should we put
> it to a vote or
> something ?
>
> IMO, not stopping traffic when #ASPs < n makes more
> sense to me ...
> however since different people are reading it in
> different ways ...
> the IG should clarify this point.

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 04:14:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA21434
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 04:14:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1P9Nrl11385
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 04:23:53 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1P9Ndp11361;
	Tue, 25 Feb 2003 04:23:39 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1P9Kdp11262
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 04:20:39 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA21382
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 04:11:11 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:e795+Kw1gT6K99GxCeRm3UaRPn+JXgIh@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1P9F4D16681;
	Tue, 25 Feb 2003 02:15:04 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1P9F4o12111;
	Tue, 25 Feb 2003 02:15:04 -0700
Date: Tue, 25 Feb 2003 02:15:03 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030225021503.A11817@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OFAAC2F093.20202130-ONC2256CD8.00226C1A@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OFAAC2F093.20202130-ONC2256CD8.00226C1A@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Tue, Feb 25, 2003 at 08:39:11AM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

TOLGA,

TOLGA ASVEREN wrote:                                                                 (Tue, 25 Feb 2003 08:39:11)
> 
> Not necessarily because I disagree with some change, but just to clarify
> where the changes might take us:

If we cannot agree, we let the RFC stand.

> NTFY was used till now only as an advisory message, this will be changed.

No it will not.  An ASP is not forbidden from acting on any Notify message.
All Notify messages are still advisory.  The ASP can take into consideration
Notify messages, information that it has about other ASPs, the time of day,
direction of the wind, etc.

> Starting/stopping(till now we focused more on starting but why not stopping
> also? So SG can decide based on load/other dynamic conditions when to stop
> traffic to/from AS) traffic will be controlled by SGP with NTFY.

The SG has always been permitted to start or stop traffic when it considers
it necessary and has always been permitted to perform its necessary functions
with the SS7 network.  The SGP can send ASPIA Ack or drop associations any
time it wishes to stop traffic.

> IMO, this requires still a change to the original AS state machine for
> transitions from INACTIVE to ACTIVE and possibly from ACTIVE to PENDING.

It does not.  The SG is only withholding a Notify message as proscribed
by the RFC.

> In general all this will change the meaning of ASPAC/ASPAC-ACK exchange, it
> will mean that ASP is ready to be triggered to send/receive traffic for a
> specific AS (something to be determined just by ASP, when the triggering
> will happen will be decided by SGP)rather than ASP can start to
> send/receive traffic for a specific AS(something to be determined by both
> ASP and SGP). I believe this migth require some other changes throughout
> the document. For example instead of unsolicited ASPIA-ACK, I would think
> SGP should send NTFY(INACTIVE). My opinion is, if we are changing
> something, let's do it clean and correct.

No nothing changes here because withholding NTFY(ACTIVE) is already part
of the RFC.  Nothing in the RFC changes.

> 
> IMO, overall this is more than just a clarification, let's decide carefully
> to prevent rollbacks (as much as possible).

Not even clarification is required.  The RFC details the necessary actions.

The only thing that needs to be done is strike section 3.15 from the IG
because it conflicts with the RFC (which is fine as it is).

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 04:55:26 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA21992
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 04:55:26 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1PA4O713607
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 05:04:24 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PA4Fp13590;
	Tue, 25 Feb 2003 05:04:15 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PA3gp13512
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 05:03:42 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA21967
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 04:54:13 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1P9ufeW013175;
	Tue, 25 Feb 2003 03:56:42 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h1P9u7F04423;
	Tue, 25 Feb 2003 03:56:07 -0600 (CST)
Date: Tue, 25 Feb 2003 03:56:07 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
In-Reply-To: <20030225021503.A11817@openss7.org>
Message-ID: <Pine.SOL.4.10.10302250329010.3454-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Tue, 25 Feb 2003, Brian F. G. Bidulock wrote:

> TOLGA,
> 
> TOLGA ASVEREN wrote:                                                                 (Tue, 25 Feb 2003 08:39:11)
> > 
> > Not necessarily because I disagree with some change, but just to clarify
> > where the changes might take us:
> 
> If we cannot agree, we let the RFC stand.
> 
> > NTFY was used till now only as an advisory message, this will be changed.
> 
> No it will not.  An ASP is not forbidden from acting on any Notify message.
> All Notify messages are still advisory.  The ASP can take into consideration
> Notify messages, information that it has about other ASPs, the time of day,
> direction of the wind, etc.
> 
> > Starting/stopping(till now we focused more on starting but why not stopping
> > also? So SG can decide based on load/other dynamic conditions when to stop
> > traffic to/from AS) traffic will be controlled by SGP with NTFY.
> 
> The SG has always been permitted to start or stop traffic when it considers
> it necessary and has always been permitted to perform its necessary functions
> with the SS7 network.  The SGP can send ASPIA Ack or drop associations any
> time it wishes to stop traffic.

Please let me know what are we achieving by stopping the traffic in the
following case(n+k - reduntancy)

When 'n' ASPs are active
------------------------
SLS-X  ---->ASP1
SLS-Y  ---->ASP2

When ASP1 goes down
-------------------
SLS-(Y,X)  ----->ASP2

Here 3 cases can be dealt

case 1) If SG receives  message with SLS Y - this traffic can be
        successfully (100%) handled by ASP2 according to the configured
        databse.
 
case 2) If SG receives messages with SLS X and Y - this traffic can be
        successfully (100%) handled by ASP2, provided congestion has not
        occured(according to q.704 section 3.8.2)

case 3) Don't allow SG to process the traffic when n<2 ?????????.
        SG disallows traffic when n=0.

IMO, SG has capability to reroute the traffic to ASP1 and there is no
meaning if SG stops traffic when n < 2.

This is just like a MTP3 link failure concept when link1 goes down ,
traffic carried by link1 will be rerouted to available link. 
All we need to do at SG is just change the SLS-ASP dynamic table to carry
SLS X.


--Samuel
> 
> > IMO, this requires still a change to the original AS state machine for
> > transitions from INACTIVE to ACTIVE and possibly from ACTIVE to PENDING.
> 
> It does not.  The SG is only withholding a Notify message as proscribed
> by the RFC.
> 
> > In general all this will change the meaning of ASPAC/ASPAC-ACK exchange, it
> > will mean that ASP is ready to be triggered to send/receive traffic for a
> > specific AS (something to be determined just by ASP, when the triggering
> > will happen will be decided by SGP)rather than ASP can start to
> > send/receive traffic for a specific AS(something to be determined by both
> > ASP and SGP). I believe this migth require some other changes throughout
> > the document. For example instead of unsolicited ASPIA-ACK, I would think
> > SGP should send NTFY(INACTIVE). My opinion is, if we are changing
> > something, let's do it clean and correct.
> 
> No nothing changes here because withholding NTFY(ACTIVE) is already part
> of the RFC.  Nothing in the RFC changes.
> 
> > 
> > IMO, overall this is more than just a clarification, let's decide carefully
> > to prevent rollbacks (as much as possible).
> 
> Not even clarification is required.  The RFC details the necessary actions.
> 
> The only thing that needs to be done is strike section 3.15 from the IG
> because it conflicts with the RFC (which is fine as it is).
> 
> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 05:16:29 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA22479
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 05:16:28 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1PAPQ615419
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 05:25:26 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PAP8p15390;
	Tue, 25 Feb 2003 05:25:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PANMp15299
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 05:23:22 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA22440
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 05:13:53 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:zuukTbuQ3Hgt7JoGAuh0yXrWts1UN4EA@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1PAHlD17569;
	Tue, 25 Feb 2003 03:17:47 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1PAHlW13099;
	Tue, 25 Feb 2003 03:17:47 -0700
Date: Tue, 25 Feb 2003 03:17:47 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030225031747.B13055@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <20030225021503.A11817@openss7.org> <Pine.SOL.4.10.10302250329010.3454-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302250329010.3454-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Tue, Feb 25, 2003 at 03:56:07AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Tue, 25 Feb 2003 03:56:07)
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Tue, 25 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > TOLGA,
> > 
> > TOLGA ASVEREN wrote:                                                                 (Tue, 25 Feb 2003 08:39:11)
> > > 
> > > Not necessarily because I disagree with some change, but just to clarify
> > > where the changes might take us:
> > 
> > If we cannot agree, we let the RFC stand.
> > 
> > > NTFY was used till now only as an advisory message, this will be changed.
> > 
> > No it will not.  An ASP is not forbidden from acting on any Notify message.
> > All Notify messages are still advisory.  The ASP can take into consideration
> > Notify messages, information that it has about other ASPs, the time of day,
> > direction of the wind, etc.
> > 
> > > Starting/stopping(till now we focused more on starting but why not stopping
> > > also? So SG can decide based on load/other dynamic conditions when to stop
> > > traffic to/from AS) traffic will be controlled by SGP with NTFY.
> > 
> > The SG has always been permitted to start or stop traffic when it considers
> > it necessary and has always been permitted to perform its necessary functions
> > with the SS7 network.  The SGP can send ASPIA Ack or drop associations any
> > time it wishes to stop traffic.
> 
> Please let me know what are we achieving by stopping the traffic in the
> following case(n+k - reduntancy)
> 
> When 'n' ASPs are active
> ------------------------
> SLS-X  ---->ASP1
> SLS-Y  ---->ASP2
> 
> When ASP1 goes down
> -------------------
> SLS-(Y,X)  ----->ASP2
> 
> Here 3 cases can be dealt
> 
> case 1) If SG receives  message with SLS Y - this traffic can be
>         successfully (100%) handled by ASP2 according to the configured
>         databse.
>  
> case 2) If SG receives messages with SLS X and Y - this traffic can be
>         successfully (100%) handled by ASP2, provided congestion has not
>         occured(according to q.704 section 3.8.2)
> 
> case 3) Don't allow SG to process the traffic when n<2 ?????????.
>         SG disallows traffic when n=0.
> 
> IMO, SG has capability to reroute the traffic to ASP1 and there is no
> meaning if SG stops traffic when n < 2.
> 
> This is just like a MTP3 link failure concept when link1 goes down ,
> traffic carried by link1 will be rerouted to available link. 
> All we need to do at SG is just change the SLS-ASP dynamic table to carry
> SLS X.
> 
> 
> --Samuel

I agree.   That is what is provided for in the RFC.  However, section 3.15 of
the IG would require the SG to stop traffic when n < 2.  So, it must be
struck.

--brian


-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 06:34:21 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA23882
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 06:34:20 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1PBhJs20761
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 06:43:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PBh5p20751;
	Tue, 25 Feb 2003 06:43:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PBfEp20674
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 06:41:14 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA23837
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 06:31:41 -0500 (EST)
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OFABC085F1.F1638CD1-ONC2256CD8.003E6D05@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Tue, 25 Feb 2003 13:32:04 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/25/2003
 01:32:10 PM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Samuel,

Why not let SG decide, what to do and leave it flexible in the
specification? SG will inform ASPs about its decision with NTFYs anyhow.

      Tolga



Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Tue, 25 Feb 2003, Brian F. G. Bidulock wrote:

> TOLGA,
>
> TOLGA ASVEREN wrote:
(Tue, 25 Feb 2003 08:39:11)
> >
> > Not necessarily because I disagree with some change, but just to
clarify
> > where the changes might take us:
>
> If we cannot agree, we let the RFC stand.
>
> > NTFY was used till now only as an advisory message, this will be
changed.
>
> No it will not.  An ASP is not forbidden from acting on any Notify
message.
> All Notify messages are still advisory.  The ASP can take into
consideration
> Notify messages, information that it has about other ASPs, the time of
day,
> direction of the wind, etc.
>
> > Starting/stopping(till now we focused more on starting but why not
stopping
> > also? So SG can decide based on load/other dynamic conditions when to
stop
> > traffic to/from AS) traffic will be controlled by SGP with NTFY.
>
> The SG has always been permitted to start or stop traffic when it
considers
> it necessary and has always been permitted to perform its necessary
functions
> with the SS7 network.  The SGP can send ASPIA Ack or drop associations
any
> time it wishes to stop traffic.

Please let me know what are we achieving by stopping the traffic in the
following case(n+k - reduntancy)

When 'n' ASPs are active
------------------------
SLS-X  ---->ASP1
SLS-Y  ---->ASP2

When ASP1 goes down
-------------------
SLS-(Y,X)  ----->ASP2

Here 3 cases can be dealt

case 1) If SG receives  message with SLS Y - this traffic can be
        successfully (100%) handled by ASP2 according to the configured
        databse.

case 2) If SG receives messages with SLS X and Y - this traffic can be
        successfully (100%) handled by ASP2, provided congestion has not
        occured(according to q.704 section 3.8.2)

case 3) Don't allow SG to process the traffic when n<2 ?????????.
        SG disallows traffic when n=0.

IMO, SG has capability to reroute the traffic to ASP1 and there is no
meaning if SG stops traffic when n < 2.

This is just like a MTP3 link failure concept when link1 goes down ,
traffic carried by link1 will be rerouted to available link.
All we need to do at SG is just change the SLS-ASP dynamic table to carry
SLS X.


--Samuel
>
> > IMO, this requires still a change to the original AS state machine for
> > transitions from INACTIVE to ACTIVE and possibly from ACTIVE to
PENDING.
>
> It does not.  The SG is only withholding a Notify message as proscribed
> by the RFC.
>
> > In general all this will change the meaning of ASPAC/ASPAC-ACK
exchange, it
> > will mean that ASP is ready to be triggered to send/receive traffic for
a
> > specific AS (something to be determined just by ASP, when the
triggering
> > will happen will be decided by SGP)rather than ASP can start to
> > send/receive traffic for a specific AS(something to be determined by
both
> > ASP and SGP). I believe this migth require some other changes
throughout
> > the document. For example instead of unsolicited ASPIA-ACK, I would
think
> > SGP should send NTFY(INACTIVE). My opinion is, if we are changing
> > something, let's do it clean and correct.
>
> No nothing changes here because withholding NTFY(ACTIVE) is already part
> of the RFC.  Nothing in the RFC changes.
>
> >
> > IMO, overall this is more than just a clarification, let's decide
carefully
> > to prevent rollbacks (as much as possible).
>
> Not even clarification is required.  The RFC details the necessary
actions.
>
> The only thing that needs to be done is strike section 3.15 from the IG
> because it conflicts with the RFC (which is fine as it is).
>
> --brian
>
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 06:47:37 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA25119
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 06:47:37 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1PBua821350
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 06:56:36 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PBuQp21312;
	Tue, 25 Feb 2003 06:56:26 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PBrjp21148
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 06:53:45 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA24456
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 06:44:15 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:36qABeCRG4bWkcPvlgjNBsC7cHTwAFze@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1PBm8D19882;
	Tue, 25 Feb 2003 04:48:08 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1PBm8l14736;
	Tue, 25 Feb 2003 04:48:08 -0700
Date: Tue, 25 Feb 2003 04:48:08 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030225044807.A14553@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OFABC085F1.F1638CD1-ONC2256CD8.003E6D05@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OFABC085F1.F1638CD1-ONC2256CD8.003E6D05@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Tue, Feb 25, 2003 at 01:32:04PM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

TOLGA,

TOLGA ASVEREN wrote:                                                                 (Tue, 25 Feb 2003 13:32:04)
> 
> Samuel,
> 
> Why not let SG decide, what to do and leave it flexible in the
> specification? SG will inform ASPs about its decision with NTFYs anyhow.
> 

That's what it already says in the RFC.  Strike IG 3.15.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 07:06:26 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA26503
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 07:06:25 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1PCFOs23398
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 07:15:24 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PCFBp23387;
	Tue, 25 Feb 2003 07:15:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PCEMp23346
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 07:14:22 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA26306
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 07:04:51 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1PC8geY024537;
	Tue, 25 Feb 2003 06:08:43 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h1PC8dF07568;
	Tue, 25 Feb 2003 06:08:39 -0600 (CST)
Date: Tue, 25 Feb 2003 06:08:39 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
In-Reply-To: <OFABC085F1.F1638CD1-ONC2256CD8.003E6D05@LocalDomain>
Message-ID: <Pine.SOL.4.10.10302250543510.3454-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Tolga,

From this thread we can end up with two kinds of SGs.

The SG which decides to stop taking the traffic when 'n' changed
from '2' to '1'.

The other SG which decides to continue the traffic  when 'n' 
changed from '2' to '1'.

When performance is considered, i think everyone will prefer second case.
Second case follows the existing network standards.But first case does
not follow any standards.

If at all change is required following things to be made clear.

  According to RFC 'k' ASP is NOTIFIED only when #ASPs < n - means
  existing ASP1 has been failed so SG sends NTFY to 'k' ASP in order to 
  maintain the traffic stability. - This is what RFC says but when traffic
  increases and SG is unable to handle the traffic with 'n=2' ASPs , now
  SG can request 'k' ASP to become active- this is new case to be decided.

--Samuel

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Tue, 25 Feb 2003, TOLGA ASVEREN wrote:

> 
> Samuel,
> 
> Why not let SG decide, what to do and leave it flexible in the
> specification? SG will inform ASPs about its decision with NTFYs anyhow.
> 
>       Tolga
> 
> 
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Tue, 25 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > TOLGA,
> >
> > TOLGA ASVEREN wrote:
> (Tue, 25 Feb 2003 08:39:11)
> > >
> > > Not necessarily because I disagree with some change, but just to
> clarify
> > > where the changes might take us:
> >
> > If we cannot agree, we let the RFC stand.
> >
> > > NTFY was used till now only as an advisory message, this will be
> changed.
> >
> > No it will not.  An ASP is not forbidden from acting on any Notify
> message.
> > All Notify messages are still advisory.  The ASP can take into
> consideration
> > Notify messages, information that it has about other ASPs, the time of
> day,
> > direction of the wind, etc.
> >
> > > Starting/stopping(till now we focused more on starting but why not
> stopping
> > > also? So SG can decide based on load/other dynamic conditions when to
> stop
> > > traffic to/from AS) traffic will be controlled by SGP with NTFY.
> >
> > The SG has always been permitted to start or stop traffic when it
> considers
> > it necessary and has always been permitted to perform its necessary
> functions
> > with the SS7 network.  The SGP can send ASPIA Ack or drop associations
> any
> > time it wishes to stop traffic.
> 
> Please let me know what are we achieving by stopping the traffic in the
> following case(n+k - reduntancy)
> 
> When 'n' ASPs are active
> ------------------------
> SLS-X  ---->ASP1
> SLS-Y  ---->ASP2
> 
> When ASP1 goes down
> -------------------
> SLS-(Y,X)  ----->ASP2
> 
> Here 3 cases can be dealt
> 
> case 1) If SG receives  message with SLS Y - this traffic can be
>         successfully (100%) handled by ASP2 according to the configured
>         databse.
> 
> case 2) If SG receives messages with SLS X and Y - this traffic can be
>         successfully (100%) handled by ASP2, provided congestion has not
>         occured(according to q.704 section 3.8.2)
> 
> case 3) Don't allow SG to process the traffic when n<2 ?????????.
>         SG disallows traffic when n=0.
> 
> IMO, SG has capability to reroute the traffic to ASP1 and there is no
> meaning if SG stops traffic when n < 2.
> 
> This is just like a MTP3 link failure concept when link1 goes down ,
> traffic carried by link1 will be rerouted to available link.
> All we need to do at SG is just change the SLS-ASP dynamic table to carry
> SLS X.
> 
> 
> --Samuel
> >
> > > IMO, this requires still a change to the original AS state machine for
> > > transitions from INACTIVE to ACTIVE and possibly from ACTIVE to
> PENDING.
> >
> > It does not.  The SG is only withholding a Notify message as proscribed
> > by the RFC.
> >
> > > In general all this will change the meaning of ASPAC/ASPAC-ACK
> exchange, it
> > > will mean that ASP is ready to be triggered to send/receive traffic for
> a
> > > specific AS (something to be determined just by ASP, when the
> triggering
> > > will happen will be decided by SGP)rather than ASP can start to
> > > send/receive traffic for a specific AS(something to be determined by
> both
> > > ASP and SGP). I believe this migth require some other changes
> throughout
> > > the document. For example instead of unsolicited ASPIA-ACK, I would
> think
> > > SGP should send NTFY(INACTIVE). My opinion is, if we are changing
> > > something, let's do it clean and correct.
> >
> > No nothing changes here because withholding NTFY(ACTIVE) is already part
> > of the RFC.  Nothing in the RFC changes.
> >
> > >
> > > IMO, overall this is more than just a clarification, let's decide
> carefully
> > > to prevent rollbacks (as much as possible).
> >
> > Not even clarification is required.  The RFC details the necessary
> actions.
> >
> > The only thing that needs to be done is strike section 3.15 from the IG
> > because it conflicts with the RFC (which is fine as it is).
> >
> > --brian
> >
> > --
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 
> 
> 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 08:04:37 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA29604
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 08:04:37 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1PDDbv27744
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 08:13:37 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PDDLp27726;
	Tue, 25 Feb 2003 08:13:23 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PDBap27625
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 08:11:36 -0500
Received: from albatross.wise.edt.ericsson.se (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA29452
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 08:02:03 -0500 (EST)
Received: from esealnt612.al.sw.ericsson.se (esealnt612.al.sw.ericsson.se [153.88.254.71])
	by albatross.wise.edt.ericsson.se (8.12.1/8.12.1/WIREfire-1.4) with ESMTP id h1PD5unS007595;
	Tue, 25 Feb 2003 14:05:56 +0100 (MET)
Received: from ESEALNT745.al.sw.ericsson.se ([153.88.251.5]) by esealnt612.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55)
	id FSMCD4J7; Tue, 25 Feb 2003 14:05:55 +0100
Received: by ESEALNT745.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <DW9DFTWF>; Tue, 25 Feb 2003 14:05:55 +0100
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B659C6@EESTQNT105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>
To: "'Samuel Dur D. Jeyaseelan'" <sjeyasee@ssd.usa.alcatel.com>,
        "Brian F. G. Bidulock" <bidulock@openss7.org>,
        TOLGA ASVEREN
	 <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: RE: [Sigtran] Re: M3UA Notify message
Date: Tue, 25 Feb 2003 14:03:32 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Hi Samuel et al., 

Take into account that if you define two ASes:
- AS1 DPC+SLS1  with two ASPs (ASP1 and ASP2) and redundancy: n=1 k=1
- AS2 DPC+SLS2  with two ASPs (ASP1 and ASP2) and redundancy: n=1 k=1

And 
- ASP1 active for AS1 and backup for AS2
- ASP2 active for AS2 and backup for AS1

You get the desired behaviour. This case can be solved with a right network configuration without changing the current "required" qualification for "n".



Apart from this, 

I would say that a compromise option would consist of considering "n" as the normal working case, but make it implementation dependent (maybe configurable) whether an implementation may have more than "n" ASPs working at the same time for an AS, and whether the SG keeps working when the number of ASP is under "n". We could make a recommendation with "MAY" allowing both options.

But now, the problem comes with the AS state machine. If we agree upon here, we could say the AS is ACTIVE in the range n_min to n_max, where these two values are configurable.


Could this be considered as a compromise solution about n+k behaviour?


Cheers // Javier.





> -----Original Message-----
> From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> Sent: martes, 25 de febrero de 2003 10:56
> To: Brian F. G. Bidulock
> Cc: TOLGA ASVEREN; sigtran@ietf.org
> Subject: Re: [Sigtran] Re: M3UA Notify message
> 
> 
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: 
> <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel 
> USA, Inc. *******
> 
> On Tue, 25 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > TOLGA,
> > 
> > TOLGA ASVEREN wrote:                                        
>                          (Tue, 25 Feb 2003 08:39:11)
> > > 
> > > Not necessarily because I disagree with some change, but 
> just to clarify
> > > where the changes might take us:
> > 
> > If we cannot agree, we let the RFC stand.
> > 
> > > NTFY was used till now only as an advisory message, this 
> will be changed.
> > 
> > No it will not.  An ASP is not forbidden from acting on any 
> Notify message.
> > All Notify messages are still advisory.  The ASP can take 
> into consideration
> > Notify messages, information that it has about other ASPs, 
> the time of day,
> > direction of the wind, etc.
> > 
> > > Starting/stopping(till now we focused more on starting 
> but why not stopping
> > > also? So SG can decide based on load/other dynamic 
> conditions when to stop
> > > traffic to/from AS) traffic will be controlled by SGP with NTFY.
> > 
> > The SG has always been permitted to start or stop traffic 
> when it considers
> > it necessary and has always been permitted to perform its 
> necessary functions
> > with the SS7 network.  The SGP can send ASPIA Ack or drop 
> associations any
> > time it wishes to stop traffic.
> 
> Please let me know what are we achieving by stopping the 
> traffic in the
> following case(n+k - reduntancy)
> 
> When 'n' ASPs are active
> ------------------------
> SLS-X  ---->ASP1
> SLS-Y  ---->ASP2
> 
> When ASP1 goes down
> -------------------
> SLS-(Y,X)  ----->ASP2
> 
> Here 3 cases can be dealt
> 
> case 1) If SG receives  message with SLS Y - this traffic can be
>         successfully (100%) handled by ASP2 according to the 
> configured
>         databse.
>  
> case 2) If SG receives messages with SLS X and Y - this traffic can be
>         successfully (100%) handled by ASP2, provided 
> congestion has not
>         occured(according to q.704 section 3.8.2)
> 
> case 3) Don't allow SG to process the traffic when n<2 ?????????.
>         SG disallows traffic when n=0.
> 
> IMO, SG has capability to reroute the traffic to ASP1 and there is no
> meaning if SG stops traffic when n < 2.
> 
> This is just like a MTP3 link failure concept when link1 goes down ,
> traffic carried by link1 will be rerouted to available link. 
> All we need to do at SG is just change the SLS-ASP dynamic 
> table to carry
> SLS X.
> 
> 
> --Samuel
> > 
> > > IMO, this requires still a change to the original AS 
> state machine for
> > > transitions from INACTIVE to ACTIVE and possibly from 
> ACTIVE to PENDING.
> > 
> > It does not.  The SG is only withholding a Notify message 
> as proscribed
> > by the RFC.
> > 
> > > In general all this will change the meaning of 
> ASPAC/ASPAC-ACK exchange, it
> > > will mean that ASP is ready to be triggered to 
> send/receive traffic for a
> > > specific AS (something to be determined just by ASP, when 
> the triggering
> > > will happen will be decided by SGP)rather than ASP can start to
> > > send/receive traffic for a specific AS(something to be 
> determined by both
> > > ASP and SGP). I believe this migth require some other 
> changes throughout
> > > the document. For example instead of unsolicited 
> ASPIA-ACK, I would think
> > > SGP should send NTFY(INACTIVE). My opinion is, if we are changing
> > > something, let's do it clean and correct.
> > 
> > No nothing changes here because withholding NTFY(ACTIVE) is 
> already part
> > of the RFC.  Nothing in the RFC changes.
> > 
> > > 
> > > IMO, overall this is more than just a clarification, 
> let's decide carefully
> > > to prevent rollbacks (as much as possible).
> > 
> > Not even clarification is required.  The RFC details the 
> necessary actions.
> > 
> > The only thing that needs to be done is strike section 3.15 
> from the IG
> > because it conflicts with the RFC (which is fine as it is).
> > 
> > --brian
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 08:27:59 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA00590
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 08:27:59 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1PDb0w28775
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 08:37:00 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PDSup28445;
	Tue, 25 Feb 2003 08:28:56 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PDNjp28217
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 08:23:45 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA00129
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 08:14:10 -0500 (EST)
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF2E63E59B.0981E8F2-ONC2256CD8.003D347A@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Tue, 25 Feb 2003 13:20:59 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/25/2003
 03:14:38 PM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,



TOLGA,

TOLGA ASVEREN wrote:
(Tue, 25 Feb 2003 08:39:11)
>
> Not necessarily because I disagree with some change, but just to clarify
> where the changes might take us:

If we cannot agree, we let the RFC stand.

> NTFY was used till now only as an advisory message, this will be changed.

No it will not.  An ASP is not forbidden from acting on any Notify message.
All Notify messages are still advisory.  The ASP can take into
consideration
Notify messages, information that it has about other ASPs, the time of day,
direction of the wind, etc.
[TOLGA]IMO important is, what ASP SHOULD do after receiving NTFY, if we
want to avoid interoperability problems. So I think to change handling of
NTFY is necessary.

> Starting/stopping(till now we focused more on starting but why not
stopping
> also? So SG can decide based on load/other dynamic conditions when to
stop
> traffic to/from AS) traffic will be controlled by SGP with NTFY.

The SG has always been permitted to start or stop traffic when it considers
it necessary and has always been permitted to perform its necessary
functions
with the SS7 network.  The SGP can send ASPIA Ack or drop associations any
time it wishes to stop traffic.
[TOLGA]If ASP is not ACTIVE for the AS, SGP should better not send traffic
to it. For stopping I agree that SGP can decide for this. My point is, this
better be reflected as a state change of AS and notified to ASP with NTFY,
which afterwards should stop traffic.

> IMO, this requires still a change to the original AS state machine for
> transitions from INACTIVE to ACTIVE and possibly from ACTIVE to PENDING.

It does not.  The SG is only withholding a Notify message as proscribed
by the RFC.
[TOLGA]Without state change, this is really ugly. Why do we have the state
machine then? In fact, another thing which is needed is probably the satet
machine from ASP point of view.

> In general all this will change the meaning of ASPAC/ASPAC-ACK exchange,
it
> will mean that ASP is ready to be triggered to send/receive traffic for a
> specific AS (something to be determined just by ASP, when the triggering
> will happen will be decided by SGP)rather than ASP can start to
> send/receive traffic for a specific AS(something to be determined by both
> ASP and SGP). I believe this migth require some other changes throughout
> the document. For example instead of unsolicited ASPIA-ACK, I would think
> SGP should send NTFY(INACTIVE). My opinion is, if we are changing
> something, let's do it clean and correct.

No nothing changes here because withholding NTFY(ACTIVE) is already part
of the RFC.  Nothing in the RFC changes.
[TOLGA]It is clear that it is not mandatory to take any action after
receiving NTFY. The only thing related with starting traffic from ASP is
that ASPAC/ASPAC-ACK exchange should be done and that ASP should wait till
number of ACTIVE ASPs reaches n.

>
> IMO, overall this is more than just a clarification, let's decide
carefully
> to prevent rollbacks (as much as possible).

Not even clarification is required.  The RFC details the necessary actions.
[TOLGA]IMO it doesn't. For example it says that ASPs should wait for n
ACTIVE ASPs (n interpreted as mandatory) and SGP may decide not to start
traffic and withhold NTFY (n not interpreted as mandatory)

The only thing that needs to be done is strike section 3.15 from the IG
because it conflicts with the RFC (which is fine as it is).
--brian

--
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 08:50:42 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA01074
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 08:50:42 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1PDxi130406
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 08:59:44 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PDxOp30356;
	Tue, 25 Feb 2003 08:59:24 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PDw2p30297
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 08:58:02 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA01018
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 08:48:28 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:oThGWG5YbTy3nmIFHVGykB8zP/gR9yzZ@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1PDqMD22027;
	Tue, 25 Feb 2003 06:52:22 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1PDqMF16315;
	Tue, 25 Feb 2003 06:52:22 -0700
Date: Tue, 25 Feb 2003 06:52:22 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030225065222.C15879@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OF2E63E59B.0981E8F2-ONC2256CD8.003D347A@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF2E63E59B.0981E8F2-ONC2256CD8.003D347A@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Tue, Feb 25, 2003 at 01:20:59PM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

TOLGA,

TOLGA ASVEREN wrote:                                                                 (Tue, 25 Feb 2003 13:20:59)
> 
> Brian,
> 
> 
> 
> TOLGA,
> 
> TOLGA ASVEREN wrote:
> (Tue, 25 Feb 2003 08:39:11)
> >
> > Not necessarily because I disagree with some change, but just to clarify
> > where the changes might take us:
> 
> If we cannot agree, we let the RFC stand.
> 
> > NTFY was used till now only as an advisory message, this will be changed.
> 
> No it will not.  An ASP is not forbidden from acting on any Notify message.
> All Notify messages are still advisory.  The ASP can take into
> consideration
> Notify messages, information that it has about other ASPs, the time of day,
> direction of the wind, etc.
> [TOLGA]IMO important is, what ASP SHOULD do after receiving NTFY, if we
> want to avoid interoperability problems. So I think to change handling of
> NTFY is necessary.

As the RFC says, the ASP SHOULD wait for "n" ASPs to become active.  How
it does that is its own business.  Nowhere does it say that the ASP should
wait for NTFY.  But the ASP is welcome to act on this information.  There
is no change to the advisory nature of NTFY.

There is no change to the RFC needed here.

> 
> > Starting/stopping(till now we focused more on starting but why not
> stopping
> > also? So SG can decide based on load/other dynamic conditions when to
> stop
> > traffic to/from AS) traffic will be controlled by SGP with NTFY.
> 
> The SG has always been permitted to start or stop traffic when it considers
> it necessary and has always been permitted to perform its necessary
> functions
> with the SS7 network.  The SGP can send ASPIA Ack or drop associations any
> time it wishes to stop traffic.
> [TOLGA]If ASP is not ACTIVE for the AS, SGP should better not send traffic
> to it. For stopping I agree that SGP can decide for this. My point is, this
> better be reflected as a state change of AS and notified to ASP with NTFY,
> which afterwards should stop traffic.

If the SGP drops the association it will not be sending any NTFY messages until
it reestablishes it.  If the SGP sends ASPIA Ack and considers the AS in the
AS-INACTIVE state, it is already required by the RFC to send NTFY(AS_Inactive)
to all ASPs in the ASP-INACTIVE state for the AS.

Again, there is no change to the RFC needed here.

> 
> > IMO, this requires still a change to the original AS state machine for
> > transitions from INACTIVE to ACTIVE and possibly from ACTIVE to PENDING.
> 
> It does not.  The SG is only withholding a Notify message as proscribed
> by the RFC.
> [TOLGA]Without state change, this is really ugly. Why do we have the state
> machine then? In fact, another thing which is needed is probably the satet
> machine from ASP point of view.

It is not ugly.  It is an optional procedure.

> 
> > In general all this will change the meaning of ASPAC/ASPAC-ACK exchange,
> it
> > will mean that ASP is ready to be triggered to send/receive traffic for a
> > specific AS (something to be determined just by ASP, when the triggering
> > will happen will be decided by SGP)rather than ASP can start to
> > send/receive traffic for a specific AS(something to be determined by both
> > ASP and SGP). I believe this migth require some other changes throughout
> > the document. For example instead of unsolicited ASPIA-ACK, I would think
> > SGP should send NTFY(INACTIVE). My opinion is, if we are changing
> > something, let's do it clean and correct.
> 
> No nothing changes here because withholding NTFY(ACTIVE) is already part
> of the RFC.  Nothing in the RFC changes.
> [TOLGA]It is clear that it is not mandatory to take any action after
> receiving NTFY. The only thing related with starting traffic from ASP is
> that ASPAC/ASPAC-ACK exchange should be done and that ASP should wait till
> number of ACTIVE ASPs reaches n.

Yes, nothing needs to change in the RFC.

> 
> >
> > IMO, overall this is more than just a clarification, let's decide
> carefully
> > to prevent rollbacks (as much as possible).
> 
> Not even clarification is required.  The RFC details the necessary actions.
> [TOLGA]IMO it doesn't. For example it says that ASPs should wait for n
> ACTIVE ASPs (n interpreted as mandatory) and SGP may decide not to start
> traffic and withhold NTFY (n not interpreted as mandatory)

That's right: there is no clarfication required.  The reasons behind the
MAY and the SHOULD have already been explained.

It does not change the fact that IG section 3.15 is in error and must be
struck.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 08:55:21 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA01224
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 08:55:21 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1PE4MM30744
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 09:04:22 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PE44p30728;
	Tue, 25 Feb 2003 09:04:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PE39p30697
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 09:03:09 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA01156
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 08:53:35 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:hMzmwQDjVhkPgVTbxJAYgAqKcNOEuNyb@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1PDvTD22084;
	Tue, 25 Feb 2003 06:57:29 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1PDvTE16366;
	Tue, 25 Feb 2003 06:57:29 -0700
Date: Tue, 25 Feb 2003 06:57:28 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>
Cc: "'Samuel Dur D. Jeyaseelan'" <sjeyasee@ssd.usa.alcatel.com>,
        TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030225065728.D15879@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>,
	"'Samuel Dur D. Jeyaseelan'" <sjeyasee@ssd.usa.alcatel.com>,
	TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B659C6@EESTQNT105.es.eu.ericsson.se>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <1AB3D30B989BF141BBD5C70057B2EF7C01B659C6@EESTQNT105.es.eu.ericsson.se>; from javier.pastor-balbas@ece.ericsson.se on Tue, Feb 25, 2003 at 02:03:32PM +0100
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Javier,

The state machine is fine as it appears in the RFC.  AS are not considered
PENDING or INACTIVE until all ASPs are down or inactive for the AS.

Anything else is merely an expansion of the ACTIVE state, which the
implementer is already welcome to expand in any manner they like.

No change is needed to the state machine or text of the RFC.

--brian

Javier Pastor-Balbas (ECE) wrote:                                                   (Tue, 25 Feb 2003 14:03:32)
> Hi Samuel et al., 
> 
> Take into account that if you define two ASes:
> - AS1 DPC+SLS1  with two ASPs (ASP1 and ASP2) and redundancy: n=1 k=1
> - AS2 DPC+SLS2  with two ASPs (ASP1 and ASP2) and redundancy: n=1 k=1
> 
> And 
> - ASP1 active for AS1 and backup for AS2
> - ASP2 active for AS2 and backup for AS1
> 
> You get the desired behaviour. This case can be solved with a right network configuration without changing the current "required" qualification for "n".
> 
> 
> 
> Apart from this, 
> 
> I would say that a compromise option would consist of considering "n" as the normal working case, but make it implementation dependent (maybe configurable) whether an implementation may have more than "n" ASPs working at the same time for an AS, and whether the SG keeps working when the number of ASP is under "n". We could make a recommendation with "MAY" allowing both options.
> 
> But now, the problem comes with the AS state machine. If we agree upon here, we could say the AS is ACTIVE in the range n_min to n_max, where these two values are configurable.
> 
> 
> Could this be considered as a compromise solution about n+k behaviour?
> 
> 
> Cheers // Javier.
> 
> 
> 
> 
> 
> > -----Original Message-----
> > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > Sent: martes, 25 de febrero de 2003 10:56
> > To: Brian F. G. Bidulock
> > Cc: TOLGA ASVEREN; sigtran@ietf.org
> > Subject: Re: [Sigtran] Re: M3UA Notify message
> > 
> > 
> > 
> > Brian,
> > 
> >   Alcatel USA, Inc.                  Internet: 
> > <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel 
> > USA, Inc. *******
> > 
> > On Tue, 25 Feb 2003, Brian F. G. Bidulock wrote:
> > 
> > > TOLGA,
> > > 
> > > TOLGA ASVEREN wrote:                                        
> >                          (Tue, 25 Feb 2003 08:39:11)
> > > > 
> > > > Not necessarily because I disagree with some change, but 
> > just to clarify
> > > > where the changes might take us:
> > > 
> > > If we cannot agree, we let the RFC stand.
> > > 
> > > > NTFY was used till now only as an advisory message, this 
> > will be changed.
> > > 
> > > No it will not.  An ASP is not forbidden from acting on any 
> > Notify message.
> > > All Notify messages are still advisory.  The ASP can take 
> > into consideration
> > > Notify messages, information that it has about other ASPs, 
> > the time of day,
> > > direction of the wind, etc.
> > > 
> > > > Starting/stopping(till now we focused more on starting 
> > but why not stopping
> > > > also? So SG can decide based on load/other dynamic 
> > conditions when to stop
> > > > traffic to/from AS) traffic will be controlled by SGP with NTFY.
> > > 
> > > The SG has always been permitted to start or stop traffic 
> > when it considers
> > > it necessary and has always been permitted to perform its 
> > necessary functions
> > > with the SS7 network.  The SGP can send ASPIA Ack or drop 
> > associations any
> > > time it wishes to stop traffic.
> > 
> > Please let me know what are we achieving by stopping the 
> > traffic in the
> > following case(n+k - reduntancy)
> > 
> > When 'n' ASPs are active
> > ------------------------
> > SLS-X  ---->ASP1
> > SLS-Y  ---->ASP2
> > 
> > When ASP1 goes down
> > -------------------
> > SLS-(Y,X)  ----->ASP2
> > 
> > Here 3 cases can be dealt
> > 
> > case 1) If SG receives  message with SLS Y - this traffic can be
> >         successfully (100%) handled by ASP2 according to the 
> > configured
> >         databse.
> >  
> > case 2) If SG receives messages with SLS X and Y - this traffic can be
> >         successfully (100%) handled by ASP2, provided 
> > congestion has not
> >         occured(according to q.704 section 3.8.2)
> > 
> > case 3) Don't allow SG to process the traffic when n<2 ?????????.
> >         SG disallows traffic when n=0.
> > 
> > IMO, SG has capability to reroute the traffic to ASP1 and there is no
> > meaning if SG stops traffic when n < 2.
> > 
> > This is just like a MTP3 link failure concept when link1 goes down ,
> > traffic carried by link1 will be rerouted to available link. 
> > All we need to do at SG is just change the SLS-ASP dynamic 
> > table to carry
> > SLS X.
> > 
> > 
> > --Samuel
> > > 
> > > > IMO, this requires still a change to the original AS 
> > state machine for
> > > > transitions from INACTIVE to ACTIVE and possibly from 
> > ACTIVE to PENDING.
> > > 
> > > It does not.  The SG is only withholding a Notify message 
> > as proscribed
> > > by the RFC.
> > > 
> > > > In general all this will change the meaning of 
> > ASPAC/ASPAC-ACK exchange, it
> > > > will mean that ASP is ready to be triggered to 
> > send/receive traffic for a
> > > > specific AS (something to be determined just by ASP, when 
> > the triggering
> > > > will happen will be decided by SGP)rather than ASP can start to
> > > > send/receive traffic for a specific AS(something to be 
> > determined by both
> > > > ASP and SGP). I believe this migth require some other 
> > changes throughout
> > > > the document. For example instead of unsolicited 
> > ASPIA-ACK, I would think
> > > > SGP should send NTFY(INACTIVE). My opinion is, if we are changing
> > > > something, let's do it clean and correct.
> > > 
> > > No nothing changes here because withholding NTFY(ACTIVE) is 
> > already part
> > > of the RFC.  Nothing in the RFC changes.
> > > 
> > > > 
> > > > IMO, overall this is more than just a clarification, 
> > let's decide carefully
> > > > to prevent rollbacks (as much as possible).
> > > 
> > > Not even clarification is required.  The RFC details the 
> > necessary actions.
> > > 
> > > The only thing that needs to be done is strike section 3.15 
> > from the IG
> > > because it conflicts with the RFC (which is fine as it is).
> > > 
> > > --brian
> > > 
> > > -- 
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 09:09:52 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA01527
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 09:09:52 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1PEIsX32195
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 09:18:54 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PEIZp32178;
	Tue, 25 Feb 2003 09:18:36 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PEHOp32141
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 09:17:24 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA01512
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 09:07:50 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1PEBgJT016607;
	Tue, 25 Feb 2003 08:11:42 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h1PEBVF29633;
	Tue, 25 Feb 2003 08:11:31 -0600 (CST)
Date: Tue, 25 Feb 2003 08:11:31 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>
cc: "Brian F. G. Bidulock" <bidulock@openss7.org>,
        TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
Subject: RE: [Sigtran] Re: M3UA Notify message
In-Reply-To: <1AB3D30B989BF141BBD5C70057B2EF7C01B659C6@EESTQNT105.es.eu.ericsson.se>
Message-ID: <Pine.SOL.4.10.10302250745130.3454-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Javier,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Tue, 25 Feb 2003, Javier Pastor-Balbas (ECE) wrote:

> Hi Samuel et al., 
> 
> Take into account that if you define two ASes:
> - AS1 DPC+SLS1  with two ASPs (ASP1 and ASP2) and redundancy: n=1 k=1
> - AS2 DPC+SLS2  with two ASPs (ASP1 and ASP2) and redundancy: n=1 k=1
> 
> And 
> - ASP1 active for AS1 and backup for AS2
> - ASP2 active for AS2 and backup for AS1
> 
> You get the desired behaviour. This case can be solved with a right 
  network configuration without changing the current "required"
  qualification for "n".
> 
> 
> 
> Apart from this, 
> 
> I would say that a compromise option would consist of considering "n" as 
  the normal working case, but make it implementation dependent (maybe
configurable) whether an implementation may have more than "n" ASPs 
 working at the same time for an AS, and whether the SG keeps working when
the number of ASP is under "n". We could make a recommendation with "MAY" 
allowing both options.
>
[Samuel] Yes .SG keeps working when n=1 or n >=2.
but current RFC says that SG NOTIFIES 'k' ASP to become active only
when n<2 and it does not say what will happen when SG requires more than
'n' ASP to handle the traffic 't+1'.
 
> But now, the problem comes with the AS state machine. If we agree upon here,
 we could say the AS is ACTIVE in the range n_min to n_max, where these two
 values are configurable.
> 
[Samuel] To handle the 'normal(best) traffic T' SG requires 'n=2' ASPs.
In the course of traffic if 'n became 1'(means ASP1 failed), here SG can
handle the traffic with 'n=1' ASP. 
So only intially SG requires 'n=2' ASPs to make the AS state as ACTIVE
and 'n=2' is fed during configuration.But while handling traffic due to 
failure of ASP1 'n' becomes '1', here still AS is ACTIVE with single ASP.

--Samuel
> 
> Could this be considered as a compromise solution about n+k behaviour?
> 
> 
> Cheers // Javier.
> 
> 
> 
> 
> 
> > -----Original Message-----
> > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > Sent: martes, 25 de febrero de 2003 10:56
> > To: Brian F. G. Bidulock
> > Cc: TOLGA ASVEREN; sigtran@ietf.org
> > Subject: Re: [Sigtran] Re: M3UA Notify message
> > 
> > 
> > 
> > Brian,
> > 
> >   Alcatel USA, Inc.                  Internet: 
> > <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel 
> > USA, Inc. *******
> > 
> > On Tue, 25 Feb 2003, Brian F. G. Bidulock wrote:
> > 
> > > TOLGA,
> > > 
> > > TOLGA ASVEREN wrote:                                        
> >                          (Tue, 25 Feb 2003 08:39:11)
> > > > 
> > > > Not necessarily because I disagree with some change, but 
> > just to clarify
> > > > where the changes might take us:
> > > 
> > > If we cannot agree, we let the RFC stand.
> > > 
> > > > NTFY was used till now only as an advisory message, this 
> > will be changed.
> > > 
> > > No it will not.  An ASP is not forbidden from acting on any 
> > Notify message.
> > > All Notify messages are still advisory.  The ASP can take 
> > into consideration
> > > Notify messages, information that it has about other ASPs, 
> > the time of day,
> > > direction of the wind, etc.
> > > 
> > > > Starting/stopping(till now we focused more on starting 
> > but why not stopping
> > > > also? So SG can decide based on load/other dynamic 
> > conditions when to stop
> > > > traffic to/from AS) traffic will be controlled by SGP with NTFY.
> > > 
> > > The SG has always been permitted to start or stop traffic 
> > when it considers
> > > it necessary and has always been permitted to perform its 
> > necessary functions
> > > with the SS7 network.  The SGP can send ASPIA Ack or drop 
> > associations any
> > > time it wishes to stop traffic.
> > 
> > Please let me know what are we achieving by stopping the 
> > traffic in the
> > following case(n+k - reduntancy)
> > 
> > When 'n' ASPs are active
> > ------------------------
> > SLS-X  ---->ASP1
> > SLS-Y  ---->ASP2
> > 
> > When ASP1 goes down
> > -------------------
> > SLS-(Y,X)  ----->ASP2
> > 
> > Here 3 cases can be dealt
> > 
> > case 1) If SG receives  message with SLS Y - this traffic can be
> >         successfully (100%) handled by ASP2 according to the 
> > configured
> >         databse.
> >  
> > case 2) If SG receives messages with SLS X and Y - this traffic can be
> >         successfully (100%) handled by ASP2, provided 
> > congestion has not
> >         occured(according to q.704 section 3.8.2)
> > 
> > case 3) Don't allow SG to process the traffic when n<2 ?????????.
> >         SG disallows traffic when n=0.
> > 
> > IMO, SG has capability to reroute the traffic to ASP1 and there is no
> > meaning if SG stops traffic when n < 2.
> > 
> > This is just like a MTP3 link failure concept when link1 goes down ,
> > traffic carried by link1 will be rerouted to available link. 
> > All we need to do at SG is just change the SLS-ASP dynamic 
> > table to carry
> > SLS X.
> > 
> > 
> > --Samuel
> > > 
> > > > IMO, this requires still a change to the original AS 
> > state machine for
> > > > transitions from INACTIVE to ACTIVE and possibly from 
> > ACTIVE to PENDING.
> > > 
> > > It does not.  The SG is only withholding a Notify message 
> > as proscribed
> > > by the RFC.
> > > 
> > > > In general all this will change the meaning of 
> > ASPAC/ASPAC-ACK exchange, it
> > > > will mean that ASP is ready to be triggered to 
> > send/receive traffic for a
> > > > specific AS (something to be determined just by ASP, when 
> > the triggering
> > > > will happen will be decided by SGP)rather than ASP can start to
> > > > send/receive traffic for a specific AS(something to be 
> > determined by both
> > > > ASP and SGP). I believe this migth require some other 
> > changes throughout
> > > > the document. For example instead of unsolicited 
> > ASPIA-ACK, I would think
> > > > SGP should send NTFY(INACTIVE). My opinion is, if we are changing
> > > > something, let's do it clean and correct.
> > > 
> > > No nothing changes here because withholding NTFY(ACTIVE) is 
> > already part
> > > of the RFC.  Nothing in the RFC changes.
> > > 
> > > > 
> > > > IMO, overall this is more than just a clarification, 
> > let's decide carefully
> > > > to prevent rollbacks (as much as possible).
> > > 
> > > Not even clarification is required.  The RFC details the 
> > necessary actions.
> > > 
> > > The only thing that needs to be done is strike section 3.15 
> > from the IG
> > > because it conflicts with the RFC (which is fine as it is).
> > > 
> > > --brian
> > > 
> > > -- 
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > 
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 09:32:29 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA02009
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 09:32:29 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1PEfVp01510
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 09:41:31 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PEfDp01495;
	Tue, 25 Feb 2003 09:41:13 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PEegp01469
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 09:40:42 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA01958
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 09:31:06 -0500 (EST)
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF8B358F12.5F0F7166-ONC2256CD8.004EDB28@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Tue, 25 Feb 2003 16:31:24 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/25/2003
 04:31:35 PM
MIME-Version: 1.0
Content-type: text/plain; charset=Windows-1254
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h1PEegp01470
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit


Brian,


TOLGA,

TOLGA ASVEREN wrote:
(Tue, 25 Feb 2003 13:20:59)
>
> Brian,
>
>
>
> TOLGA,
>
> TOLGA ASVEREN wrote:
> (Tue, 25 Feb 2003 08:39:11)
> >
> > Not necessarily because I disagree with some change, but just to
clarify
> > where the changes might take us:
>
> If we cannot agree, we let the RFC stand.
>
> > NTFY was used till now only as an advisory message, this will be
changed.
>
> No it will not.  An ASP is not forbidden from acting on any Notify
message.
> All Notify messages are still advisory.  The ASP can take into
> consideration
> Notify messages, information that it has about other ASPs, the time of
day,
> direction of the wind, etc.
> [TOLGA]IMO important is, what ASP SHOULD do after receiving NTFY, if we
> want to avoid interoperability problems. So I think to change handling of
> NTFY is necessary.

As the RFC says, the ASP SHOULD wait for "n" ASPs to become active.  How
it does that is its own business.  Nowhere does it say that the ASP should
wait for NTFY.  But the ASP is welcome to act on this information.  There
is no change to the advisory nature of NTFY.
[TOLGA]You also agree that RFC does interprete NTFY as an advisory message
but it is mandatory that ASP decides to start traffic based on NTFY(AS
ACTIVE) if we want SG to decide, when traffic flow between SGP/ASP is to be
opened. This needs to be clarified in IG. And IMO this also requires a
chnage in the AS state machine, for transition from INACTIVE to ACTIVE.
This transition won't occure with the first ACTIVE ASP but SG will decide
for it based on number of ACTIVE ASPs.
BTW, RFC says that ASPs coordinate among themselves number of ACTIVE ASPs
to decide to start traffic, again needs to be changed -if there is
consensus for the change-.



There is no change to the RFC needed here.

>
> > Starting/stopping(till now we focused more on starting but why not
> stopping
> > also? So SG can decide based on load/other dynamic conditions when to
> stop
> > traffic to/from AS) traffic will be controlled by SGP with NTFY.
>
> The SG has always been permitted to start or stop traffic when it
considers
> it necessary and has always been permitted to perform its necessary
> functions
> with the SS7 network.  The SGP can send ASPIA Ack or drop associations
any
> time it wishes to stop traffic.
> [TOLGA]If ASP is not ACTIVE for the AS, SGP should better not send
traffic
> to it. For stopping I agree that SGP can decide for this. My point is,
this
> better be reflected as a state change of AS and notified to ASP with
NTFY,
> which afterwards should stop traffic.

If the SGP drops the association it will not be sending any NTFY messages
until
it reestablishes it.  If the SGP sends ASPIA Ack and considers the AS in
the
AS-INACTIVE state, it is already required by the RFC to send
NTFY(AS_Inactive)
to all ASPs in the ASP-INACTIVE state for the AS.
[TOLGA]Yes, one needs to send NTFY(AS Inactive) in such a case, but still
advisory. And I would think, if ASPs rely on NTFY for activation, they
better rely on NTFY also for deactivation.


Again, there is no change to the RFC needed here.

>
> > IMO, this requires still a change to the original AS state machine for
> > transitions from INACTIVE to ACTIVE and possibly from ACTIVE to
PENDING.
>
> It does not.  The SG is only withholding a Notify message as proscribed
> by the RFC.
> [TOLGA]Without state change, this is really ugly. Why do we have the
state
> machine then? In fact, another thing which is needed is probably the
satet
> machine from ASP point of view.

It is not ugly.  It is an optional procedure.

>
> > In general all this will change the meaning of ASPAC/ASPAC-ACK
exchange,
> it
> > will mean that ASP is ready to be triggered to send/receive traffic for
a
> > specific AS (something to be determined just by ASP, when the
triggering
> > will happen will be decided by SGP)rather than ASP can start to
> > send/receive traffic for a specific AS(something to be determined by
both
> > ASP and SGP). I believe this migth require some other changes
throughout
> > the document. For example instead of unsolicited ASPIA-ACK, I would
think
> > SGP should send NTFY(INACTIVE). My opinion is, if we are changing
> > something, let's do it clean and correct.
>
> No nothing changes here because withholding NTFY(ACTIVE) is already part
> of the RFC.  Nothing in the RFC changes.
> [TOLGA]It is clear that it is not mandatory to take any action after
> receiving NTFY. The only thing related with starting traffic from ASP is
> that ASPAC/ASPAC-ACK exchange should be done and that ASP should wait
till
> number of ACTIVE ASPs reaches n.

Yes, nothing needs to change in the RFC.
[TOLGA]Then it won't work like you propose, because ASPs won't care about
NTFY(AS Actýve) but they will coordinate number of ACTIVE ASPs and will
start traffic when this number reaches n.

>
> >
> > IMO, overall this is more than just a clarification, let's decide
> carefully
> > to prevent rollbacks (as much as possible).
>
> Not even clarification is required.  The RFC details the necessary
actions.
> [TOLGA]IMO it doesn't. For example it says that ASPs should wait for n
> ACTIVE ASPs (n interpreted as mandatory) and SGP may decide not to start
> traffic and withhold NTFY (n not interpreted as mandatory)

That's right: there is no clarfication required.  The reasons behind the
MAY and the SHOULD have already been explained.
[TOLGA]That SGP MAY withhold traffic/Notify is what you propose, this is in
contradiction with the statement that ASPs SHOULD wait for n ACTIVE ASPs.
For example n=5. SG decides to start traffic with the 2nd ACTIVE ASP. But
according to RFC ASPs will wait for the 5th ACTIVE ASP before starting
traffic. In my opinion, this is a problem.


It does not change the fact that IG section 3.15 is in error and must be
struck.

--brian

--
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 09:47:20 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA02248
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 09:47:20 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1PEuNU02114
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 09:56:23 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PEuAp02106;
	Tue, 25 Feb 2003 09:56:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PEu0p02076
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 09:56:00 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA02231
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 09:46:25 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:VYc87MlK8fFsHeiWqnJzaseVuIaPLRUa@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1PEoJD23120;
	Tue, 25 Feb 2003 07:50:19 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1PEoJK17449;
	Tue, 25 Feb 2003 07:50:19 -0700
Date: Tue, 25 Feb 2003 07:50:18 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030225075018.B16811@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OF8B358F12.5F0F7166-ONC2256CD8.004EDB28@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF8B358F12.5F0F7166-ONC2256CD8.004EDB28@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Tue, Feb 25, 2003 at 04:31:24PM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: 8bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

TOLGA,

TOLGA ASVEREN wrote:                                                                 (Tue, 25 Feb 2003 16:31:24)
> 
> Brian,
> 
> 
> TOLGA,
> 
> TOLGA ASVEREN wrote:
> (Tue, 25 Feb 2003 13:20:59)
> >
> > Brian,
> >
> >
> >
> > TOLGA,
> >
> > TOLGA ASVEREN wrote:
> > (Tue, 25 Feb 2003 08:39:11)
> > >
> > > Not necessarily because I disagree with some change, but just to
> clarify
> > > where the changes might take us:
> >
> > If we cannot agree, we let the RFC stand.
> >
> > > NTFY was used till now only as an advisory message, this will be
> changed.
> >
> > No it will not.  An ASP is not forbidden from acting on any Notify
> message.
> > All Notify messages are still advisory.  The ASP can take into
> > consideration
> > Notify messages, information that it has about other ASPs, the time of
> day,
> > direction of the wind, etc.
> > [TOLGA]IMO important is, what ASP SHOULD do after receiving NTFY, if we
> > want to avoid interoperability problems. So I think to change handling of
> > NTFY is necessary.
> 
> As the RFC says, the ASP SHOULD wait for "n" ASPs to become active.  How
> it does that is its own business.  Nowhere does it say that the ASP should
> wait for NTFY.  But the ASP is welcome to act on this information.  There
> is no change to the advisory nature of NTFY.
> [TOLGA]You also agree that RFC does interprete NTFY as an advisory message
> but it is mandatory that ASP decides to start traffic based on NTFY(AS
> ACTIVE) if we want SG to decide, when traffic flow between SGP/ASP is to be
> opened.

I disagree.  The ASP can use whatever means it has at its disposal to decide
when "n" ASPs are active.  That includes advisory messages.  It is not
(and cannot be) required that the ASP start traffic based on NTFY(AS_ACTIVE),
but that does not forbid the ASP from using NTFY(AS_ACTIVE) in any fashion
it desires.

This is why the RFC says that the ASP SHOULD wait for "n" ASPs to start
traffic.  How the ASP decides when "n" has been reached is the ASP's own
business.  One of the things it may consider is NTFY.

This is no different than an ASP using (or not) NTFY("Insufficent ASPs
Active in AS") as one of a number of things to decide to go active.

> This needs to be clarified in IG.

It does not need to be clarified.

> And IMO this also requires a chnage in the AS state machine, for transition
> from INACTIVE to ACTIVE.  This transition won't occure with the first ACTIVE
> ASP but SG will decide for it based on number of ACTIVE ASPs.

No, no, no.  The SG moves the AS to AS-ACTIVE.  It only decides when to send
NTFY based on an OPTIONAL procedure.

> BTW, RFC says that ASPs coordinate among themselves number of ACTIVE ASPs
> to decide to start traffic, again needs to be changed -if there is
> consensus for the change-.

They SHOULD coordinate until "n" are active.  This is already stated.

> 
> 
> There is no change to the RFC needed here.
> 
> >
> > > Starting/stopping(till now we focused more on starting but why not
> > stopping
> > > also? So SG can decide based on load/other dynamic conditions when to
> > stop
> > > traffic to/from AS) traffic will be controlled by SGP with NTFY.
> >
> > The SG has always been permitted to start or stop traffic when it
> considers
> > it necessary and has always been permitted to perform its necessary
> > functions
> > with the SS7 network.  The SGP can send ASPIA Ack or drop associations
> any
> > time it wishes to stop traffic.
> > [TOLGA]If ASP is not ACTIVE for the AS, SGP should better not send
> traffic
> > to it. For stopping I agree that SGP can decide for this. My point is,
> this
> > better be reflected as a state change of AS and notified to ASP with
> NTFY,
> > which afterwards should stop traffic.
> 
> If the SGP drops the association it will not be sending any NTFY messages
> until
> it reestablishes it.  If the SGP sends ASPIA Ack and considers the AS in
> the
> AS-INACTIVE state, it is already required by the RFC to send
> NTFY(AS_Inactive)
> to all ASPs in the ASP-INACTIVE state for the AS.
> [TOLGA]Yes, one needs to send NTFY(AS Inactive) in such a case, but still
> advisory. And I would think, if ASPs rely on NTFY for activation, they
> better rely on NTFY also for deactivation.

The ASP can rely (or not) on NTFY until the cows come home.  It does not
change the advisory nature of NTFY.  The point is that the ASP is never
REQUIRED to act on NTFY messages.  They are advice: not commands.  This has
not changed.  The ASP is welcome to use whatever other means it has at its
disposal (e.g. coordinating between ASPs, but NTFY too if it so desires)
to determine when "n" ASPs are active.

Still there is no change required.  That is the way it is now in the RFC.

> 
> 
> Again, there is no change to the RFC needed here.
> 
> >
> > > IMO, this requires still a change to the original AS state machine for
> > > transitions from INACTIVE to ACTIVE and possibly from ACTIVE to
> PENDING.
> >
> > It does not.  The SG is only withholding a Notify message as proscribed
> > by the RFC.
> > [TOLGA]Without state change, this is really ugly. Why do we have the
> state
> > machine then? In fact, another thing which is needed is probably the
> satet
> > machine from ASP point of view.
> 
> It is not ugly.  It is an optional procedure.
> 
> >
> > > In general all this will change the meaning of ASPAC/ASPAC-ACK
> exchange,
> > it
> > > will mean that ASP is ready to be triggered to send/receive traffic for
> a
> > > specific AS (something to be determined just by ASP, when the
> triggering
> > > will happen will be decided by SGP)rather than ASP can start to
> > > send/receive traffic for a specific AS(something to be determined by
> both
> > > ASP and SGP). I believe this migth require some other changes
> throughout
> > > the document. For example instead of unsolicited ASPIA-ACK, I would
> think
> > > SGP should send NTFY(INACTIVE). My opinion is, if we are changing
> > > something, let's do it clean and correct.
> >
> > No nothing changes here because withholding NTFY(ACTIVE) is already part
> > of the RFC.  Nothing in the RFC changes.
> > [TOLGA]It is clear that it is not mandatory to take any action after
> > receiving NTFY. The only thing related with starting traffic from ASP is
> > that ASPAC/ASPAC-ACK exchange should be done and that ASP should wait
> till
> > number of ACTIVE ASPs reaches n.
> 
> Yes, nothing needs to change in the RFC.
> [TOLGA]Then it won't work like you propose, because ASPs won't care about
> NTFY(AS Actýve) but they will coordinate number of ACTIVE ASPs and will
> start traffic when this number reaches n.

So?  The ASPs SHOULD NOT start traffic until "n" ASPs are active regardless
of whether they receive NTFY(AS_Active) or not.  The RFC is fine as it stands.

> >
> > >
> > > IMO, overall this is more than just a clarification, let's decide
> > carefully
> > > to prevent rollbacks (as much as possible).
> >
> > Not even clarification is required.  The RFC details the necessary
> actions.
> > [TOLGA]IMO it doesn't. For example it says that ASPs should wait for n
> > ACTIVE ASPs (n interpreted as mandatory) and SGP may decide not to start
> > traffic and withhold NTFY (n not interpreted as mandatory)
> 
> That's right: there is no clarfication required.  The reasons behind the
> MAY and the SHOULD have already been explained.
> [TOLGA]That SGP MAY withhold traffic/Notify is what you propose, this is in
> contradiction with the statement that ASPs SHOULD wait for n ACTIVE ASPs.

It is not.  The SG MAY withhold traffic and the ASPs SHOULD wait regardless.

> For example n=5. SG decides to start traffic with the 2nd ACTIVE ASP. But
> according to RFC ASPs will wait for the 5th ACTIVE ASP before starting
> traffic. In my opinion, this is a problem.

There is no problem.  The ASPs should wait for the 5th ASP to be active before
starting traffic regardless of when the SG starts traffic.

> 
> 
> It does not change the fact that IG section 3.15 is in error and must be
> struck.


Section 3.15 still needs to be struck because it is plain wrong.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 15:16:01 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA13560
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 15:16:01 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1PKPAw27653
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 15:25:10 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PKOup27612;
	Tue, 25 Feb 2003 15:24:56 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PKLAp27353
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 15:21:10 -0500
Received: from web12805.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id PAA13334
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 15:11:29 -0500 (EST)
Message-ID: <20030225201522.39283.qmail@web12805.mail.yahoo.com>
Received: from [62.29.65.115] by web12805.mail.yahoo.com via HTTP; Tue, 25 Feb 2003 12:15:22 PST
Date: Tue, 25 Feb 2003 12:15:22 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
In-Reply-To: <20030225075018.B16811@openss7.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Brian,

[..snip..]

> For example n=5. SG decides to start traffic with
>the 2nd ACTIVE ASP. 
>But
> according to RFC ASPs will wait for the 5th ACTIVE
>ASP before 
>starting
> traffic. In my opinion, this is a problem.

There is no problem.  The ASPs should wait for the 5th
ASP to be active 
before
starting traffic regardless of when the SG starts
traffic.
[TOLGA]O.K. finally there we are, the source of the
disagreement. IMO, it is not a good idea to open/close
the traffic path between ASP/SGP unidirectionaly. 
And secondly, if you prefer SG to decide dynamically
to start traffic, why do you want ASPs to decide for
this statically by interpreting n as mandatory?

__________________________________________________
Do you Yahoo!?
Yahoo! Tax Center - forms, calculators, tips, more
http://taxes.yahoo.com/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 15:29:53 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA14038
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 15:29:53 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1PKd3o29024
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 15:39:03 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PKcop28994;
	Tue, 25 Feb 2003 15:38:52 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PKahp28161
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 15:36:43 -0500
Received: from web12802.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id PAA13916
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 15:27:01 -0500 (EST)
Message-ID: <20030225203056.80100.qmail@web12802.mail.yahoo.com>
Received: from [62.29.65.115] by web12802.mail.yahoo.com via HTTP; Tue, 25 Feb 2003 12:30:56 PST
Date: Tue, 25 Feb 2003 12:30:56 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: RE: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
In-Reply-To: <1AB3D30B989BF141BBD5C70057B2EF7C01B659C6@EESTQNT105.es.eu.ericsson.se>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Javier,

> Apart from this, 
> 
> I would say that a compromise option would consist
> of considering "n" as the normal working case, but
> make it implementation dependent (maybe
> configurable) whether an implementation may have
> more than "n" ASPs working at the same time for an
> AS, and whether the SG keeps working when the number
> of ASP is under "n". We could make a recommendation
> with "MAY" allowing both options.
> 
> But now, the problem comes with the AS state
> machine. If we agree upon here, we could say the AS
> is ACTIVE in the range n_min to n_max, where these
> two values are configurable.
> 
> 
> Could this be considered as a compromise solution
> about n+k behaviour?
[TOLGA]My opinion after the discussions in this thread
is, let the SG decide when to start/stop traffic
between ASP/SGP. So, SG can decide how it will
interprete n and there is no need to expose the
concept of n to ASPs, they can act based on NTFYs sent
by SGP. I would think this should be suitable for
various configurations. But before proposing text
let's see, whether the WG is happy with this solution.


__________________________________________________
Do you Yahoo!?
Yahoo! Tax Center - forms, calculators, tips, more
http://taxes.yahoo.com/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 15:30:09 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA14055
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 15:30:09 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1PKdJa29055
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 15:39:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PKd5p29042;
	Tue, 25 Feb 2003 15:39:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PCsrp26026
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 07:54:53 -0500
Received: from hrtades9.atea.be (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA28372
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 07:45:19 -0500 (EST)
Received: from hrtades10.atea.be (siemens.atea.be [139.10.143.141]) by hrtades9.atea.be with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2653.13)
	id 1RFV2H1F; Tue, 25 Feb 2003 13:53:17 +0100
Received: by siemens.atea.be with Internet Mail Service (5.5.2653.19)
	id <FNAN6PHA>; Tue, 25 Feb 2003 13:51:49 +0100
Message-ID: <57FD2C3A246F76438CA6FDAD8FE9F195A7F38C@hrtades7.atea.be>
From: Coene Lode <Lode.Coene@siemens.com>
To: "'Javier Pastor-Balbas (ECE)'" <javier.pastor-balbas@ece.ericsson.se>,
        "'bidulock@openss7.org'" <bidulock@openss7.org>,
        Coene Lode
	 <Lode.Coene@siemens.com>
Cc: sigtran@ietf.org
Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-signalling-over-sctp-
	 applic-04.txt
Date: Tue, 25 Feb 2003 13:43:49 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h1PCsrp26027
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

i've looked through the list of Brian and there were still some stuff we did
miss. 
I've updated the draft and I will send the list of points with their
resolution to the list. 
Sorry for taking so long, but I experienced a minor system
malfunction(translation: I fell sick) and there are still some yellow
warning lights flashing at my end. I hope to get it out at the end of this
week.

yours sincerly,
Lode

Siemens atea
atealaan 34          2200 Herentals, Belgium
E-mail: lode.coene@siemens.atea.be
Tel: +32-14-252081
Fax: +32-14-253212


> -----Original Message-----
> From: Javier Pastor-Balbas (ECE)
> [mailto:javier.pastor-balbas@ece.ericsson.se]
> Sent: Monday, February 24, 2003 12:35 PM
> To: 'bidulock@openss7.org'; Coene Lode
> Cc: sigtran@ietf.org
> Subject: RE: [Sigtran] I-D
> ACTION:draft-ietf-sigtran-signalling-over-sctp- applic-04.txt
> 
> 
> Hi Brian, 
> 
> All the points you refer here were taken into account during 
> the session edition in Minneapolis. And the result was 
> version 5. If there is any particular point that is not there 
> now, and you want to discuss it, just point it, and we'll 
> show the reasons why we included or excluded it.
> 
> Thanks // Javier.
> 
> 
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: miércoles, 19 de febrero de 2003 10:57
> To: Coene Lode
> Cc: sigtran@ietf.org
> Subject: Re: [Sigtran] I-D
> ACTION:draft-ietf-sigtran-signalling-over-sctp- applic-04.txt
> 
> 
> Coene,
> 
> Although many of these comments were handled between draft-04 
> and draft-07,
> there are a number that have not been addressed.
> 
> Please consider the unresolved comments in this list as 
> comments for LC.
> 
> --brian
> 
> Coene Lode wrote:                                             
>                        (Thu, 07 Mar 2002 09:10:05)
> > Brian,
> > 
> > Thank you for the detailed list.
> > I will not answer them point for point as the document 
> itself was/is not
> > fully baked yet. 
> > With your input It looks like we'll get much nearer that goal.
> > We will take into account all the points you mentioned 
> during our editing in
> > Minneapolis.
> > I hope that after a new release all your points will have 
> taken care of.
> > 
> > yours sincerly,
> > Lode
> > 
> > Siemens atea
> > atealaan 34          2200 Herentals, Belgium
> > E-mail: lode.coene@siemens.atea.be
> > Tel: +32-14-252081
> > Fax: +32-14-253212
> > 
> > 
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: Wednesday, March 06, 2002 11:47 AM
> > > To: Coene Lode
> > > Cc: sigtran@ietf.org
> > > Subject: Re: [Sigtran] I-D
> > > ACTION:draft-ietf-sigtran-signalling-over-sctp-applic-04.txt
> > > 
> > > 
> > > Lode,
> > > 
> > > A few comments on this draft:
> > > 
> > > Section 2:
> > > 
> > > >   ...            Message boundaries are preserved during data
> > > >   transport and so no message delineation is needed. ...
> > > 
> > > Yet, each UA message has a common header which contains a label
> > > and a length which indicates the length of the message.  This is
> > > a message delimitation.  Why is it present in each UA if it is
> > > not required?
> > > 
> > > >                                                  ... The 
> > > user data can
> > > >   be delivered by the order of transmission within a stream(in
> > > >   sequence delivery) or the order of arrival.
> > > 
> > > Unordered delivery in SCTP and "order of arrival" delivery 
> > > are different
> > > things.  SCTP is permitted to place a priority on out of 
> order packets
> > > from various streams.  Preference for delivery of stream 
> "0" unordered
> > > packets ahead of other stream unordered packets is of 
> > > significant advantage
> > > to most UAs (regarless of order of arrival).  Please use the term
> > > "unordered delivery" instead of "order of arrival".
> > > 
> > > >   SCTP can be used to provide redundancy and fault 
> tolerance at the
> > > >   transport layer and below. Telephony applications needing 
> > > this level
> > > >   of fault tolerance can make use of SCTP's 
> multi-homing support.
> > > 
> > > "Fault tolerance" usually has a different connotation when used in
> > > telephony related computing.  It indicates a level of reliability
> > > (greater than 5 9's).  Please reword as follows:
> > > 
> > > <   SCTP can be used to provide redundancy at the 
> transport layer and
> > > <   below. Telephony applications needing this level of 
> redundancy can
> > > <   make use of SCTP's multi-homing support.
> > > 
> > > Section 3.1:
> > > 
> > > >                                               ...  Additionally,
> > > >   telephony applications may use SCTP stream based flow control
> > > >   [SCTPFLOW].
> > > 
> > > I believe that stream based flow controls were dropped 
> (in favor of
> > > application level flow controls).  Please remove this sentence.
> > > 
> > > Section 3.2:
> > > 
> > > >   Telephony systems often must achieve high availability in 
> > > operation.
> > > 
> > > Again, "high availability" has the connotation of a specific level
> > > of reliability in telephony related computing (in this 
> case less than
> > > 5 9's).  Please reword as:
> > > 
> > > <   Telephony systems often must achieve redundancy in operation.
> > > 
> > > or
> > > 
> > > <   Telephony systems often must have no single point of 
> failure in
> > > <   operation.
> > > 
> > > >   transport layer and below must be detected very fast 
> so that the
> > > 
> > > grammar, reword as:
> > > 
> > > <   transport layer and below must be detected quickly so that the
> > > 
> > > >   For example, they are often required to be able to 
> preserve stable
> > > >   calls during a component failure. Therefore error 
> > > situations at the
> > > >   transport layer and below must be detected very fast 
> so that the
> > > >   application can take approriate steps to recover and 
> preserve the
> > > >   stable calls. ...
> > > 
> > > This example is a poor one and shows an utter lack of 
> understanding
> > > for telephony.  Stable calls have *no* signalling associated 
> > > with them.
> > > It is only a call in transition that has signalling.  Signalling
> > > transport need not worry about affecting stable calls.  
> Only redundant
> > > systems maintaining call state need worry about this.  
> While this is
> > > a system requirement for an MG or MGC, it has nothing to do with
> > > signalling components.  Also, sharing of call state 
> between redundant
> > > systems is explicitly excluded from the scope of the UAs.  Please
> > > remove this example: it makes the applicability statement 
> > > sound foolish.
> > > 
> > > On the other hand, telephony applications pose requirements on the
> > > availability of signalling paths to meet service availability and
> > > performance requirements.  SS7 has such signalling relation 
> > > availability
> > > requirements.  These are the telephony requirements which 
> impose on
> > > SCTP's ability to detect failures.  (Again, nothing to do 
> with call
> > > state.)  Please replace this example in section 3.2 with 
> words to this
> > > effect.
> > > 
> > > Section 3.2.2:
> > > 
> > > >   For faster detection of (un)availability of idle paths, the
> > > >   telephony application may consider lowering the SCTP parameter
> > > >   HB.interval. It should be noted this will result in a 
> > > higher traffic
> > > >   load.
> > > 
> > > It is not always true that lowered HB.interval will 
> incresae traffic
> > > load.  Please change "will result" to "might result".
> > > 
> > > Section 3.2.3:
> > > 
> > > >   Setting Path.Max.Retrans and Association.Max.Retrans SCTP 
> > > parameters
> > > >   to lower values will speed up both destination 
> address and peer
> > > >   failure detection. However, if these values are set 
> too low, the
> > > >   probability of false detections will increase.
> > > 
> > > What false detection?  A retransmission is a 
> retransmission.  There is
> > > no "false" retransmission or "false detection" mentioned anywhere.
> > > In tests we have shown that Path.Max.Retrans of zero (0) (changing
> > > destination address on every retransmission) works quite 
> > > fine.  I don't
> > > understand what you are trying to warn against here...
> > > 
> > > Section 3.3:
> > > 
> > > >   Telephony applications often require short end-to-end message
> > > >   delays.  The methods described in section 3.2.1 on 
> > > lowering RTO and
> > > >   not using delayed SACK may be considered.
> > > 
> > > Not using delayed SACK does nothing to shorten end-to-end message
> > > delay.  Please remove it.
> > > 
> > > Section 4:
> > > 
> > > >   Users Adaptation Layers are defined to substitute the 
> telephony
> > > >   signaling protocol that is below of the telephony 
> > > signaling protocol
> > > >   to be relayed.
> > > 
> > > "that is below of the"????  This sentence is so malformed 
> I don't know
> > > what it means.  Please correct.
> > > 
> > > Section 4:
> > > 
> > > >      - Support management of SCTP transport 
> associations and traffic
> > > >         between SGs and ISEPs or two ISEPs
> > > 
> > > It is a sore point that several UAs do *not* adequately support
> > > management between "two ISEPs."
> > > 
> > > Section 4:
> > > 
> > > >                 SP:   Signaling Protocol
> > > 
> > > What is shown is not a signalling protocol, but a 
> signalling protocol
> > > user.  Please change to SPU (Signalling Protocol User).
> > > 
> > > >                 TTST: Traditional Telephony Signaling Transport
> > > 
> > > The only traditional SS7 signalling that qualifies as a transport
> > > is TCAP.  I suggest you use "Protocol" instead of "Transport".
> > > Please replace TTST with TTSP (Traditional Telephony Signalling
> > > Protocol).
> > > 
> > > >   It is also referred as SG to AS communication. AS I 
> the name that
> > > 
> > > Typo: change "AS I the name" to "AS is the name"
> > > 
> > > >   It is also referred as IPSP communication. IPSP is 
> the name of the
> > > >   role that an IP-based node plays UAL usually gives to the ISEP
> > > >   nodes. It stands for IP Signaling Point.
> > > 
> > > English usage.  Please reword this entire paragraph 
> (especially the
> > > middle line "IP-based node plays UAL usually").
> > > 
> > > Section 4.1:
> > > 
> > > >                                             ... The default 
> > > orientation
> > > >   would be for the SG to take on the role of server while 
> > > the ISEP is
> > > >   the client. ...
> > > 
> > > Grammar: change "would be for" to "is for".
> > > 
> > > >   Examples of the upper layers to be transported would 
> be Q.931 and
> > > >   QSIG.
> > > 
> > > Grammar: change "would be" to "are".
> > > 
> > > >   communication where the ISEP role is typically played 
> by a node
> > > >   called MGC defined in [RFC2719].
> > > 
> > > Grammar: change "called MGC defined" to "called an MGC, 
> as defined".
> > > 
> > > >   The value assigned by IANA for the Payload Protocol 
> > > Identifier in the
> > > >   SCTP Payload Data chunk is ô1ö
> > > 
> > > Non-ascii characters at the end of this sentence.
> > > 
> > > Section 4.2:
> > > 
> > > >                                                   ... It is 
> > > basically
> > > >   implemented in an interworking scenario with SG.
> > > 
> > > Usage: change "basically" to "typically".
> > > 
> > > >           AN    û Access Network
> > > >           NIF   û Nodal Interworking Function
> > > >           LAPV5 û Link Access Protocol for the V5 channel
> > > 
> > > Non-ascii characters in text.
> > > 
> > > Section 4.3:
> > > 
> > > >   The DUA is built on top of IUA defining the necessary 
> > > extensions to
> > > >   IUA for a DPNSS/DASS2 transport. ...
> > > 
> > > grammar: change "top of IAU defining" to "top of IUA and defines"
> > > 
> > > >                                ... DPNSS stands for 
> Digital Private
> > > >   Network Signaling System and DASS2 for Digital Access 
> Signaling
> > > >   System No 2
> > > 
> > > Missing period at end of sentence.
> > > 
> > > >   The value assigned by IANA for the Payload Protocol 
> > > Identifier in the
> > > >   SCTP Payload Data chunk is ôTBDö.
> > > 
> > > Non-ascii text at end of sentence.
> > > 
> > > Does not DUA have a PPI assigned yet?
> > > 
> > > Section 4.4:
> > > 
> > > >   This protocol would be mainly used between a Signaling 
> > > Gateway (SG)
> > > >   and Media Gateway Controler (MGC). ...
> > > 
> > > Grammar: change "would be" to "is"; change "mainly" to 
> "typically" or
> > > "largely" or "usually".
> > > 
> > > >   The only SS7 MTP2 User is MTP3 that is the protocol 
> transported by
> > > >   this UAL.
> > > 
> > > Really bad sentence formation:  change to:
> > > 
> > > <   MTP3 is the only MTP2-User that is transport by this UAL.
> > > 
> > > Besides, the statement is incorrect: MTP3b could also be 
> transported.
> > > 
> > > >   The SG provides a interworking of transport functions 
> with the IP
> > > >   transport, in order to transfer the MTP2-User signaling 
> > > messages to
> > > >   and from an Application Server (e.g. MGC) where the peer MTP2-
> > > >   User protocol layer exists.
> > > 
> > > This is a relly awkward paragraph.  I think you mean 
> "interworking of
> > > SS7 transport functions with IP transport functions".  Change 
> > > "in order
> > > to" to "to", change "the MTP2-User" to "MTP2-User"; change 
> > > "to and from"
> > > to "with"; change "Application Server (e.g., MGC) where the peer
> > > MTP2-User protocol layer exists" to "MTP2-User at an 
> > > Application Server
> > > (e.g., MGC)."
> > > 
> > > >   The value assigned by IANA for the Payload Protocol 
> > > Identifier in the
> > > >   SCTP Payload Data chunk is ô2ö
> > > 
> > > Non-ascii text at end of sentence.
> > > 
> > > Section 4.5:
> > > 
> > > >   This protocol is used between SS7 Signaling Points 
> > > employing the MTP
> > > >   Level 3 protocol. The SS7 Signaling Points may also 
> > > employ standard
> > > >   SS7 links using the SS7 MTP Layer 2 to provide transport 
> > > of MTP Layer
> > > >   3 signaling messages.
> > > 
> > > Word usage: change "employing" to "using" and "employ" to "use".
> > > 
> > > There is no such thing as "MTP Layer 2" or "MTP Layer 3"; 
> in SS7 it is
> > > referred to as "MTP Level 2" and "MTP Level 3".
> > > 
> > > >   Both configurations: intercommunication of SS7 and IP 
> with SG and
> > > >   communication between ISEPs are possible.
> > > 
> > > There is no such word as "intercommunication".  "Communication" is
> > > already between two things ("inter").
> > > 
> > > Besides, M2PA only provides for communication between 
> SPs.  There is
> > > no need for the two diagrams for M2PA.  There is no 
> > > distinction between
> > > the first diagram and the right-hand side of the second 
> diagram for
> > > M2PA.
> > > 
> > > >   These figures are only an example. Other configurations 
> > > are possible.
> > > >   For example, IPSPs without traditional SS7 links could use the
> > > >   protocol layers MTP3/M2PA/SCTP/IP to route SS7 messages 
> > > in a network
> > > >   with all IP links.
> > > >
> > > >   Another example is that two SGs could be connected over 
> > > an IP network
> > > >   to form an SG mated pair similar to the way STPs are 
> > > provisioned in
> > > >   traditional SS7 networks.
> > > 
> > > These statements sound foolish.  M2PA is really just a 
> different sort
> > > of SS7 link.  Plesae remove.
> > > 
> > > >   The value assigned by IANA for the Payload Protocol 
> > > Identifier in the
> > > >   SCTP Payload Data chunk is TBD
> > > 
> > > Does not M2PA have a PPI yet?
> > > 
> > > >      b. M2PA: SG-IPSP connection is an SS7 link.
> > > >         M2UA: SG-MGC connection is not an SS7 link. It is an
> > > >               extension of MTP to a remote entity.
> > > 
> > > There is nothing stopping M2UA from being used in a peer-to-peer
> > > fashion.
> > > 
> > > >      c. M2PA: SG is an SS7 node with a point code.
> > > >         M2UA: SG is not an SS7 node and has no point code.
> > > 
> > > This is not necessarily a correct statement.  There is nothing
> > > restricting an M2UA SG from having a point code or being an
> > > SS7 node in its own right.
> > > 
> > > >      d. M2PA: SG can have upper SS7 layers, e.g., SCCP.
> > > >         M2UA: SG does not have upper SS7 layers since it 
> > > has no MTP3.
> > > 
> > > Same is true here.  The above statement is false.
> > > 
> > > >      e. M2PA: relies on MTP3 for management procedures.
> > > >         M2UA: uses M2UA management procedures.
> > > 
> > > If need explanation of the absurdity of this statement, consider
> > > this SG:
> > >                      ____
> > >                     |
> > >                     |
> > >                 +------+
> > >                 | M2UA |
> > >                 +------+
> > >                 | M2PA |
> > >                 +------+
> > >                     |
> > >              _______|
> > > 
> > > Which is referred to in the M2UA document.
> > > 
> > > This SG is an M2PA SG, but it has no point code, has no 
> local MTP3,
> > > has no local upper layers (SCCP), and relies on M2UA for 
> management
> > > procedures.
> > > 
> > > Section 4.6:
> > > 
> > > >      - Interconnection of SS7 and IP nodes
> > > 
> > > Grammar: replace "Interconection" with "Connection".
> > > 
> > > 
> > > >   It works using the client-server philosophy. ...
> > > 
> > > This might be ok in German, but is awkward in English.  Reword as:
> > > 
> > > <   M3UA uses a client-server architecture.
> > > 
> > > >                                            ... ISEP is 
> > > recommended to
> > > >   be client when talking with a SG.
> > > 
> > > <   It is recommended that the ISEP act as the client and initiate
> > > <   SCTP associations with the SG.
> > > 
> > > >                                 ... The reserved port by 
> > > IANA is 2905
> > > >   to listen to possible client connections.
> > > 
> > > <   The port reserved by IANA is port number 2905.  This 
> is the port
> > > <   upon which the SG should listen for client connections.
> > > 
> > > >   The assigned payload protocol identifier for the SCTP 
> > > DATA chunks is
> > > >   ô3ö.
> > > 
> > > Non-ascii text at end of sentence.
> > > 
> > > 
> > > Section 4.7:
> > > 
> > > >                 ... SUA can support only non-call 
> related signaling.
> > > 
> > > Almost all SUA signalling is call related.  I think you mean
> > > non-call-associated signalling; but, then, *all* SS7 signalling is
> > > non-call-associated signalling.  Please strike this sentence.
> > > 
> > > >   SUA does not pose stringent timing constraints on 
> SCTP due to the
> > > >   fact that SUA applications have broad timing requirement 
> > > (from 10 of
> > > >   seconds to hours) which the applications guard 
> themselves and the
> > > >   timing supervision of the application is end-to-end, 
> not hop-by-
> > > >   hop(as with ISUP).
> > > 
> > > This statement is false and misleading.  SCCP, 
> particularly for the
> > > cited applications: MAP, INAP, SMS, BSSAP, RANAP, has 
> very stringent
> > > timing requirements, often more stringent than ISUP due 
> to the fact
> > > that these represent transaction which are made during ISUP 
> > > call setup.
> > > 
> > > Tens of seconds to hours is false; and atypical.  
> Transaction timers
> > > an performance requirements for INAP, for example, are 1 
> > > second maximum
> > > with 100 ms nominal query/response times.  Please remove this 
> > > paragraph.
> > > 
> > > >      - Interconnection of SS7 and IP
> > > 
> > > Change "Interconnection" to "Connection".
> > > 
> > > >   connections. The payload protocol identifier for the SCTP 
> > > DATA chunks
> > > >   is ô4ö.
> > > 
> > > Non-ascii text.
> > > 
> > > 
> > > General:
> > > 
> > > There is a mixture of "signalling" (two l's 
> British/Canadian spelling)
> > > and "signaling" (American spelling).  As the title 
> contains signalling
> > > with two l's, please use signalling with two l's throughout.
> > > 
> > > --brian
> > > 
> > > -- 
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > 
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 15:46:15 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA15319
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 15:46:15 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1PKtQX29653
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 15:55:26 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PKtDp29646;
	Tue, 25 Feb 2003 15:55:13 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PKsVp29612
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 15:54:31 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA15002
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 15:44:49 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:YY4MG5iOIuQryhdiRlsnuh3QNsw2Ldlt@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1PKmhD29070;
	Tue, 25 Feb 2003 13:48:43 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1PKmhT28411;
	Tue, 25 Feb 2003 13:48:43 -0700
Date: Tue, 25 Feb 2003 13:48:42 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tolga Asveren <tolga_asveren@yahoo.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030225134842.A27461@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030225075018.B16811@openss7.org> <20030225201522.39283.qmail@web12805.mail.yahoo.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20030225201522.39283.qmail@web12805.mail.yahoo.com>; from tolga_asveren@yahoo.com on Tue, Feb 25, 2003 at 12:15:22PM -0800
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Tolga,

Tolga Asveren wrote:                                                                 (Tue, 25 Feb 2003 12:15:22)
> Brian,
> 
> [..snip..]
> 
> > For example n=5. SG decides to start traffic with
> >the 2nd ACTIVE ASP. 
> >But
> > according to RFC ASPs will wait for the 5th ACTIVE
> >ASP before 
> >starting
> > traffic. In my opinion, this is a problem.
> 
> There is no problem.  The ASPs should wait for the 5th
> ASP to be active 
> before
> starting traffic regardless of when the SG starts
> traffic.
> [TOLGA]O.K. finally there we are, the source of the
> disagreement. IMO, it is not a good idea to open/close
> the traffic path between ASP/SGP unidirectionaly. 

Nowhere does it say that the traffic path need be
opened or closed unidirectionally.  The RFC talks about
the "start" of traffic, not the establishment of
unidirectional flows.

> And secondly, if you prefer SG to decide dynamically
> to start traffic, why do you want ASPs to decide for
> this statically by interpreting n as mandatory?

The SG is neither required nor prohibited by the RFC to
decide dynamically.  Nor is the ASP required to decide
statically.

The RFC permits the SG to do what it needs to do with
regards to the SS7 network because of the "MAY".

The RFC recommends that the ASPs wait until "n" because
they SHOULD.

I explained the rationale behind both already on this
thread.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 16:00:17 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA16109
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 16:00:17 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1PL9Se30987
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 16:09:28 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PL9Ep30975;
	Tue, 25 Feb 2003 16:09:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PL8Zp30940
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 16:08:35 -0500
Received: from web12803.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id PAA15887
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 15:58:53 -0500 (EST)
Message-ID: <20030225210247.88118.qmail@web12803.mail.yahoo.com>
Received: from [62.29.65.115] by web12803.mail.yahoo.com via HTTP; Tue, 25 Feb 2003 13:02:47 PST
Date: Tue, 25 Feb 2003 13:02:47 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
In-Reply-To: <20030225134842.A27461@openss7.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Brian,
--- "Brian F. G. Bidulock" <bidulock@openss7.org>
wrote:
> Tolga,
> 
> Tolga Asveren wrote:                                
>                                 (Tue, 25 Feb 2003
> 12:15:22)
> > Brian,
> > 
> > [..snip..]
> > 
> > > For example n=5. SG decides to start traffic
> with
> > >the 2nd ACTIVE ASP. 
> > >But
> > > according to RFC ASPs will wait for the 5th
> ACTIVE
> > >ASP before 
> > >starting
> > > traffic. In my opinion, this is a problem.
> > 
> > There is no problem.  The ASPs should wait for the
> 5th
> > ASP to be active 
> > before
> > starting traffic regardless of when the SG starts
> > traffic.
> > [TOLGA]O.K. finally there we are, the source of
> the
> > disagreement. IMO, it is not a good idea to
> open/close
> > the traffic path between ASP/SGP unidirectionaly. 
> 
> Nowhere does it say that the traffic path need be
> opened or closed unidirectionally.  The RFC talks
> about
> the "start" of traffic, not the establishment of
> unidirectional flows.
[TOLGA]IMO, the path between ASP/SGP will be opened
unidirectionslly, if ASPs will wait for n of them to
be ACTIVE whereas SG decides to not withhold traffic
when number of ACTIVE ASPs is less than n.
> 
> > And secondly, if you prefer SG to decide
> dynamically
> > to start traffic, why do you want ASPs to decide
> for
> > this statically by interpreting n as mandatory?
> 
> The SG is neither required nor prohibited by the RFC
> to
> decide dynamically.  Nor is the ASP required to
> decide
> statically.
[TOLGA]For SG part I agree.But for ASP I disagree, RFC
says that ASPs SHOULD not start traffic before n of
them are ACTIVE, and this n is a static attribute. If
ASPs rely on NTFY for this, starting traffic will be
decided by SG and will be decided dynamically (or
better said by SG, which can base this decision also
to the configured value n statically, this IMO is
totally responsibility of SG)
> 
> The RFC permits the SG to do what it needs to do
> with
> regards to the SS7 network because of the "MAY".
[TOLGA]The MAY is about withholding traffic to ASP
destination. I don't interprete it directly related
with interaction of SG with SS7 network.
> 
> The RFC recommends that the ASPs wait until "n"
> because
> they SHOULD.
[TOLGA]So it is static according to RFC.
> 
> I explained the rationale behind both already on
> this
> thread.
> 
> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/


__________________________________________________
Do you Yahoo!?
Yahoo! Tax Center - forms, calculators, tips, more
http://taxes.yahoo.com/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 16:01:08 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA16239
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 16:01:08 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1PLAJW31079
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 16:10:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PLA7p31051;
	Tue, 25 Feb 2003 16:10:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PL9Wp31000
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 16:09:32 -0500
Received: from web12806.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id PAA16044
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 15:59:51 -0500 (EST)
Message-ID: <20030225210320.55689.qmail@web12806.mail.yahoo.com>
Received: from [62.29.65.115] by web12806.mail.yahoo.com via HTTP; Tue, 25 Feb 2003 13:03:20 PST
Date: Tue, 25 Feb 2003 13:03:20 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
In-Reply-To: <20030225134842.A27461@openss7.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Brian,
--- "Brian F. G. Bidulock" <bidulock@openss7.org>
wrote:
> Tolga,
> 
> Tolga Asveren wrote:                                
>                                 (Tue, 25 Feb 2003
> 12:15:22)
> > Brian,
> > 
> > [..snip..]
> > 
> > > For example n=5. SG decides to start traffic
> with
> > >the 2nd ACTIVE ASP. 
> > >But
> > > according to RFC ASPs will wait for the 5th
> ACTIVE
> > >ASP before 
> > >starting
> > > traffic. In my opinion, this is a problem.
> > 
> > There is no problem.  The ASPs should wait for the
> 5th
> > ASP to be active 
> > before
> > starting traffic regardless of when the SG starts
> > traffic.
> > [TOLGA]O.K. finally there we are, the source of
> the
> > disagreement. IMO, it is not a good idea to
> open/close
> > the traffic path between ASP/SGP unidirectionaly. 
> 
> Nowhere does it say that the traffic path need be
> opened or closed unidirectionally.  The RFC talks
> about
> the "start" of traffic, not the establishment of
> unidirectional flows.
[TOLGA]IMO, the path between ASP/SGP will be opened
unidirectionslly, if ASPs will wait for n of them to
be ACTIVE whereas SG decides to not withhold traffic
when number of ACTIVE ASPs is less than n.
> 
> > And secondly, if you prefer SG to decide
> dynamically
> > to start traffic, why do you want ASPs to decide
> for
> > this statically by interpreting n as mandatory?
> 
> The SG is neither required nor prohibited by the RFC
> to
> decide dynamically.  Nor is the ASP required to
> decide
> statically.
[TOLGA]For SG part I agree.But for ASP I disagree, RFC
says that ASPs SHOULD not start traffic before n of
them are ACTIVE, and this n is a static attribute. If
ASPs rely on NTFY for this, starting traffic will be
decided by SG and will be decided dynamically (or
better said by SG, which can base this decision also
to the configured value n statically, this IMO is
totally responsibility of SG)
> 
> The RFC permits the SG to do what it needs to do
> with
> regards to the SS7 network because of the "MAY".
[TOLGA]The MAY is about withholding traffic to ASP
destination. I don't interprete it directly related
with interaction of SG with SS7 network.
> 
> The RFC recommends that the ASPs wait until "n"
> because
> they SHOULD.
[TOLGA]So it is static according to RFC.
> 
> I explained the rationale behind both already on
> this
> thread.
> 
> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/


__________________________________________________
Do you Yahoo!?
Yahoo! Tax Center - forms, calculators, tips, more
http://taxes.yahoo.com/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 16:01:21 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA16272
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 16:01:21 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1PLAWu31105
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 16:10:32 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PLALp31094;
	Tue, 25 Feb 2003 16:10:21 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PL91p30962
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 16:09:01 -0500
Received: from web12806.mail.yahoo.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id PAA15948
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 15:59:19 -0500 (EST)
Message-ID: <20030225210220.55173.qmail@web12806.mail.yahoo.com>
Received: from [62.29.65.115] by web12806.mail.yahoo.com via HTTP; Tue, 25 Feb 2003 13:02:20 PST
Date: Tue, 25 Feb 2003 13:02:20 -0800 (PST)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
In-Reply-To: <20030225134842.A27461@openss7.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Brian,
--- "Brian F. G. Bidulock" <bidulock@openss7.org>
wrote:
> Tolga,
> 
> Tolga Asveren wrote:                                
>                                 (Tue, 25 Feb 2003
> 12:15:22)
> > Brian,
> > 
> > [..snip..]
> > 
> > > For example n=5. SG decides to start traffic
> with
> > >the 2nd ACTIVE ASP. 
> > >But
> > > according to RFC ASPs will wait for the 5th
> ACTIVE
> > >ASP before 
> > >starting
> > > traffic. In my opinion, this is a problem.
> > 
> > There is no problem.  The ASPs should wait for the
> 5th
> > ASP to be active 
> > before
> > starting traffic regardless of when the SG starts
> > traffic.
> > [TOLGA]O.K. finally there we are, the source of
> the
> > disagreement. IMO, it is not a good idea to
> open/close
> > the traffic path between ASP/SGP unidirectionaly. 
> 
> Nowhere does it say that the traffic path need be
> opened or closed unidirectionally.  The RFC talks
> about
> the "start" of traffic, not the establishment of
> unidirectional flows.
[TOLGA]IMO, the path between ASP/SGP will be opened
unidirectionslly, if ASPs will wait for n of them to
be ACTIVE whereas SG decides to not withhold traffic
when number of ACTIVE ASPs is less than n.
> 
> > And secondly, if you prefer SG to decide
> dynamically
> > to start traffic, why do you want ASPs to decide
> for
> > this statically by interpreting n as mandatory?
> 
> The SG is neither required nor prohibited by the RFC
> to
> decide dynamically.  Nor is the ASP required to
> decide
> statically.
[TOLGA]For SG part I agree.But for ASP I disagree, RFC
says that ASPs SHOULD not start traffic before n of
them are ACTIVE, and this n is a static attribute. If
ASPs rely on NTFY for this, starting traffic will be
decided by SG and will be decided dynamically (or
better said by SG, which can base this decision also
to the configured value n statically, this IMO is
totally responsibility of SG)
> 
> The RFC permits the SG to do what it needs to do
> with
> regards to the SS7 network because of the "MAY".
[TOLGA]The MAY is about withholding traffic to ASP
destination. I don't interprete it directly related
with interaction of SG with SS7 network.
> 
> The RFC recommends that the ASPs wait until "n"
> because
> they SHOULD.
[TOLGA]So it is static according to RFC.
> 
> I explained the rationale behind both already on
> this
> thread.
> 
> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/


__________________________________________________
Do you Yahoo!?
Yahoo! Tax Center - forms, calculators, tips, more
http://taxes.yahoo.com/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Tue Feb 25 16:33:14 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA18399
	for <sigtran-archive@odin.ietf.org>; Tue, 25 Feb 2003 16:33:14 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1PLgQr00808
	for sigtran-archive@odin.ietf.org; Tue, 25 Feb 2003 16:42:26 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PLg3p00757;
	Tue, 25 Feb 2003 16:42:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1PLeEp00683
	for <sigtran@optimus.ietf.org>; Tue, 25 Feb 2003 16:40:14 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA18301
	for <sigtran@ietf.org>; Tue, 25 Feb 2003 16:30:30 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:LyhMUUUZh8YNs1k3PO1yqLs6PpWhCBDt@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1PLYPD29878;
	Tue, 25 Feb 2003 14:34:25 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1PLYPK30263;
	Tue, 25 Feb 2003 14:34:25 -0700
Date: Tue, 25 Feb 2003 14:34:24 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tolga Asveren <tolga_asveren@yahoo.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030225143424.A29556@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030225134842.A27461@openss7.org> <20030225210220.55173.qmail@web12806.mail.yahoo.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20030225210220.55173.qmail@web12806.mail.yahoo.com>; from tolga_asveren@yahoo.com on Tue, Feb 25, 2003 at 01:02:20PM -0800
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Tolga,

Tolga Asveren wrote:                                                                 (Tue, 25 Feb 2003 13:02:20)
> Brian,
> --- "Brian F. G. Bidulock" <bidulock@openss7.org>
> wrote:
> > Tolga,
> > 
> > Tolga Asveren wrote:                                
> >                                 (Tue, 25 Feb 2003
> > 12:15:22)
> > > Brian,
> > > 
> > > [..snip..]
> > > 
> > > > For example n=5. SG decides to start traffic
> > with
> > > >the 2nd ACTIVE ASP. 
> > > >But
> > > > according to RFC ASPs will wait for the 5th
> > ACTIVE
> > > >ASP before 
> > > >starting
> > > > traffic. In my opinion, this is a problem.
> > > 
> > > There is no problem.  The ASPs should wait for the
> > 5th
> > > ASP to be active 
> > > before
> > > starting traffic regardless of when the SG starts
> > > traffic.
> > > [TOLGA]O.K. finally there we are, the source of
> > the
> > > disagreement. IMO, it is not a good idea to
> > open/close
> > > the traffic path between ASP/SGP unidirectionaly. 
> > 
> > Nowhere does it say that the traffic path need be
> > opened or closed unidirectionally.  The RFC talks
> > about
> > the "start" of traffic, not the establishment of
> > unidirectional flows.
> [TOLGA]IMO, the path between ASP/SGP will be opened
> unidirectionslly, if ASPs will wait for n of them to
> be ACTIVE whereas SG decides to not withhold traffic
> when number of ACTIVE ASPs is less than n.

| For the n+k redundancy case, ASPs which are in that AS should
| coordianate amoung themselves the number of active ASPs in the AS, and
| should start sending traffic only after n ASPs are active.

This does not say that the ASP should not respond to traffic
delivered to it by the SG: it says that the ASP should not
"start" sending traffic.

It is just saying that the ASPs should coordinate going into
service and shouldn't start sending traffic until they are
capable of handling the expected load (and they shouldn't).

> > 
> > > And secondly, if you prefer SG to decide
> > dynamically
> > > to start traffic, why do you want ASPs to decide
> > for
> > > this statically by interpreting n as mandatory?
> > 
> > The SG is neither required nor prohibited by the RFC
> > to
> > decide dynamically.  Nor is the ASP required to
> > decide
> > statically.
> [TOLGA]For SG part I agree.But for ASP I disagree, RFC
> says that ASPs SHOULD not start traffic before n of
> them are ACTIVE, and this n is a static attribute. If
> ASPs rely on NTFY for this, starting traffic will be
> decided by SG and will be decided dynamically (or
> better said by SG, which can base this decision also
> to the configured value n statically, this IMO is
> totally responsibility of SG)

You are making many assuptions, most of which are not valid.
The SG does not "start" traffic because there is no stimulus
at the SG: the stimulus is either at the ASP or in the SS7
network.  All the SG can do is decide to "direct" traffic.
That does not mean that traffic actually exists, or that it
will exist any time in the near future after.

This is not a switch that you throw and current starts flowing.


> > 
> > The RFC permits the SG to do what it needs to do
> > with
> > regards to the SS7 network because of the "MAY".
> [TOLGA]The MAY is about withholding traffic to ASP
> destination. I don't interprete it directly related
> with interaction of SG with SS7 network.

But it is directly related to the SS7 network and that was
the reason for the MAY.  This is an AS which is newly active:
not an AS recovering from AS-PENDING.  Regardless of the
granularity of the RK, the SG has performed necessary management
functions towards the SS7 network with regards to the AS.
This could be UPU response, TFC response, or even TFP response
if the SG is acting as STP.  Directing traffic back to the
ASP for the AS requires that something be done to undo whatever
was told to the SS7 network.

As the SG's primary purpose in life is to transer traffic between
ASPs and the SS7 network, what did you think the SG was going to
do to "direct" traffic if it is not going to interact with the
SS7 network to do so?


> > 
> > The RFC recommends that the ASPs wait until "n"
> > because
> > they SHOULD.
> [TOLGA]So it is static according to RFC.

Where does it say static?  It says "coordinate" which sounds
pretty dynamic to me.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 00:53:22 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA29765
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 00:53:22 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1Q62ja32429
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 01:02:45 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1Q62Pp32421;
	Wed, 26 Feb 2003 01:02:27 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1Q610p32355
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 01:01:00 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA29737
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 00:51:05 -0500 (EST)
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF4D09FADD.6A771A15-ONC2256CD9.001E8449@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Wed, 26 Feb 2003 07:51:30 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/26/2003
 07:51:34 AM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,
> Tolga,
>
> Tolga Asveren wrote:
>                                 (Tue, 25 Feb 2003
> 13:02:20)
> > Brian,
> > --- "Brian F. G. Bidulock" <bidulock@openss7.org>
> > wrote:
> > > Tolga,
> > >
> > > Tolga Asveren wrote:
>
> > >                                 (Tue, 25 Feb
> 2003
> > > 12:15:22)
> > > > Brian,
> > > >
> > > > [..snip..]
> > > >
> > > > > For example n=5. SG decides to start traffic
> > > with
> > > > >the 2nd ACTIVE ASP.
> > > > >But
> > > > > according to RFC ASPs will wait for the 5th
> > > ACTIVE
> > > > >ASP before
> > > > >starting
> > > > > traffic. In my opinion, this is a problem.
> > > >
> > > > There is no problem.  The ASPs should wait for
> the
> > > 5th
> > > > ASP to be active
> > > > before
> > > > starting traffic regardless of when the SG
> starts
> > > > traffic.
> > > > [TOLGA]O.K. finally there we are, the source
> of
> > > the
> > > > disagreement. IMO, it is not a good idea to
> > > open/close
> > > > the traffic path between ASP/SGP
> unidirectionaly.
> > >
> > > Nowhere does it say that the traffic path need
> be
> > > opened or closed unidirectionally.  The RFC
> talks
> > > about
> > > the "start" of traffic, not the establishment of
> > > unidirectional flows.
> > [TOLGA]IMO, the path between ASP/SGP will be
> opened
> > unidirectionslly, if ASPs will wait for n of them
> to
> > be ACTIVE whereas SG decides to not withhold
> traffic
> > when number of ACTIVE ASPs is less than n.
>
> | For the n+k redundancy case, ASPs which are in
> that AS should
> | coordianate amoung themselves the number of active
> ASPs in the AS, and
> | should start sending traffic only after n ASPs are
> active.
>
> This does not say that the ASP should not respond to
> traffic
> delivered to it by the SG: it says that the ASP
> should not
> "start" sending traffic.
>
> It is just saying that the ASPs should coordinate
> going into
> service and shouldn't start sending traffic until
> they are
> capable of handling the expected load (and they
> shouldn't).
[TOLGA]What is the reason behind allowing ASP to respond to messages from
SG but not starting traffic? How does ASP-M3UA know whether a message
coming from user parts is a response or newly initiated traffic(whatever
the definition of newly initiated traffic is)? If SG decides that there are
enough number of ACTIVE ASPs, why should we prevent ASPs to initiate
traffic?
BTW, I proposed that text to the RFC and I never thougth that somebody can
interprete the word "start" like you do. If you are really uncomfortable
and think that the sentence is not clear, will changing "traffic" to
"messages" make the ambiguity disappear?
>
> > >
> > > > And secondly, if you prefer SG to decide
> > > dynamically
> > > > to start traffic, why do you want ASPs to
> decide
> > > for
> > > > this statically by interpreting n as
> mandatory?
> > >
> > > The SG is neither required nor prohibited by the
> RFC
> > > to
> > > decide dynamically.  Nor is the ASP required to
> > > decide
> > > statically.
> > [TOLGA]For SG part I agree.But for ASP I disagree,
> RFC
> > says that ASPs SHOULD not start traffic before n
> of
> > them are ACTIVE, and this n is a static attribute.
> If
> > ASPs rely on NTFY for this, starting traffic will
> be
> > decided by SG and will be decided dynamically (or
> > better said by SG, which can base this decision
> also
> > to the configured value n statically, this IMO is
> > totally responsibility of SG)
>
> You are making many assuptions, most of which are
> not valid.
> The SG does not "start" traffic because there is no
> stimulus
> at the SG: the stimulus is either at the ASP or in
> the SS7
> network.  All the SG can do is decide to "direct"
> traffic.
> That does not mean that traffic actually exists, or
> that it
> will exist any time in the near future after.
>
> This is not a switch that you throw and current
> starts flowing.
[TOLGA]I think several times I explained what I mean with "SG starting
traffic", but O.K. I will use "directing", it is more clear, but doesn't
change too much from this discussion point of view IMO.
>
>
> > >
> > > The RFC permits the SG to do what it needs to do
> > > with
> > > regards to the SS7 network because of the "MAY".
> > [TOLGA]The MAY is about withholding traffic to ASP
> > destination. I don't interprete it directly
> related
> > with interaction of SG with SS7 network.
>
> But it is directly related to the SS7 network and
> that was
> the reason for the MAY.  This is an AS which is
> newly active:
> not an AS recovering from AS-PENDING.  Regardless of
> the
> granularity of the RK, the SG has performed
> necessary management
> functions towards the SS7 network with regards to
> the AS.
> This could be UPU response, TFC response, or even
> TFP response
> if the SG is acting as STP.  Directing traffic back
> to the
> ASP for the AS requires that something be done to
> undo whatever
> was told to the SS7 network.
[TOLGA] Of course SG will execute necessary SS7 procedures. And if you
interprete it as undoing something told to SS7 network, I think SG should
do this, whenever it decides that there are enough number of ACTIVE ASPs.
>
> As the SG's primary purpose in life is to transer
> traffic between
> ASPs and the SS7 network, what did you think the SG
> was going to
> do to "direct" traffic if it is not going to
> interact with the
> SS7 network to do so?
[TOLGA]I didn't say that SG won't interact with SS7 network. But the word
"MAY" about withholding traffic is mainly related for communication between
SGP/ASP.
BTW, before SG decides that AS can be considered ACTIVE, it won't execute
necessary SS7 procedures. So, where does the traffic, SG needs to withhold
comes from? So, I would think, this withholding traffic will be meaningfull
only for PENDING case(?).
>
>
> > >
> > > The RFC recommends that the ASPs wait until "n"
> > > because
> > > they SHOULD.
> > [TOLGA]So it is static according to RFC.
>
> Where does it say static?  It says "coordinate"
> which sounds
> pretty dynamic to me.
[TOLGA]Because it relies on the statically configured value n. IMO, only SG
can decide dynamically, when to open the communication path between
ASP/SGP. And for that we need to make ASPs to decide based on received NTFY
from SG, not to count number of ACTIVE ASPs by themselves.

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 01:02:29 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA29906
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 01:02:29 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1Q6BpX01055
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 01:11:51 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1Q6Bcp01044;
	Wed, 26 Feb 2003 01:11:38 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1Q6Ahp01015
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 01:10:43 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA29882
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 01:00:49 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1Q64iqe019145
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 00:04:44 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h1Q64ZF19012
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 00:04:35 -0600 (CST)
Date: Wed, 26 Feb 2003 00:04:35 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: sigtran@ietf.org
Subject: RE: [Sigtran] Re: M3UA Notify message
In-Reply-To: <Pine.SOL.4.10.10302250745130.3454-100000@sun3271.ssd.usa.alcatel.com>
Message-ID: <Pine.SOL.4.10.10302252359470.12860-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Javier,


               SG                        +------+  AS1
 normal SS7  +-----+             +-------|      |
-----------> |     |------------/--------| ASP1 |      
 traffic(T)  |     |           /         +------+        'n' ASPs
             | SGP1|          /          +------+
             |     |---------/-----------| ASP2 |
             +-----+        /     +------|      |
               *           /     /       +------+
               *          /     /        +------+  
               **************************| ASP3 |  spare 'k' ASP
                        /     /          +------+
                       /     /              *
             +---- +  /     /               *
             |     |-+     /                *
             |     |      /                 *
             |SGP2 |-----+                  *
             |     |                        *
             +-----+                        *
               *                            *
               *                            *
               ******************************
               

At SGP1 SLS-ASP table
--------------------
SLS             ASP
---             ---
X               ASP1
Y               ASP2
-               ASP3 (spare ASP)

AT AS1 SLS-ASP table
-------------------
SLS             ASP        PRIMARY-RT         SECND-RT
---             ---        ----------         --------
A               ASP1        SGP1               SGP2
B               ASP2        SGP1               SGP2


Initially AS1 state at SGP is AS-DOWN.

SGP requires 'n' ASPs to handle the normal SS7 Traffic 'T', this can be
called as 'best traffic'.AS1 state is changed to AS-ACTIVE when 'n' ASPs
have become active.

Suppose ASP1 is failed so 'n' becomes '1'.

Now the question.

1) If AS1 state is changed from ACTIVE to PENDING(as per IG)
   then no DATA messages will be sent to AS.
   
   IMO, we should keep the AS state as ACTIVE.
   
   According to q.704 section 3.8.2 we should reroute the 
   traffic to ASP2 provided congestion has not occured.

   Just reroute the SLS(X,Y) to available ASP1 , i beleive this is
   what we follow in legacy SS7 network.
   
   When ASP2 transmission buffer reaches congestion level then
   we can send NTFY to spare 'k' ASP and change the AS1 state to
   PENDING along with timer running.
   
 2)What will happen when the traffic reaches from 'T' to 'T+1'?
 
   At SG 'n' ASPs are not sufficient to handle this 'worst case'.
   So SG is required to send NTFY to spare 'k' ASP.
   But RFC and IG ignores to handle this case.We should enhance IG to
   support this case.
   
 IG section 3.15 misguides and should be re-writen to handle the above two 
 points.
 
--Samuel

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 02:19:32 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA16072
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 02:19:32 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1Q7SsK15851
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 02:28:54 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1Q7Sbp15828;
	Wed, 26 Feb 2003 02:28:38 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1Q7Rtp15777
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 02:27:55 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA14461
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 02:18:00 -0500 (EST)
Subject: RE: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF2A12AF66.13F56D33-ONC2256CD9.00277FCF@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Wed, 26 Feb 2003 09:18:23 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/26/2003
 09:18:27 AM
MIME-Version: 1.0
Content-type: text/plain; charset=Windows-1254
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h1Q7Rtp15778
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit


Samuel,
--- "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com> wrote:
>
> Javier,
>
>
>                SG                        +------+
> AS1
>  normal SS7  +-----+             +-------|      |
> -----------> |     |------------/--------| ASP1 |
>
>  traffic(T)  |     |           /         +------+
>     'n' ASPs
>              | SGP1|          /          +------+
>              |     |---------/-----------| ASP2 |
>              +-----+        /     +------|      |
>                *           /     /       +------+
>                *          /     /        +------+
>                **************************| ASP3 |
> spare 'k' ASP
>                         /     /          +------+
>                        /     /              *
>              +---- +  /     /               *
>              |     |-+     /                *
>              |     |      /                 *
>              |SGP2 |-----+                  *
>              |     |                        *
>              +-----+                        *
>                *                            *
>                *                            *
>                ******************************
>
>
> At SGP1 SLS-ASP table
> --------------------
> SLS             ASP
> ---             ---
> X               ASP1
> Y               ASP2
> -               ASP3 (spare ASP)
>
> AT AS1 SLS-ASP table
> -------------------
> SLS             ASP        PRIMARY-RT
> SECND-RT
> ---             ---        ----------
> --------
> A               ASP1        SGP1               SGP2
> B               ASP2        SGP1               SGP2
>
>
> Initially AS1 state at SGP is AS-DOWN.
>
> SGP requires 'n' ASPs to handle the normal SS7
> Traffic 'T', this can be
> called as 'best traffic'.AS1 state is changed to
> AS-ACTIVE when 'n' ASPs
> have become active.
>
> Suppose ASP1 is failed so 'n' becomes '1'.
>
> Now the question.
>
> 1) If AS1 state is changed from ACTIVE to PENDING(as
> per IG)
>    then no DATA messages will be sent to AS.
>
>    IMO, we should keep the AS state as ACTIVE.
>
>    According to q.704 section 3.8.2 we should
> reroute the
>    traffic to ASP2 provided congestion has not
> occured.
[TOLGA]IMO Q.704 can't be binding for this decision. The interface between
ASP/SGP (especially from SGP to ASP direction) is a MTP3/MTP3-User
interface, not a MTP3/MTP3 interface. But I agree that SG behaves like you
describe is a clever choice for most of the times. Still my opinion is to
leave the decision to SG. SGPs can inform ASPs about this decision with
NTFY. This logic can be applied also to other transitions, like the
transitions from ACTIVE to PENDING and from PENDING to ACTIVE. Does this
count as a solution ýn your opinion?
>
>    Just reroute the SLS(X,Y) to available ASP1 , i
> beleive this is
>    what we follow in legacy SS7 network.
>
>    When ASP2 transmission buffer reaches congestion
> level then
>    we can send NTFY to spare 'k' ASP and change the
> AS1 state to
>    PENDING along with timer running.
>
>  2)What will happen when the traffic reaches from
> 'T' to 'T+1'?
>
>    At SG 'n' ASPs are not sufficient to handle this
> 'worst case'.
>    So SG is required to send NTFY to spare 'k' ASP.
>    But RFC and IG ignores to handle this case.We
> should enhance IG to
>    support this case.
>
>  IG section 3.15 misguides and should be re-writen
> to handle the above two
>  points.

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 02:36:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA27165
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 02:36:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1Q7kKd17543
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 02:46:20 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1Q7k7p17533;
	Wed, 26 Feb 2003 02:46:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1Q7j3p17468
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 02:45:03 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA27052
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 02:35:09 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:3gJQYxyzxCYgu8egbBXY61zWkjrlZd2+@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1Q7d3D05853;
	Wed, 26 Feb 2003 00:39:03 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1Q7d2008854;
	Wed, 26 Feb 2003 00:39:02 -0700
Date: Wed, 26 Feb 2003 00:39:02 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030226003902.A8776@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OF2A12AF66.13F56D33-ONC2256CD9.00277FCF@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF2A12AF66.13F56D33-ONC2256CD9.00277FCF@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Wed, Feb 26, 2003 at 09:18:23AM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: 8bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

TOLGA,

TOLGA ASVEREN wrote:                                                                 (Wed, 26 Feb 2003 09:18:23)
> 
> Samuel,
> --- "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com> wrote:
> >
> > Javier,
> >
> >
> >                SG                        +------+
> > AS1
> >  normal SS7  +-----+             +-------|      |
> > -----------> |     |------------/--------| ASP1 |
> >
> >  traffic(T)  |     |           /         +------+
> >     'n' ASPs
> >              | SGP1|          /          +------+
> >              |     |---------/-----------| ASP2 |
> >              +-----+        /     +------|      |
> >                *           /     /       +------+
> >                *          /     /        +------+
> >                **************************| ASP3 |
> > spare 'k' ASP
> >                         /     /          +------+
> >                        /     /              *
> >              +---- +  /     /               *
> >              |     |-+     /                *
> >              |     |      /                 *
> >              |SGP2 |-----+                  *
> >              |     |                        *
> >              +-----+                        *
> >                *                            *
> >                *                            *
> >                ******************************
> >
> >
> > At SGP1 SLS-ASP table
> > --------------------
> > SLS             ASP
> > ---             ---
> > X               ASP1
> > Y               ASP2
> > -               ASP3 (spare ASP)
> >
> > AT AS1 SLS-ASP table
> > -------------------
> > SLS             ASP        PRIMARY-RT
> > SECND-RT
> > ---             ---        ----------
> > --------
> > A               ASP1        SGP1               SGP2
> > B               ASP2        SGP1               SGP2
> >
> >
> > Initially AS1 state at SGP is AS-DOWN.
> >
> > SGP requires 'n' ASPs to handle the normal SS7
> > Traffic 'T', this can be
> > called as 'best traffic'.AS1 state is changed to
> > AS-ACTIVE when 'n' ASPs
> > have become active.
> >
> > Suppose ASP1 is failed so 'n' becomes '1'.
> >
> > Now the question.
> >
> > 1) If AS1 state is changed from ACTIVE to PENDING(as
> > per IG)
> >    then no DATA messages will be sent to AS.
> >
> >    IMO, we should keep the AS state as ACTIVE.
> >
> >    According to q.704 section 3.8.2 we should
> > reroute the
> >    traffic to ASP2 provided congestion has not
> > occured.
> [TOLGA]IMO Q.704 can't be binding for this decision. The interface between
> ASP/SGP (especially from SGP to ASP direction) is a MTP3/MTP3-User
> interface, not a MTP3/MTP3 interface. But I agree that SG behaves like you
> describe is a clever choice for most of the times. Still my opinion is to
> leave the decision to SG. SGPs can inform ASPs about this decision with
> NTFY. This logic can be applied also to other transitions, like the
> transitions from ACTIVE to PENDING and from PENDING to ACTIVE. Does this
> count as a solution ýn your opinion?

The RFC already provides for the above.  It is IG 3.15 that does not.
Striking IG 3.15 is the best start.

--brian

> >
> >    Just reroute the SLS(X,Y) to available ASP1 , i
> > beleive this is
> >    what we follow in legacy SS7 network.
> >
> >    When ASP2 transmission buffer reaches congestion
> > level then
> >    we can send NTFY to spare 'k' ASP and change the
> > AS1 state to
> >    PENDING along with timer running.
> >
> >  2)What will happen when the traffic reaches from
> > 'T' to 'T+1'?
> >
> >    At SG 'n' ASPs are not sufficient to handle this
> > 'worst case'.
> >    So SG is required to send NTFY to spare 'k' ASP.
> >    But RFC and IG ignores to handle this case.We
> > should enhance IG to
> >    support this case.
> >
> >  IG section 3.15 misguides and should be re-writen
> > to handle the above two
> >  points.
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 02:59:09 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA27587
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 02:59:08 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1Q88WR19170
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 03:08:32 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1Q88Gp19152;
	Wed, 26 Feb 2003 03:08:16 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1Q87Ip18733
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 03:07:18 -0500
Received: from albatross.wise.edt.ericsson.se (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA27572
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 02:57:23 -0500 (EST)
Received: from esealnt613.al.sw.ericsson.se (esealnt613.al.sw.ericsson.se [153.88.254.72])
	by albatross.wise.edt.ericsson.se (8.12.1/8.12.1/WIREfire-1.4) with ESMTP id h1Q81HnS019281;
	Wed, 26 Feb 2003 09:01:17 +0100 (MET)
Received: from ESEALNT745.al.sw.ericsson.se ([153.88.251.5]) by esealnt613.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55)
	id F4Q8P642; Wed, 26 Feb 2003 09:01:16 +0100
Received: by ESEALNT745.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <DW9DF5HC>; Wed, 26 Feb 2003 09:01:16 +0100
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B659CD@EESTQNT105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>,
        TOLGA ASVEREN
	 <TOLGA.ASVEREN@o2.com.tr>,
        "'sjeyasee@ssd.usa.alcatel.com'"
	 <sjeyasee@ssd.usa.alcatel.com>
Cc: sigtran@ietf.org
Subject: RE: [Sigtran] Re: M3UA Notify message
Date: Wed, 26 Feb 2003 09:01:07 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h1Q87Ip18734
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Hi, 

There is no problem in striking sec 3.15 from the IG if everybody agrees. We just have to make an agreement about what is the best way of doing things. The IG should reflect that agreement. So I would focus on the agreement instead. Let's work as if section 3.15 did not exist anymore...

I also agree that an SG handling traffic with #ASPs<n might be an intelligent approach. Of course, all kind of alarms should show this situation and the best configuration should be recovered as soon as possible.

Cheers // Javier.

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: miércoles, 26 de febrero de 2003 8:39
> To: TOLGA ASVEREN
> Cc: sigtran@ietf.org
> Subject: Re: [Sigtran] Re: M3UA Notify message
> 
> 
> TOLGA,
> 
> TOLGA ASVEREN wrote:                                          
>                        (Wed, 26 Feb 2003 09:18:23)
> > 
> > Samuel,
> > --- "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com> wrote:
> > >
> > > Javier,
> > >
> > >
> > >                SG                        +------+
> > > AS1
> > >  normal SS7  +-----+             +-------|      |
> > > -----------> |     |------------/--------| ASP1 |
> > >
> > >  traffic(T)  |     |           /         +------+
> > >     'n' ASPs
> > >              | SGP1|          /          +------+
> > >              |     |---------/-----------| ASP2 |
> > >              +-----+        /     +------|      |
> > >                *           /     /       +------+
> > >                *          /     /        +------+
> > >                **************************| ASP3 |
> > > spare 'k' ASP
> > >                         /     /          +------+
> > >                        /     /              *
> > >              +---- +  /     /               *
> > >              |     |-+     /                *
> > >              |     |      /                 *
> > >              |SGP2 |-----+                  *
> > >              |     |                        *
> > >              +-----+                        *
> > >                *                            *
> > >                *                            *
> > >                ******************************
> > >
> > >
> > > At SGP1 SLS-ASP table
> > > --------------------
> > > SLS             ASP
> > > ---             ---
> > > X               ASP1
> > > Y               ASP2
> > > -               ASP3 (spare ASP)
> > >
> > > AT AS1 SLS-ASP table
> > > -------------------
> > > SLS             ASP        PRIMARY-RT
> > > SECND-RT
> > > ---             ---        ----------
> > > --------
> > > A               ASP1        SGP1               SGP2
> > > B               ASP2        SGP1               SGP2
> > >
> > >
> > > Initially AS1 state at SGP is AS-DOWN.
> > >
> > > SGP requires 'n' ASPs to handle the normal SS7
> > > Traffic 'T', this can be
> > > called as 'best traffic'.AS1 state is changed to
> > > AS-ACTIVE when 'n' ASPs
> > > have become active.
> > >
> > > Suppose ASP1 is failed so 'n' becomes '1'.
> > >
> > > Now the question.
> > >
> > > 1) If AS1 state is changed from ACTIVE to PENDING(as
> > > per IG)
> > >    then no DATA messages will be sent to AS.
> > >
> > >    IMO, we should keep the AS state as ACTIVE.
> > >
> > >    According to q.704 section 3.8.2 we should
> > > reroute the
> > >    traffic to ASP2 provided congestion has not
> > > occured.
> > [TOLGA]IMO Q.704 can't be binding for this decision. The 
> interface between
> > ASP/SGP (especially from SGP to ASP direction) is a MTP3/MTP3-User
> > interface, not a MTP3/MTP3 interface. But I agree that SG 
> behaves like you
> > describe is a clever choice for most of the times. Still my 
> opinion is to
> > leave the decision to SG. SGPs can inform ASPs about this 
> decision with
> > NTFY. This logic can be applied also to other transitions, like the
> > transitions from ACTIVE to PENDING and from PENDING to 
> ACTIVE. Does this
> > count as a solution ýn your opinion?
> 
> The RFC already provides for the above.  It is IG 3.15 that does not.
> Striking IG 3.15 is the best start.
> 
> --brian
> 
> > >
> > >    Just reroute the SLS(X,Y) to available ASP1 , i
> > > beleive this is
> > >    what we follow in legacy SS7 network.
> > >
> > >    When ASP2 transmission buffer reaches congestion
> > > level then
> > >    we can send NTFY to spare 'k' ASP and change the
> > > AS1 state to
> > >    PENDING along with timer running.
> > >
> > >  2)What will happen when the traffic reaches from
> > > 'T' to 'T+1'?
> > >
> > >    At SG 'n' ASPs are not sufficient to handle this
> > > 'worst case'.
> > >    So SG is required to send NTFY to spare 'k' ASP.
> > >    But RFC and IG ignores to handle this case.We
> > > should enhance IG to
> > >    support this case.
> > >
> > >  IG section 3.15 misguides and should be re-writen
> > > to handle the above two
> > >  points.
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 03:07:03 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA27763
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 03:07:03 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1Q8GR419482
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 03:16:27 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1Q8G9p19472;
	Wed, 26 Feb 2003 03:16:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1Q8Fvp19442
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 03:15:57 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA27750
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 03:05:56 -0500 (EST)
Subject: RE: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF54CDA0BE.51445E83-ONC2256CD9.002BEB06@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Wed, 26 Feb 2003 10:06:15 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/26/2003
 10:06:27 AM
MIME-Version: 1.0
Content-type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h1Q8Fvp19443
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit


Javier,

I also agree that there is no problem with removing IG 3.15. My proposal
for the solution is to make ASPs decide based on NTFY to start send/receive
DATA. If we have consensus about this strategy, we can go into the details
or we can consider other strategies.

     Tolga



Hi,

There is no problem in striking sec 3.15 from the IG if everybody agrees.
We just have to make an agreement about what is the best way of doing
things. The IG should reflect that agreement. So I would focus on the
agreement instead. Let's work as if section 3.15 did not exist anymore...

I also agree that an SG handling traffic with #ASPs<n might be an
intelligent approach. Of course, all kind of alarms should show this
situation and the best configuration should be recovered as soon as
possible.

Cheers // Javier.

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: miércoles, 26 de febrero de 2003 8:39
> To: TOLGA ASVEREN
> Cc: sigtran@ietf.org
> Subject: Re: [Sigtran] Re: M3UA Notify message
>
>
> TOLGA,
>
> TOLGA ASVEREN wrote:
>                        (Wed, 26 Feb 2003 09:18:23)
> >
> > Samuel,
> > --- "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com> wrote:
> > >
> > > Javier,
> > >
> > >
> > >                SG                        +------+
> > > AS1
> > >  normal SS7  +-----+             +-------|      |
> > > -----------> |     |------------/--------| ASP1 |
> > >
> > >  traffic(T)  |     |           /         +------+
> > >     'n' ASPs
> > >              | SGP1|          /          +------+
> > >              |     |---------/-----------| ASP2 |
> > >              +-----+        /     +------|      |
> > >                *           /     /       +------+
> > >                *          /     /        +------+
> > >                **************************| ASP3 |
> > > spare 'k' ASP
> > >                         /     /          +------+
> > >                        /     /              *
> > >              +---- +  /     /               *
> > >              |     |-+     /                *
> > >              |     |      /                 *
> > >              |SGP2 |-----+                  *
> > >              |     |                        *
> > >              +-----+                        *
> > >                *                            *
> > >                *                            *
> > >                ******************************
> > >
> > >
> > > At SGP1 SLS-ASP table
> > > --------------------
> > > SLS             ASP
> > > ---             ---
> > > X               ASP1
> > > Y               ASP2
> > > -               ASP3 (spare ASP)
> > >
> > > AT AS1 SLS-ASP table
> > > -------------------
> > > SLS             ASP        PRIMARY-RT
> > > SECND-RT
> > > ---             ---        ----------
> > > --------
> > > A               ASP1        SGP1               SGP2
> > > B               ASP2        SGP1               SGP2
> > >
> > >
> > > Initially AS1 state at SGP is AS-DOWN.
> > >
> > > SGP requires 'n' ASPs to handle the normal SS7
> > > Traffic 'T', this can be
> > > called as 'best traffic'.AS1 state is changed to
> > > AS-ACTIVE when 'n' ASPs
> > > have become active.
> > >
> > > Suppose ASP1 is failed so 'n' becomes '1'.
> > >
> > > Now the question.
> > >
> > > 1) If AS1 state is changed from ACTIVE to PENDING(as
> > > per IG)
> > >    then no DATA messages will be sent to AS.
> > >
> > >    IMO, we should keep the AS state as ACTIVE.
> > >
> > >    According to q.704 section 3.8.2 we should
> > > reroute the
> > >    traffic to ASP2 provided congestion has not
> > > occured.
> > [TOLGA]IMO Q.704 can't be binding for this decision. The
> interface between
> > ASP/SGP (especially from SGP to ASP direction) is a MTP3/MTP3-User
> > interface, not a MTP3/MTP3 interface. But I agree that SG
> behaves like you
> > describe is a clever choice for most of the times. Still my
> opinion is to
> > leave the decision to SG. SGPs can inform ASPs about this
> decision with
> > NTFY. This logic can be applied also to other transitions, like the
> > transitions from ACTIVE to PENDING and from PENDING to
> ACTIVE. Does this
> > count as a solution ýn your opinion?
>
> The RFC already provides for the above.  It is IG 3.15 that does not.
> Striking IG 3.15 is the best start.
>
> --brian
>
> > >
> > >    Just reroute the SLS(X,Y) to available ASP1 , i
> > > beleive this is
> > >    what we follow in legacy SS7 network.
> > >
> > >    When ASP2 transmission buffer reaches congestion
> > > level then
> > >    we can send NTFY to spare 'k' ASP and change the
> > > AS1 state to
> > >    PENDING along with timer running.
> > >
> > >  2)What will happen when the traffic reaches from
> > > 'T' to 'T+1'?
> > >
> > >    At SG 'n' ASPs are not sufficient to handle this
> > > 'worst case'.
> > >    So SG is required to send NTFY to spare 'k' ASP.
> > >    But RFC and IG ignores to handle this case.We
> > > should enhance IG to
> > >    support this case.
> > >
> > >  IG section 3.15 misguides and should be re-writen
> > > to handle the above two
> > >  points.
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
>
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 03:09:54 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA27857
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 03:09:54 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1Q8JI019589
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 03:19:18 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1Q8J7p19578;
	Wed, 26 Feb 2003 03:19:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1Q8Isp19561
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 03:18:54 -0500
Received: from penguin.wise.edt.ericsson.se (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA27811
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 03:08:59 -0500 (EST)
Received: from esealnt612.al.sw.ericsson.se (esealnt612.al.sw.ericsson.se [153.88.254.71])
	by penguin.wise.edt.ericsson.se (8.12.1/8.12.1/WIREfire-1.4) with ESMTP id h1Q8CrYe016003;
	Wed, 26 Feb 2003 09:12:53 +0100 (MET)
Received: from ESEALNT745.al.sw.ericsson.se ([153.88.251.5]) by esealnt612.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55)
	id F4SN1HD0; Wed, 26 Feb 2003 09:12:52 +0100
Received: by ESEALNT745.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <DW9DF5M1>; Wed, 26 Feb 2003 09:12:52 +0100
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B659CE@EESTQNT105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>
To: "'Samuel Dur D. Jeyaseelan'" <sjeyasee@ssd.usa.alcatel.com>,
        sigtran@ietf.org
Subject: RE: [Sigtran] Re: M3UA Notify message
Date: Wed, 26 Feb 2003 09:12:43 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id h1Q8Isp19562
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit

Hi Samuel, 

Reading again your proposed change #1. I would agree to change n as recommended in traffic mode while mandatory at start time, as I said before.
But it seems to me that you are routing inside a RK, which is not disallowed, but why don't you use separate RKs? I mean, the RK should be sufficient to base the routing decissions on the SG...

Cheers // Javier.

> -----Original Message-----
> From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> Sent: miércoles, 26 de febrero de 2003 7:05
> To: sigtran@ietf.org
> Subject: RE: [Sigtran] Re: M3UA Notify message
> 
> 
> 
> Javier,
> 
> 
>                SG                        +------+  AS1
>  normal SS7  +-----+             +-------|      |
> -----------> |     |------------/--------| ASP1 |      
>  traffic(T)  |     |           /         +------+        'n' ASPs
>              | SGP1|          /          +------+
>              |     |---------/-----------| ASP2 |
>              +-----+        /     +------|      |
>                *           /     /       +------+
>                *          /     /        +------+  
>                **************************| ASP3 |  spare 'k' ASP
>                         /     /          +------+
>                        /     /              *
>              +---- +  /     /               *
>              |     |-+     /                *
>              |     |      /                 *
>              |SGP2 |-----+                  *
>              |     |                        *
>              +-----+                        *
>                *                            *
>                *                            *
>                ******************************
>                
> 
> At SGP1 SLS-ASP table
> --------------------
> SLS             ASP
> ---             ---
> X               ASP1
> Y               ASP2
> -               ASP3 (spare ASP)
> 
> AT AS1 SLS-ASP table
> -------------------
> SLS             ASP        PRIMARY-RT         SECND-RT
> ---             ---        ----------         --------
> A               ASP1        SGP1               SGP2
> B               ASP2        SGP1               SGP2
> 
> 
> Initially AS1 state at SGP is AS-DOWN.
> 
> SGP requires 'n' ASPs to handle the normal SS7 Traffic 'T', 
> this can be
> called as 'best traffic'.AS1 state is changed to AS-ACTIVE 
> when 'n' ASPs
> have become active.
> 
> Suppose ASP1 is failed so 'n' becomes '1'.
> 
> Now the question.
> 
> 1) If AS1 state is changed from ACTIVE to PENDING(as per IG)
>    then no DATA messages will be sent to AS.
>    
>    IMO, we should keep the AS state as ACTIVE.
>    
>    According to q.704 section 3.8.2 we should reroute the 
>    traffic to ASP2 provided congestion has not occured.
> 
>    Just reroute the SLS(X,Y) to available ASP1 , i beleive this is
>    what we follow in legacy SS7 network.
>    
>    When ASP2 transmission buffer reaches congestion level then
>    we can send NTFY to spare 'k' ASP and change the AS1 state to
>    PENDING along with timer running.
>    
>  2)What will happen when the traffic reaches from 'T' to 'T+1'?
>  
>    At SG 'n' ASPs are not sufficient to handle this 'worst case'.
>    So SG is required to send NTFY to spare 'k' ASP.
>    But RFC and IG ignores to handle this case.We should enhance IG to
>    support this case.
>    
>  IG section 3.15 misguides and should be re-writen to handle 
> the above two 
>  points.
>  
> --Samuel
> 
>   Alcatel USA, Inc.                  Internet: 
> <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel 
> USA, Inc. *******
> 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 03:50:45 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA28411
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 03:50:45 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1Q909F21800
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 04:00:09 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1Q8xop21755;
	Wed, 26 Feb 2003 03:59:51 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1Q8wmp21725
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 03:58:48 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA28392
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 03:48:52 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1Q8qjqe003134;
	Wed, 26 Feb 2003 02:52:45 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h1Q8pvF01247;
	Wed, 26 Feb 2003 02:51:57 -0600 (CST)
Date: Wed, 26 Feb 2003 02:51:57 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Javier Pastor-Balbas (ECE)" <javier.pastor-balbas@ece.ericsson.se>
cc: sigtran@ietf.org
Subject: RE: [Sigtran] Re: M3UA Notify message
In-Reply-To: <1AB3D30B989BF141BBD5C70057B2EF7C01B659CE@EESTQNT105.es.eu.ericsson.se>
Message-ID: <Pine.SOL.4.10.10302260239410.12860-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=X-UNKNOWN
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from QUOTED-PRINTABLE to 8bit by www1.ietf.org id h1Q8wnp21726
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit


Javier,

In n+k laodsharing case 'n' APSs are sharing the traffic for the
particular RK.In REG message there is no parameter, based on which
load is shared.So i have used SLS statically for loadsharing on RK.

If i use RK1-ASP1 and RK2-ASP2 , Can this be called loadsharing?

--Samuel


  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Wed, 26 Feb 2003, Javier Pastor-Balbas (ECE) wrote:

> Hi Samuel, 
> 
> Reading again your proposed change #1. I would agree to change n as recommended in traffic mode while mandatory at start time, as I said before.
> But it seems to me that you are routing inside a RK, which is not disallowed, but why don't you use separate RKs? I mean, the RK should be sufficient to base the routing decissions on the SG...
> 
> Cheers // Javier.
> 
> > -----Original Message-----
> > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > Sent: miércoles, 26 de febrero de 2003 7:05
> > To: sigtran@ietf.org
> > Subject: RE: [Sigtran] Re: M3UA Notify message
> > 
> > 
> > 
> > Javier,
> > 
> > 
> >                SG                        +------+  AS1
> >  normal SS7  +-----+             +-------|      |
> > -----------> |     |------------/--------| ASP1 |      
> >  traffic(T)  |     |           /         +------+        'n' ASPs
> >              | SGP1|          /          +------+
> >              |     |---------/-----------| ASP2 |
> >              +-----+        /     +------|      |
> >                *           /     /       +------+
> >                *          /     /        +------+  
> >                **************************| ASP3 |  spare 'k' ASP
> >                         /     /          +------+
> >                        /     /              *
> >              +---- +  /     /               *
> >              |     |-+     /                *
> >              |     |      /                 *
> >              |SGP2 |-----+                  *
> >              |     |                        *
> >              +-----+                        *
> >                *                            *
> >                *                            *
> >                ******************************
> >                
> > 
> > At SGP1 SLS-ASP table
> > --------------------
> > SLS             ASP
> > ---             ---
> > X               ASP1
> > Y               ASP2
> > -               ASP3 (spare ASP)
> > 
> > AT AS1 SLS-ASP table
> > -------------------
> > SLS             ASP        PRIMARY-RT         SECND-RT
> > ---             ---        ----------         --------
> > A               ASP1        SGP1               SGP2
> > B               ASP2        SGP1               SGP2
> > 
> > 
> > Initially AS1 state at SGP is AS-DOWN.
> > 
> > SGP requires 'n' ASPs to handle the normal SS7 Traffic 'T', 
> > this can be
> > called as 'best traffic'.AS1 state is changed to AS-ACTIVE 
> > when 'n' ASPs
> > have become active.
> > 
> > Suppose ASP1 is failed so 'n' becomes '1'.
> > 
> > Now the question.
> > 
> > 1) If AS1 state is changed from ACTIVE to PENDING(as per IG)
> >    then no DATA messages will be sent to AS.
> >    
> >    IMO, we should keep the AS state as ACTIVE.
> >    
> >    According to q.704 section 3.8.2 we should reroute the 
> >    traffic to ASP2 provided congestion has not occured.
> > 
> >    Just reroute the SLS(X,Y) to available ASP1 , i beleive this is
> >    what we follow in legacy SS7 network.
> >    
> >    When ASP2 transmission buffer reaches congestion level then
> >    we can send NTFY to spare 'k' ASP and change the AS1 state to
> >    PENDING along with timer running.
> >    
> >  2)What will happen when the traffic reaches from 'T' to 'T+1'?
> >  
> >    At SG 'n' ASPs are not sufficient to handle this 'worst case'.
> >    So SG is required to send NTFY to spare 'k' ASP.
> >    But RFC and IG ignores to handle this case.We should enhance IG to
> >    support this case.
> >    
> >  IG section 3.15 misguides and should be re-writen to handle 
> > the above two 
> >  points.
> >  
> > --Samuel
> > 
> >   Alcatel USA, Inc.                  Internet: 
> > <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel 
> > USA, Inc. *******
> > 
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 04:56:44 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA29465
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 04:56:44 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QA6B026124
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 05:06:11 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QA5pp26084;
	Wed, 26 Feb 2003 05:05:52 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QA4gp26032
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 05:04:42 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA29423
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 04:54:43 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:gOV0pgE1urAdO+YwjKSW8gOFz+EOAi8+@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1Q9wcD07920;
	Wed, 26 Feb 2003 02:58:38 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1Q9wbW10906;
	Wed, 26 Feb 2003 02:58:37 -0700
Date: Wed, 26 Feb 2003 02:58:37 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030226025837.B8776@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OF4D09FADD.6A771A15-ONC2256CD9.001E8449@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF4D09FADD.6A771A15-ONC2256CD9.001E8449@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Wed, Feb 26, 2003 at 07:51:30AM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

TOLGA,

All in all nothing has changed from the original discussions:

o   The ASP should coordinate with other ASPs the number of active ASPs.

o   An ASP should not start traffic until sufficient ASPs are active.

o   The SG may choose not to direct traffic to an ASP until sufficient ASPs
    are active.

o   The SG should withhold NTFY(AS_Active) until it does direct traffic.

This is just as it is in the RFC.

--brian

TOLGA ASVEREN wrote:                                                                 (Wed, 26 Feb 2003 07:51:30)
> 
> Brian,
> > Tolga,
> >
> > Tolga Asveren wrote:
> >                                 (Tue, 25 Feb 2003
> > 13:02:20)
> > > Brian,
> > > --- "Brian F. G. Bidulock" <bidulock@openss7.org>
> > > wrote:
> > > > Tolga,
> > > >
> > > > Tolga Asveren wrote:
> >
> > > >                                 (Tue, 25 Feb
> > 2003
> > > > 12:15:22)
> > > > > Brian,
> > > > >
> > > > > [..snip..]
> > > > >
> > > > > > For example n=5. SG decides to start traffic
> > > > with
> > > > > >the 2nd ACTIVE ASP.
> > > > > >But
> > > > > > according to RFC ASPs will wait for the 5th
> > > > ACTIVE
> > > > > >ASP before
> > > > > >starting
> > > > > > traffic. In my opinion, this is a problem.
> > > > >
> > > > > There is no problem.  The ASPs should wait for
> > the
> > > > 5th
> > > > > ASP to be active
> > > > > before
> > > > > starting traffic regardless of when the SG
> > starts
> > > > > traffic.
> > > > > [TOLGA]O.K. finally there we are, the source
> > of
> > > > the
> > > > > disagreement. IMO, it is not a good idea to
> > > > open/close
> > > > > the traffic path between ASP/SGP
> > unidirectionaly.
> > > >
> > > > Nowhere does it say that the traffic path need
> > be
> > > > opened or closed unidirectionally.  The RFC
> > talks
> > > > about
> > > > the "start" of traffic, not the establishment of
> > > > unidirectional flows.
> > > [TOLGA]IMO, the path between ASP/SGP will be
> > opened
> > > unidirectionslly, if ASPs will wait for n of them
> > to
> > > be ACTIVE whereas SG decides to not withhold
> > traffic
> > > when number of ACTIVE ASPs is less than n.
> >
> > | For the n+k redundancy case, ASPs which are in
> > that AS should
> > | coordianate amoung themselves the number of active
> > ASPs in the AS, and
> > | should start sending traffic only after n ASPs are
> > active.
> >
> > This does not say that the ASP should not respond to
> > traffic
> > delivered to it by the SG: it says that the ASP
> > should not
> > "start" sending traffic.
> >
> > It is just saying that the ASPs should coordinate
> > going into
> > service and shouldn't start sending traffic until
> > they are
> > capable of handling the expected load (and they
> > shouldn't).
> [TOLGA]What is the reason behind allowing ASP to respond to messages from
> SG but not starting traffic?

When the stimulus is at the ASP, this provides for more graceful recovery
regardless of what the SG does or does not do.

But the ASP need not respond.  There is no requirement to respond and no
requirement not to respond.  The ASP designer could chose to discard messages,
buffer them, or activate the local MTP User and deliver them.

> How does ASP-M3UA know whether a message coming from user parts is a
> response or newly initiated traffic(whatever the definition of newly
> initiated traffic is)?

It is used for "starting" traffic at the ASP.  The MTP User at the ASP
should not be informed that MTP is available to it until "n" ASPs are
available.

> If SG decides that there are enough number of ACTIVE ASPs, why should we
> prevent ASPs to initiate traffic?

Because of the "start decision" criteria explained earlier on this thread.
The ASPs should wait, otherwise, they stand a chance of failure and wasting
network resources during an already critical network outage condition with
oscillating congestive failures.  SGs will typically always be able to handle
the traffic and have higher availability requirements than ASPs, particularly
in the SG as STP configuration.

> BTW, I proposed that text to the RFC and I never thougth that somebody can
> interprete the word "start" like you do. If you are really uncomfortable
> and think that the sentence is not clear, will changing "traffic" to
> "messages" make the ambiguity disappear?

I know the history of the passage: I started the "M3UA-10: Example 5.1.4"
thread.  It was one of my LC issues.

It was Greg's original suggestion from his note of Jan 2, 2002:

	"Do we need to add a hint that an ASP should wait for the
	Notify(AS_Active) (or via internal knowledge that sufficient
	ASPs are ASP Active)."

I have no problems with changing "n" in your text to "sufficient" (but its
really the same thing now isn't it?)

> >
> > > >
> > > > > And secondly, if you prefer SG to decide
> > > > dynamically
> > > > > to start traffic, why do you want ASPs to
> > decide
> > > > for
> > > > > this statically by interpreting n as
> > mandatory?
> > > >
> > > > The SG is neither required nor prohibited by the
> > RFC
> > > > to
> > > > decide dynamically.  Nor is the ASP required to
> > > > decide
> > > > statically.
> > > [TOLGA]For SG part I agree.But for ASP I disagree,
> > RFC
> > > says that ASPs SHOULD not start traffic before n
> > of
> > > them are ACTIVE, and this n is a static attribute.
> > If
> > > ASPs rely on NTFY for this, starting traffic will
> > be
> > > decided by SG and will be decided dynamically (or
> > > better said by SG, which can base this decision
> > also
> > > to the configured value n statically, this IMO is
> > > totally responsibility of SG)
> >
> > You are making many assuptions, most of which are
> > not valid.
> > The SG does not "start" traffic because there is no
> > stimulus
> > at the SG: the stimulus is either at the ASP or in
> > the SS7
> > network.  All the SG can do is decide to "direct"
> > traffic.
> > That does not mean that traffic actually exists, or
> > that it
> > will exist any time in the near future after.
> >
> > This is not a switch that you throw and current
> > starts flowing.
> [TOLGA]I think several times I explained what I mean with "SG starting
> traffic", but O.K. I will use "directing", it is more clear, but doesn't
> change too much from this discussion point of view IMO.
> >
> >
> > > >
> > > > The RFC permits the SG to do what it needs to do
> > > > with
> > > > regards to the SS7 network because of the "MAY".
> > > [TOLGA]The MAY is about withholding traffic to ASP
> > > destination. I don't interprete it directly related
> > > with interaction of SG with SS7 network.
> >
> > But it is directly related to the SS7 network and
> > that was
> > the reason for the MAY.  This is an AS which is
> > newly active:
> > not an AS recovering from AS-PENDING.  Regardless of
> > the
> > granularity of the RK, the SG has performed
> > necessary management
> > functions towards the SS7 network with regards to
> > the AS.
> > This could be UPU response, TFC response, or even
> > TFP response
> > if the SG is acting as STP.  Directing traffic back
> > to the
> > ASP for the AS requires that something be done to
> > undo whatever
> > was told to the SS7 network.
> [TOLGA] Of course SG will execute necessary SS7 procedures. And if you
> interprete it as undoing something told to SS7 network, I think SG should
> do this, whenever it decides that there are enough number of ACTIVE ASPs.

It is germane to the discussion because the "MAY" permits the SG
to take any necessary actions with respect to the SS7 network.
To say "SHOULD" would be inappropriate (placing a requirement on how
the SG interacts with the SS7 network).

> >
> > As the SG's primary purpose in life is to transer
> > traffic between
> > ASPs and the SS7 network, what did you think the SG
> > was going to
> > do to "direct" traffic if it is not going to
> > interact with the
> > SS7 network to do so?
> [TOLGA]I didn't say that SG won't interact with SS7 network. But the word
> "MAY" about withholding traffic is mainly related for communication between
> SGP/ASP.

Here I disagree.  There is no mention of buffering data in the SG as there
is for the AS-PENDING state.  The statement is directly with regard to SPMC
management.

> BTW, before SG decides that AS can be considered ACTIVE, it won't execute
> necessary SS7 procedures. So, where does the traffic, SG needs to withhold
> comes from? So, I would think, this withholding traffic will be meaningfull
> only for PENDING case(?).

Now your going around in circles...  The text does not say that the SG
withholds traffic.  It says that it "MAY choose not to direct traffic to
an newly active ASP".  Withholding traffic (placing it into a buffer) is
what happens in AS-PENDING state.  Choosing not to direct traffic to a
newly active ASP (for a newly active AS) consists of not telling the SS7
network (or local MTP Users) that the AS is accessible yet.  What *is*
"withheld" is the NTFY(AS_Active).

Consider the following example because it appears that you don't understand
what is involved:

AS1 RK1 = PCA/ISUP.  n = 2, k = 1.  3 ASPs: ASP1, ASP2, ASP3, 1 SGP: SGP1

AS1 is inactive.  ASP1 comes up and goes active for AS1.  SGP1 knows that
n=2 and also knows that ASP2 and ASP3 are in the ASP-DOWN state.  Until
now it has been responding to local MTP User MTP-TRANSFER requests to PCA,
SI=5 with MTP-STATUS(Inaccessible User).  Messages from the SS7 network
have been replied to with UPU(Inaccessible).  Now, SGP1 MAY decide to not
direct traffic to ASP1, but continue to respond to local MTP-TRANSFER
requests destined to AS1 with MTP-STATUS(Inaccessible User) and respond to
received SS7 messages for AS1 with UPU(Inaccessible).  If it does so, it
SHOULD withhold the NTFY(AS_Active).  If it decides to direct traffic to
ASP1, it MUST send NTFY(AS_Active) and will mark the MTP User as accessible
(and stop responding with MTP-STATUS and UPU).  The RFC cannot tell the
SGP designer what they should or should not do here because it is intricately
involved with SS7 conformance and performance.

ASP2 and ASP3 SHOULD coordinate with ASP1 the number of active ASPs.  That
is, ASP2 and ASP3 should know that "n" are required and should take action
to establish associations, go ASP-UP and at least one of them should go
ASP-ACTIVE for AS1.  There is no way for SGP1 to "tell" an ASP to go active
for an AS, so this is something the ASPs need to do on their own.  Also,
ASP1 SHOULD wait until one of ASP2 or ASP3 is active for AS1 before informing
the local ISUP (MTP User at PCA/SI=5) that it can start traffic.

Any questions?

There is absolutely nothing the matter with the statements in the RFC.

> >
> >
> > > >
> > > > The RFC recommends that the ASPs wait until "n"
> > > > because
> > > > they SHOULD.
> > > [TOLGA]So it is static according to RFC.
> >
> > Where does it say static?  It says "coordinate"
> > which sounds
> > pretty dynamic to me.
> [TOLGA]Because it relies on the statically configured value n. IMO, only SG
> can decide dynamically, when to open the communication path between
> ASP/SGP. And for that we need to make ASPs to decide based on received NTFY
> from SG, not to count number of ACTIVE ASPs by themselves.

In the above example, when ASP1 starts traffic is based on the situation with
regard to itself, ASP2, ASP3 and SGP1 and from the view of the M3UA protocol
is no different than the basis on which the SG decides.  SGP1 decides on its
own status, and its knowledge of the status of ASP1, ASP2 and ASP3.

In fact in the example the SGP has less information than ASP1 (because it does
not coordinate with ASP2 and ASP3 which are in the down state as far as the
SGP is concerned).

The SGP has some information about MTP users in the SS7 network, but then ASP1
has information about its local MTP users to which the SGP is not privy.

What kinds of dynamic information did you think the SGP has at its disposal
that the ASP does not?

--brian

> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 05:00:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA29532
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 05:00:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QAAOR27090
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 05:10:24 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QAABp27075;
	Wed, 26 Feb 2003 05:10:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QA9pp27052
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 05:09:51 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA29509
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 04:59:53 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:tStqurym5CP1GMXzVcfPLBBN5qd4LjXb@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1QA3mD08047;
	Wed, 26 Feb 2003 03:03:48 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1QA3m511127;
	Wed, 26 Feb 2003 03:03:48 -0700
Date: Wed, 26 Feb 2003 03:03:48 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030226030348.C8776@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OF54CDA0BE.51445E83-ONC2256CD9.002BEB06@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF54CDA0BE.51445E83-ONC2256CD9.002BEB06@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Wed, Feb 26, 2003 at 10:06:15AM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

TOLGA,

I agree with John Schantz' original objection to this proposal.  The Notify
message is advisory only and that requiring a response from the ASP to a
Notify message is too drastic a change to the protocol.

Besides, as the ASPs need to coordinate things so that n are active (the SG
cannot activate an ASP), they should surely know when n of them are active and
can use that to determine when to start traffic.  But, this is what the RFC
already says...

--brian

TOLGA ASVEREN wrote:                                                                 (Wed, 26 Feb 2003 10:06:15)
> 
> Javier,
> 
> I also agree that there is no problem with removing IG 3.15. My proposal
> for the solution is to make ASPs decide based on NTFY to start send/receive
> DATA. If we have consensus about this strategy, we can go into the details
> or we can consider other strategies.
> 
>      Tolga
> 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 05:13:54 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA29775
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 05:13:54 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QANLT27665
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 05:23:21 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QANAp27648;
	Wed, 26 Feb 2003 05:23:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QAMjp27629
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 05:22:45 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA29756
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 05:12:46 -0500 (EST)
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OFE6B4DE34.EBDAE7AC-ONC2256CD9.00373788@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Wed, 26 Feb 2003 12:12:22 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/26/2003
 12:13:16 PM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,



Consider the following example because it appears that you don't understand
what is involved:

AS1 RK1 = PCA/ISUP.  n = 2, k = 1.  3 ASPs: ASP1, ASP2, ASP3, 1 SGP: SGP1

AS1 is inactive.  ASP1 comes up and goes active for AS1.  SGP1 knows that
n=2 and also knows that ASP2 and ASP3 are in the ASP-DOWN state.  Until
now it has been responding to local MTP User MTP-TRANSFER requests to PCA,
SI=5 with MTP-STATUS(Inaccessible User).  Messages from the SS7 network
have been replied to with UPU(Inaccessible).  Now, SGP1 MAY decide to not
direct traffic to ASP1, but continue to respond to local MTP-TRANSFER
requests destined to AS1 with MTP-STATUS(Inaccessible User) and respond to
received SS7 messages for AS1 with UPU(Inaccessible).  If it does so, it
SHOULD withhold the NTFY(AS_Active).  If it decides to direct traffic to
ASP1, it MUST send NTFY(AS_Active) and will mark the MTP User as accessible
(and stop responding with MTP-STATUS and UPU).  The RFC cannot tell the
SGP designer what they should or should not do here because it is
intricately
involved with SS7 conformance and performance.

ASP2 and ASP3 SHOULD coordinate with ASP1 the number of active ASPs.  That
is, ASP2 and ASP3 should know that "n" are required and should take action
to establish associations, go ASP-UP and at least one of them should go
ASP-ACTIVE for AS1.  There is no way for SGP1 to "tell" an ASP to go active
for an AS, so this is something the ASPs need to do on their own.  Also,
ASP1 SHOULD wait until one of ASP2 or ASP3 is active for AS1 before
informing
the local ISUP (MTP User at PCA/SI=5) that it can start traffic.

Any questions?
[TOLGA] Sure. I don't get the reasoning behind letting wait ASPs for n of
them to be ACTIVE with a SHOULD and give freedom to SG to direct or not to
direct the traffic with MAY. For example in your example if SG decides to
direct traffic to AS with the first ACTIVE ASP, ASPs will still wait for
the nth ASP to be ACTIVE. I don't think this is a good choice. If ASPs
don't rely on an indication from SGP (NTFY), they can't know about the
decision of SGP, when it is going to direct traffic (unless this is
predetermined, which I call interpreting n as mandatory).








_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 05:14:47 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA29818
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 05:14:47 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QAOET27730
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 05:24:14 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QAO3p27717;
	Wed, 26 Feb 2003 05:24:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QANop27696
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 05:23:50 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA29772
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 05:13:51 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1QAHkeW029263
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 04:17:46 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h1QAHVF08402
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 04:17:31 -0600 (CST)
Date: Wed, 26 Feb 2003 04:17:31 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: sigtran@ietf.org
Subject: RE: [Sigtran] Re: M3UA Notify message
In-Reply-To: <Pine.SOL.4.10.10302190313420.8070-100000@sun3271.ssd.usa.alcatel.com>
Message-ID: <Pine.SOL.4.10.10302260413510.12860-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hi,

Why there is no discussion when traffic reaches from 'T' to 'T+1'?
(as mail below).
Is it not possible to have 'T+1' taffic?

--Samuel

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Wed, 19 Feb 2003, Samuel Dur D. Jeyaseelan wrote:

> 
> Hi,
> 
> I beleive 'n' is minimum number of ASPs to handle the 'best traffic T'.
> To handle the 'worst traffic T+1', SG may require 'n+1' ASPs.
> 
> 'n'is the derived value based on the traffic report given by Traffic
> Metering and Measurement.
> 
> So 'n' is fed during configuration of stack.When traffic increases from
> 'T' to 'T+1' or # APSs < n - this case SG is required to send NOTIFY to
> 'k' ASP.
> 
> And one more thing when # APSs < n, SG still can handle the traffic with
> single ASP.
> 
> When 'n' ASPs are active
> ------------------------
> 
> SLS-X  ---->ASP1
> SLS-Y  ---->ASP2
> 
> When ASP1 goes down
> -------------------
> 
> SLS-(Y,X)  ----->ASP2
> 
> Instead of banning the complete traffic ,it can be handled (not 100%) by
> ASP2.
> 
> 
> --Samuel
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Tue, 18 Feb 2003, Javier Pastor-Balbas (ECE) wrote:
> 
> > Brian, 
> > 
> > I agree 
> > 
> > RFC simply says: 
> > 
> > 	"n ASPs is the minimum number of redundant ASPs required to handle traffic"
> > 
> > As this is the minimum number, it can be inferred that for #ASPs < n no traffic can be handled. Nothing is mentioned in the RFC about engineered load although it may be thought that n value calculation is based on that.  
> > 
> > I would say that the current model does not support the use of extra ASPs for loads bigger than the engineered nor the use of less than n for smaller loads... The message is clear saying that #ASPs<n implies no traffic, i.e. whenever you can not support the engineered load, stop the system. This last sentence is not in the RFC but it can be deduced as the engineered load for n can... 
> > 
> > Cheers // Javier. 
> > 
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: Tuesday, February 18, 2003 12:49 PM
> > > To: Javier Pastor-Balbas (ECE)
> > > Cc: sigtran@ietf.org
> > > Subject: Re: [Sigtran] Re: M3UA Notify message
> > > 
> > > 
> > > Javier,
> > > 
> > > Javier Pastor-Balbas (ECE) wrote:             (Tue, 18 Feb 
> > > 2003 12:17:54)
> > > > > Javier Pastor-Balbas (ECE) wrote:             (Tue, 18 
> > > Feb 2003 07:57:23)
> > > > > > 
> > > > > > Hi Brian, 
> > > > > > 
> > > > > > According to the RFC "n" is required to handle traffic 
> > > with minimum
> > > > > > losses, what is expected from the SS7 point of view in 
> > > an interworking
> > > > > > scenario.
> > > 
> > > Please cite this passage, because I cannot find it.  I find 
> > > no mention of
> > > minimum number of ASPs and "message loss".  The only 
> > > reference to message loss
> > > that I know of is unrelated.
> > > 
> > 
> > 
> > 
> > > > > 
> > > > > That would be the numbe required to handle maximum 
> > > engineered load, which
> > > > > is the same thing.
> > > > > 
> > > > 
> > > > If you refer to the compromise load engineered to be 
> > > available 99% (aprox.)
> > > > of the time, then I agree. But this is a statically value fed at
> > > > configuration time. And the system won't work till n number 
> > > is reach since
> > > > the system will likely have quality contrains.
> > > 
> > > The maximum engineered load is a figure established at 
> > > configuration time.
> > > 
> > > Carrier-grade systems must always work with 1 processor.  The 
> > > system will just
> > > not necessarily be able to handle the offered load.  (But it might:
> > > particularly during quiet periods, off-peak, normal-day, etc.)
> > > 
> > > The offered load cannot be established at configuration time 
> > > and varies widely
> > > for telephony systems.
> > > 
> > > You must also realized that the system might, at any time,  
> > > be offered a load
> > > greater than the maximum engineered load (mass calling, busy 
> > > periods, etc.):
> > > so for carrier-grade systems even "n" processors must 
> > > implement overload
> > > controls (and they could not claim SS7 compliance if they didn't).
> > > 
> > > Look back in this thread and you will see the factors 
> > > contributing to the
> > > anticipated offered load for the start, restart and stop decision.
> > > 
> > > The decisions are not based on less than "n" processors not 
> > > allowing the
> > > system to work (where n > 1).  They are based on the negative 
> > > effects of
> > > delivering pent up traffic to processors that will likely 
> > > just congest.
> > > 
> > > --brian
> > > 
> > > -- 
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > 
> 
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 05:30:56 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA00128
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 05:30:56 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QAeOL29173
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 05:40:24 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QAe9p29126;
	Wed, 26 Feb 2003 05:40:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QAXTp28126
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 05:33:29 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA00017
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 05:23:30 -0500 (EST)
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF7DCF7760.B36F608D-ONC2256CD9.0038143A@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Wed, 26 Feb 2003 12:23:56 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/26/2003
 12:23:59 PM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,




TOLGA,

I agree with John Schantz' original objection to this proposal.  The Notify
message is advisory only and that requiring a response from the ASP to a
Notify message is too drastic a change to the protocol.
[TOLGA]I agree that it is not a minor change, especially just before the
text was fozen it was not a good option. That is the reason, why the
sentence saying that ASPs SHOULD coordinate among each other number of
ACTIVE ASPs is included at the last minute. If people still don't want to
change advisory meaning of NTFY, then for starting traffic, both ASP and
SGP SHOULD wait till n ASPs are ACTIVE. So, MAY for SG should be changed to
SHOULD. This change is already done in IG.









_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 06:51:58 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA01693
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 06:51:58 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QC1RB01555
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 07:01:27 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QC10p01509;
	Wed, 26 Feb 2003 07:01:00 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QBx7p01414
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 06:59:07 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA01664
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 06:49:06 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:MC8AGVJ3CvB3jKcoq/7J7o9QpUbwz2Hn@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1QBr2D10175;
	Wed, 26 Feb 2003 04:53:02 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1QBr1p13154;
	Wed, 26 Feb 2003 04:53:01 -0700
Date: Wed, 26 Feb 2003 04:53:01 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030226045301.F8776@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OFE6B4DE34.EBDAE7AC-ONC2256CD9.00373788@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OFE6B4DE34.EBDAE7AC-ONC2256CD9.00373788@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Wed, Feb 26, 2003 at 12:12:22PM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

TOLGA,

TOLGA ASVEREN wrote:                                                                 (Wed, 26 Feb 2003 12:12:22)
> 
> Brian,
> 
> 
> 
> Consider the following example because it appears that you don't understand
> what is involved:
> 
> AS1 RK1 = PCA/ISUP.  n = 2, k = 1.  3 ASPs: ASP1, ASP2, ASP3, 1 SGP: SGP1
> 
> AS1 is inactive.  ASP1 comes up and goes active for AS1.  SGP1 knows that
> n=2 and also knows that ASP2 and ASP3 are in the ASP-DOWN state.  Until
> now it has been responding to local MTP User MTP-TRANSFER requests to PCA,
> SI=5 with MTP-STATUS(Inaccessible User).  Messages from the SS7 network
> have been replied to with UPU(Inaccessible).  Now, SGP1 MAY decide to not
> direct traffic to ASP1, but continue to respond to local MTP-TRANSFER
> requests destined to AS1 with MTP-STATUS(Inaccessible User) and respond to
> received SS7 messages for AS1 with UPU(Inaccessible).  If it does so, it
> SHOULD withhold the NTFY(AS_Active).  If it decides to direct traffic to
> ASP1, it MUST send NTFY(AS_Active) and will mark the MTP User as accessible
> (and stop responding with MTP-STATUS and UPU).  The RFC cannot tell the
> SGP designer what they should or should not do here because it is
> intricately
> involved with SS7 conformance and performance.
> 
> ASP2 and ASP3 SHOULD coordinate with ASP1 the number of active ASPs.  That
> is, ASP2 and ASP3 should know that "n" are required and should take action
> to establish associations, go ASP-UP and at least one of them should go
> ASP-ACTIVE for AS1.  There is no way for SGP1 to "tell" an ASP to go active
> for an AS, so this is something the ASPs need to do on their own.  Also,
> ASP1 SHOULD wait until one of ASP2 or ASP3 is active for AS1 before
> informing
> the local ISUP (MTP User at PCA/SI=5) that it can start traffic.
> 
> Any questions?

> [TOLGA] Sure. I don't get the reasoning behind letting wait ASPs for n of
> them to be ACTIVE with a SHOULD and give freedom to SG to direct or not to
> direct the traffic with MAY.

Because M3UA cannot say what the SG "SHOULD" do with regard to the SS7
network.  It can only talk about what it "MAY" do with regard to the SS7
network.

> For example in your example if SG decides to direct traffic to AS with the
> first ACTIVE ASP, ASPs will still wait for the nth ASP to be ACTIVE. I don't
> think this is a good choice.

It is the best choice.

> If ASPs don't rely on an indication from SGP (NTFY), they can't know about
> the decision of SGP, when it is going to direct traffic (unless this is
> predetermined, which I call interpreting n as mandatory).

The ASPs can reply on an indication from the SGP (NTFY) because it says
"SHOULD" in the RFC and, if the NTFY is not withheld it is a "MUST" to
send NTFY(AS_Active) when the first ASP goes active.  Therefore, the ASP
can rely upon the fact that a correct SGP will always send NTFY(AS_Active)
before or when it directs traffic.

But regardless of the SGP's decision, the ASP SHOULD still wait.  It is up
to the ASP to determine what to do with any traffic it might receive while
it is waiting (which might be none, even if the SGP "directs" traffic).

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 06:52:50 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA01727
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 06:52:50 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QC2KW01606
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 07:02:20 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QC24p01588;
	Wed, 26 Feb 2003 07:02:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QC0kp01499
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 07:00:46 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA01680
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 06:50:44 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:c+T25G6oaY/KlXAk4FR0LguLYPTBdD+m@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1QBseD10192;
	Wed, 26 Feb 2003 04:54:40 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1QBses13161;
	Wed, 26 Feb 2003 04:54:40 -0700
Date: Wed, 26 Feb 2003 04:54:40 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030226045440.G8776@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	sigtran@ietf.org
References: <Pine.SOL.4.10.10302190313420.8070-100000@sun3271.ssd.usa.alcatel.com> <Pine.SOL.4.10.10302260413510.12860-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302260413510.12860-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Wed, Feb 26, 2003 at 04:17:31AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

The RFC says that the ASPs should coordinate between themselves the number
of active ASPs.  A k ASP is welcome to activate whenever it wants.  The
SGP cannot invoke the activation of an ASP.

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Wed, 26 Feb 2003 04:17:31)
> 
> Hi,
> 
> Why there is no discussion when traffic reaches from 'T' to 'T+1'?
> (as mail below).
> Is it not possible to have 'T+1' taffic?
> 
> --Samuel
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Wed, 19 Feb 2003, Samuel Dur D. Jeyaseelan wrote:
> 
> > 
> > Hi,
> > 
> > I beleive 'n' is minimum number of ASPs to handle the 'best traffic T'.
> > To handle the 'worst traffic T+1', SG may require 'n+1' ASPs.
> > 
> > 'n'is the derived value based on the traffic report given by Traffic
> > Metering and Measurement.
> > 
> > So 'n' is fed during configuration of stack.When traffic increases from
> > 'T' to 'T+1' or # APSs < n - this case SG is required to send NOTIFY to
> > 'k' ASP.
> > 
> > And one more thing when # APSs < n, SG still can handle the traffic with
> > single ASP.
> > 
> > When 'n' ASPs are active
> > ------------------------
> > 
> > SLS-X  ---->ASP1
> > SLS-Y  ---->ASP2
> > 
> > When ASP1 goes down
> > -------------------
> > 
> > SLS-(Y,X)  ----->ASP2
> > 
> > Instead of banning the complete traffic ,it can be handled (not 100%) by
> > ASP2.
> > 
> > 
> > --Samuel
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Tue, 18 Feb 2003, Javier Pastor-Balbas (ECE) wrote:
> > 
> > > Brian, 
> > > 
> > > I agree 
> > > 
> > > RFC simply says: 
> > > 
> > > 	"n ASPs is the minimum number of redundant ASPs required to handle traffic"
> > > 
> > > As this is the minimum number, it can be inferred that for #ASPs < n no traffic can be handled. Nothing is mentioned in the RFC about engineered load although it may be thought that n value calculation is based on that.  
> > > 
> > > I would say that the current model does not support the use of extra ASPs for loads bigger than the engineered nor the use of less than n for smaller loads... The message is clear saying that #ASPs<n implies no traffic, i.e. whenever you can not support the engineered load, stop the system. This last sentence is not in the RFC but it can be deduced as the engineered load for n can... 
> > > 
> > > Cheers // Javier. 
> > > 
> > > > -----Original Message-----
> > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > Sent: Tuesday, February 18, 2003 12:49 PM
> > > > To: Javier Pastor-Balbas (ECE)
> > > > Cc: sigtran@ietf.org
> > > > Subject: Re: [Sigtran] Re: M3UA Notify message
> > > > 
> > > > 
> > > > Javier,
> > > > 
> > > > Javier Pastor-Balbas (ECE) wrote:             (Tue, 18 Feb 
> > > > 2003 12:17:54)
> > > > > > Javier Pastor-Balbas (ECE) wrote:             (Tue, 18 
> > > > Feb 2003 07:57:23)
> > > > > > > 
> > > > > > > Hi Brian, 
> > > > > > > 
> > > > > > > According to the RFC "n" is required to handle traffic 
> > > > with minimum
> > > > > > > losses, what is expected from the SS7 point of view in 
> > > > an interworking
> > > > > > > scenario.
> > > > 
> > > > Please cite this passage, because I cannot find it.  I find 
> > > > no mention of
> > > > minimum number of ASPs and "message loss".  The only 
> > > > reference to message loss
> > > > that I know of is unrelated.
> > > > 
> > > 
> > > 
> > > 
> > > > > > 
> > > > > > That would be the numbe required to handle maximum 
> > > > engineered load, which
> > > > > > is the same thing.
> > > > > > 
> > > > > 
> > > > > If you refer to the compromise load engineered to be 
> > > > available 99% (aprox.)
> > > > > of the time, then I agree. But this is a statically value fed at
> > > > > configuration time. And the system won't work till n number 
> > > > is reach since
> > > > > the system will likely have quality contrains.
> > > > 
> > > > The maximum engineered load is a figure established at 
> > > > configuration time.
> > > > 
> > > > Carrier-grade systems must always work with 1 processor.  The 
> > > > system will just
> > > > not necessarily be able to handle the offered load.  (But it might:
> > > > particularly during quiet periods, off-peak, normal-day, etc.)
> > > > 
> > > > The offered load cannot be established at configuration time 
> > > > and varies widely
> > > > for telephony systems.
> > > > 
> > > > You must also realized that the system might, at any time,  
> > > > be offered a load
> > > > greater than the maximum engineered load (mass calling, busy 
> > > > periods, etc.):
> > > > so for carrier-grade systems even "n" processors must 
> > > > implement overload
> > > > controls (and they could not claim SS7 compliance if they didn't).
> > > > 
> > > > Look back in this thread and you will see the factors 
> > > > contributing to the
> > > > anticipated offered load for the start, restart and stop decision.
> > > > 
> > > > The decisions are not based on less than "n" processors not 
> > > > allowing the
> > > > system to work (where n > 1).  They are based on the negative 
> > > > effects of
> > > > delivering pent up traffic to processors that will likely 
> > > > just congest.
> > > > 
> > > > --brian
> > > > 
> > > > -- 
> > > > Brian F. G. Bidulock
> > > > bidulock@openss7.org
> > > > http://www.openss7.org/
> > > > 
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> > 
> > 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 06:54:46 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA01789
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 06:54:46 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QC4FR01727
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 07:04:15 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QC44p01719;
	Wed, 26 Feb 2003 07:04:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QC37p01657
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 07:03:07 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA01744
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 06:53:06 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:Ws533T7jMLEhY9jsCNUTH+Nm6SZFRmmd@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1QBv1D10252;
	Wed, 26 Feb 2003 04:57:01 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1QBv1U13299;
	Wed, 26 Feb 2003 04:57:01 -0700
Date: Wed, 26 Feb 2003 04:57:01 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030226045701.H8776@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OF7DCF7760.B36F608D-ONC2256CD9.0038143A@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF7DCF7760.B36F608D-ONC2256CD9.0038143A@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Wed, Feb 26, 2003 at 12:23:56PM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

TOLGA,

TOLGA ASVEREN wrote:                                                                 (Wed, 26 Feb 2003 12:23:56)
> 
> Brian,
> 
> 
> 
> 
> TOLGA,
> 
> I agree with John Schantz' original objection to this proposal.  The Notify
> message is advisory only and that requiring a response from the ASP to a
> Notify message is too drastic a change to the protocol.
> [TOLGA]I agree that it is not a minor change, especially just before the
> text was fozen it was not a good option. That is the reason, why the
> sentence saying that ASPs SHOULD coordinate among each other number of
> ACTIVE ASPs is included at the last minute. If people still don't want to
> change advisory meaning of NTFY, then for starting traffic, both ASP and
> SGP SHOULD wait till n ASPs are ACTIVE. So, MAY for SG should be changed to
> SHOULD. This change is already done in IG.

You cannot place a "SHOULD" on the SG's interaction with the SS7 network
(directing traffic).  That is why it says "MAY" and must remain as "MAY".

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 07:03:50 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA02005
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 07:03:50 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QCDJo02873
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 07:13:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QCD8p02865;
	Wed, 26 Feb 2003 07:13:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QCCup02847
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 07:12:56 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA01993
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 07:02:54 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1QC6jeY007089;
	Wed, 26 Feb 2003 06:06:46 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h1QC6TF01305;
	Wed, 26 Feb 2003 06:06:29 -0600 (CST)
Date: Wed, 26 Feb 2003 06:06:29 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Brian F. G. Bidulock" <bidulock@openss7.org>
cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
In-Reply-To: <20030226045440.G8776@openss7.org>
Message-ID: <Pine.SOL.4.10.10302260559210.12860-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,
I understood the same.But there may be a time where SG can request k ASP
to become active in order to handle the heavy traffic.
Because configured 'n' ASPs are sufficient to handle the normal traffic
but when traffic increases than 'normal traffic' from SS7 network towards
SG- this case SG may request k ASP to become active.

--Samuel


  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Wed, 26 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> The RFC says that the ASPs should coordinate between themselves the number
> of active ASPs.  A k ASP is welcome to activate whenever it wants.  The
> SGP cannot invoke the activation of an ASP.
> 
> --brian
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Wed, 26 Feb 2003 04:17:31)
> > 
> > Hi,
> > 
> > Why there is no discussion when traffic reaches from 'T' to 'T+1'?
> > (as mail below).
> > Is it not possible to have 'T+1' taffic?
> > 
> > --Samuel
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Wed, 19 Feb 2003, Samuel Dur D. Jeyaseelan wrote:
> > 
> > > 
> > > Hi,
> > > 
> > > I beleive 'n' is minimum number of ASPs to handle the 'best traffic T'.
> > > To handle the 'worst traffic T+1', SG may require 'n+1' ASPs.
> > > 
> > > 'n'is the derived value based on the traffic report given by Traffic
> > > Metering and Measurement.
> > > 
> > > So 'n' is fed during configuration of stack.When traffic increases from
> > > 'T' to 'T+1' or # APSs < n - this case SG is required to send NOTIFY to
> > > 'k' ASP.
> > > 
> > > And one more thing when # APSs < n, SG still can handle the traffic with
> > > single ASP.
> > > 
> > > When 'n' ASPs are active
> > > ------------------------
> > > 
> > > SLS-X  ---->ASP1
> > > SLS-Y  ---->ASP2
> > > 
> > > When ASP1 goes down
> > > -------------------
> > > 
> > > SLS-(Y,X)  ----->ASP2
> > > 
> > > Instead of banning the complete traffic ,it can be handled (not 100%) by
> > > ASP2.
> > > 
> > > 
> > > --Samuel
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Tue, 18 Feb 2003, Javier Pastor-Balbas (ECE) wrote:
> > > 
> > > > Brian, 
> > > > 
> > > > I agree 
> > > > 
> > > > RFC simply says: 
> > > > 
> > > > 	"n ASPs is the minimum number of redundant ASPs required to handle traffic"
> > > > 
> > > > As this is the minimum number, it can be inferred that for #ASPs < n no traffic can be handled. Nothing is mentioned in the RFC about engineered load although it may be thought that n value calculation is based on that.  
> > > > 
> > > > I would say that the current model does not support the use of extra ASPs for loads bigger than the engineered nor the use of less than n for smaller loads... The message is clear saying that #ASPs<n implies no traffic, i.e. whenever you can not support the engineered load, stop the system. This last sentence is not in the RFC but it can be deduced as the engineered load for n can... 
> > > > 
> > > > Cheers // Javier. 
> > > > 
> > > > > -----Original Message-----
> > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > Sent: Tuesday, February 18, 2003 12:49 PM
> > > > > To: Javier Pastor-Balbas (ECE)
> > > > > Cc: sigtran@ietf.org
> > > > > Subject: Re: [Sigtran] Re: M3UA Notify message
> > > > > 
> > > > > 
> > > > > Javier,
> > > > > 
> > > > > Javier Pastor-Balbas (ECE) wrote:             (Tue, 18 Feb 
> > > > > 2003 12:17:54)
> > > > > > > Javier Pastor-Balbas (ECE) wrote:             (Tue, 18 
> > > > > Feb 2003 07:57:23)
> > > > > > > > 
> > > > > > > > Hi Brian, 
> > > > > > > > 
> > > > > > > > According to the RFC "n" is required to handle traffic 
> > > > > with minimum
> > > > > > > > losses, what is expected from the SS7 point of view in 
> > > > > an interworking
> > > > > > > > scenario.
> > > > > 
> > > > > Please cite this passage, because I cannot find it.  I find 
> > > > > no mention of
> > > > > minimum number of ASPs and "message loss".  The only 
> > > > > reference to message loss
> > > > > that I know of is unrelated.
> > > > > 
> > > > 
> > > > 
> > > > 
> > > > > > > 
> > > > > > > That would be the numbe required to handle maximum 
> > > > > engineered load, which
> > > > > > > is the same thing.
> > > > > > > 
> > > > > > 
> > > > > > If you refer to the compromise load engineered to be 
> > > > > available 99% (aprox.)
> > > > > > of the time, then I agree. But this is a statically value fed at
> > > > > > configuration time. And the system won't work till n number 
> > > > > is reach since
> > > > > > the system will likely have quality contrains.
> > > > > 
> > > > > The maximum engineered load is a figure established at 
> > > > > configuration time.
> > > > > 
> > > > > Carrier-grade systems must always work with 1 processor.  The 
> > > > > system will just
> > > > > not necessarily be able to handle the offered load.  (But it might:
> > > > > particularly during quiet periods, off-peak, normal-day, etc.)
> > > > > 
> > > > > The offered load cannot be established at configuration time 
> > > > > and varies widely
> > > > > for telephony systems.
> > > > > 
> > > > > You must also realized that the system might, at any time,  
> > > > > be offered a load
> > > > > greater than the maximum engineered load (mass calling, busy 
> > > > > periods, etc.):
> > > > > so for carrier-grade systems even "n" processors must 
> > > > > implement overload
> > > > > controls (and they could not claim SS7 compliance if they didn't).
> > > > > 
> > > > > Look back in this thread and you will see the factors 
> > > > > contributing to the
> > > > > anticipated offered load for the start, restart and stop decision.
> > > > > 
> > > > > The decisions are not based on less than "n" processors not 
> > > > > allowing the
> > > > > system to work (where n > 1).  They are based on the negative 
> > > > > effects of
> > > > > delivering pent up traffic to processors that will likely 
> > > > > just congest.
> > > > > 
> > > > > --brian
> > > > > 
> > > > > -- 
> > > > > Brian F. G. Bidulock
> > > > > bidulock@openss7.org
> > > > > http://www.openss7.org/
> > > > > 
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > 
> > > 
> > > 
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 07:10:48 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA02105
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 07:10:48 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QCKIj03136
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 07:20:18 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QCK5p03128;
	Wed, 26 Feb 2003 07:20:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QCJ8p03071
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 07:19:08 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA02080
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 07:09:06 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:dxXmCDmUO68lYXCHZ3/uFFnAEJHCLX5R@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1QCD2D10547;
	Wed, 26 Feb 2003 05:13:02 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1QCD1p13707;
	Wed, 26 Feb 2003 05:13:01 -0700
Date: Wed, 26 Feb 2003 05:13:01 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030226051301.I8776@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	sigtran@ietf.org
References: <20030226045440.G8776@openss7.org> <Pine.SOL.4.10.10302260559210.12860-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302260559210.12860-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Wed, Feb 26, 2003 at 06:06:29AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Wed, 26 Feb 2003 06:06:29)
> 
> Brian,
> I understood the same.But there may be a time where SG can request k ASP
> to become active in order to handle the heavy traffic.
> Because configured 'n' ASPs are sufficient to handle the normal traffic
> but when traffic increases than 'normal traffic' from SS7 network towards
> SG- this case SG may request k ASP to become active.

The SG has no understanding of the traffic handling capability of the ASPs.

The only thing it has in the number "n".

"n" is the numer of ASPs required to handle the maximum engineered (expected)
load, (not the "normal" load).

The SG does not (and cannot be required to) monitor any traffic load and even
if it did, it cannot translate this into ASP handling capacity.

The ASPs should coordinate how many of them are active with consideration for
their own process loading.

The RFC is fine as it is.

--brian

> 
> --Samuel
> 
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Wed, 26 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > The RFC says that the ASPs should coordinate between themselves the number
> > of active ASPs.  A k ASP is welcome to activate whenever it wants.  The
> > SGP cannot invoke the activation of an ASP.
> > 
> > --brian
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Wed, 26 Feb 2003 04:17:31)
> > > 
> > > Hi,
> > > 
> > > Why there is no discussion when traffic reaches from 'T' to 'T+1'?
> > > (as mail below).
> > > Is it not possible to have 'T+1' taffic?
> > > 
> > > --Samuel
> > > 
> > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > 
> > > On Wed, 19 Feb 2003, Samuel Dur D. Jeyaseelan wrote:
> > > 
> > > > 
> > > > Hi,
> > > > 
> > > > I beleive 'n' is minimum number of ASPs to handle the 'best traffic T'.
> > > > To handle the 'worst traffic T+1', SG may require 'n+1' ASPs.
> > > > 
> > > > 'n'is the derived value based on the traffic report given by Traffic
> > > > Metering and Measurement.
> > > > 
> > > > So 'n' is fed during configuration of stack.When traffic increases from
> > > > 'T' to 'T+1' or # APSs < n - this case SG is required to send NOTIFY to
> > > > 'k' ASP.
> > > > 
> > > > And one more thing when # APSs < n, SG still can handle the traffic with
> > > > single ASP.
> > > > 
> > > > When 'n' ASPs are active
> > > > ------------------------
> > > > 
> > > > SLS-X  ---->ASP1
> > > > SLS-Y  ---->ASP2
> > > > 
> > > > When ASP1 goes down
> > > > -------------------
> > > > 
> > > > SLS-(Y,X)  ----->ASP2
> > > > 
> > > > Instead of banning the complete traffic ,it can be handled (not 100%) by
> > > > ASP2.
> > > > 
> > > > 
> > > > --Samuel
> > > > 
> > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > 
> > > > On Tue, 18 Feb 2003, Javier Pastor-Balbas (ECE) wrote:
> > > > 
> > > > > Brian, 
> > > > > 
> > > > > I agree 
> > > > > 
> > > > > RFC simply says: 
> > > > > 
> > > > > 	"n ASPs is the minimum number of redundant ASPs required to handle traffic"
> > > > > 
> > > > > As this is the minimum number, it can be inferred that for #ASPs < n no traffic can be handled. Nothing is mentioned in the RFC about engineered load although it may be thought that n value calculation is based on that.  
> > > > > 
> > > > > I would say that the current model does not support the use of extra ASPs for loads bigger than the engineered nor the use of less than n for smaller loads... The message is clear saying that #ASPs<n implies no traffic, i.e. whenever you can not support the engineered load, stop the system. This last sentence is not in the RFC but it can be deduced as the engineered load for n can... 
> > > > > 
> > > > > Cheers // Javier. 
> > > > > 
> > > > > > -----Original Message-----
> > > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > > Sent: Tuesday, February 18, 2003 12:49 PM
> > > > > > To: Javier Pastor-Balbas (ECE)
> > > > > > Cc: sigtran@ietf.org
> > > > > > Subject: Re: [Sigtran] Re: M3UA Notify message
> > > > > > 
> > > > > > 
> > > > > > Javier,
> > > > > > 
> > > > > > Javier Pastor-Balbas (ECE) wrote:             (Tue, 18 Feb 
> > > > > > 2003 12:17:54)
> > > > > > > > Javier Pastor-Balbas (ECE) wrote:             (Tue, 18 
> > > > > > Feb 2003 07:57:23)
> > > > > > > > > 
> > > > > > > > > Hi Brian, 
> > > > > > > > > 
> > > > > > > > > According to the RFC "n" is required to handle traffic 
> > > > > > with minimum
> > > > > > > > > losses, what is expected from the SS7 point of view in 
> > > > > > an interworking
> > > > > > > > > scenario.
> > > > > > 
> > > > > > Please cite this passage, because I cannot find it.  I find 
> > > > > > no mention of
> > > > > > minimum number of ASPs and "message loss".  The only 
> > > > > > reference to message loss
> > > > > > that I know of is unrelated.
> > > > > > 
> > > > > 
> > > > > 
> > > > > 
> > > > > > > > 
> > > > > > > > That would be the numbe required to handle maximum 
> > > > > > engineered load, which
> > > > > > > > is the same thing.
> > > > > > > > 
> > > > > > > 
> > > > > > > If you refer to the compromise load engineered to be 
> > > > > > available 99% (aprox.)
> > > > > > > of the time, then I agree. But this is a statically value fed at
> > > > > > > configuration time. And the system won't work till n number 
> > > > > > is reach since
> > > > > > > the system will likely have quality contrains.
> > > > > > 
> > > > > > The maximum engineered load is a figure established at 
> > > > > > configuration time.
> > > > > > 
> > > > > > Carrier-grade systems must always work with 1 processor.  The 
> > > > > > system will just
> > > > > > not necessarily be able to handle the offered load.  (But it might:
> > > > > > particularly during quiet periods, off-peak, normal-day, etc.)
> > > > > > 
> > > > > > The offered load cannot be established at configuration time 
> > > > > > and varies widely
> > > > > > for telephony systems.
> > > > > > 
> > > > > > You must also realized that the system might, at any time,  
> > > > > > be offered a load
> > > > > > greater than the maximum engineered load (mass calling, busy 
> > > > > > periods, etc.):
> > > > > > so for carrier-grade systems even "n" processors must 
> > > > > > implement overload
> > > > > > controls (and they could not claim SS7 compliance if they didn't).
> > > > > > 
> > > > > > Look back in this thread and you will see the factors 
> > > > > > contributing to the
> > > > > > anticipated offered load for the start, restart and stop decision.
> > > > > > 
> > > > > > The decisions are not based on less than "n" processors not 
> > > > > > allowing the
> > > > > > system to work (where n > 1).  They are based on the negative 
> > > > > > effects of
> > > > > > delivering pent up traffic to processors that will likely 
> > > > > > just congest.
> > > > > > 
> > > > > > --brian
> > > > > > 
> > > > > > -- 
> > > > > > Brian F. G. Bidulock
> > > > > > bidulock@openss7.org
> > > > > > http://www.openss7.org/
> > > > > > 
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > 
> > > > 
> > > > 
> > > 
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 07:22:08 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA02316
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 07:22:08 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QCVd903579
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 07:31:39 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QCVQp03566;
	Wed, 26 Feb 2003 07:31:26 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QCUNp03488
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 07:30:23 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA02301
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 07:20:19 -0500 (EST)
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OFFADF9483.A9175FCB-ONC2256CD9.0041C7B8@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Wed, 26 Feb 2003 14:20:39 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/26/2003
 02:20:50 PM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,


TOLGA,

TOLGA ASVEREN wrote:
(Wed, 26 Feb 2003 12:23:56)
>
> Brian,
>
>
>
>
> TOLGA,
>
> I agree with John Schantz' original objection to this proposal.  The
Notify
> message is advisory only and that requiring a response from the ASP to a
> Notify message is too drastic a change to the protocol.
> [TOLGA]I agree that it is not a minor change, especially just before the
> text was fozen it was not a good option. That is the reason, why the
> sentence saying that ASPs SHOULD coordinate among each other number of
> ACTIVE ASPs is included at the last minute. If people still don't want to
> change advisory meaning of NTFY, then for starting traffic, both ASP and
> SGP SHOULD wait till n ASPs are ACTIVE. So, MAY for SG should be changed
to
> SHOULD. This change is already done in IG.

You cannot place a "SHOULD" on the SG's interaction with the SS7 network
(directing traffic).  That is why it says "MAY" and must remain as "MAY".
[TOLGA]The sentence is, "An SGP or IPSP, upon reception of an ASP Active
message for the first
   ASP in a Loadshare AS, MAY choose not to direct traffic to a newly
active ASP until it determines that there are sufficient resources to
handle the expected load (e.g., until there are "n" ASPs in state
ASP-ACTIVE in the AS)." (In IG MAY is changed to SHOULD) It is not about
directing traffic to SS7 network but to ASPs. Why is this considered as an
interaction of SG with SS7 network? Or how is this different than saying
that for non n+k case SGP should not direct traffic unless there is one
ACTIVE ASP? And as I said, opening the path unidirectionally is not the
best choice in my opinion and should be avoided.






_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 08:38:29 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA04009
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 08:38:28 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QDlxK09144
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 08:47:59 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QDlip09130;
	Wed, 26 Feb 2003 08:47:45 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QDknp09084
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 08:46:49 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA03977
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 08:36:47 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:q73YDxoONbRHl9M1bjgkjCXW0i7JLQ2O@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1QDegD11851;
	Wed, 26 Feb 2003 06:40:42 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1QDefI15034;
	Wed, 26 Feb 2003 06:40:41 -0700
Date: Wed, 26 Feb 2003 06:40:41 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030226064041.L8776@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OFFADF9483.A9175FCB-ONC2256CD9.0041C7B8@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OFFADF9483.A9175FCB-ONC2256CD9.0041C7B8@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Wed, Feb 26, 2003 at 02:20:39PM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

TOLGA,

TOLGA ASVEREN wrote:                                                                 (Wed, 26 Feb 2003 14:20:39)
> 
> Brian,
> 
> 
> TOLGA,
> 
> TOLGA ASVEREN wrote:
> (Wed, 26 Feb 2003 12:23:56)
> >
> > Brian,
> >
> >
> >
> >
> > TOLGA,
> >
> > I agree with John Schantz' original objection to this proposal.  The
> Notify
> > message is advisory only and that requiring a response from the ASP to a
> > Notify message is too drastic a change to the protocol.
> > [TOLGA]I agree that it is not a minor change, especially just before the
> > text was fozen it was not a good option. That is the reason, why the
> > sentence saying that ASPs SHOULD coordinate among each other number of
> > ACTIVE ASPs is included at the last minute. If people still don't want to
> > change advisory meaning of NTFY, then for starting traffic, both ASP and
> > SGP SHOULD wait till n ASPs are ACTIVE. So, MAY for SG should be changed
> to
> > SHOULD. This change is already done in IG.
> 
> You cannot place a "SHOULD" on the SG's interaction with the SS7 network
> (directing traffic).  That is why it says "MAY" and must remain as "MAY".
> [TOLGA]The sentence is, "An SGP or IPSP, upon reception of an ASP Active
> message for the first
>    ASP in a Loadshare AS, MAY choose not to direct traffic to a newly
> active ASP until it determines that there are sufficient resources to
> handle the expected load (e.g., until there are "n" ASPs in state
> ASP-ACTIVE in the AS)." (In IG MAY is changed to SHOULD) It is not about
> directing traffic to SS7 network but to ASPs. Why is this considered as an
> interaction of SG with SS7 network? Or how is this different than saying

See example from previous note.  The SG needs to interact with SS7 to direct
traffic to an ASP when an AS first becomes active.  Where else is it to get
the traffic that was previously suppressed?

Also, the SG could take immediate action to direct traffic that will not
appear for some time depending on the RK the AS is configured for.  For
example, ISUP will only test remote user part availability periodically.
The SG could direct traffic immediately and the traffic could show up several
minutes later.

It is even impossible to talk about the traffic that the SG is "directing"
without giving consideration to SS7 users involved.  The only way the SG
can "direct" any SS7 traffic in this situation is by taking actions within
the scope of SS7 and outside the scope of M3UA.

You cannot say SHOULD.

Leave the RFC as it stands.

> that for non n+k case SGP should not direct traffic unless there is one
> ACTIVE ASP? And as I said, opening the path unidirectionally is not the
> best choice in my opinion and should be avoided.

I already explained to you: there is no "unidirectional path".  This
is not a switch that you turn on an the current flows.

--brian


-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 09:02:31 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA05424
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 09:02:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QEC1P11036
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 09:12:01 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QEBpp10973;
	Wed, 26 Feb 2003 09:11:51 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QEAEp10822
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 09:10:14 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA04882
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 09:00:10 -0500 (EST)
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OFF843FE7A.129828A1-ONC2256CD9.004B7333@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Wed, 26 Feb 2003 16:00:29 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/26/2003
 04:00:38 PM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,




TOLGA,

TOLGA ASVEREN wrote:
(Wed, 26 Feb 2003 14:20:39)
>
> Brian,
>
>
> TOLGA,
>
> TOLGA ASVEREN wrote:
> (Wed, 26 Feb 2003 12:23:56)
> >
> > Brian,
> >
> >
> >
> >
> > TOLGA,
> >
> > I agree with John Schantz' original objection to this proposal.  The
> Notify
> > message is advisory only and that requiring a response from the ASP to
a
> > Notify message is too drastic a change to the protocol.
> > [TOLGA]I agree that it is not a minor change, especially just before
the
> > text was fozen it was not a good option. That is the reason, why the
> > sentence saying that ASPs SHOULD coordinate among each other number of
> > ACTIVE ASPs is included at the last minute. If people still don't want
to
> > change advisory meaning of NTFY, then for starting traffic, both ASP
and
> > SGP SHOULD wait till n ASPs are ACTIVE. So, MAY for SG should be
changed
> to
> > SHOULD. This change is already done in IG.
>
> You cannot place a "SHOULD" on the SG's interaction with the SS7 network
> (directing traffic).  That is why it says "MAY" and must remain as "MAY".
> [TOLGA]The sentence is, "An SGP or IPSP, upon reception of an ASP Active
> message for the first
>    ASP in a Loadshare AS, MAY choose not to direct traffic to a newly
> active ASP until it determines that there are sufficient resources to
> handle the expected load (e.g., until there are "n" ASPs in state
> ASP-ACTIVE in the AS)." (In IG MAY is changed to SHOULD) It is not about
> directing traffic to SS7 network but to ASPs. Why is this considered as
an
> interaction of SG with SS7 network? Or how is this different than saying

See example from previous note.  The SG needs to interact with SS7 to
direct
traffic to an ASP when an AS first becomes active.  Where else is it to get
the traffic that was previously suppressed?
[TOLGA]Of course SG needs to interact with SS7 network. My point is the
corresponding sentence is about directing traffic to ASP. For example
should it be considered as an intervention to the SG/SS7 relationship that
SG shouldn't divert traffic unless there is one ACTIVE ASP for non n+k
case? I don't think so.

Also, the SG could take immediate action to direct traffic that will not
appear for some time depending on the RK the AS is configured for.  For
example, ISUP will only test remote user part availability periodically.
The SG could direct traffic immediately and the traffic could show up
several
minutes later.
[TOLGA]Yes. But so what?

It is even impossible to talk about the traffic that the SG is "directing"
without giving consideration to SS7 users involved.  The only way the SG
can "direct" any SS7 traffic in this situation is by taking actions within
the scope of SS7 and outside the scope of M3UA.

You cannot say SHOULD.

Leave the RFC as it stands.

> that for non n+k case SGP should not direct traffic unless there is one
> ACTIVE ASP? And as I said, opening the path unidirectionally is not the
> best choice in my opinion and should be avoided.

I already explained to you: there is no "unidirectional path".  This
is not a switch that you turn on an the current flows.
[TOLGA]If SG can send traffic to ASP but ASP can't send to SG, I consider
this as an unidirectional path.

--brian


--
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 09:06:43 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA05671
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 09:06:43 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QEGEr11413
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 09:16:14 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QEFlp11368;
	Wed, 26 Feb 2003 09:15:47 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QECwp11175
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 09:12:58 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA05461
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 09:02:51 -0500 (EST)
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF6D1BB2B3.ADA68F4A-ONC2256CD9.004CFC97@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Wed, 26 Feb 2003 16:03:15 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/26/2003
 04:03:23 PM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,



Samuel,

Samuel Dur D. Jeyaseelan wrote:
(Wed, 26 Feb 2003 06:06:29)
>
> Brian,
> I understood the same.But there may be a time where SG can request k ASP
> to become active in order to handle the heavy traffic.
> Because configured 'n' ASPs are sufficient to handle the normal traffic
> but when traffic increases than 'normal traffic' from SS7 network towards
> SG- this case SG may request k ASP to become active.

The SG has no understanding of the traffic handling capability of the ASPs.

The only thing it has in the number "n".
[TOLGA]If SG knows only about "n" (and nothing else) how can it decide to
direct traffic or not? In such a case, isn't the only option for SG to wait
for n ASPs to be ACTIVE? Otherwise, which other paramaters are playing a
role for that decision?








_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 09:22:06 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA06458
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 09:22:06 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QEVbB12194
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 09:31:37 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QEVMp12178;
	Wed, 26 Feb 2003 09:31:22 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QEUqp12105
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 09:30:52 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA06206
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 09:20:48 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:Pb6RprL0qvHRACFEcybeTM3hZIuRuZrV@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1QEOgD12597;
	Wed, 26 Feb 2003 07:24:42 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1QEOgk15941;
	Wed, 26 Feb 2003 07:24:42 -0700
Date: Wed, 26 Feb 2003 07:24:42 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030226072442.B15127@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OFF843FE7A.129828A1-ONC2256CD9.004B7333@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OFF843FE7A.129828A1-ONC2256CD9.004B7333@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Wed, Feb 26, 2003 at 04:00:29PM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

TOLGA,

TOLGA ASVEREN wrote:                                                                 (Wed, 26 Feb 2003 16:00:29)
> 
> Brian,
> 
> 
> 
> 
> TOLGA,
> 
> TOLGA ASVEREN wrote:
> (Wed, 26 Feb 2003 14:20:39)
> >
> > Brian,
> >
> >
> > TOLGA,
> >
> > TOLGA ASVEREN wrote:
> > (Wed, 26 Feb 2003 12:23:56)
> > >
> > > Brian,
> > >
> > >
> > >
> > >
> > > TOLGA,
> > >
> > > I agree with John Schantz' original objection to this proposal.  The
> > Notify
> > > message is advisory only and that requiring a response from the ASP to
> a
> > > Notify message is too drastic a change to the protocol.
> > > [TOLGA]I agree that it is not a minor change, especially just before
> the
> > > text was fozen it was not a good option. That is the reason, why the
> > > sentence saying that ASPs SHOULD coordinate among each other number of
> > > ACTIVE ASPs is included at the last minute. If people still don't want
> to
> > > change advisory meaning of NTFY, then for starting traffic, both ASP
> and
> > > SGP SHOULD wait till n ASPs are ACTIVE. So, MAY for SG should be
> changed
> > to
> > > SHOULD. This change is already done in IG.
> >
> > You cannot place a "SHOULD" on the SG's interaction with the SS7 network
> > (directing traffic).  That is why it says "MAY" and must remain as "MAY".
> > [TOLGA]The sentence is, "An SGP or IPSP, upon reception of an ASP Active
> > message for the first
> >    ASP in a Loadshare AS, MAY choose not to direct traffic to a newly
> > active ASP until it determines that there are sufficient resources to
> > handle the expected load (e.g., until there are "n" ASPs in state
> > ASP-ACTIVE in the AS)." (In IG MAY is changed to SHOULD) It is not about
> > directing traffic to SS7 network but to ASPs. Why is this considered as
> an
> > interaction of SG with SS7 network? Or how is this different than saying
> 
> See example from previous note.  The SG needs to interact with SS7 to
> direct
> traffic to an ASP when an AS first becomes active.  Where else is it to get
> the traffic that was previously suppressed?
> [TOLGA]Of course SG needs to interact with SS7 network. My point is the
> corresponding sentence is about directing traffic to ASP. For example
> should it be considered as an intervention to the SG/SS7 relationship that
> SG shouldn't divert traffic unless there is one ACTIVE ASP for non n+k
> case? I don't think so.
> 
> Also, the SG could take immediate action to direct traffic that will not
> appear for some time depending on the RK the AS is configured for.  For
> example, ISUP will only test remote user part availability periodically.
> The SG could direct traffic immediately and the traffic could show up
> several
> minutes later.
> [TOLGA]Yes. But so what?

A requirement cannot be placed on the SG that is not visible to the ASP
because it is not a requirement for interoperability.  The ASP cannot
tell when the SG directs traffic, so it cannot care.

> 
> It is even impossible to talk about the traffic that the SG is "directing"
> without giving consideration to SS7 users involved.  The only way the SG
> can "direct" any SS7 traffic in this situation is by taking actions within
> the scope of SS7 and outside the scope of M3UA.
> 
> You cannot say SHOULD.
> 
> Leave the RFC as it stands.
> 
> > that for non n+k case SGP should not direct traffic unless there is one
> > ACTIVE ASP? And as I said, opening the path unidirectionally is not the
> > best choice in my opinion and should be avoided.
> 
> I already explained to you: there is no "unidirectional path".  This
> is not a switch that you turn on an the current flows.
> [TOLGA]If SG can send traffic to ASP but ASP can't send to SG, I consider
> this as an unidirectional path.

But the SG is not sending anything; it is only establishing the capability
to send traffic some time in the future after SS7 decides to actually send
something.  If the stimulus is from the ASP, this won't be until the ASP
sends something first.  Also, if the stimulus is from SS7, the ASP is still
under no obligation to process any DATA message received while it is waiting
for n ASPs to start.

What horrible things do you think would happen anyway?

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 09:23:47 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA06635
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 09:23:47 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QEXJq12317
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 09:33:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QEX3p12288;
	Wed, 26 Feb 2003 09:33:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QEWtp12254
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 09:32:55 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA06579
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 09:22:53 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:LZogShtmzbNjHh9VXOXxQNkzeqYbjcdn@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1QEQlD12637;
	Wed, 26 Feb 2003 07:26:47 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1QEQl416019;
	Wed, 26 Feb 2003 07:26:47 -0700
Date: Wed, 26 Feb 2003 07:26:47 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030226072647.C15127@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OF6D1BB2B3.ADA68F4A-ONC2256CD9.004CFC97@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF6D1BB2B3.ADA68F4A-ONC2256CD9.004CFC97@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Wed, Feb 26, 2003 at 04:03:15PM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

TOLGA,

TOLGA ASVEREN wrote:                                                                 (Wed, 26 Feb 2003 16:03:15)
> 
> Brian,
> 
> 
> 
> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:
> (Wed, 26 Feb 2003 06:06:29)
> >
> > Brian,
> > I understood the same.But there may be a time where SG can request k ASP
> > to become active in order to handle the heavy traffic.
> > Because configured 'n' ASPs are sufficient to handle the normal traffic
> > but when traffic increases than 'normal traffic' from SS7 network towards
> > SG- this case SG may request k ASP to become active.
> 
> The SG has no understanding of the traffic handling capability of the ASPs.
> 
> The only thing it has in the number "n".
> [TOLGA]If SG knows only about "n" (and nothing else) how can it decide to
> direct traffic or not? In such a case, isn't the only option for SG to wait
> for n ASPs to be ACTIVE? Otherwise, which other paramaters are playing a
> role for that decision?

The other option is for the SG not to wait.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 09:35:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA07397
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 09:35:56 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QEjSB13974
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 09:45:28 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QEj8p13946;
	Wed, 26 Feb 2003 09:45:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QEimp13903
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 09:44:48 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA07369
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 09:34:45 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:EmPPxKyXk3gnUaw4J0f5PyWAtycniC3K@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1QEcdD12859;
	Wed, 26 Feb 2003 07:38:39 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1QEccg16340;
	Wed, 26 Feb 2003 07:38:38 -0700
Date: Wed, 26 Feb 2003 07:38:38 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030226073838.D15127@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OF6D1BB2B3.ADA68F4A-ONC2256CD9.004CFC97@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF6D1BB2B3.ADA68F4A-ONC2256CD9.004CFC97@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Wed, Feb 26, 2003 at 04:03:15PM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

TOLGA,

Please note that the option in the RFC states that the "SGP" MAY wait,
not that the "SG" MAY wait.  The SG might have the AS active for more
than n ASPs with other SGPs, or more than n ASPs combined across the
SGPs.

Even if there wasn't SS7 considerations, saying that an SGP "SHOULD"
wait would be foolish, particularly if all the APSs are merrily
handling AS traffic via other SGPs in the same SG.

Saying the "SG" "SHOULD" wait would place an undue requirement on the
SGP to make its decision based upon their peer SGP's (possibly incorrect
or untimely) view all ASPs state for the AS.

--brian

TOLGA ASVEREN wrote:                            (Wed, 26 Feb 2003 16:03:15)
> 
> Brian,
> 
> 
> 
> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:
> (Wed, 26 Feb 2003 06:06:29)
> >
> > Brian,
> > I understood the same.But there may be a time where SG can request k ASP
> > to become active in order to handle the heavy traffic.
> > Because configured 'n' ASPs are sufficient to handle the normal traffic
> > but when traffic increases than 'normal traffic' from SS7 network towards
> > SG- this case SG may request k ASP to become active.
> 
> The SG has no understanding of the traffic handling capability of the ASPs.
> 
> The only thing it has in the number "n".
> [TOLGA]If SG knows only about "n" (and nothing else) how can it decide to
> direct traffic or not? In such a case, isn't the only option for SG to wait
> for n ASPs to be ACTIVE? Otherwise, which other paramaters are playing a
> role for that decision?
> 
> 
> 
> 
> 
> 
> 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 09:54:27 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA08578
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 09:54:27 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QF3xD15120
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 10:03:59 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QF3bp15103;
	Wed, 26 Feb 2003 10:03:38 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QF2Ip15036
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 10:02:18 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA08360
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 09:52:13 -0500 (EST)
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OFD7C5E742.85B8434E-ONC2256CD9.00513854@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Wed, 26 Feb 2003 16:52:37 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/26/2003
 04:52:41 PM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,


TOLGA,

Please note that the option in the RFC states that the "SGP" MAY wait,
not that the "SG" MAY wait.  The SG might have the AS active for more
than n ASPs with other SGPs, or more than n ASPs combined across the
SGPs.
[TOLGA]Yes, it is SGP not SG.

Even if there wasn't SS7 considerations, saying that an SGP "SHOULD"
wait would be foolish, particularly if all the APSs are merrily
handling AS traffic via other SGPs in the same SG.
[TOLGA]With SHOULD, other SGPs also SHOULD NOT direct traffic unless there
are n ACTIVE ASPs(for that specific AS). For other AS, of course any SGP
can divert traffic -depending on the traffic mode of the corresponding AS-.
So, why should a "SHOULD" create an inconsistant behaviour among SGPs
belonging to same SG if all of them follow the rule?

Saying the "SG" "SHOULD" wait would place an undue requirement on the
SGP to make its decision based upon their peer SGP's (possibly incorrect
or untimely) view all ASPs state for the AS.
[TOLGA]As you pointed out at the beginning it should be SGP.

--brian

TOLGA ASVEREN wrote:                            (Wed, 26 Feb 2003 16:03:15)
>
> Brian,
>
>
>
> Samuel,
>
> Samuel Dur D. Jeyaseelan wrote:
> (Wed, 26 Feb 2003 06:06:29)
> >
> > Brian,
> > I understood the same.But there may be a time where SG can request k
ASP
> > to become active in order to handle the heavy traffic.
> > Because configured 'n' ASPs are sufficient to handle the normal traffic
> > but when traffic increases than 'normal traffic' from SS7 network
towards
> > SG- this case SG may request k ASP to become active.
>
> The SG has no understanding of the traffic handling capability of the
ASPs.
>
> The only thing it has in the number "n".
> [TOLGA]If SG knows only about "n" (and nothing else) how can it decide to
> direct traffic or not? In such a case, isn't the only option for SG to
wait
> for n ASPs to be ACTIVE? Otherwise, which other paramaters are playing a
> role for that decision?
>
>
>
>
>
>
>
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 10:09:48 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA09868
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 10:09:47 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QFJK816624
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 10:19:20 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QFJ9p16602;
	Wed, 26 Feb 2003 10:19:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QFIFp16560
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 10:18:15 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA09636
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 10:08:11 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:tKbIwalaI0sUNqSF+FiX1keXlJ1na4r7@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1QFC5D13394;
	Wed, 26 Feb 2003 08:12:05 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1QFC5l17216;
	Wed, 26 Feb 2003 08:12:05 -0700
Date: Wed, 26 Feb 2003 08:12:05 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030226081205.A17011@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OFD7C5E742.85B8434E-ONC2256CD9.00513854@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OFD7C5E742.85B8434E-ONC2256CD9.00513854@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Wed, Feb 26, 2003 at 04:52:37PM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

TOLGA,

TOLGA ASVEREN wrote:                                                                 (Wed, 26 Feb 2003 16:52:37)
> 
> Brian,
> 
> 
> TOLGA,
> 
> Please note that the option in the RFC states that the "SGP" MAY wait,
> not that the "SG" MAY wait.  The SG might have the AS active for more
> than n ASPs with other SGPs, or more than n ASPs combined across the
> SGPs.
> [TOLGA]Yes, it is SGP not SG.
> 
> Even if there wasn't SS7 considerations, saying that an SGP "SHOULD"
> wait would be foolish, particularly if all the APSs are merrily
> handling AS traffic via other SGPs in the same SG.
> [TOLGA]With SHOULD, other SGPs also SHOULD NOT direct traffic unless there
> are n ACTIVE ASPs(for that specific AS).

Huh?  The decision then cannot possibly be then related to ASP load (and
therefore unrelated to 'n').  This certainly cannot be recommended.


> For other AS, of course any SGP can divert traffic -depending on the traffic
> mode of the corresponding AS-.  So, why should a "SHOULD" create an
> inconsistant behaviour among SGPs belonging to same SG if all of them follow
> the rule?

Because there is no point for any SGP but the first one to follow the rule,
even if one could find justification to impose it (which you cannot).
Placing a requirement to make your rule look pretty is not in fitting with
RFC 2119.

> 
> Saying the "SG" "SHOULD" wait would place an undue requirement on the
> SGP to make its decision based upon their peer SGP's (possibly incorrect
> or untimely) view all ASPs state for the AS.
> [TOLGA]As you pointed out at the beginning it should be SGP.

Huh? Did you mean SG?

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 10:26:54 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA11232
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 10:26:54 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QFaR317828
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 10:36:27 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QFaBp17820;
	Wed, 26 Feb 2003 10:36:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QFZop17783
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 10:35:50 -0500
Received: from rainier.illuminet.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA11106
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 10:25:44 -0500 (EST)
Received: from olwinexsmtp01.corp.illuminet.com (olwinexsmtp01.corp.illuminet.com [10.55.13.9]) by rainier.illuminet.com (8.8.8/8.8.8) with ESMTP id HAA18483; Wed, 26 Feb 2003 07:29:37 -0800 (PST)
Received: by olwinexsmtp01.corp.illuminet.com with Internet Mail Service (5.5.2656.59)
	id <192Q7JT3>; Wed, 26 Feb 2003 07:29:37 -0800
Message-ID: <90B62898D7B43B448D1A7297980A8C5F6CE600@opwinex01.corp.illuminet.com>
From: Kevin McCandless <KMcCandless@verisign.com>
To: "'Ong, Lyndon'" <LyOng@ciena.com>,
        "'sigtran@ietf.org'"
	 <sigtran@ietf.org>
Subject: RE: [Sigtran] meeting time moved
Date: Wed, 26 Feb 2003 07:29:30 -0800
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2656.59)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Lyndon:

Some of us are leaving Thursday night.  Any chance of having the meeting
earlier in the week?

Kevin.....

> -----Original Message-----
> From: Ong, Lyndon [mailto:LyOng@ciena.com]
> Sent: Monday, February 24, 2003 5:33 PM
> To: 'sigtran@ietf.org'
> Subject: [Sigtran] meeting time moved
> 
> 
> Hi Folks,
> 
> The Sigtran WG Meeting time has been moved to Friday,
> please be advised.
> 
> Cheers,
> 
> L. Ong
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 14:40:38 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA21290
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 14:40:38 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QJoG505037
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 14:50:16 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QJnxp05011;
	Wed, 26 Feb 2003 14:49:59 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QJk7p04868
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 14:46:07 -0500
Received: from revere.sonusnet.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA21121
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 14:35:58 -0500 (EST)
Received: from sonusdc3.sonusnet.com (sonusdc3.sonusnet.com [10.128.32.53])
	by revere.sonusnet.com (Switch-2.1.0/Switch-2.1.0) with ESMTP id h1QJdc715517
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 14:39:38 -0500 (EST)
Received: by sonusdc3.sonusnet.com with Internet Mail Service (5.5.2653.19)
	id <FWSKDKRY>; Wed, 26 Feb 2003 14:39:52 -0500
Message-ID: <4CF48720B6E9D4118D040060CF2074E9072F6C4F@sonusdc3.sonusnet.com>
From: "Wung, Joseph" <JWung@sonusnet.com>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Date: Wed, 26 Feb 2003 14:39:51 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Sigtran] m3ua: ASP Active w/o RC
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

In section 4.3.4.3 of M3UA, ASP Active Procedures

	"In the case where an ASP Active message does 
	not contain a Routing Context parameter, the receiver 
	must know, via configuration data, which Application 
	Server(s) the ASP is a member." 

Does this rule apply to the dynamically registered routing keys?

Joe
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 14:46:46 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA21487
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 14:46:46 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QJuNj05283
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 14:56:23 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QJuBp05267;
	Wed, 26 Feb 2003 14:56:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QJtep05234
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 14:55:40 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA21444
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 14:45:31 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:FF2lbMXYMTQ2VX15iU+bE2JCXURRncmN@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1QJnQD18007;
	Wed, 26 Feb 2003 12:49:26 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1QJnPK26604;
	Wed, 26 Feb 2003 12:49:25 -0700
Date: Wed, 26 Feb 2003 12:49:25 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Wung, Joseph" <JWung@sonusnet.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] m3ua: ASP Active w/o RC
Message-ID: <20030226124925.B26366@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Wung, Joseph" <JWung@sonusnet.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <4CF48720B6E9D4118D040060CF2074E9072F6C4F@sonusdc3.sonusnet.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <4CF48720B6E9D4118D040060CF2074E9072F6C4F@sonusdc3.sonusnet.com>; from JWung@sonusnet.com on Wed, Feb 26, 2003 at 02:39:51PM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Joeseph,

Wung, Joseph wrote:                                                                  (Wed, 26 Feb 2003 14:39:51)
> In section 4.3.4.3 of M3UA, ASP Active Procedures
> 
> 	"In the case where an ASP Active message does 
> 	not contain a Routing Context parameter, the receiver 
> 	must know, via configuration data, which Application 
> 	Server(s) the ASP is a member." 
> 
> Does this rule apply to the dynamically registered routing keys?

What has dynamically registered routing keys go to do with it?

I think it is pretty clear that if there is no RC in the ASPAC that
the SG has to know somehow which AS the message corresonds to.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 14:56:47 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA21842
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 14:56:47 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1QK6PF05793
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 15:06:25 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QK6Ap05771;
	Wed, 26 Feb 2003 15:06:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1QK53p05728
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 15:05:03 -0500
Received: from revere.sonusnet.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA21810
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 14:54:54 -0500 (EST)
Received: from sonusdc3.sonusnet.com (sonusdc3.sonusnet.com [10.128.32.53])
	by revere.sonusnet.com (Switch-2.1.0/Switch-2.1.0) with ESMTP id h1QJwR717376;
	Wed, 26 Feb 2003 14:58:27 -0500 (EST)
Received: by sonusdc3.sonusnet.com with Internet Mail Service (5.5.2653.19)
	id <FWSKDKVR>; Wed, 26 Feb 2003 14:58:41 -0500
Message-ID: <4CF48720B6E9D4118D040060CF2074E9072F6C50@sonusdc3.sonusnet.com>
From: "Wung, Joseph" <JWung@sonusnet.com>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: RE: [Sigtran] m3ua: ASP Active w/o RC
Date: Wed, 26 Feb 2003 14:58:41 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Brian,

The problem I have is that "via configuration data" in the rule.
My interpretation of "configuration data" in this case is that
it only refers to the routing keys that have been statically 
registered, and does NOT refer to those already dynamically 
registered.  Am I being wrong?

Joe

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Wednesday, February 26, 2003 2:49 PM
> To: Wung, Joseph
> Cc: 'sigtran@ietf.org'
> Subject: Re: [Sigtran] m3ua: ASP Active w/o RC
> 
> 
> Joeseph,
> 
> Wung, Joseph wrote:                                           
>                        (Wed, 26 Feb 2003 14:39:51)
> > In section 4.3.4.3 of M3UA, ASP Active Procedures
> > 
> > 	"In the case where an ASP Active message does 
> > 	not contain a Routing Context parameter, the receiver 
> > 	must know, via configuration data, which Application 
> > 	Server(s) the ASP is a member." 
> > 
> > Does this rule apply to the dynamically registered routing keys?
> 
> What has dynamically registered routing keys go to do with it?
> 
> I think it is pretty clear that if there is no RC in the ASPAC that
> the SG has to know somehow which AS the message corresonds to.
> 
> --brian
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 20:03:16 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA04550
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 20:03:16 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1R1D2V27186
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 20:13:02 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1R1Cgp27148;
	Wed, 26 Feb 2003 20:12:44 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1R1B8p27091
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 20:11:08 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA04504
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 20:00:50 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:65pc41RYWdrqBhfS2znAZxfEf7KZznt5@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1R14lD22658;
	Wed, 26 Feb 2003 18:04:47 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1R14ki01385;
	Wed, 26 Feb 2003 18:04:46 -0700
Date: Wed, 26 Feb 2003 18:04:46 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Wung, Joseph" <JWung@sonusnet.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] m3ua: ASP Active w/o RC
Message-ID: <20030226180446.B1059@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Wung, Joseph" <JWung@sonusnet.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <4CF48720B6E9D4118D040060CF2074E9072F6C50@sonusdc3.sonusnet.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <4CF48720B6E9D4118D040060CF2074E9072F6C50@sonusdc3.sonusnet.com>; from JWung@sonusnet.com on Wed, Feb 26, 2003 at 02:58:41PM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Joseph,

Configuration of the SG is largely outside the scope of the RFC (but would
be within the scope of the M3UA-MIB).  Whether a key is statically or
dynamically registered allocated shouldn't matter.  The ASP is not going
to use the RC value in either case.

Was the ASP going to register a single RK and then activate it without
giving reference to the returned RC?  What would that gain?

--brian

Wung, Joseph wrote:                                                                  (Wed, 26 Feb 2003 14:58:41)
> Brian,
> 
> The problem I have is that "via configuration data" in the rule.
> My interpretation of "configuration data" in this case is that
> it only refers to the routing keys that have been statically 
> registered, and does NOT refer to those already dynamically 
> registered.  Am I being wrong?
> 
> Joe
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: Wednesday, February 26, 2003 2:49 PM
> > To: Wung, Joseph
> > Cc: 'sigtran@ietf.org'
> > Subject: Re: [Sigtran] m3ua: ASP Active w/o RC
> > 
> > 
> > Joeseph,
> > 
> > Wung, Joseph wrote:                                           
> >                        (Wed, 26 Feb 2003 14:39:51)
> > > In section 4.3.4.3 of M3UA, ASP Active Procedures
> > > 
> > > 	"In the case where an ASP Active message does 
> > > 	not contain a Routing Context parameter, the receiver 
> > > 	must know, via configuration data, which Application 
> > > 	Server(s) the ASP is a member." 
> > > 
> > > Does this rule apply to the dynamically registered routing keys?
> > 
> > What has dynamically registered routing keys go to do with it?
> > 
> > I think it is pretty clear that if there is no RC in the ASPAC that
> > the SG has to know somehow which AS the message corresonds to.
> > 
> > --brian
> > 
> > -- 
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > 

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 23:13:39 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA07548
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 23:13:39 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1R4NSs05528
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 23:23:28 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1R4N4p05515;
	Wed, 26 Feb 2003 23:23:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1R4LGp05475
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 23:21:16 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA07519
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 23:10:55 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1R4EmeY025297
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 22:14:50 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h1R4DvF21351
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 22:13:57 -0600 (CST)
Date: Wed, 26 Feb 2003 22:13:57 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
In-Reply-To: <20030226051301.I8776@openss7.org>
Message-ID: <Pine.SOL.4.10.10302262151430.12860-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

On Wed, 26 Feb 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Wed, 26 Feb 2003 06:06:29)
> > 
> > Brian,
> > I understood the same.But there may be a time where SG can request k ASP
> > to become active in order to handle the heavy traffic.
> > Because configured 'n' ASPs are sufficient to handle the normal traffic
> > but when traffic increases than 'normal traffic' from SS7 network towards
> > SG- this case SG may request k ASP to become active.
> 
> The SG has no understanding of the traffic handling capability of the ASPs.
I beleive SGP uses a single transmission buffer and re-transmission buffer
towards 'n' ASPs.
SGP measures these buffers to understand 100% traffic handling
capability of 'n' ASPS .When these bufers are overflowing, SGP can send
NTFY to 'k' ASP.

From ASP point of view, ASP sends SCON when congestion occurs, assume at
SGP transmission buffer and re-transmission buffers are overflowing, in
this case Should SGP stop traffic towards 'n' ASPs ? or Can SGP send NTFY 
to 'k' ASP to become active. Sending NTFY to 'k' ASP is the best
approach-this is my understanding.

--Samuel
> 
> The only thing it has in the number "n".
> 
> "n" is the numer of ASPs required to handle the maximum engineered (expected)
> load, (not the "normal" load).
> 
> The SG does not (and cannot be required to) monitor any traffic load and even
> if it did, it cannot translate this into ASP handling capacity.
> 
> The ASPs should coordinate how many of them are active with consideration for
> their own process loading.
> 
> The RFC is fine as it is.
> 
> --brian
> 
> > 
> > --Samuel
> > 
> > 
> >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > 
> > On Wed, 26 Feb 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > The RFC says that the ASPs should coordinate between themselves the number
> > > of active ASPs.  A k ASP is welcome to activate whenever it wants.  The
> > > SGP cannot invoke the activation of an ASP.
> > > 
> > > --brian
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                                                     (Wed, 26 Feb 2003 04:17:31)
> > > > 
> > > > Hi,
> > > > 
> > > > Why there is no discussion when traffic reaches from 'T' to 'T+1'?
> > > > (as mail below).
> > > > Is it not possible to have 'T+1' taffic?
> > > > 
> > > > --Samuel
> > > > 
> > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > >   1000 Coit Road, Plano, Texas 75075
> > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > 
> > > > On Wed, 19 Feb 2003, Samuel Dur D. Jeyaseelan wrote:
> > > > 
> > > > > 
> > > > > Hi,
> > > > > 
> > > > > I beleive 'n' is minimum number of ASPs to handle the 'best traffic T'.
> > > > > To handle the 'worst traffic T+1', SG may require 'n+1' ASPs.
> > > > > 
> > > > > 'n'is the derived value based on the traffic report given by Traffic
> > > > > Metering and Measurement.
> > > > > 
> > > > > So 'n' is fed during configuration of stack.When traffic increases from
> > > > > 'T' to 'T+1' or # APSs < n - this case SG is required to send NOTIFY to
> > > > > 'k' ASP.
> > > > > 
> > > > > And one more thing when # APSs < n, SG still can handle the traffic with
> > > > > single ASP.
> > > > > 
> > > > > When 'n' ASPs are active
> > > > > ------------------------
> > > > > 
> > > > > SLS-X  ---->ASP1
> > > > > SLS-Y  ---->ASP2
> > > > > 
> > > > > When ASP1 goes down
> > > > > -------------------
> > > > > 
> > > > > SLS-(Y,X)  ----->ASP2
> > > > > 
> > > > > Instead of banning the complete traffic ,it can be handled (not 100%) by
> > > > > ASP2.
> > > > > 
> > > > > 
> > > > > --Samuel
> > > > > 
> > > > >   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
> > > > >   1000 Coit Road, Plano, Texas 75075
> > > > >   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> > > > > 
> > > > > On Tue, 18 Feb 2003, Javier Pastor-Balbas (ECE) wrote:
> > > > > 
> > > > > > Brian, 
> > > > > > 
> > > > > > I agree 
> > > > > > 
> > > > > > RFC simply says: 
> > > > > > 
> > > > > > 	"n ASPs is the minimum number of redundant ASPs required to handle traffic"
> > > > > > 
> > > > > > As this is the minimum number, it can be inferred that for #ASPs < n no traffic can be handled. Nothing is mentioned in the RFC about engineered load although it may be thought that n value calculation is based on that.  
> > > > > > 
> > > > > > I would say that the current model does not support the use of extra ASPs for loads bigger than the engineered nor the use of less than n for smaller loads... The message is clear saying that #ASPs<n implies no traffic, i.e. whenever you can not support the engineered load, stop the system. This last sentence is not in the RFC but it can be deduced as the engineered load for n can... 
> > > > > > 
> > > > > > Cheers // Javier. 
> > > > > > 
> > > > > > > -----Original Message-----
> > > > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > > > Sent: Tuesday, February 18, 2003 12:49 PM
> > > > > > > To: Javier Pastor-Balbas (ECE)
> > > > > > > Cc: sigtran@ietf.org
> > > > > > > Subject: Re: [Sigtran] Re: M3UA Notify message
> > > > > > > 
> > > > > > > 
> > > > > > > Javier,
> > > > > > > 
> > > > > > > Javier Pastor-Balbas (ECE) wrote:             (Tue, 18 Feb 
> > > > > > > 2003 12:17:54)
> > > > > > > > > Javier Pastor-Balbas (ECE) wrote:             (Tue, 18 
> > > > > > > Feb 2003 07:57:23)
> > > > > > > > > > 
> > > > > > > > > > Hi Brian, 
> > > > > > > > > > 
> > > > > > > > > > According to the RFC "n" is required to handle traffic 
> > > > > > > with minimum
> > > > > > > > > > losses, what is expected from the SS7 point of view in 
> > > > > > > an interworking
> > > > > > > > > > scenario.
> > > > > > > 
> > > > > > > Please cite this passage, because I cannot find it.  I find 
> > > > > > > no mention of
> > > > > > > minimum number of ASPs and "message loss".  The only 
> > > > > > > reference to message loss
> > > > > > > that I know of is unrelated.
> > > > > > > 
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > > > > > 
> > > > > > > > > That would be the numbe required to handle maximum 
> > > > > > > engineered load, which
> > > > > > > > > is the same thing.
> > > > > > > > > 
> > > > > > > > 
> > > > > > > > If you refer to the compromise load engineered to be 
> > > > > > > available 99% (aprox.)
> > > > > > > > of the time, then I agree. But this is a statically value fed at
> > > > > > > > configuration time. And the system won't work till n number 
> > > > > > > is reach since
> > > > > > > > the system will likely have quality contrains.
> > > > > > > 
> > > > > > > The maximum engineered load is a figure established at 
> > > > > > > configuration time.
> > > > > > > 
> > > > > > > Carrier-grade systems must always work with 1 processor.  The 
> > > > > > > system will just
> > > > > > > not necessarily be able to handle the offered load.  (But it might:
> > > > > > > particularly during quiet periods, off-peak, normal-day, etc.)
> > > > > > > 
> > > > > > > The offered load cannot be established at configuration time 
> > > > > > > and varies widely
> > > > > > > for telephony systems.
> > > > > > > 
> > > > > > > You must also realized that the system might, at any time,  
> > > > > > > be offered a load
> > > > > > > greater than the maximum engineered load (mass calling, busy 
> > > > > > > periods, etc.):
> > > > > > > so for carrier-grade systems even "n" processors must 
> > > > > > > implement overload
> > > > > > > controls (and they could not claim SS7 compliance if they didn't).
> > > > > > > 
> > > > > > > Look back in this thread and you will see the factors 
> > > > > > > contributing to the
> > > > > > > anticipated offered load for the start, restart and stop decision.
> > > > > > > 
> > > > > > > The decisions are not based on less than "n" processors not 
> > > > > > > allowing the
> > > > > > > system to work (where n > 1).  They are based on the negative 
> > > > > > > effects of
> > > > > > > delivering pent up traffic to processors that will likely 
> > > > > > > just congest.
> > > > > > > 
> > > > > > > --brian
> > > > > > > 
> > > > > > > -- 
> > > > > > > Brian F. G. Bidulock
> > > > > > > bidulock@openss7.org
> > > > > > > http://www.openss7.org/
> > > > > > > 
> > > > > > _______________________________________________
> > > > > > Sigtran mailing list
> > > > > > Sigtran@ietf.org
> > > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > > > 
> > > > > 
> > > > > 
> > > > 
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> > > -- 
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > 
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Wed Feb 26 23:44:02 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA07886
	for <sigtran-archive@odin.ietf.org>; Wed, 26 Feb 2003 23:44:02 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1R4rqC07038
	for sigtran-archive@odin.ietf.org; Wed, 26 Feb 2003 23:53:52 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1R4rBp07026;
	Wed, 26 Feb 2003 23:53:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1R4qSp06988
	for <sigtran@optimus.ietf.org>; Wed, 26 Feb 2003 23:52:28 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA07872
	for <sigtran@ietf.org>; Wed, 26 Feb 2003 23:42:08 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:at09k8JrfWbu9zDX5LcF2UnLPmzL9eLH@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1R4k3D26100;
	Wed, 26 Feb 2003 21:46:03 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1R4k3h04131;
	Wed, 26 Feb 2003 21:46:03 -0700
Date: Wed, 26 Feb 2003 21:46:02 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030226214602.A4042@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	sigtran@ietf.org
References: <20030226051301.I8776@openss7.org> <Pine.SOL.4.10.10302262151430.12860-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302262151430.12860-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Wed, Feb 26, 2003 at 10:13:57PM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Wed, 26 Feb 2003 22:13:57)
> 
> Brian,
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> On Wed, 26 Feb 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Wed, 26 Feb 2003 06:06:29)
> > > 
> > > Brian,
> > > I understood the same.But there may be a time where SG can request k ASP
> > > to become active in order to handle the heavy traffic.
> > > Because configured 'n' ASPs are sufficient to handle the normal traffic
> > > but when traffic increases than 'normal traffic' from SS7 network towards
> > > SG- this case SG may request k ASP to become active.
> > 
> > The SG has no understanding of the traffic handling capability of the ASPs.
> I beleive SGP uses a single transmission buffer and re-transmission buffer
> towards 'n' ASPs.
> SGP measures these buffers to understand 100% traffic handling
> capability of 'n' ASPS .When these bufers are overflowing, SGP can send
> NTFY to 'k' ASP.

By monitoring its send buffer depth (which is difficult on socket based
SCTP APIs in the first place) the SGP cannot distinguish between queueing
due to IP network congestion, retransmission, inter-stream blocking,
undetected outage, or peer congestion.

Regardless, the SGP cannot and should not activate an ASP.  The ASPs should
coordinate activation between themselves in the n+k scenario.

> 
> >From ASP point of view, ASP sends SCON when congestion occurs, assume at
> SGP transmission buffer and re-transmission buffers are overflowing, in
> this case Should SGP stop traffic towards 'n' ASPs ? or Can SGP send NTFY 
> to 'k' ASP to become active. Sending NTFY to 'k' ASP is the best
> approach-this is my understanding.

The ASP is not required to send SCON at any time.  Nor is the SGP required
to do anthing with SCON received from the ASP.

The SG is responsible for translating its view of SCTP congestion events into
any necessary SS7 events.

--brian


-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 27 00:47:39 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA09042
	for <sigtran-archive@odin.ietf.org>; Thu, 27 Feb 2003 00:47:39 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1R5vUk10306
	for sigtran-archive@odin.ietf.org; Thu, 27 Feb 2003 00:57:30 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1R5vJp10286;
	Thu, 27 Feb 2003 00:57:19 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1R5oYp10098
	for <sigtran@optimus.ietf.org>; Thu, 27 Feb 2003 00:50:34 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA08916
	for <sigtran@ietf.org>; Thu, 27 Feb 2003 00:40:11 -0500 (EST)
Subject: Re: [Sigtran] Re: M3UA Notify message
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF2B15DBA8.C6542F6A-ONC2256CDA.001D06D8@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Thu, 27 Feb 2003 07:40:37 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/27/2003
 07:40:40 AM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Brian,

Briefly what I think is(not to convince you, just the story from my side):
I don't see a problem with interpreting n as mandatory on SGP side while
starting traffic, IMO functionally it is not different from non n+k case
(or from the case where n=1). I would think SGP will decide to expose AS
available to SS7 network (and I need to add SG will provide the unique view
to the SS7 network, but this is again no different than from n=1 case) when
the path between ASP and SGP is to be opened in both directions. So that
SGP SHOULD wait for n ACTIVE ASPs makes sense for me as it stands in IG.
And I don't think to have an unidirectional open path between ASP and SGP
is nice (even while writing this I am scared :-) ). This interface is
MTP3/MTP3 User interface and IMO should be open in both directions or not
at all. Otherwise you let other nodes in SS7 network send messages to you,
accept their messages and then don't let your user part reply to them.

If we wanted SGP decide dynamically for the number of ACTIVE ASPs to
declare AS as available, IMO the only solution was to use NTFY at ASP side
to start traffic, because as far as I see this is the only way to make both
ASP and SGP see the path between them open at the same time. But seems to
interprete n dynamically is not desired(?).

It seems the real concern was more about transition from ACTIVE to PENDING.
My opinion is rather than saying "the last ASP goes INACTIVE/DOWN", it is
better to let SGP decide by itself based on the number of ACTIVE ASPs.
Probably we need also to think more about other transitions(?).

(And one small answer below, just to clarify)

    Tolga



>
> Saying the "SG" "SHOULD" wait would place an undue requirement on the
> SGP to make its decision based upon their peer SGP's (possibly incorrect
> or untimely) view all ASPs state for the AS.
> [TOLGA]As you pointed out at the beginning it should be SGP.

Huh? Did you mean SG?
[TOLGA]No.





_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 27 01:04:55 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA09317
	for <sigtran-archive@odin.ietf.org>; Thu, 27 Feb 2003 01:04:50 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1R6EfR11734
	for sigtran-archive@odin.ietf.org; Thu, 27 Feb 2003 01:14:41 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1R6EQp11716;
	Thu, 27 Feb 2003 01:14:26 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1R6D8p11640
	for <sigtran@optimus.ietf.org>; Thu, 27 Feb 2003 01:13:08 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA09287
	for <sigtran@ietf.org>; Thu, 27 Feb 2003 01:02:45 -0500 (EST)
Subject: RE: [Sigtran] m3ua: ASP Active w/o RC
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF0FF1E378.66993A5F-ONC2256CDA.0020CA20@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Thu, 27 Feb 2003 08:03:12 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/27/2003
 08:03:14 AM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Joe,


Brian,

The problem I have is that "via configuration data" in the rule.
My interpretation of "configuration data" in this case is that
it only refers to the routing keys that have been statically
registered, and does NOT refer to those already dynamically
registered.  Am I being wrong?
[TOLGA]I think ASPAC with no RC should also activate dynamically registered
RKs (at least I don't see a reason why it shouldn't). The sentence you are
referring to might be interpreted differently (for example like in your
question, that it is applciable only statically configured RKs) but I
believe this sentence is put to the draft quite long time ago(maybe before
dynmic registration was added (?) )or not enough attentaion was paid to
make it clear and nobody complained till now. As I said, I would think
ASPAC with no RC should activate all AS configured for ASP, static and
dynamic.

Joe

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Wednesday, February 26, 2003 2:49 PM
> To: Wung, Joseph
> Cc: 'sigtran@ietf.org'
> Subject: Re: [Sigtran] m3ua: ASP Active w/o RC
>
>
> Joeseph,
>
> Wung, Joseph wrote:
>                        (Wed, 26 Feb 2003 14:39:51)
> > In section 4.3.4.3 of M3UA, ASP Active Procedures
> >
> >        "In the case where an ASP Active message does
> >        not contain a Routing Context parameter, the receiver
> >        must know, via configuration data, which Application
> >        Server(s) the ASP is a member."
> >
> > Does this rule apply to the dynamically registered routing keys?
>
> What has dynamically registered routing keys go to do with it?
>
> I think it is pretty clear that if there is no RC in the ASPAC that
> the SG has to know somehow which AS the message corresonds to.
>
> --brian
>
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
>
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 27 02:28:29 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA21520
	for <sigtran-archive@odin.ietf.org>; Thu, 27 Feb 2003 02:28:29 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1R7cKO27217
	for sigtran-archive@odin.ietf.org; Thu, 27 Feb 2003 02:38:20 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1R7c4p27201;
	Thu, 27 Feb 2003 02:38:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1R7Xtp26360
	for <sigtran@optimus.ietf.org>; Thu, 27 Feb 2003 02:33:55 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA21475
	for <sigtran@ietf.org>; Thu, 27 Feb 2003 02:23:29 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:LhlgjNaHv/PhRIYiTObfJWa1WILAi/V6@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1R7RPD28677;
	Thu, 27 Feb 2003 00:27:25 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1R7RPe06158;
	Thu, 27 Feb 2003 00:27:25 -0700
Date: Thu, 27 Feb 2003 00:27:25 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re: M3UA Notify message
Message-ID: <20030227002725.A6115@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OF2B15DBA8.C6542F6A-ONC2256CDA.001D06D8@LocalDomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OF2B15DBA8.C6542F6A-ONC2256CDA.001D06D8@LocalDomain>; from TOLGA.ASVEREN@o2.com.tr on Thu, Feb 27, 2003 at 07:40:37AM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

TOLGA,

TOLGA ASVEREN wrote:                                                                 (Thu, 27 Feb 2003 07:40:37)
> 
> Brian,
> 
> Briefly what I think is(not to convince you, just the story from my side):
> I don't see a problem with interpreting n as mandatory on SGP side while
> starting traffic, IMO functionally it is not different from non n+k case
> (or from the case where n=1). I would think SGP will decide to expose AS
> available to SS7 network (and I need to add SG will provide the unique view
> to the SS7 network, but this is again no different than from n=1 case) when
> the path between ASP and SGP is to be opened in both directions. So that
> SGP SHOULD wait for n ACTIVE ASPs makes sense for me as it stands in IG.
> And I don't think to have an unidirectional open path between ASP and SGP
> is nice (even while writing this I am scared :-) ).

What bad thing do you think will happpen?

> This interface is MTP3/MTP3 User interface and IMO should be open in both
> directions or not at all. Otherwise you let other nodes in SS7 network send
> messages to you, accept their messages and then don't let your user part
> reply to them.

Discard them if you wish.

> 
> If we wanted SGP decide dynamically for the number of ACTIVE ASPs to
> declare AS as available, IMO the only solution was to use NTFY at ASP side
> to start traffic, because as far as I see this is the only way to make both
> ASP and SGP see the path between them open at the same time. But seems to
> interprete n dynamically is not desired(?).

Your ASP is welcome to discard any messages received from the SG while
waiting for N to become active.

> 
> It seems the real concern was more about transition from ACTIVE to PENDING.
> My opinion is rather than saying "the last ASP goes INACTIVE/DOWN", it is
> better to let SGP decide by itself based on the number of ACTIVE ASPs.
> Probably we need also to think more about other transitions(?).

The SGP can move the AS to the INACTIVE or DOWN state whever it wants.
It needs to perform ASPIA Ack or ASPDN Ack procedures to do so, or drop
the association.

> 
> (And one small answer below, just to clarify)
> 
>     Tolga
> 
> 
> 
> >
> > Saying the "SG" "SHOULD" wait would place an undue requirement on the
> > SGP to make its decision based upon their peer SGP's (possibly incorrect
> > or untimely) view all ASPs state for the AS.
> > [TOLGA]As you pointed out at the beginning it should be SGP.
> 
> Huh? Did you mean SG?
> [TOLGA]No.

Then I couldn't diagree more.

--brian

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 27 06:38:02 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA25774
	for <sigtran-archive@odin.ietf.org>; Thu, 27 Feb 2003 06:38:02 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1RBm0V10280
	for sigtran-archive@odin.ietf.org; Thu, 27 Feb 2003 06:48:00 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1RBlmp10271;
	Thu, 27 Feb 2003 06:47:48 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1RBkPp10237
	for <sigtran@optimus.ietf.org>; Thu, 27 Feb 2003 06:46:25 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA25764
	for <sigtran@ietf.org>; Thu, 27 Feb 2003 06:35:55 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1RBdlqe022639;
	Thu, 27 Feb 2003 05:39:48 -0600 (CST)
Received: from axsppc68 by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id QAA12549; Thu, 27 Feb 2003 16:08:29 -0500 (GMT)
Message-ID: <00d401c2deb5$d09a86b0$7601ff0a@axsppc68>
From: "Subhashini" <sksubha@axes-mach01.usa.alcatel.com>
To: <bidulock@openss7.org>, "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Cc: <sigtran@ietf.org>
References: <OF24FAFD42.687DD8CF-ONC2256CD4.003A2896@LocalDomain> <20030221042755.A25531@openss7.org>
Subject: Re: [Sigtran] Repost - More M2PA 7 questions] - LS OOS during alignment.
Date: Thu, 27 Feb 2003 17:12:54 -0600
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 5.50.4133.2400
X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

The following text in section 4.1.3 and the diagram in section 3 are
contradicting to each other.

Note: Between the time M2PA sends Link Status Alignment to its peer
and receives Link Status Alignment from its peer, M2PA may receive
Link Status Out of Service from its peer. This message is
ignored.
As per diagram, on getting LS OOS (while waiting for LS Alignment), instead
of ignoring the message, the state is changed to OOS.

Which one is correct? Please clarify.

 thanks,

Subhashini ...

----- Original Message -----
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Cc: <sigtran@ietf.org>
Sent: Friday, February 21, 2003 5:27 AM
Subject: Re: [Sigtran] Repost - More M2PA 7 questions]


> TOLGA,
>
> TOLGA ASVEREN wrote:
(Fri, 21 Feb 2003 12:44:30)
> >
> > Brian,
> > --- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
> > > TOLGA,
> > >
> > > TOLGA ASVEREN wrote:
> > >                                 (Fri, 21 Feb 2003
> > > 09:43:22)
> > > >
> > > > Thanks , now it is clear, so that thread was
> > > concluded(sometimes hard to
> > > > decide for this) and we will have a slightly
> > > modified version of Tom's
> > > > initial interpretation, where data is not
> > > discarded but not acknowledged
> > > > either. Then I also would think stopping T7 is
> > > fine.
> > > >
> > > > One more question:
> > > > During changeover, will data received during RPO
> > > but not acknowledged,
> > > > considered as received successfully or not, i.e.
> > > in case of changeover,
> > > > will they be delivered to MTP3 or not?
> > >
> > > Unacknowledged messages must never be delivered to
> > > MTP3 or duplication
> > > will result on changeover.  This is true for Q.703
> > > as well as M2PA.
> > [TOLGA]Why so? I would think, when Retrieve BSTN is received, it might
be
> > good to deliver data received successfully but not delivered to MTP3
> > because of RPO, and then reply to Retrieve BSTN request. This migth save
us
> > to retranmit them on another link (I am not sure though, whether this
will
> > create a problem for MTP3, Q.704 Figure30 Sheet 2 of 6 does not show
what
> > to do in such a case, but I would think this should be fine, I assume it
is
> > possible that MTP2 is delivering a message to MTP3, while MTP3 is
sending
> > Retrieve BSTN. So, MTP3 should be prepared to get something different
than
> > BSTN after sending Retrive BSTN (?) )
>
> Bad idea.
>
> --brian
>
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 27 06:47:22 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA25886
	for <sigtran-archive@odin.ietf.org>; Thu, 27 Feb 2003 06:47:22 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1RBvK210545
	for sigtran-archive@odin.ietf.org; Thu, 27 Feb 2003 06:57:20 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1RBv8p10535;
	Thu, 27 Feb 2003 06:57:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1RBv0p10521
	for <sigtran@optimus.ietf.org>; Thu, 27 Feb 2003 06:57:00 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA25861
	for <sigtran@ietf.org>; Thu, 27 Feb 2003 06:46:30 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:Un1fnN4CKdIYXm3KNnAY6YPYyPUH9ZeP@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1RBoPD01270;
	Thu, 27 Feb 2003 04:50:25 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1RBoPL10409;
	Thu, 27 Feb 2003 04:50:25 -0700
Date: Thu, 27 Feb 2003 04:50:25 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Subhashini <sksubha@axes-mach01.usa.alcatel.com>
Cc: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
Subject: Re: [Sigtran] Repost - More M2PA 7 questions] - LS OOS during alignment.
Message-ID: <20030227045025.A10316@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Subhashini <sksubha@axes-mach01.usa.alcatel.com>,
	TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OF24FAFD42.687DD8CF-ONC2256CD4.003A2896@LocalDomain> <20030221042755.A25531@openss7.org> <00d401c2deb5$d09a86b0$7601ff0a@axsppc68>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <00d401c2deb5$d09a86b0$7601ff0a@axsppc68>; from sksubha@axes-mach01.usa.alcatel.com on Thu, Feb 27, 2003 at 05:12:54PM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Subhashini,

The text is correct, the diagram is not.

--brian

Subhashini wrote:                                                               (Thu, 27 Feb 2003 17:12:54)
> The following text in section 4.1.3 and the diagram in section 3 are
> contradicting to each other.
> 
> Note: Between the time M2PA sends Link Status Alignment to its peer
> and receives Link Status Alignment from its peer, M2PA may receive
> Link Status Out of Service from its peer. This message is
> ignored.
> As per diagram, on getting LS OOS (while waiting for LS Alignment), instead
> of ignoring the message, the state is changed to OOS.
> 
> Which one is correct? Please clarify.
> 
>  thanks,
> 
> Subhashini ...
> 
> ----- Original Message -----
> From: "Brian F. G. Bidulock" <bidulock@openss7.org>
> To: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
> Cc: <sigtran@ietf.org>
> Sent: Friday, February 21, 2003 5:27 AM
> Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
> 
> 
> > TOLGA,
> >
> > TOLGA ASVEREN wrote:
> (Fri, 21 Feb 2003 12:44:30)
> > >
> > > Brian,
> > > --- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
> > > > TOLGA,
> > > >
> > > > TOLGA ASVEREN wrote:
> > > >                                 (Fri, 21 Feb 2003
> > > > 09:43:22)
> > > > >
> > > > > Thanks , now it is clear, so that thread was
> > > > concluded(sometimes hard to
> > > > > decide for this) and we will have a slightly
> > > > modified version of Tom's
> > > > > initial interpretation, where data is not
> > > > discarded but not acknowledged
> > > > > either. Then I also would think stopping T7 is
> > > > fine.
> > > > >
> > > > > One more question:
> > > > > During changeover, will data received during RPO
> > > > but not acknowledged,
> > > > > considered as received successfully or not, i.e.
> > > > in case of changeover,
> > > > > will they be delivered to MTP3 or not?
> > > >
> > > > Unacknowledged messages must never be delivered to
> > > > MTP3 or duplication
> > > > will result on changeover.  This is true for Q.703
> > > > as well as M2PA.
> > > [TOLGA]Why so? I would think, when Retrieve BSTN is received, it might
> be
> > > good to deliver data received successfully but not delivered to MTP3
> > > because of RPO, and then reply to Retrieve BSTN request. This migth save
> us
> > > to retranmit them on another link (I am not sure though, whether this
> will
> > > create a problem for MTP3, Q.704 Figure30 Sheet 2 of 6 does not show
> what
> > > to do in such a case, but I would think this should be fine, I assume it
> is
> > > possible that MTP2 is delivering a message to MTP3, while MTP3 is
> sending
> > > Retrieve BSTN. So, MTP3 should be prepared to get something different
> than
> > > BSTN after sending Retrive BSTN (?) )
> >
> > Bad idea.
> >
> > --brian
> >
> > --
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> >

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 27 07:02:53 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA26146
	for <sigtran-archive@odin.ietf.org>; Thu, 27 Feb 2003 07:02:53 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1RCCpw11872
	for sigtran-archive@odin.ietf.org; Thu, 27 Feb 2003 07:12:51 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1RCCXp11846;
	Thu, 27 Feb 2003 07:12:33 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1RCBFp11809
	for <sigtran@optimus.ietf.org>; Thu, 27 Feb 2003 07:11:15 -0500
Received: from auds952.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA26093
	for <sigtran@ietf.org>; Thu, 27 Feb 2003 07:00:45 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1RC4cqe024629;
	Thu, 27 Feb 2003 06:04:39 -0600 (CST)
Received: from axsppc68 by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id QAA13209; Thu, 27 Feb 2003 16:33:24 -0500 (GMT)
Message-ID: <012301c2deb9$48e94b80$7601ff0a@axsppc68>
From: "Subhashini" <sksubha@axes-mach01.usa.alcatel.com>
To: <bidulock@openss7.org>
Cc: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>, <sigtran@ietf.org>
References: <OF24FAFD42.687DD8CF-ONC2256CD4.003A2896@LocalDomain> <20030221042755.A25531@openss7.org> <00d401c2deb5$d09a86b0$7601ff0a@axsppc68> <20030227045025.A10316@openss7.org>
Subject: Re: [Sigtran] Repost - More M2PA 7 questions] - LS OOS during alignment.
Date: Thu, 27 Feb 2003 17:37:47 -0600
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 5.50.4133.2400
X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

In section 4.1.3, the text "M2PA MAY send additional Link Status Ready
messages while timer T1 is running" is not clear. At what interval this LS
Ready message is to be sent?

thanks,
Subhashini ...

----- Original Message -----
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Subhashini" <sksubha@axes-mach01.usa.alcatel.com>
Cc: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>; <sigtran@ietf.org>
Sent: Thursday, February 27, 2003 5:50 AM
Subject: Re: [Sigtran] Repost - More M2PA 7 questions] - LS OOS during
alignment.


> Subhashini,
>
> The text is correct, the diagram is not.
>
> --brian
>
> Subhashini wrote:
(Thu, 27 Feb 2003 17:12:54)
> > The following text in section 4.1.3 and the diagram in section 3 are
> > contradicting to each other.
> >
> > Note: Between the time M2PA sends Link Status Alignment to its peer
> > and receives Link Status Alignment from its peer, M2PA may receive
> > Link Status Out of Service from its peer. This message is
> > ignored.
> > As per diagram, on getting LS OOS (while waiting for LS Alignment),
instead
> > of ignoring the message, the state is changed to OOS.
> >
> > Which one is correct? Please clarify.
> >
> >  thanks,
> >
> > Subhashini ...
> >
> > ----- Original Message -----
> > From: "Brian F. G. Bidulock" <bidulock@openss7.org>
> > To: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
> > Cc: <sigtran@ietf.org>
> > Sent: Friday, February 21, 2003 5:27 AM
> > Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
> >
> >
> > > TOLGA,
> > >
> > > TOLGA ASVEREN wrote:
> > (Fri, 21 Feb 2003 12:44:30)
> > > >
> > > > Brian,
> > > > --- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
> > > > > TOLGA,
> > > > >
> > > > > TOLGA ASVEREN wrote:
> > > > >                                 (Fri, 21 Feb 2003
> > > > > 09:43:22)
> > > > > >
> > > > > > Thanks , now it is clear, so that thread was
> > > > > concluded(sometimes hard to
> > > > > > decide for this) and we will have a slightly
> > > > > modified version of Tom's
> > > > > > initial interpretation, where data is not
> > > > > discarded but not acknowledged
> > > > > > either. Then I also would think stopping T7 is
> > > > > fine.
> > > > > >
> > > > > > One more question:
> > > > > > During changeover, will data received during RPO
> > > > > but not acknowledged,
> > > > > > considered as received successfully or not, i.e.
> > > > > in case of changeover,
> > > > > > will they be delivered to MTP3 or not?
> > > > >
> > > > > Unacknowledged messages must never be delivered to
> > > > > MTP3 or duplication
> > > > > will result on changeover.  This is true for Q.703
> > > > > as well as M2PA.
> > > > [TOLGA]Why so? I would think, when Retrieve BSTN is received, it
might
> > be
> > > > good to deliver data received successfully but not delivered to MTP3
> > > > because of RPO, and then reply to Retrieve BSTN request. This migth
save
> > us
> > > > to retranmit them on another link (I am not sure though, whether
this
> > will
> > > > create a problem for MTP3, Q.704 Figure30 Sheet 2 of 6 does not show
> > what
> > > > to do in such a case, but I would think this should be fine, I
assume it
> > is
> > > > possible that MTP2 is delivering a message to MTP3, while MTP3 is
> > sending
> > > > Retrieve BSTN. So, MTP3 should be prepared to get something
different
> > than
> > > > BSTN after sending Retrive BSTN (?) )
> > >
> > > Bad idea.
> > >
> > > --brian
> > >
> > > --
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >
>
> --
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 27 08:31:46 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA01339
	for <sigtran-archive@odin.ietf.org>; Thu, 27 Feb 2003 08:31:46 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1RDflT19275
	for sigtran-archive@odin.ietf.org; Thu, 27 Feb 2003 08:41:47 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1RDfWp19256;
	Thu, 27 Feb 2003 08:41:32 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1RDe3p19182
	for <sigtran@optimus.ietf.org>; Thu, 27 Feb 2003 08:40:03 -0500
Received: from domino2.nc.tekelec.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA01243
	for <sigtran@ietf.org>; Thu, 27 Feb 2003 08:29:30 -0500 (EST)
From: jeffrey.craig@tekelec.com
Received: from domino4.nc.tekelec.com ([198.89.40.9])
          by domino2.nc.tekelec.com (Lotus Domino Release 5.0.10)
          with ESMTP id 2003022708265581:19355 ;
          Thu, 27 Feb 2003 08:26:55 -0500 
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.7  March 21, 2001
Message-ID: <OFCFA30EA8.AB765A49-ON85256CDA.004929B0@nc.tekelec.com>
Date: Thu, 27 Feb 2003 08:29:08 -0500
MIME-Version: 1.0
X-MIMETrack: Serialize by Router on Domino4/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/27/2003 08:29:08 AM,
	Itemize by SMTP Server on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/27/2003 08:26:55 AM,
	Serialize by Router on Domino2/Raleigh/TEKELEC(Release 5.0.10 |March 22, 2002) at
 02/27/2003 08:26:58 AM,
	Serialize complete at 02/27/2003 08:26:58 AM
Content-type: text/plain; charset=us-ascii
Subject: [Sigtran] (M2PA) Another change to the state machine diagram -- Rcv Rdy or PO during
 PROVING
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Hello All,

I suggest that another change be made to the M2PA state machine
diagram to clarify treatment of received LSPN, LSPE, LSR, and
LSPO during PROVING. Receiving any of these messages during
PROVING results in no state change.

The diagram below includes this change, along with all of the
other changes I've proposed recently.

Regards,

Jeff


                   +-----------+
                   |   POWER   |
                   |    OFF    |
                   +-----------+
                         |
                         | Power On
                         |
       Flush Buffers     |  +-----------------------------------------+
       OR Retrieval      |  |                                         |
              +-------+  |  |                                         |
              |       |  V  V                                         |
              |    +-----------+                                      |
              +--->|  OUT OF   |                                      |
  +--------------->|  SERVICE  |<--+                                  |
  |                +-----------+   | Link Configured                  |
  |                      |   |     | (Associate)                      |
  |                      |   +-----+                                  |
  |     Receive LS OOS   |                                            |
  |           +-------+  | MTP3 Start                                 |
  |           |       |  V                                            |
  |           |    +-----------+                                      |
  |           +--->|    NOT    |                                      |
  +<---------------|  ALIGNED  |------------------------------------->+
  |     MTP3 Stop  +-----------+     SCTP Comm Error                  |
  |  OR T2 Expiry        |        OR SCTP Comm Lost                   |
  |                      |                                            |
  |                      |    Receive LS Align                        |
  |                      | OR Receive LS Prov Norm                    |
  |     Receive LS Align | OR Receive LS Prov Emer                    |
  |           +-------+  |                                            |
  |           |       |  V                                            |
  |           |    +-----------+                                      |
  |           +--->|  ALIGNED  |                                      |
  +<---------------|           |------------------------------------->+
  |     MTP3 Stop  +-----------+     SCTP Comm Error                  |
  |  OR T3 Expiry      ^    |     OR SCTP Comm Lost                   |
  |  OR Receive LS OOS |    |                                         |
  |                    |    |                                         |
  |                    |    |    Receive LS Prov Norm                 |
  |   Receive LS Align |    | OR Receive LS Prov Emer                 |
  |                    |    |                                         |
  |                    |    |                                         |
  |      Rcv LS Prov X |    |                                         |
  |   OR Rcv LS Ready  |    |                                         |
  |   OR Rcv LS PO     |    |                                         |
  |           +------+ |    |                                         |
  |           |      | |    V                                         |
  |           |    +-----------+                                      |
  |           +--->|  PROVING  |------------------------------------->+
  |<---------------|           |------+          SCTP Comm Error      |
  |     MTP3 Stop  +-----------+      |         OR SCTP Comm Lost     |
  |  OR Receive LS OOS   |            |                               |
  |                      |  T4 Expiry |                               |
  |                      |   (LPO)    |                               |
  |            T4 Expiry |            |                               |
  |             (No LPO) |            | MTP3 Stop            +------->+
  |                      |            | OR Receive LS OOS    |        |
  |                      |            | OR SCTP Comm Error   |        |
  |                      |            | OR SCTP Comm Lost    |        |
  |                      |            | OR T1 Expiry         |        |
  |       Rcv LS Prov X  |            | OR M2PA Link Failure |        |
  |           +-------+  |            | OR Receive LS Align  |        |
  |           |       |  V            |                      |        |
  |           |    +-----------+ LPO  V            +-----------+      |
  |           +--->|  ALIGNED  |------+----------->|  ALIGNED  |      |
  +<---------------|   READY   |<------------------| NOT READY |      |
  |     MTP3 Stop  +-----------+       LPR         +-----------+      |
  |  OR T1 Expiry        |   |                         | Rcv LS Ready |
  |  OR Receive LS OOS   |   |                         | OR Rcv Data  |
  |  OR Receive LS Align |   +------------------------>+ OR Rcv LS PO |
  |  OR SCTP Comm Error  |   Receive LS PO             |              |
  |  OR SCTP Comm Lost   |                             |              |
  |                      |    Receive LS Ready         |              |
  |                      | OR Receive User Data        |              |
  |                      V                             V              |
  |                +-----------+ LPO or Rcv LS PO  +-----------+      |
  |                |    IN     |------------------>| PROCESSOR |      |
  +<---------------|  SERVICE  |<------------------|  OUTAGE   |      |
                   +-----------+    MTP3 Flush     +-----------+      |
       MTP3 Stop                 OR MTP3 Continue            |        |
    OR Receive LS OOS                                        |        |
    OR SCTP Comm Error                     MTP3 Stop         |        |
    OR SCTP Comm Lost                   OR Receive LS OOS    |        |
    OR T6 Expiry                        OR SCTP Comm Error   |        |
    OR T7 Expiry                        OR SCTP Comm Lost    |        |
    OR M2PA Link Failure                OR M2PA Link Failure |        |
    OR Receive LS Align                 OR Receive LS Align  +------->+


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 27 09:09:29 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA03624
	for <sigtran-archive@odin.ietf.org>; Thu, 27 Feb 2003 09:09:29 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1REJTU22098
	for sigtran-archive@odin.ietf.org; Thu, 27 Feb 2003 09:19:29 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1REJGp22088;
	Thu, 27 Feb 2003 09:19:17 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1REIep22064
	for <sigtran@optimus.ietf.org>; Thu, 27 Feb 2003 09:18:40 -0500
Received: from hindon.hss.co.in (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA03547
	for <sigtran@ietf.org>; Thu, 27 Feb 2003 09:08:05 -0500 (EST)
From: sharmam@hss.hns.com
Received: from hindon.hss.co.in (localhost [127.0.0.1])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h1REAMs19493
	for <sigtran@ietf.org>; Thu, 27 Feb 2003 19:40:22 +0530 (IST)
Received: from ultra.hss.co.in (ultra [192.168.100.5] (may be forged))
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h1REAL219485
	for <sigtran@ietf.org>; Thu, 27 Feb 2003 19:40:21 +0530 (IST)
Received: from sandesh.hss.hns.com (localhost [127.0.0.1])
	by ultra.hss.co.in (8.10.0/8.10.0) with SMTP id h1RECu008890
	for <sigtran@ietf.org>; Thu, 27 Feb 2003 19:42:57 +0530 (IST)
Received: by sandesh.hss.hns.com(Lotus SMTP MTA v4.6.3  (733.2 10-16-1998))  id 65256CDA.004D45BE ; Thu, 27 Feb 2003 19:34:01 +0530
X-Lotus-FromDomain: HSS
To: sigtran@ietf.org
Message-ID: <65256CDA.004D43EA.00@sandesh.hss.hns.com>
Date: Thu, 27 Feb 2003 19:37:10 +0530
Mime-Version: 1.0
Content-type: text/plain; charset=us-ascii
Content-Disposition: inline
Subject: [Sigtran] to handle handle misconfiguration in IUA
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>



Hi,
   There is an example shown in IG to handle misconfiguration

   "This scenario shows the example IUA message flows for the
   establishment of traffic between an SG and an ASP in which some
   of the Interface Identifiers have been misconfigured on the
   ASP side.  The SG in this case has Interface Identifers 1-5
   configured for ASP1."

                SG                               ASP1
                 |                                |
                 |                                |
                 |<----ASP Active (IIDs 1-10)-----|
                 |---ASP Active Ack (IIDs 1-5)--->|
                 |-------Error (IIDs 6)---------->|
                 |-------Error (IIDs 7)---------->|
                 |-------Error (IIDs 8)---------->|
                 |-------Error (IIDs 9)---------->|
                 |-------Error (IIDs 10)--------->|


    The other alternative which i am showing is correct or not , as this is
implementation dependant

                 SG                               ASP1
                 |                                |
                 |-------Error (IIDs 1)---------->|
                 |-------Error (IIDs 2)---------->|
                 |-------Error (IIDs 3)---------->|
                 |-------Error (IIDs 4)---------->|
                 |-------Error (IIDs 5)---------->|
                 |-------Error (IIDs 6)---------->|
                 |-------Error (IIDs 7)---------->|
                 |-------Error (IIDs 8)---------->|
                 |-------Error (IIDs 9)---------->|
                 |-------Error (IIDs 10)--------->|


 Regards
 Manish Sharma
 www.hssworld.com









_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 27 09:25:35 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA04714
	for <sigtran-archive@odin.ietf.org>; Thu, 27 Feb 2003 09:25:35 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1REZZJ23185
	for sigtran-archive@odin.ietf.org; Thu, 27 Feb 2003 09:35:35 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1REZDp23160;
	Thu, 27 Feb 2003 09:35:13 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1REYqp23111
	for <sigtran@optimus.ietf.org>; Thu, 27 Feb 2003 09:34:52 -0500
Received: from rtp-core-1.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA04686
	for <sigtran@ietf.org>; Thu, 27 Feb 2003 09:24:20 -0500 (EST)
Received: from cia.cisco.com (IDENT:mirapoint@cia.cisco.com [161.44.85.200])
	by rtp-core-1.cisco.com (8.12.6/8.12.6) with ESMTP id h1RES6JR014798;
	Thu, 27 Feb 2003 09:28:07 -0500 (EST)
Received: from kmorneau-w2k.cisco.com (rtp-vpn1-981.cisco.com [10.82.227.213])
	by cia.cisco.com (Mirapoint)
	with ESMTP id ABQ28195;
	Thu, 27 Feb 2003 09:18:08 -0500 (EST)
Message-Id: <4.3.2.7.2.20030227092736.0545dea8@cia.cisco.com>
X-Sender: kmorneau@cia.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Thu, 27 Feb 2003 09:28:04 -0500
To: sharmam@hss.hns.com, sigtran@ietf.org
From: Ken Morneault <kmorneau@cisco.com>
Subject: Re: [Sigtran] to handle handle misconfiguration in IUA
In-Reply-To: <65256CDA.004D43EA.00@sandesh.hss.hns.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>



Manish,

Why are you proposing that the SG return Error (Interface Identifer)
for valid Interface Identifiers?

Ken

At 07:37 PM 2/27/2003 +0530, sharmam@hss.hns.com wrote:


>Hi,
>   There is an example shown in IG to handle misconfiguration
>
>   "This scenario shows the example IUA message flows for the
>   establishment of traffic between an SG and an ASP in which some
>   of the Interface Identifiers have been misconfigured on the
>   ASP side.  The SG in this case has Interface Identifers 1-5
>   configured for ASP1."
>
>                SG                               ASP1
>                 |                                |
>                 |                                |
>                 |<----ASP Active (IIDs 1-10)-----|
>                 |---ASP Active Ack (IIDs 1-5)--->|
>                 |-------Error (IIDs 6)---------->|
>                 |-------Error (IIDs 7)---------->|
>                 |-------Error (IIDs 8)---------->|
>                 |-------Error (IIDs 9)---------->|
>                 |-------Error (IIDs 10)--------->|
>
>
>    The other alternative which i am showing is correct or not , as this is
>implementation dependant
>
>                 SG                               ASP1
>                 |                                |
>                 |-------Error (IIDs 1)---------->|
>                 |-------Error (IIDs 2)---------->|
>                 |-------Error (IIDs 3)---------->|
>                 |-------Error (IIDs 4)---------->|
>                 |-------Error (IIDs 5)---------->|
>                 |-------Error (IIDs 6)---------->|
>                 |-------Error (IIDs 7)---------->|
>                 |-------Error (IIDs 8)---------->|
>                 |-------Error (IIDs 9)---------->|
>                 |-------Error (IIDs 10)--------->|
>
>
> Regards
> Manish Sharma
> www.hssworld.com
>
>
>
>
>
>
>
>
>
>_______________________________________________
>Sigtran mailing list
>Sigtran@ietf.org
>https://www1.ietf.org/mailman/listinfo/sigtran 

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 27 09:38:55 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA05503
	for <sigtran-archive@odin.ietf.org>; Thu, 27 Feb 2003 09:38:55 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1REmuv24516
	for sigtran-archive@odin.ietf.org; Thu, 27 Feb 2003 09:48:56 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1REmdp24506;
	Thu, 27 Feb 2003 09:48:41 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1REl7p24472
	for <sigtran@optimus.ietf.org>; Thu, 27 Feb 2003 09:47:07 -0500
Received: from hindon.hss.co.in (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA05410
	for <sigtran@ietf.org>; Thu, 27 Feb 2003 09:36:33 -0500 (EST)
From: sharmam@hss.hns.com
Received: from hindon.hss.co.in (localhost [127.0.0.1])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h1REcps23184
	for <sigtran@ietf.org>; Thu, 27 Feb 2003 20:08:51 +0530 (IST)
Received: from ultra.hss.co.in (ultra [192.168.100.5] (may be forged))
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h1REco223178;
	Thu, 27 Feb 2003 20:08:50 +0530 (IST)
Received: from sandesh.hss.hns.com (localhost [127.0.0.1])
	by ultra.hss.co.in (8.10.0/8.10.0) with SMTP id h1REfPX10829;
	Thu, 27 Feb 2003 20:11:26 +0530 (IST)
Received: by sandesh.hss.hns.com(Lotus SMTP MTA v4.6.3  (733.2 10-16-1998))  id 65256CDA.004FE1C6 ; Thu, 27 Feb 2003 20:02:31 +0530
X-Lotus-FromDomain: HSS
To: Ken Morneault <kmorneau@cisco.com>
cc: sigtran@ietf.org
Message-ID: <65256CDA.004FE064.00@sandesh.hss.hns.com>
Date: Thu, 27 Feb 2003 20:05:41 +0530
Subject: Re: [Sigtran] to handle handle misconfiguration in IUA
Mime-Version: 1.0
Content-type: text/plain; charset=us-ascii
Content-Disposition: inline
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>



Hi Ken,
   The objective of this example to handle the misconfiguration, So ASP side
stack  has to update its configuration based on peer SG configuration
   which is not correct. It is responsibility of the system management entities
to make the ASP and SG configuration alike. Why to update database on
   misconfiguration of SG side.

   Please correct me if i am wrong.


   Regards
   Manish S
   www.hssworld.com





Ken Morneault <kmorneau@cisco.com> on 02/27/2003 07:58:04 PM

To:   Manish Sharma/HSS@HSS, sigtran@ietf.org
cc:

Subject:  Re: [Sigtran] to handle handle misconfiguration in IUA






Manish,

Why are you proposing that the SG return Error (Interface Identifer)
for valid Interface Identifiers?

Ken

At 07:37 PM 2/27/2003 +0530, sharmam@hss.hns.com wrote:


>Hi,
>   There is an example shown in IG to handle misconfiguration
>
>   "This scenario shows the example IUA message flows for the
>   establishment of traffic between an SG and an ASP in which some
>   of the Interface Identifiers have been misconfigured on the
>   ASP side.  The SG in this case has Interface Identifers 1-5
>   configured for ASP1."
>
>                SG                               ASP1
>                 |                                |
>                 |                                |
>                 |<----ASP Active (IIDs 1-10)-----|
>                 |---ASP Active Ack (IIDs 1-5)--->|
>                 |-------Error (IIDs 6)---------->|
>                 |-------Error (IIDs 7)---------->|
>                 |-------Error (IIDs 8)---------->|
>                 |-------Error (IIDs 9)---------->|
>                 |-------Error (IIDs 10)--------->|
>
>
>    The other alternative which i am showing is correct or not , as this is
>implementation dependant
>
>                 SG                               ASP1
>                 |                                |
>                 |-------Error (IIDs 1)---------->|
>                 |-------Error (IIDs 2)---------->|
>                 |-------Error (IIDs 3)---------->|
>                 |-------Error (IIDs 4)---------->|
>                 |-------Error (IIDs 5)---------->|
>                 |-------Error (IIDs 6)---------->|
>                 |-------Error (IIDs 7)---------->|
>                 |-------Error (IIDs 8)---------->|
>                 |-------Error (IIDs 9)---------->|
>                 |-------Error (IIDs 10)--------->|
>
>
> Regards
> Manish Sharma
> www.hssworld.com
>
>
>
>
>
>
>
>
>
>_______________________________________________
>Sigtran mailing list
>Sigtran@ietf.org
>https://www1.ietf.org/mailman/listinfo/sigtran





_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 27 13:44:46 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA16230
	for <sigtran-archive@odin.ietf.org>; Thu, 27 Feb 2003 13:44:46 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1RIsrv12217
	for sigtran-archive@odin.ietf.org; Thu, 27 Feb 2003 13:54:53 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1RIpap11825;
	Thu, 27 Feb 2003 13:51:37 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1RIo9p11745
	for <sigtran@optimus.ietf.org>; Thu, 27 Feb 2003 13:50:09 -0500
Received: from rtp-core-2.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA15731
	for <sigtran@ietf.org>; Thu, 27 Feb 2003 13:39:31 -0500 (EST)
Received: from cia.cisco.com (IDENT:mirapoint@cia.cisco.com [161.44.85.200])
	by rtp-core-2.cisco.com (8.12.6/8.12.6) with ESMTP id h1RIhKNh013909;
	Thu, 27 Feb 2003 13:43:21 -0500 (EST)
Received: from kmorneau-w2k.cisco.com (rtp-vpn1-981.cisco.com [10.82.227.213])
	by cia.cisco.com (Mirapoint)
	with ESMTP id ABQ31420;
	Thu, 27 Feb 2003 13:33:22 -0500 (EST)
Message-Id: <4.3.2.7.2.20030227134121.04a34008@cia.cisco.com>
X-Sender: kmorneau@cia.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Thu, 27 Feb 2003 13:43:15 -0500
To: <sharmam@hss.hns.com>
From: Ken Morneault <kmorneau@cisco.com>
Subject: Re: [Sigtran] to handle handle misconfiguration in IUA
Cc: sigtran@ietf.org
In-Reply-To: <65256CDA.004FE064.00@sandesh.hss.hns.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>



Manish,

Who's to say which side has been configured incorrectly.  Could be
the SG or ASP side.  

The first message flow allows traffic to flow for IIDs 1-5 and provides
a means for an alarm (by way of the Error msg) to be generated
to notify the operator of the misconfiguration of IIDs 6-10. 

Ken

At 08:05 PM 2/27/2003 +0530, sharmam@hss.hns.com wrote:


>Hi Ken,
>   The objective of this example to handle the misconfiguration, So ASP side
>stack  has to update its configuration based on peer SG configuration
>   which is not correct. It is responsibility of the system management entities
>to make the ASP and SG configuration alike. Why to update database on
>   misconfiguration of SG side.
>
>   Please correct me if i am wrong.
>
>
>   Regards
>   Manish S
>   www.hssworld.com
>
>
>
>
>
>Ken Morneault <kmorneau@cisco.com> on 02/27/2003 07:58:04 PM
>
>To:   Manish Sharma/HSS@HSS, sigtran@ietf.org
>cc:
>
>Subject:  Re: [Sigtran] to handle handle misconfiguration in IUA
>
>
>
>
>
>
>Manish,
>
>Why are you proposing that the SG return Error (Interface Identifer)
>for valid Interface Identifiers?
>
>Ken
>
>At 07:37 PM 2/27/2003 +0530, sharmam@hss.hns.com wrote:
>
>
>>Hi,
>>   There is an example shown in IG to handle misconfiguration
>>
>>   "This scenario shows the example IUA message flows for the
>>   establishment of traffic between an SG and an ASP in which some
>>   of the Interface Identifiers have been misconfigured on the
>>   ASP side.  The SG in this case has Interface Identifers 1-5
>>   configured for ASP1."
>>
>>                SG                               ASP1
>>                 |                                |
>>                 |                                |
>>                 |<----ASP Active (IIDs 1-10)-----|
>>                 |---ASP Active Ack (IIDs 1-5)--->|
>>                 |-------Error (IIDs 6)---------->|
>>                 |-------Error (IIDs 7)---------->|
>>                 |-------Error (IIDs 8)---------->|
>>                 |-------Error (IIDs 9)---------->|
>>                 |-------Error (IIDs 10)--------->|
>>
>>
>>    The other alternative which i am showing is correct or not , as this is
>>implementation dependant
>>
>>                 SG                               ASP1
>>                 |                                |
>>                 |-------Error (IIDs 1)---------->|
>>                 |-------Error (IIDs 2)---------->|
>>                 |-------Error (IIDs 3)---------->|
>>                 |-------Error (IIDs 4)---------->|
>>                 |-------Error (IIDs 5)---------->|
>>                 |-------Error (IIDs 6)---------->|
>>                 |-------Error (IIDs 7)---------->|
>>                 |-------Error (IIDs 8)---------->|
>>                 |-------Error (IIDs 9)---------->|
>>                 |-------Error (IIDs 10)--------->|
>>
>>
>> Regards
>> Manish Sharma
>> www.hssworld.com
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>_______________________________________________
>>Sigtran mailing list
>>Sigtran@ietf.org
>>https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Thu Feb 27 23:58:11 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA03594
	for <sigtran-archive@odin.ietf.org>; Thu, 27 Feb 2003 23:58:11 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1S58Uo17192
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 00:08:30 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1S58Dp17176;
	Fri, 28 Feb 2003 00:08:13 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1S54ip16258
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 00:04:44 -0500
Received: from hindon.hss.co.in (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA03448
	for <sigtran@ietf.org>; Thu, 27 Feb 2003 23:53:47 -0500 (EST)
From: sharmam@hss.hns.com
Received: from hindon.hss.co.in (localhost [127.0.0.1])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h1S4u0s06076
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 10:26:04 +0530 (IST)
Received: from ultra.hss.co.in (ultra.hss.hns.com [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h1S4tw206069;
	Fri, 28 Feb 2003 10:25:58 +0530 (IST)
Received: from sandesh.hss.hns.com (localhost [127.0.0.1])
	by ultra.hss.co.in (8.10.0/8.10.0) with SMTP id h1S4wXr23118;
	Fri, 28 Feb 2003 10:28:34 +0530 (IST)
Received: by sandesh.hss.hns.com(Lotus SMTP MTA v4.6.3  (733.2 10-16-1998))  id 65256CDB.001A8488 ; Fri, 28 Feb 2003 10:19:38 +0530
X-Lotus-FromDomain: HSS
To: Ken Morneault <kmorneau@cisco.com>
cc: sigtran@ietf.org
Message-ID: <65256CDB.001A8444.00@sandesh.hss.hns.com>
Date: Fri, 28 Feb 2003 10:22:49 +0530
Subject: Re: [Sigtran] to handle handle misconfiguration in IUA and Stream
	 0 for ASPM messages
Mime-Version: 1.0
Content-type: text/plain; charset=us-ascii
Content-Disposition: inline
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>



Hi ken,
 Ok i agree that we don't know that which side is configured incorrectly but
 if there is misconfiguration then its work of System management entity to
correct that.
 Its not proper to change configuration based on peer side database( SG/ASP).
 We can't say to an implementation to give   error for selected IID's.

 Second is
  " All ASPM messages are sent on a sequenced
    stream to ensure ordering.SCTP stream '0' SHOULD be used"
  This is written in RFC 3057 and there is no text regarding this to make it
  inline with other UA's. Other UA's have no such restriction for ASPM messages.


 M3UA imp guide says

 "The following rules MUST to be followed (see section 3.1.2):
 1. DATA is never sent on stream 0.
 2. ASPSM, MGMT, RKM should be sent on stream 0 (Other than BEAT and BEAT ACK)
 3. SSNM , ASPTM , BEAT and BEAT ACK can be sent on any stream."

 Please see this, Thanks in advance


  Regards
  Manish Sharma
  www.hssworld.com






Ken Morneault <kmorneau@cisco.com> on 02/28/2003 12:13:15 AM

To:   Manish Sharma/HSS@HSS
cc:   sigtran@ietf.org

Subject:  Re: [Sigtran] to handle handle misconfiguration in IUA






Manish,

Who's to say which side has been configured incorrectly.  Could be
the SG or ASP side.

The first message flow allows traffic to flow for IIDs 1-5 and provides
a means for an alarm (by way of the Error msg) to be generated
to notify the operator of the misconfiguration of IIDs 6-10.

Ken

At 08:05 PM 2/27/2003 +0530, sharmam@hss.hns.com wrote:


>Hi Ken,
>   The objective of this example to handle the misconfiguration, So ASP side
>stack  has to update its configuration based on peer SG configuration
>   which is not correct. It is responsibility of the system management entities
>to make the ASP and SG configuration alike. Why to update database on
>   misconfiguration of SG side.
>
>   Please correct me if i am wrong.
>
>
>   Regards
>   Manish S
>   www.hssworld.com
>
>
>
>
>
>Ken Morneault <kmorneau@cisco.com> on 02/27/2003 07:58:04 PM
>
>To:   Manish Sharma/HSS@HSS, sigtran@ietf.org
>cc:
>
>Subject:  Re: [Sigtran] to handle handle misconfiguration in IUA
>
>
>
>
>
>
>Manish,
>
>Why are you proposing that the SG return Error (Interface Identifer)
>for valid Interface Identifiers?
>
>Ken
>
>At 07:37 PM 2/27/2003 +0530, sharmam@hss.hns.com wrote:
>
>
>>Hi,
>>   There is an example shown in IG to handle misconfiguration
>>
>>   "This scenario shows the example IUA message flows for the
>>   establishment of traffic between an SG and an ASP in which some
>>   of the Interface Identifiers have been misconfigured on the
>>   ASP side.  The SG in this case has Interface Identifers 1-5
>>   configured for ASP1."
>>
>>                SG                               ASP1
>>                 |                                |
>>                 |                                |
>>                 |<----ASP Active (IIDs 1-10)-----|
>>                 |---ASP Active Ack (IIDs 1-5)--->|
>>                 |-------Error (IIDs 6)---------->|
>>                 |-------Error (IIDs 7)---------->|
>>                 |-------Error (IIDs 8)---------->|
>>                 |-------Error (IIDs 9)---------->|
>>                 |-------Error (IIDs 10)--------->|
>>
>>
>>    The other alternative which i am showing is correct or not , as this is
>>implementation dependant
>>
>>                 SG                               ASP1
>>                 |                                |
>>                 |-------Error (IIDs 1)---------->|
>>                 |-------Error (IIDs 2)---------->|
>>                 |-------Error (IIDs 3)---------->|
>>                 |-------Error (IIDs 4)---------->|
>>                 |-------Error (IIDs 5)---------->|
>>                 |-------Error (IIDs 6)---------->|
>>                 |-------Error (IIDs 7)---------->|
>>                 |-------Error (IIDs 8)---------->|
>>                 |-------Error (IIDs 9)---------->|
>>                 |-------Error (IIDs 10)--------->|
>>
>>
>> Regards
>> Manish Sharma
>> www.hssworld.com
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>_______________________________________________
>>Sigtran mailing list
>>Sigtran@ietf.org
>>https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 28 00:10:12 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA03719
	for <sigtran-archive@odin.ietf.org>; Fri, 28 Feb 2003 00:10:12 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1S5KVX17731
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 00:20:31 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1S5KDp17709;
	Fri, 28 Feb 2003 00:20:13 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1S5Hmp17606
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 00:17:48 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA03697
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 00:06:52 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:BVJfXM+p9ol7qqVAID8P9cOJIHpy6ct2@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1S5AmD18717;
	Thu, 27 Feb 2003 22:10:48 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1S5Amc31208;
	Thu, 27 Feb 2003 22:10:48 -0700
Date: Thu, 27 Feb 2003 22:10:47 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: sharmam@hss.hns.com
Cc: Ken Morneault <kmorneau@cisco.com>, sigtran@ietf.org
Subject: Re: [Sigtran] to handle handle misconfiguration in IUA and Stream 0 for ASPM messages
Message-ID: <20030227221047.A31141@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: sharmam@hss.hns.com, Ken Morneault <kmorneau@cisco.com>,
	sigtran@ietf.org
References: <65256CDB.001A8444.00@sandesh.hss.hns.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <65256CDB.001A8444.00@sandesh.hss.hns.com>; from sharmam@hss.hns.com on Fri, Feb 28, 2003 at 10:22:49AM +0530
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

sharmam,

If you look at the IUA imp guide I think you will see that it has similar
text to the other UAs with regard to ordering of messages in streams.

--brian

sharmam@hss.hns.com wrote:                                                         (Fri, 28 Feb 2003 10:22:49)
> 
> 
> Hi ken,
>  Ok i agree that we don't know that which side is configured incorrectly but
>  if there is misconfiguration then its work of System management entity to
> correct that.
>  Its not proper to change configuration based on peer side database( SG/ASP).
>  We can't say to an implementation to give   error for selected IID's.
> 
>  Second is
>   " All ASPM messages are sent on a sequenced
>     stream to ensure ordering.SCTP stream '0' SHOULD be used"
>   This is written in RFC 3057 and there is no text regarding this to make it
>   inline with other UA's. Other UA's have no such restriction for ASPM messages.
> 
> 
>  M3UA imp guide says
> 
>  "The following rules MUST to be followed (see section 3.1.2):
>  1. DATA is never sent on stream 0.
>  2. ASPSM, MGMT, RKM should be sent on stream 0 (Other than BEAT and BEAT ACK)
>  3. SSNM , ASPTM , BEAT and BEAT ACK can be sent on any stream."
> 
>  Please see this, Thanks in advance
> 
> 
>   Regards
>   Manish Sharma
>   www.hssworld.com
> 
> 
> 
> 
> 
> 
> Ken Morneault <kmorneau@cisco.com> on 02/28/2003 12:13:15 AM
> 
> To:   Manish Sharma/HSS@HSS
> cc:   sigtran@ietf.org
> 
> Subject:  Re: [Sigtran] to handle handle misconfiguration in IUA
> 
> 
> 
> 
> 
> 
> Manish,
> 
> Who's to say which side has been configured incorrectly.  Could be
> the SG or ASP side.
> 
> The first message flow allows traffic to flow for IIDs 1-5 and provides
> a means for an alarm (by way of the Error msg) to be generated
> to notify the operator of the misconfiguration of IIDs 6-10.
> 
> Ken
> 
> At 08:05 PM 2/27/2003 +0530, sharmam@hss.hns.com wrote:
> 
> 
> >Hi Ken,
> >   The objective of this example to handle the misconfiguration, So ASP side
> >stack  has to update its configuration based on peer SG configuration
> >   which is not correct. It is responsibility of the system management entities
> >to make the ASP and SG configuration alike. Why to update database on
> >   misconfiguration of SG side.
> >
> >   Please correct me if i am wrong.
> >
> >
> >   Regards
> >   Manish S
> >   www.hssworld.com
> >
> >
> >
> >
> >
> >Ken Morneault <kmorneau@cisco.com> on 02/27/2003 07:58:04 PM
> >
> >To:   Manish Sharma/HSS@HSS, sigtran@ietf.org
> >cc:
> >
> >Subject:  Re: [Sigtran] to handle handle misconfiguration in IUA
> >
> >
> >
> >
> >
> >
> >Manish,
> >
> >Why are you proposing that the SG return Error (Interface Identifer)
> >for valid Interface Identifiers?
> >
> >Ken
> >
> >At 07:37 PM 2/27/2003 +0530, sharmam@hss.hns.com wrote:
> >
> >
> >>Hi,
> >>   There is an example shown in IG to handle misconfiguration
> >>
> >>   "This scenario shows the example IUA message flows for the
> >>   establishment of traffic between an SG and an ASP in which some
> >>   of the Interface Identifiers have been misconfigured on the
> >>   ASP side.  The SG in this case has Interface Identifers 1-5
> >>   configured for ASP1."
> >>
> >>                SG                               ASP1
> >>                 |                                |
> >>                 |                                |
> >>                 |<----ASP Active (IIDs 1-10)-----|
> >>                 |---ASP Active Ack (IIDs 1-5)--->|
> >>                 |-------Error (IIDs 6)---------->|
> >>                 |-------Error (IIDs 7)---------->|
> >>                 |-------Error (IIDs 8)---------->|
> >>                 |-------Error (IIDs 9)---------->|
> >>                 |-------Error (IIDs 10)--------->|
> >>
> >>
> >>    The other alternative which i am showing is correct or not , as this is
> >>implementation dependant
> >>
> >>                 SG                               ASP1
> >>                 |                                |
> >>                 |-------Error (IIDs 1)---------->|
> >>                 |-------Error (IIDs 2)---------->|
> >>                 |-------Error (IIDs 3)---------->|
> >>                 |-------Error (IIDs 4)---------->|
> >>                 |-------Error (IIDs 5)---------->|
> >>                 |-------Error (IIDs 6)---------->|
> >>                 |-------Error (IIDs 7)---------->|
> >>                 |-------Error (IIDs 8)---------->|
> >>                 |-------Error (IIDs 9)---------->|
> >>                 |-------Error (IIDs 10)--------->|
> >>
> >>
> >> Regards
> >> Manish Sharma
> >> www.hssworld.com
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >>_______________________________________________
> >>Sigtran mailing list
> >>Sigtran@ietf.org
> >>https://www1.ietf.org/mailman/listinfo/sigtran
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 
> 
> 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 28 00:21:15 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA03910
	for <sigtran-archive@odin.ietf.org>; Fri, 28 Feb 2003 00:21:15 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1S5VYJ18215
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 00:31:34 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1S5VKp18199;
	Fri, 28 Feb 2003 00:31:20 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1S5Sop18087
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 00:28:50 -0500
Received: from hindon.hss.co.in (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA03802
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 00:17:29 -0500 (EST)
From: sharmam@hss.hns.com
Received: from hindon.hss.co.in (localhost [127.0.0.1])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h1S5JRs10681
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 10:49:27 +0530 (IST)
Received: from ultra.hss.co.in (ultra.hss.hns.com [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h1S5JH210569;
	Fri, 28 Feb 2003 10:49:17 +0530 (IST)
Received: from sandesh.hss.hns.com (localhost [127.0.0.1])
	by ultra.hss.co.in (8.10.0/8.10.0) with SMTP id h1S5LgR25258;
	Fri, 28 Feb 2003 10:51:43 +0530 (IST)
Received: by sandesh.hss.hns.com(Lotus SMTP MTA v4.6.3  (733.2 10-16-1998))  id 65256CDB.001C9DB1 ; Fri, 28 Feb 2003 10:42:33 +0530
X-Lotus-FromDomain: HSS
To: bidulock@openss7.org
cc: Ken Morneault <kmorneau@cisco.com>, sigtran@ietf.org
Message-ID: <65256CDB.001C8570.00@sandesh.hss.hns.com>
Date: Fri, 28 Feb 2003 10:44:44 +0530
Subject: Re: [Sigtran] to handle handle misconfiguration in IUA and Stream
	 0 for ASPM messages
Mime-Version: 1.0
Content-type: text/plain; charset=us-ascii
Content-Disposition: inline
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>



Hi Brian,
   Can you please provide me the text stated in IUA IG ? It might possible that
i have missed that.

  Thanks
  Manish S
  www.hssworld.com





"Brian F. G. Bidulock" <bidulock@openss7.org> on 02/28/2003 10:40:47 AM

Please respond to bidulock@openss7.org

To:   Manish Sharma/HSS@HSS
cc:   Ken Morneault <kmorneau@cisco.com>, sigtran@ietf.org

Subject:  Re: [Sigtran] to handle handle misconfiguration in IUA and Stream 0
      for ASPM messages




sharmam,

If you look at the IUA imp guide I think you will see that it has similar
text to the other UAs with regard to ordering of messages in streams.

--brian

sharmam@hss.hns.com wrote:
(Fri, 28 Feb 2003 10:22:49)
>
>
> Hi ken,
>  Ok i agree that we don't know that which side is configured incorrectly but
>  if there is misconfiguration then its work of System management entity to
> correct that.
>  Its not proper to change configuration based on peer side database( SG/ASP).
>  We can't say to an implementation to give   error for selected IID's.
>
>  Second is
>   " All ASPM messages are sent on a sequenced
>     stream to ensure ordering.SCTP stream '0' SHOULD be used"
>   This is written in RFC 3057 and there is no text regarding this to make it
>   inline with other UA's. Other UA's have no such restriction for ASPM
messages.
>
>
>  M3UA imp guide says
>
>  "The following rules MUST to be followed (see section 3.1.2):
>  1. DATA is never sent on stream 0.
>  2. ASPSM, MGMT, RKM should be sent on stream 0 (Other than BEAT and BEAT ACK)
>  3. SSNM , ASPTM , BEAT and BEAT ACK can be sent on any stream."
>
>  Please see this, Thanks in advance
>
>
>   Regards
>   Manish Sharma
>   www.hssworld.com
>
>
>
>
>
>
> Ken Morneault <kmorneau@cisco.com> on 02/28/2003 12:13:15 AM
>
> To:   Manish Sharma/HSS@HSS
> cc:   sigtran@ietf.org
>
> Subject:  Re: [Sigtran] to handle handle misconfiguration in IUA
>
>
>
>
>
>
> Manish,
>
> Who's to say which side has been configured incorrectly.  Could be
> the SG or ASP side.
>
> The first message flow allows traffic to flow for IIDs 1-5 and provides
> a means for an alarm (by way of the Error msg) to be generated
> to notify the operator of the misconfiguration of IIDs 6-10.
>
> Ken
>
> At 08:05 PM 2/27/2003 +0530, sharmam@hss.hns.com wrote:
>
>
> >Hi Ken,
> >   The objective of this example to handle the misconfiguration, So ASP side
> >stack  has to update its configuration based on peer SG configuration
> >   which is not correct. It is responsibility of the system management
entities
> >to make the ASP and SG configuration alike. Why to update database on
> >   misconfiguration of SG side.
> >
> >   Please correct me if i am wrong.
> >
> >
> >   Regards
> >   Manish S
> >   www.hssworld.com
> >
> >
> >
> >
> >
> >Ken Morneault <kmorneau@cisco.com> on 02/27/2003 07:58:04 PM
> >
> >To:   Manish Sharma/HSS@HSS, sigtran@ietf.org
> >cc:
> >
> >Subject:  Re: [Sigtran] to handle handle misconfiguration in IUA
> >
> >
> >
> >
> >
> >
> >Manish,
> >
> >Why are you proposing that the SG return Error (Interface Identifer)
> >for valid Interface Identifiers?
> >
> >Ken
> >
> >At 07:37 PM 2/27/2003 +0530, sharmam@hss.hns.com wrote:
> >
> >
> >>Hi,
> >>   There is an example shown in IG to handle misconfiguration
> >>
> >>   "This scenario shows the example IUA message flows for the
> >>   establishment of traffic between an SG and an ASP in which some
> >>   of the Interface Identifiers have been misconfigured on the
> >>   ASP side.  The SG in this case has Interface Identifers 1-5
> >>   configured for ASP1."
> >>
> >>                SG                               ASP1
> >>                 |                                |
> >>                 |                                |
> >>                 |<----ASP Active (IIDs 1-10)-----|
> >>                 |---ASP Active Ack (IIDs 1-5)--->|
> >>                 |-------Error (IIDs 6)---------->|
> >>                 |-------Error (IIDs 7)---------->|
> >>                 |-------Error (IIDs 8)---------->|
> >>                 |-------Error (IIDs 9)---------->|
> >>                 |-------Error (IIDs 10)--------->|
> >>
> >>
> >>    The other alternative which i am showing is correct or not , as this is
> >>implementation dependant
> >>
> >>                 SG                               ASP1
> >>                 |                                |
> >>                 |-------Error (IIDs 1)---------->|
> >>                 |-------Error (IIDs 2)---------->|
> >>                 |-------Error (IIDs 3)---------->|
> >>                 |-------Error (IIDs 4)---------->|
> >>                 |-------Error (IIDs 5)---------->|
> >>                 |-------Error (IIDs 6)---------->|
> >>                 |-------Error (IIDs 7)---------->|
> >>                 |-------Error (IIDs 8)---------->|
> >>                 |-------Error (IIDs 9)---------->|
> >>                 |-------Error (IIDs 10)--------->|
> >>
> >>
> >> Regards
> >> Manish Sharma
> >> www.hssworld.com
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >>_______________________________________________
> >>Sigtran mailing list
> >>Sigtran@ietf.org
> >>https://www1.ietf.org/mailman/listinfo/sigtran
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>
>
>
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/






_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 28 02:13:36 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA16188
	for <sigtran-archive@odin.ietf.org>; Fri, 28 Feb 2003 02:13:35 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1S7Nwa03046
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 02:23:58 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1S7Ngp03034;
	Fri, 28 Feb 2003 02:23:44 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1S7Ffp02324
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 02:15:41 -0500
Received: from hotmail.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA11994
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 02:04:48 -0500 (EST)
Received: from mail pickup service by hotmail.com with Microsoft SMTPSVC;
	 Thu, 27 Feb 2003 23:08:44 -0800
Received: from 164.164.96.102 by lw11fd.law11.hotmail.msn.com with HTTP;
	Fri, 28 Feb 2003 07:08:44 GMT
X-Originating-IP: [164.164.96.102]
From: "bijoy palathunkal" <b_palathunkal@hotmail.com>
To: sigtran@ietf.org
Date: Fri, 28 Feb 2003 12:38:44 +0530
Mime-Version: 1.0
Content-Type: text/plain; format=flowed
Message-ID: <F139o2R1z6fieNwcbWM0002699c@hotmail.com>
X-OriginalArrivalTime: 28 Feb 2003 07:08:44.0820 (UTC) FILETIME=[3B50C540:01C2DEF8]
Subject: [Sigtran] M3UA:SG's for same SGP??
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

hi all,
  What is the meaning of "two SG's being served
by the same SGP" ?Can anyone give us a SS7 network
scenario corresponding to this situation.


thanks
palathunkal







_________________________________________________________________


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 28 02:33:09 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA16420
	for <sigtran-archive@odin.ietf.org>; Fri, 28 Feb 2003 02:33:09 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1S7hWn04598
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 02:43:32 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1S7hFp04588;
	Fri, 28 Feb 2003 02:43:15 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1S7e5p04421
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 02:40:05 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA16329
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 02:29:10 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1S7WwJT029104;
	Fri, 28 Feb 2003 01:32:59 -0600 (CST)
Received: from axsppc68 by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id MAA02528; Fri, 28 Feb 2003 12:01:43 -0500 (GMT)
Message-ID: <004d01c2df5c$7bb526d0$7601ff0a@axsppc68>
From: "Subhashini" <sksubha@axes-mach01.usa.alcatel.com>
To: <bidulock@openss7.org>
Cc: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>, <sigtran@ietf.org>
Subject: Re: [Sigtran] Repost - More M2PA 7 questions] - LS OOS during alignment.
Date: Fri, 28 Feb 2003 13:06:02 -0600
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 5.50.4133.2400
X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

brian,

Still I did not get any reply for the mail below. I am referring to version
0.7, 17 Jan 2003. In this version, the Status Interval timer is totally
removed.
Please clarify.

thanks,
Subhashini ...

----- Original Message -----
From: "Subhashini" <sksubha@axes-mach01.usa.alcatel.com>
To: <bidulock@openss7.org>
Cc: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>; <sigtran@ietf.org>
Sent: Thursday, February 27, 2003 5:37 PM
Subject: Re: [Sigtran] Repost - More M2PA 7 questions] - LS OOS during
alignment.


> In section 4.1.3, the text "M2PA MAY send additional Link Status Ready
> messages while timer T1 is running" is not clear. At what interval this LS
> Ready message is to be sent?
>
> thanks,
> Subhashini ...
>
> ----- Original Message -----
> From: "Brian F. G. Bidulock" <bidulock@openss7.org>
> To: "Subhashini" <sksubha@axes-mach01.usa.alcatel.com>
> Cc: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>; <sigtran@ietf.org>
> Sent: Thursday, February 27, 2003 5:50 AM
> Subject: Re: [Sigtran] Repost - More M2PA 7 questions] - LS OOS during
> alignment.
>
>
> > Subhashini,
> >
> > The text is correct, the diagram is not.
> >
> > --brian
> >
> > Subhashini wrote:
> (Thu, 27 Feb 2003 17:12:54)
> > > The following text in section 4.1.3 and the diagram in section 3 are
> > > contradicting to each other.
> > >
> > > Note: Between the time M2PA sends Link Status Alignment to its peer
> > > and receives Link Status Alignment from its peer, M2PA may receive
> > > Link Status Out of Service from its peer. This message is
> > > ignored.
> > > As per diagram, on getting LS OOS (while waiting for LS Alignment),
> instead
> > > of ignoring the message, the state is changed to OOS.
> > >
> > > Which one is correct? Please clarify.
> > >
> > >  thanks,
> > >
> > > Subhashini ...
> > >
> > > ----- Original Message -----
> > > From: "Brian F. G. Bidulock" <bidulock@openss7.org>
> > > To: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
> > > Cc: <sigtran@ietf.org>
> > > Sent: Friday, February 21, 2003 5:27 AM
> > > Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
> > >
> > >
> > > > TOLGA,
> > > >
> > > > TOLGA ASVEREN wrote:
> > > (Fri, 21 Feb 2003 12:44:30)
> > > > >
> > > > > Brian,
> > > > > --- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
> > > > > > TOLGA,
> > > > > >
> > > > > > TOLGA ASVEREN wrote:
> > > > > >                                 (Fri, 21 Feb 2003
> > > > > > 09:43:22)
> > > > > > >
> > > > > > > Thanks , now it is clear, so that thread was
> > > > > > concluded(sometimes hard to
> > > > > > > decide for this) and we will have a slightly
> > > > > > modified version of Tom's
> > > > > > > initial interpretation, where data is not
> > > > > > discarded but not acknowledged
> > > > > > > either. Then I also would think stopping T7 is
> > > > > > fine.
> > > > > > >
> > > > > > > One more question:
> > > > > > > During changeover, will data received during RPO
> > > > > > but not acknowledged,
> > > > > > > considered as received successfully or not, i.e.
> > > > > > in case of changeover,
> > > > > > > will they be delivered to MTP3 or not?
> > > > > >
> > > > > > Unacknowledged messages must never be delivered to
> > > > > > MTP3 or duplication
> > > > > > will result on changeover.  This is true for Q.703
> > > > > > as well as M2PA.
> > > > > [TOLGA]Why so? I would think, when Retrieve BSTN is received, it
> might
> > > be
> > > > > good to deliver data received successfully but not delivered to
MTP3
> > > > > because of RPO, and then reply to Retrieve BSTN request. This
migth
> save
> > > us
> > > > > to retranmit them on another link (I am not sure though, whether
> this
> > > will
> > > > > create a problem for MTP3, Q.704 Figure30 Sheet 2 of 6 does not
show
> > > what
> > > > > to do in such a case, but I would think this should be fine, I
> assume it
> > > is
> > > > > possible that MTP2 is delivering a message to MTP3, while MTP3 is
> > > sending
> > > > > Retrieve BSTN. So, MTP3 should be prepared to get something
> different
> > > than
> > > > > BSTN after sending Retrive BSTN (?) )
> > > >
> > > > Bad idea.
> > > >
> > > > --brian
> > > >
> > > > --
> > > > Brian F. G. Bidulock
> > > > bidulock@openss7.org
> > > > http://www.openss7.org/
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > >
> >
> > --
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> >
>

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 28 02:56:15 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA16707
	for <sigtran-archive@odin.ietf.org>; Fri, 28 Feb 2003 02:56:15 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1S86cm05359
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 03:06:38 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1S86Mp05336;
	Fri, 28 Feb 2003 03:06:22 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1S83Gp05256
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 03:03:16 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA16671
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 02:52:21 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:w2b3XjHXzDd7OScmX2YqohJe4jxkPti2@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1S7uID21627;
	Fri, 28 Feb 2003 00:56:18 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1S7uHr00621;
	Fri, 28 Feb 2003 00:56:17 -0700
Date: Fri, 28 Feb 2003 00:56:17 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Subhashini <sksubha@axes-mach01.usa.alcatel.com>
Cc: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
Subject: Re: [Sigtran] Repost - More M2PA 7 questions] - LS OOS during alignment.
Message-ID: <20030228005617.A568@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Subhashini <sksubha@axes-mach01.usa.alcatel.com>,
	TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <004d01c2df5c$7bb526d0$7601ff0a@axsppc68>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <004d01c2df5c$7bb526d0$7601ff0a@axsppc68>; from sksubha@axes-mach01.usa.alcatel.com on Fri, Feb 28, 2003 at 01:06:02PM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Subhashini,

As the procedure is optional, send it as often as you like or not at all.

--brian

Subhashini wrote:                                                               (Fri, 28 Feb 2003 13:06:02)
> brian,
> 
> Still I did not get any reply for the mail below. I am referring to version
> 0.7, 17 Jan 2003. In this version, the Status Interval timer is totally
> removed.
> Please clarify.
> 
> thanks,
> Subhashini ...
> 
> ----- Original Message -----
> From: "Subhashini" <sksubha@axes-mach01.usa.alcatel.com>
> To: <bidulock@openss7.org>
> Cc: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>; <sigtran@ietf.org>
> Sent: Thursday, February 27, 2003 5:37 PM
> Subject: Re: [Sigtran] Repost - More M2PA 7 questions] - LS OOS during
> alignment.
> 
> 
> > In section 4.1.3, the text "M2PA MAY send additional Link Status Ready
> > messages while timer T1 is running" is not clear. At what interval this LS
> > Ready message is to be sent?
> >
> > thanks,
> > Subhashini ...
> >
> > ----- Original Message -----
> > From: "Brian F. G. Bidulock" <bidulock@openss7.org>
> > To: "Subhashini" <sksubha@axes-mach01.usa.alcatel.com>
> > Cc: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>; <sigtran@ietf.org>
> > Sent: Thursday, February 27, 2003 5:50 AM
> > Subject: Re: [Sigtran] Repost - More M2PA 7 questions] - LS OOS during
> > alignment.
> >
> >
> > > Subhashini,
> > >
> > > The text is correct, the diagram is not.
> > >
> > > --brian
> > >
> > > Subhashini wrote:
> > (Thu, 27 Feb 2003 17:12:54)
> > > > The following text in section 4.1.3 and the diagram in section 3 are
> > > > contradicting to each other.
> > > >
> > > > Note: Between the time M2PA sends Link Status Alignment to its peer
> > > > and receives Link Status Alignment from its peer, M2PA may receive
> > > > Link Status Out of Service from its peer. This message is
> > > > ignored.
> > > > As per diagram, on getting LS OOS (while waiting for LS Alignment),
> > instead
> > > > of ignoring the message, the state is changed to OOS.
> > > >
> > > > Which one is correct? Please clarify.
> > > >
> > > >  thanks,
> > > >
> > > > Subhashini ...
> > > >
> > > > ----- Original Message -----
> > > > From: "Brian F. G. Bidulock" <bidulock@openss7.org>
> > > > To: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
> > > > Cc: <sigtran@ietf.org>
> > > > Sent: Friday, February 21, 2003 5:27 AM
> > > > Subject: Re: [Sigtran] Repost - More M2PA 7 questions]
> > > >
> > > >
> > > > > TOLGA,
> > > > >
> > > > > TOLGA ASVEREN wrote:
> > > > (Fri, 21 Feb 2003 12:44:30)
> > > > > >
> > > > > > Brian,
> > > > > > --- "Brian F. G. Bidulock" <bidulock@openss7.org> wrote:
> > > > > > > TOLGA,
> > > > > > >
> > > > > > > TOLGA ASVEREN wrote:
> > > > > > >                                 (Fri, 21 Feb 2003
> > > > > > > 09:43:22)
> > > > > > > >
> > > > > > > > Thanks , now it is clear, so that thread was
> > > > > > > concluded(sometimes hard to
> > > > > > > > decide for this) and we will have a slightly
> > > > > > > modified version of Tom's
> > > > > > > > initial interpretation, where data is not
> > > > > > > discarded but not acknowledged
> > > > > > > > either. Then I also would think stopping T7 is
> > > > > > > fine.
> > > > > > > >
> > > > > > > > One more question:
> > > > > > > > During changeover, will data received during RPO
> > > > > > > but not acknowledged,
> > > > > > > > considered as received successfully or not, i.e.
> > > > > > > in case of changeover,
> > > > > > > > will they be delivered to MTP3 or not?
> > > > > > >
> > > > > > > Unacknowledged messages must never be delivered to
> > > > > > > MTP3 or duplication
> > > > > > > will result on changeover.  This is true for Q.703
> > > > > > > as well as M2PA.
> > > > > > [TOLGA]Why so? I would think, when Retrieve BSTN is received, it
> > might
> > > > be
> > > > > > good to deliver data received successfully but not delivered to
> MTP3
> > > > > > because of RPO, and then reply to Retrieve BSTN request. This
> migth
> > save
> > > > us
> > > > > > to retranmit them on another link (I am not sure though, whether
> > this
> > > > will
> > > > > > create a problem for MTP3, Q.704 Figure30 Sheet 2 of 6 does not
> show
> > > > what
> > > > > > to do in such a case, but I would think this should be fine, I
> > assume it
> > > > is
> > > > > > possible that MTP2 is delivering a message to MTP3, while MTP3 is
> > > > sending
> > > > > > Retrieve BSTN. So, MTP3 should be prepared to get something
> > different
> > > > than
> > > > > > BSTN after sending Retrive BSTN (?) )
> > > > >
> > > > > Bad idea.
> > > > >
> > > > > --brian
> > > > >
> > > > > --
> > > > > Brian F. G. Bidulock
> > > > > bidulock@openss7.org
> > > > > http://www.openss7.org/
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > >
> > >
> > > --
> > > Brian F. G. Bidulock
> > > bidulock@openss7.org
> > > http://www.openss7.org/
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >
> >

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 28 03:46:56 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA17567
	for <sigtran-archive@odin.ietf.org>; Fri, 28 Feb 2003 03:46:56 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1S8vJ008605
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 03:57:19 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1S8v0p08588;
	Fri, 28 Feb 2003 03:57:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1S8rsp08487
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 03:53:54 -0500
Received: from hindon.hss.co.in (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA17540
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 03:42:50 -0500 (EST)
From: sharmam@hss.hns.com
Received: from hindon.hss.co.in (localhost [127.0.0.1])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h1S8j4s15355
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 14:15:04 +0530 (IST)
Received: from ultra.hss.co.in (ultra [192.168.100.5] (may be forged))
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h1S8j2215348;
	Fri, 28 Feb 2003 14:15:03 +0530 (IST)
Received: from sandesh.hss.hns.com (localhost [127.0.0.1])
	by ultra.hss.co.in (8.10.0/8.10.0) with SMTP id h1S8lXd11936;
	Fri, 28 Feb 2003 14:17:33 +0530 (IST)
Received: by sandesh.hss.hns.com(Lotus SMTP MTA v4.6.3  (733.2 10-16-1998))  id 65256CDB.002F7993 ; Fri, 28 Feb 2003 14:08:33 +0530
X-Lotus-FromDomain: HSS
To: sharmam@hss.hns.com
cc: bidulock@openss7.org, Ken Morneault <kmorneau@cisco.com>, sigtran@ietf.org
Message-ID: <65256CDB.002F78F1.00@sandesh.hss.hns.com>
Date: Fri, 28 Feb 2003 14:11:46 +0530
Subject: Re: [Sigtran] to handle misconfiguration in IUA and Stream 0 for
	 ASPM messages (REPOST)
Mime-Version: 1.0
Content-type: text/plain; charset=us-ascii
Content-Disposition: inline
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>



<REPOST>






sharmam@hss.hns.com on 02/28/2003 10:44:44 AM

To:   bidulock@openss7.org
cc:   Ken Morneault <kmorneau@cisco.com>, sigtran@ietf.org (bcc: Manish
      Sharma/HSS)

Subject:  Re: [Sigtran] to handle handle misconfiguration in IUA and Stream  0
      for ASPM messages






Hi Brian,
   Can you please provide me the text stated in IUA IG ? It might possible that
i have missed that.

  Thanks
  Manish S
  www.hssworld.com





"Brian F. G. Bidulock" <bidulock@openss7.org> on 02/28/2003 10:40:47 AM

Please respond to bidulock@openss7.org

To:   Manish Sharma/HSS@HSS
cc:   Ken Morneault <kmorneau@cisco.com>, sigtran@ietf.org

Subject:  Re: [Sigtran] to handle handle misconfiguration in IUA and Stream 0
      for ASPM messages




sharmam,

If you look at the IUA imp guide I think you will see that it has similar
text to the other UAs with regard to ordering of messages in streams.

--brian

sharmam@hss.hns.com wrote:
(Fri, 28 Feb 2003 10:22:49)
>
>
> Hi ken,
>  Ok i agree that we don't know that which side is configured incorrectly but
>  if there is misconfiguration then its work of System management entity to
> correct that.
>  Its not proper to change configuration based on peer side database( SG/ASP).
>  We can't say to an implementation to give   error for selected IID's.
>
>  Second is
>   " All ASPM messages are sent on a sequenced
>     stream to ensure ordering.SCTP stream '0' SHOULD be used"
>   This is written in RFC 3057 and there is no text regarding this to make it
>   inline with other UA's. Other UA's have no such restriction for ASPM
messages.
>
>
>  M3UA imp guide says
>
>  "The following rules MUST to be followed (see section 3.1.2):
>  1. DATA is never sent on stream 0.
>  2. ASPSM, MGMT, RKM should be sent on stream 0 (Other than BEAT and BEAT ACK)
>  3. SSNM , ASPTM , BEAT and BEAT ACK can be sent on any stream."
>
>  Please see this, Thanks in advance
>
>
>   Regards
>   Manish Sharma
>   www.hssworld.com
>
>
>
>
>
>
> Ken Morneault <kmorneau@cisco.com> on 02/28/2003 12:13:15 AM
>
> To:   Manish Sharma/HSS@HSS
> cc:   sigtran@ietf.org
>
> Subject:  Re: [Sigtran] to handle handle misconfiguration in IUA
>
>
>
>
>
>
> Manish,
>
> Who's to say which side has been configured incorrectly.  Could be
> the SG or ASP side.
>
> The first message flow allows traffic to flow for IIDs 1-5 and provides
> a means for an alarm (by way of the Error msg) to be generated
> to notify the operator of the misconfiguration of IIDs 6-10.
>
> Ken
>
> At 08:05 PM 2/27/2003 +0530, sharmam@hss.hns.com wrote:
>
>
> >Hi Ken,
> >   The objective of this example to handle the misconfiguration, So ASP side
> >stack  has to update its configuration based on peer SG configuration
> >   which is not correct. It is responsibility of the system management
entities
> >to make the ASP and SG configuration alike. Why to update database on
> >   misconfiguration of SG side.
> >
> >   Please correct me if i am wrong.
> >
> >
> >   Regards
> >   Manish S
> >   www.hssworld.com
> >
> >
> >
> >
> >
> >Ken Morneault <kmorneau@cisco.com> on 02/27/2003 07:58:04 PM
> >
> >To:   Manish Sharma/HSS@HSS, sigtran@ietf.org
> >cc:
> >
> >Subject:  Re: [Sigtran] to handle handle misconfiguration in IUA
> >
> >
> >
> >
> >
> >
> >Manish,
> >
> >Why are you proposing that the SG return Error (Interface Identifer)
> >for valid Interface Identifiers?
> >
> >Ken
> >
> >At 07:37 PM 2/27/2003 +0530, sharmam@hss.hns.com wrote:
> >
> >
> >>Hi,
> >>   There is an example shown in IG to handle misconfiguration
> >>
> >>   "This scenario shows the example IUA message flows for the
> >>   establishment of traffic between an SG and an ASP in which some
> >>   of the Interface Identifiers have been misconfigured on the
> >>   ASP side.  The SG in this case has Interface Identifers 1-5
> >>   configured for ASP1."
> >>
> >>                SG                               ASP1
> >>                 |                                |
> >>                 |                                |
> >>                 |<----ASP Active (IIDs 1-10)-----|
> >>                 |---ASP Active Ack (IIDs 1-5)--->|
> >>                 |-------Error (IIDs 6)---------->|
> >>                 |-------Error (IIDs 7)---------->|
> >>                 |-------Error (IIDs 8)---------->|
> >>                 |-------Error (IIDs 9)---------->|
> >>                 |-------Error (IIDs 10)--------->|
> >>
> >>
> >>    The other alternative which i am showing is correct or not , as this is
> >>implementation dependant
> >>
> >>                 SG                               ASP1
> >>                 |                                |
> >>                 |-------Error (IIDs 1)---------->|
> >>                 |-------Error (IIDs 2)---------->|
> >>                 |-------Error (IIDs 3)---------->|
> >>                 |-------Error (IIDs 4)---------->|
> >>                 |-------Error (IIDs 5)---------->|
> >>                 |-------Error (IIDs 6)---------->|
> >>                 |-------Error (IIDs 7)---------->|
> >>                 |-------Error (IIDs 8)---------->|
> >>                 |-------Error (IIDs 9)---------->|
> >>                 |-------Error (IIDs 10)--------->|
> >>
> >>
> >> Regards
> >> Manish Sharma
> >> www.hssworld.com
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >>_______________________________________________
> >>Sigtran mailing list
> >>Sigtran@ietf.org
> >>https://www1.ietf.org/mailman/listinfo/sigtran
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>
>
>
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/






_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 28 04:06:40 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA17921
	for <sigtran-archive@odin.ietf.org>; Fri, 28 Feb 2003 04:06:40 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1S9H3310435
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 04:17:03 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1S9GOp10386;
	Fri, 28 Feb 2003 04:16:24 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1S9Drp10303
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 04:13:53 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA17823
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 04:02:58 -0500 (EST)
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1S96rJT018210
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 03:06:53 -0600 (CST)
Received: from sun3271.ssd.usa.alcatel.com (sun3271.ssd.usa.alcatel.com [143.209.151.211])
	by ssd.usa.alcatel.com (8.11.1/8.11.1) with ESMTP id h1S962F15958
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 03:06:02 -0600 (CST)
Date: Fri, 28 Feb 2003 03:06:02 -0600 (CST)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: sigtran@ietf.org
Message-ID: <Pine.SOL.4.10.10302280053191.12860-100000@sun3271.ssd.usa.alcatel.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Sigtran] [m3ua]-AS-SG
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Hi,

section 5.4.1.1.1

...
- Determine the correct association to the chosen SGP;

I have interpreted this statement as below.

1)ASP is selecting the correct association to the chosen SGP when SGP is
  a member of more than one SG.

 +----+   ASSOC-1
 | SG1|----------+
 +----+          |
                 |     +-----+     
 +----+  ASSOC-2 +-----|     |------ +-----+ 
 | SG2|----------------| SGP1|------ | ASP1| 
 +----+          +-----|     |------ +-----+      
                 |     +-----+     
 +----+  ASSOC-3 |
 | SG3|----------+
 +----+

In this case SGP1 will have 3 IP addresses to have 3 SCTP associations
with ASP1.When ASP1 wants to send a message to SG3 then ASP will select
ASSOC-3.


 Will the below diagram be possible?
(i think it is not possible)

  PC1
 +----+  
 | SG1|----------+
 +----+          |
  PC2            |     +-----+
 +----+          +-----|     |       +-----+
 | SG2|----------------| SGP1|------ | ASP1|
 +----+          +-----|     |       +-----+
  PC3            |     +-----+
 +----+          |
 | SG3|----------+
 +----+   

This diagram is wrong because ASP1 may not be able intstrcut SGP1 to send
a message to a particular SG.

Can SGP1 use PC (which is inside routing lable) to dintinguish SGs?


2)ASP is selecting the correct association to the chosen SGP when SGP is
  a member of single SG.

    SG1
 +-------+
 |+-----+|  ASSOC-1
 || SGP1||---------+
 |+-----+|         |
 |       |         |          +------+
 |+-----+| ASSOC-2 +----------|      | 
 || SGP2||--------------------| ASP1 |
 |+-----+|         +----------|      | 
 |       |         |          +------+     
 |+-----+|ASSOC-3  |
 || SG3 ||---------+
 |+-----+|
 +-------+


Are my interpretations all correct ?

 --Samuel

 

  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 28 04:55:51 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA18940
	for <sigtran-archive@odin.ietf.org>; Fri, 28 Feb 2003 04:55:50 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1SA6FO13130
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 05:06:15 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SA5wp13111;
	Fri, 28 Feb 2003 05:05:59 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SA2Bp13013
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 05:02:11 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA18832
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 04:51:15 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:0z/KeapxNASFUUGbypNH8cG8kfQhjvXd@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1S9tBD23697;
	Fri, 28 Feb 2003 02:55:11 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1S9tBx02332;
	Fri, 28 Feb 2003 02:55:11 -0700
Date: Fri, 28 Feb 2003 02:55:11 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] [m3ua]-AS-SG
Message-ID: <20030228025511.A2173@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	sigtran@ietf.org
References: <Pine.SOL.4.10.10302280053191.12860-100000@sun3271.ssd.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.SOL.4.10.10302280053191.12860-100000@sun3271.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Fri, Feb 28, 2003 at 03:06:02AM -0600
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Samuel,

See Figure 5 - Physical Model on Page 115 of RFC 3332.

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 28 Feb 2003 03:06:02)
> 
> Hi,
> 
> section 5.4.1.1.1
> 
> ...
> - Determine the correct association to the chosen SGP;
> 
> I have interpreted this statement as below.
> 
> 1)ASP is selecting the correct association to the chosen SGP when SGP is
>   a member of more than one SG.
> 
>  +----+   ASSOC-1
>  | SG1|----------+
>  +----+          |
>                  |     +-----+     
>  +----+  ASSOC-2 +-----|     |------ +-----+ 
>  | SG2|----------------| SGP1|------ | ASP1| 
>  +----+          +-----|     |------ +-----+      
>                  |     +-----+     
>  +----+  ASSOC-3 |
>  | SG3|----------+
>  +----+
> 
> In this case SGP1 will have 3 IP addresses to have 3 SCTP associations
> with ASP1.When ASP1 wants to send a message to SG3 then ASP will select
> ASSOC-3.
> 
> 
>  Will the below diagram be possible?
> (i think it is not possible)
> 
>   PC1
>  +----+  
>  | SG1|----------+
>  +----+          |
>   PC2            |     +-----+
>  +----+          +-----|     |       +-----+
>  | SG2|----------------| SGP1|------ | ASP1|
>  +----+          +-----|     |       +-----+
>   PC3            |     +-----+
>  +----+          |
>  | SG3|----------+
>  +----+   
> 
> This diagram is wrong because ASP1 may not be able intstrcut SGP1 to send
> a message to a particular SG.
> 
> Can SGP1 use PC (which is inside routing lable) to dintinguish SGs?
> 
> 
> 2)ASP is selecting the correct association to the chosen SGP when SGP is
>   a member of single SG.
> 
>     SG1
>  +-------+
>  |+-----+|  ASSOC-1
>  || SGP1||---------+
>  |+-----+|         |
>  |       |         |          +------+
>  |+-----+| ASSOC-2 +----------|      | 
>  || SGP2||--------------------| ASP1 |
>  |+-----+|         +----------|      | 
>  |       |         |          +------+     
>  |+-----+|ASSOC-3  |
>  || SG3 ||---------+
>  |+-----+|
>  +-------+
> 
> 
> Are my interpretations all correct ?
> 
>  --Samuel
> 
>  
> 
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 28 05:00:58 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA19033
	for <sigtran-archive@odin.ietf.org>; Fri, 28 Feb 2003 05:00:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1SABM314092
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 05:11:22 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SAB9p14063;
	Fri, 28 Feb 2003 05:11:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SA8Op13970
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 05:08:24 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA18972
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 04:57:28 -0500 (EST)
Subject: Re: [Sigtran] [m3ua]-AS-SG
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF0B363061.44C6EBC6-ONC2256CDB.0035A9A3@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Fri, 28 Feb 2003 11:57:53 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/28/2003
 11:57:58 AM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Samuel,

Please find below my interpretations.

   Tolga


Hi,

section 5.4.1.1.1

...
- Determine the correct association to the chosen SGP;

I have interpreted this statement as below.

1)ASP is selecting the correct association to the chosen SGP when SGP is
  a member of more than one SG.
[TOLGA]I don't think it is correct from modelling point of view to see one
SGP as a member of multiple SGs. I would think, choosing SCTP association
to a given SGP is just for the case, where you have more than one SCTP
association(for whatever reason) to a SGP from your ASP.

 +----+   ASSOC-1
 | SG1|----------+
 +----+          |
                 |     +-----+
 +----+  ASSOC-2 +-----|     |------ +-----+
 | SG2|----------------| SGP1|------ | ASP1|
 +----+          +-----|     |------ +-----+
                 |     +-----+
 +----+  ASSOC-3 |
 | SG3|----------+
 +----+

In this case SGP1 will have 3 IP addresses to have 3 SCTP associations
with ASP1.When ASP1 wants to send a message to SG3 then ASP will select
ASSOC-3.
[TOLGA]As Is aid above, I don't think it is a very good idea to see one SGP
as part of multiple SGs. From ASP point of view, if there are there
different SGs, then there should be three different SGPs:

  ****SG1********
  * +--------+  *
  * | SGP1   +--*-----+
  * +--------+  *     |
  ***************     |
  ****SG2********     |   +--------+
  * +--------+  *     +---+        |
  * | SGP2   +--*---------+ ASP1   |
  * +--------+  *     +---+        |
  ***************     |   +--------+
  ****SG3********     |
  * +--------+  *     |
  * | SGP3   +--*-----+
  * +--------+  *
  ***************


 Will the below diagram be possible?
(i think it is not possible)

  PC1
 +----+
 | SG1|----------+
 +----+          |
  PC2            |     +-----+
 +----+          +-----|     |       +-----+
 | SG2|----------------| SGP1|------ | ASP1|
 +----+          +-----|     |       +-----+
  PC3            |     +-----+
 +----+          |
 | SG3|----------+
 +----+

This diagram is wrong because ASP1 may not be able intstrcut SGP1 to send
a message to a particular SG.
[TOLGA]Because IMO it is not correct to see a SGP as part of multiple SGs,
this one is also not possible.

Can SGP1 use PC (which is inside routing lable) to dintinguish SGs?
[TOLGA]How can you do this?Are your messages destined only to the PC of
SGs?


2)ASP is selecting the correct association to the chosen SGP when SGP is
  a member of single SG.

    SG1
 +-------+
 |+-----+|  ASSOC-1
 || SGP1||---------+
 |+-----+|         |
 |       |         |          +------+
 |+-----+| ASSOC-2 +----------|      |
 || SGP2||--------------------| ASP1 |
 |+-----+|         +----------|      |
 |       |         |          +------+
 |+-----+|ASSOC-3  |
 || SG3 ||---------+
 |+-----+|
 +-------+
[TOLGA]In my opinion, this is, how it should be.


Are my interpretations all correct ?

 --Samuel



  Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
  1000 Coit Road, Plano, Texas 75075
  ******* The opinions expressed are not those of Alcatel USA, Inc. *******

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran




_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 28 05:32:29 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA19657
	for <sigtran-archive@odin.ietf.org>; Fri, 28 Feb 2003 05:32:29 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1SAgtG15873
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 05:42:55 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SAgIp15828;
	Fri, 28 Feb 2003 05:42:18 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SAd6p15720
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 05:39:06 -0500
Received: from auds951.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA19595
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 05:28:09 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1SAW2JT025015;
	Fri, 28 Feb 2003 04:32:04 -0600 (CST)
Received: from axes-mach01.usa.alcatel.com by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id PAA07557; Fri, 28 Feb 2003 15:00:59 -0500 (GMT)
Message-ID: <3E5F3ABA.9787BD15@axes-mach01.usa.alcatel.com>
Date: Fri, 28 Feb 2003 16:02:27 +0530
From: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
CC: sigtran@ietf.org
Subject: Re: [Sigtran] [m3ua]-AS-SG
References: <OF0B363061.44C6EBC6-ONC2256CDB.0035A9A3@LocalDomain>
Content-Type: text/plain; charset=x-user-defined
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Tolga,

    The following mail replied by Brian says, a SGP can be part of multiple
SGs. How is this different from Samuel model?.

--Srini

>Srini,

>ksrinivas
wrote:                                                                 (Thu, 30
Jan 2003 10:08:41)
>> Tolga,
>>
>> An ASP can be a part of different ASs. Similarly, Can a SGP be a part of
>> different SGs?

>Sure.  But from the viewpoint of an ASP, the SGP belongs to one SG.  If an
>SGP were to perform the function for two SGs for the same AS, the ASP would
>have to form two associations with the SGP: one for SG-A, the other for SG-B.
>The SGP would also need some way of knowing (e.g. static port allocation)
>for which SG the ASP was forming an association.

>--brian

TOLGA ASVEREN wrote:

> Samuel,
>
> Please find below my interpretations.
>
>    Tolga
>
> Hi,
>
> section 5.4.1.1.1
>
> ...
> - Determine the correct association to the chosen SGP;
>
> I have interpreted this statement as below.
>
> 1)ASP is selecting the correct association to the chosen SGP when SGP is
>   a member of more than one SG.
> [TOLGA]I don't think it is correct from modelling point of view to see one
> SGP as a member of multiple SGs. I would think, choosing SCTP association
> to a given SGP is just for the case, where you have more than one SCTP
> association(for whatever reason) to a SGP from your ASP.
>
>  +----+   ASSOC-1
>  | SG1|----------+
>  +----+          |
>                  |     +-----+
>  +----+  ASSOC-2 +-----|     |------ +-----+
>  | SG2|----------------| SGP1|------ | ASP1|
>  +----+          +-----|     |------ +-----+
>                  |     +-----+
>  +----+  ASSOC-3 |
>  | SG3|----------+
>  +----+
>
> In this case SGP1 will have 3 IP addresses to have 3 SCTP associations
> with ASP1.When ASP1 wants to send a message to SG3 then ASP will select
> ASSOC-3.
> [TOLGA]As Is aid above, I don't think it is a very good idea to see one SGP
> as part of multiple SGs. From ASP point of view, if there are there
> different SGs, then there should be three different SGPs:
>
>   ****SG1********
>   * +--------+  *
>   * | SGP1   +--*-----+
>   * +--------+  *     |
>   ***************     |
>   ****SG2********     |   +--------+
>   * +--------+  *     +---+        |
>   * | SGP2   +--*---------+ ASP1   |
>   * +--------+  *     +---+        |
>   ***************     |   +--------+
>   ****SG3********     |
>   * +--------+  *     |
>   * | SGP3   +--*-----+
>   * +--------+  *
>   ***************
>
>  Will the below diagram be possible?
> (i think it is not possible)
>
>   PC1
>  +----+
>  | SG1|----------+
>  +----+          |
>   PC2            |     +-----+
>  +----+          +-----|     |       +-----+
>  | SG2|----------------| SGP1|------ | ASP1|
>  +----+          +-----|     |       +-----+
>   PC3            |     +-----+
>  +----+          |
>  | SG3|----------+
>  +----+
>
> This diagram is wrong because ASP1 may not be able intstrcut SGP1 to send
> a message to a particular SG.
> [TOLGA]Because IMO it is not correct to see a SGP as part of multiple SGs,
> this one is also not possible.
>
> Can SGP1 use PC (which is inside routing lable) to dintinguish SGs?
> [TOLGA]How can you do this?Are your messages destined only to the PC of
> SGs?
>
> 2)ASP is selecting the correct association to the chosen SGP when SGP is
>   a member of single SG.
>
>     SG1
>  +-------+
>  |+-----+|  ASSOC-1
>  || SGP1||---------+
>  |+-----+|         |
>  |       |         |          +------+
>  |+-----+| ASSOC-2 +----------|      |
>  || SGP2||--------------------| ASP1 |
>  |+-----+|         +----------|      |
>  |       |         |          +------+
>  |+-----+|ASSOC-3  |
>  || SG3 ||---------+
>  |+-----+|
>  +-------+
> [TOLGA]In my opinion, this is, how it should be.
>
> Are my interpretations all correct ?
>
>  --Samuel
>
>   Alcatel USA, Inc.                  Internet: <userid>@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc. *******
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 28 06:18:57 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA20448
	for <sigtran-archive@odin.ietf.org>; Fri, 28 Feb 2003 06:18:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1SBTO118066
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 06:29:24 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SBSxp18036;
	Fri, 28 Feb 2003 06:29:00 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SBKjp17750
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 06:20:45 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA20243
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 06:09:46 -0500 (EST)
Subject: Re: [Sigtran] [m3ua]-AS-SG
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF56E1405A.0EB79C38-ONC2256CDB.003A1C59@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Fri, 28 Feb 2003 13:10:11 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/28/2003
 01:10:17 PM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Srini,

I don't think, what Brian says is the same with the first diagram of
Samuel's mail,  from the viewpoint of an ASP, one SGP belongs to one SG,
which is violated by the first diagram of Samuel's mail. Furthermore SGP
and SG are not different entities which are communicating with each other.
One SGP is part of a SG. That is the reason, why several SGP boxes are in
one SG box (whereas in the first figure, we even have SCTP associations
between them). I think, the figure Brian has referenced in his message
(RFC page 115) is a good one to understand the realtionship between SG and
SGP.

(And also I don't think it is a good idea to consider interpretations from
individuals -regardless of the person (Brian, no offense ;-) )-binding,
let's consider just RFC and IG).

      Tolga


Tolga,

    The following mail replied by Brian says, a SGP can be part of multiple
SGs. How is this different from Samuel model?.

--Srini

>Srini,

>ksrinivas
wrote:
(Thu, 30
Jan 2003 10:08:41)
>> Tolga,
>>
>> An ASP can be a part of different ASs. Similarly, Can a SGP be a part of
>> different SGs?

>Sure.  But from the viewpoint of an ASP, the SGP belongs to one SG.  If an
>SGP were to perform the function for two SGs for the same AS, the ASP
would
>have to form two associations with the SGP: one for SG-A, the other for
SG-B.
>The SGP would also need some way of knowing (e.g. static port allocation)
>for which SG the ASP was forming an association.

>--brian

TOLGA ASVEREN wrote:

> Samuel,
>
> Please find below my interpretations.
>
>    Tolga
>
> Hi,
>
> section 5.4.1.1.1
>
> ...
> - Determine the correct association to the chosen SGP;
>
> I have interpreted this statement as below.
>
> 1)ASP is selecting the correct association to the chosen SGP when SGP is
>   a member of more than one SG.
> [TOLGA]I don't think it is correct from modelling point of view to see
one
> SGP as a member of multiple SGs. I would think, choosing SCTP association
> to a given SGP is just for the case, where you have more than one SCTP
> association(for whatever reason) to a SGP from your ASP.
>
>  +----+   ASSOC-1
>  | SG1|----------+
>  +----+          |
>                  |     +-----+
>  +----+  ASSOC-2 +-----|     |------ +-----+
>  | SG2|----------------| SGP1|------ | ASP1|
>  +----+          +-----|     |------ +-----+
>                  |     +-----+
>  +----+  ASSOC-3 |
>  | SG3|----------+
>  +----+
>
> In this case SGP1 will have 3 IP addresses to have 3 SCTP associations
> with ASP1.When ASP1 wants to send a message to SG3 then ASP will select
> ASSOC-3.
> [TOLGA]As Is aid above, I don't think it is a very good idea to see one
SGP
> as part of multiple SGs. From ASP point of view, if there are there
> different SGs, then there should be three different SGPs:
>
>   ****SG1********
>   * +--------+  *
>   * | SGP1   +--*-----+
>   * +--------+  *     |
>   ***************     |
>   ****SG2********     |   +--------+
>   * +--------+  *     +---+        |
>   * | SGP2   +--*---------+ ASP1   |
>   * +--------+  *     +---+        |
>   ***************     |   +--------+
>   ****SG3********     |
>   * +--------+  *     |
>   * | SGP3   +--*-----+
>   * +--------+  *
>   ***************
>
>  Will the below diagram be possible?
> (i think it is not possible)
>
>   PC1
>  +----+
>  | SG1|----------+
>  +----+          |
>   PC2            |     +-----+
>  +----+          +-----|     |       +-----+
>  | SG2|----------------| SGP1|------ | ASP1|
>  +----+          +-----|     |       +-----+
>   PC3            |     +-----+
>  +----+          |
>  | SG3|----------+
>  +----+
>
> This diagram is wrong because ASP1 may not be able intstrcut SGP1 to send
> a message to a particular SG.
> [TOLGA]Because IMO it is not correct to see a SGP as part of multiple
SGs,
> this one is also not possible.
>
> Can SGP1 use PC (which is inside routing lable) to dintinguish SGs?
> [TOLGA]How can you do this?Are your messages destined only to the PC of
> SGs?
>
> 2)ASP is selecting the correct association to the chosen SGP when SGP is
>   a member of single SG.
>
>     SG1
>  +-------+
>  |+-----+|  ASSOC-1
>  || SGP1||---------+
>  |+-----+|         |
>  |       |         |          +------+
>  |+-----+| ASSOC-2 +----------|      |
>  || SGP2||--------------------| ASP1 |
>  |+-----+|         +----------|      |
>  |       |         |          +------+
>  |+-----+|ASSOC-3  |
>  || SG3 ||---------+
>  |+-----+|
>  +-------+
> [TOLGA]In my opinion, this is, how it should be.
>
> Are my interpretations all correct ?
>
>  --Samuel
>
>   Alcatel USA, Inc.                  Internet: <userid>
@ssd.usa.alcatel.com
>   1000 Coit Road, Plano, Texas 75075
>   ******* The opinions expressed are not those of Alcatel USA, Inc.
*******
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran





_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 28 07:46:54 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA24409
	for <sigtran-archive@odin.ietf.org>; Fri, 28 Feb 2003 07:46:54 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1SCvNr24578
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 07:57:23 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SCv8p24551;
	Fri, 28 Feb 2003 07:57:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SCocp24099
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 07:50:38 -0500
Received: from pt.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA23862
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 07:39:39 -0500 (EST)
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id HAA21505;
	Fri, 28 Feb 2003 07:37:33 -0500
Message-ID: <3E5F5966.50604@pt.com>
Date: Fri, 28 Feb 2003 07:43:18 -0500
From: Andrew Booth <abooth@pt.com>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.1) Gecko/20020826
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: sigtran@ietf.org, tom.george@alcatel.com
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] M2PA nits
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Hi Tom,

I don't know if this is important enough to worry about but I'll mention 
it anyway:

In section 4.2.1 there is a reference to " FSNs of the User Data 
messages in the M2PA transmit buffer".  Should that be referring to the 
retransmit buffer?

Andrew


-- 
Andrew Booth, Software Developer
Signaling Systems Group, Performance Technologies
613-237-4344

Visit www.pt.com/mailing.html to subscribe to our Packets & Signals eLetter.


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 28 07:54:43 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA24993
	for <sigtran-archive@odin.ietf.org>; Fri, 28 Feb 2003 07:54:43 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1SD5CK25404
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 08:05:12 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SCxsp24719;
	Fri, 28 Feb 2003 07:59:54 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SCqIp24164
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 07:52:18 -0500
Received: from auds953.usa.alcatel.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA24008
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 07:41:17 -0500 (EST)
Received: from axes-mach01.axes-mach01.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.6/8.12.6) with ESMTP id h1SCjAeW020687;
	Fri, 28 Feb 2003 06:45:12 -0600 (CST)
Received: from axes-mach01.usa.alcatel.com by axes-mach01.axes-mach01.usa.alcatel.com (8.9.3+Sun/SMI-SVR4)
	id RAA10996; Fri, 28 Feb 2003 17:14:08 -0500 (GMT)
Message-ID: <3E5F59F1.1C1D4685@axes-mach01.usa.alcatel.com>
Date: Fri, 28 Feb 2003 18:15:37 +0530
From: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>
CC: sigtran@ietf.org
Subject: Re: [Sigtran] [m3ua]-AS-SG
References: <OF56E1405A.0EB79C38-ONC2256CDB.003A1C59@LocalDomain>
Content-Type: text/plain; charset=x-user-defined
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit

Tolga,

An ASP can be a part of different ASs. Similarly, Can a SGP be a part of
different SGs?

As per RFC page 115, the figure has only one Association between any two
Signalling Processes(SGP/ASP). RFC (or even IG) does not say whether we can
have a single Association or multiple Association between any to Signalling
Processes. But Brain mail says we can have multiple associations. What is your
comment on this?.

--Srini

TOLGA ASVEREN wrote:

> Srini,
>
> I don't think, what Brian says is the same with the first diagram of
> Samuel's mail,  from the viewpoint of an ASP, one SGP belongs to one SG,
> which is violated by the first diagram of Samuel's mail. Furthermore SGP
> and SG are not different entities which are communicating with each other.
> One SGP is part of a SG. That is the reason, why several SGP boxes are in
> one SG box (whereas in the first figure, we even have SCTP associations
> between them). I think, the figure Brian has referenced in his message
> (RFC page 115) is a good one to understand the realtionship between SG and
> SGP.
>
> (And also I don't think it is a good idea to consider interpretations from
> individuals -regardless of the person (Brian, no offense ;-) )-binding,
> let's consider just RFC and IG).
>
>       Tolga
>
> Tolga,
>
>     The following mail replied by Brian says, a SGP can be part of multiple
> SGs. How is this different from Samuel model?.
>
> --Srini
>
> >Srini,
>
> >ksrinivas
> wrote:
> (Thu, 30
> Jan 2003 10:08:41)
> >> Tolga,
> >>
> >> An ASP can be a part of different ASs. Similarly, Can a SGP be a part of
> >> different SGs?
>
> >Sure.  But from the viewpoint of an ASP, the SGP belongs to one SG.  If an
> >SGP were to perform the function for two SGs for the same AS, the ASP
> would
> >have to form two associations with the SGP: one for SG-A, the other for
> SG-B.
> >The SGP would also need some way of knowing (e.g. static port allocation)
> >for which SG the ASP was forming an association.
>
> >--brian
>
> TOLGA ASVEREN wrote:
>
> > Samuel,
> >
> > Please find below my interpretations.
> >
> >    Tolga
> >
> > Hi,
> >
> > section 5.4.1.1.1
> >
> > ...
> > - Determine the correct association to the chosen SGP;
> >
> > I have interpreted this statement as below.
> >
> > 1)ASP is selecting the correct association to the chosen SGP when SGP is
> >   a member of more than one SG.
> > [TOLGA]I don't think it is correct from modelling point of view to see
> one
> > SGP as a member of multiple SGs. I would think, choosing SCTP association
> > to a given SGP is just for the case, where you have more than one SCTP
> > association(for whatever reason) to a SGP from your ASP.
> >
> >  +----+   ASSOC-1
> >  | SG1|----------+
> >  +----+          |
> >                  |     +-----+
> >  +----+  ASSOC-2 +-----|     |------ +-----+
> >  | SG2|----------------| SGP1|------ | ASP1|
> >  +----+          +-----|     |------ +-----+
> >                  |     +-----+
> >  +----+  ASSOC-3 |
> >  | SG3|----------+
> >  +----+
> >
> > In this case SGP1 will have 3 IP addresses to have 3 SCTP associations
> > with ASP1.When ASP1 wants to send a message to SG3 then ASP will select
> > ASSOC-3.
> > [TOLGA]As Is aid above, I don't think it is a very good idea to see one
> SGP
> > as part of multiple SGs. From ASP point of view, if there are there
> > different SGs, then there should be three different SGPs:
> >
> >   ****SG1********
> >   * +--------+  *
> >   * | SGP1   +--*-----+
> >   * +--------+  *     |
> >   ***************     |
> >   ****SG2********     |   +--------+
> >   * +--------+  *     +---+        |
> >   * | SGP2   +--*---------+ ASP1   |
> >   * +--------+  *     +---+        |
> >   ***************     |   +--------+
> >   ****SG3********     |
> >   * +--------+  *     |
> >   * | SGP3   +--*-----+
> >   * +--------+  *
> >   ***************
> >
> >  Will the below diagram be possible?
> > (i think it is not possible)
> >
> >   PC1
> >  +----+
> >  | SG1|----------+
> >  +----+          |
> >   PC2            |     +-----+
> >  +----+          +-----|     |       +-----+
> >  | SG2|----------------| SGP1|------ | ASP1|
> >  +----+          +-----|     |       +-----+
> >   PC3            |     +-----+
> >  +----+          |
> >  | SG3|----------+
> >  +----+
> >
> > This diagram is wrong because ASP1 may not be able intstrcut SGP1 to send
> > a message to a particular SG.
> > [TOLGA]Because IMO it is not correct to see a SGP as part of multiple
> SGs,
> > this one is also not possible.
> >
> > Can SGP1 use PC (which is inside routing lable) to dintinguish SGs?
> > [TOLGA]How can you do this?Are your messages destined only to the PC of
> > SGs?
> >
> > 2)ASP is selecting the correct association to the chosen SGP when SGP is
> >   a member of single SG.
> >
> >     SG1
> >  +-------+
> >  |+-----+|  ASSOC-1
> >  || SGP1||---------+
> >  |+-----+|         |
> >  |       |         |          +------+
> >  |+-----+| ASSOC-2 +----------|      |
> >  || SGP2||--------------------| ASP1 |
> >  |+-----+|         +----------|      |
> >  |       |         |          +------+
> >  |+-----+|ASSOC-3  |
> >  || SG3 ||---------+
> >  |+-----+|
> >  +-------+
> > [TOLGA]In my opinion, this is, how it should be.
> >
> > Are my interpretations all correct ?
> >
> >  --Samuel
> >
> >   Alcatel USA, Inc.                  Internet: <userid>
> @ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> *******
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 28 08:02:11 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA25584
	for <sigtran-archive@odin.ietf.org>; Fri, 28 Feb 2003 08:02:11 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1SDCet26602
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 08:12:40 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SDCNp26582;
	Fri, 28 Feb 2003 08:12:23 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SD9sp26453
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 08:09:54 -0500
Received: from postman (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA25379
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 07:58:52 -0500 (EST)
Subject: Re: [Sigtran] [m3ua]-AS-SG
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.6a  January 17, 2001
Message-ID: <OF67AEC730.8333D56C-ONC2256CDB.00465C45@LocalDomain>
From: "TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Date: Fri, 28 Feb 2003 14:59:17 +0200
X-MIMETrack: Serialize by Router on OKSIJEN/OKSIJEN(Release 5.0.8 |June 18, 2001) at 02/28/2003
 02:59:22 PM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>


Srini,


Tolga,

An ASP can be a part of different ASs. Similarly, Can a SGP be a part of
different SGs?
[TOLGA]No.

As per RFC page 115, the figure has only one Association between any two
Signalling Processes(SGP/ASP). RFC (or even IG) does not say whether we can
have a single Association or multiple Association between any to Signalling
Processes. But Brain mail says we can have multiple associations. What is
your
comment on this?.
[TOLGA]I don't think multiple associations between two signalling entities
are forbidden. Just I wouldn't consider it as very usefull. For example,
5.4.1.1.1 (the section Samuel was referring to) talks about determining the
correct association to the choosen SGP.

> Srini,
>
> I don't think, what Brian says is the same with the first diagram of
> Samuel's mail,  from the viewpoint of an ASP, one SGP belongs to one SG,
> which is violated by the first diagram of Samuel's mail. Furthermore SGP
> and SG are not different entities which are communicating with each
other.
> One SGP is part of a SG. That is the reason, why several SGP boxes are in
> one SG box (whereas in the first figure, we even have SCTP associations
> between them). I think, the figure Brian has referenced in his message
> (RFC page 115) is a good one to understand the realtionship between SG
and
> SGP.
>
> (And also I don't think it is a good idea to consider interpretations
from
> individuals -regardless of the person (Brian, no offense ;-) )-binding,
> let's consider just RFC and IG).
>
>       Tolga
>
> Tolga,
>
>     The following mail replied by Brian says, a SGP can be part of
multiple
> SGs. How is this different from Samuel model?.
>
> --Srini
>
> >Srini,
>
> >ksrinivas
> wrote:
> (Thu, 30
> Jan 2003 10:08:41)
> >> Tolga,
> >>
> >> An ASP can be a part of different ASs. Similarly, Can a SGP be a part
of
> >> different SGs?
>
> >Sure.  But from the viewpoint of an ASP, the SGP belongs to one SG.  If
an
> >SGP were to perform the function for two SGs for the same AS, the ASP
> would
> >have to form two associations with the SGP: one for SG-A, the other for
> SG-B.
> >The SGP would also need some way of knowing (e.g. static port
allocation)
> >for which SG the ASP was forming an association.
>
> >--brian
>
> TOLGA ASVEREN wrote:
>
> > Samuel,
> >
> > Please find below my interpretations.
> >
> >    Tolga
> >
> > Hi,
> >
> > section 5.4.1.1.1
> >
> > ...
> > - Determine the correct association to the chosen SGP;
> >
> > I have interpreted this statement as below.
> >
> > 1)ASP is selecting the correct association to the chosen SGP when SGP
is
> >   a member of more than one SG.
> > [TOLGA]I don't think it is correct from modelling point of view to see
> one
> > SGP as a member of multiple SGs. I would think, choosing SCTP
association
> > to a given SGP is just for the case, where you have more than one SCTP
> > association(for whatever reason) to a SGP from your ASP.
> >
> >  +----+   ASSOC-1
> >  | SG1|----------+
> >  +----+          |
> >                  |     +-----+
> >  +----+  ASSOC-2 +-----|     |------ +-----+
> >  | SG2|----------------| SGP1|------ | ASP1|
> >  +----+          +-----|     |------ +-----+
> >                  |     +-----+
> >  +----+  ASSOC-3 |
> >  | SG3|----------+
> >  +----+
> >
> > In this case SGP1 will have 3 IP addresses to have 3 SCTP associations
> > with ASP1.When ASP1 wants to send a message to SG3 then ASP will select
> > ASSOC-3.
> > [TOLGA]As Is aid above, I don't think it is a very good idea to see one
> SGP
> > as part of multiple SGs. From ASP point of view, if there are there
> > different SGs, then there should be three different SGPs:
> >
> >   ****SG1********
> >   * +--------+  *
> >   * | SGP1   +--*-----+
> >   * +--------+  *     |
> >   ***************     |
> >   ****SG2********     |   +--------+
> >   * +--------+  *     +---+        |
> >   * | SGP2   +--*---------+ ASP1   |
> >   * +--------+  *     +---+        |
> >   ***************     |   +--------+
> >   ****SG3********     |
> >   * +--------+  *     |
> >   * | SGP3   +--*-----+
> >   * +--------+  *
> >   ***************
> >
> >  Will the below diagram be possible?
> > (i think it is not possible)
> >
> >   PC1
> >  +----+
> >  | SG1|----------+
> >  +----+          |
> >   PC2            |     +-----+
> >  +----+          +-----|     |       +-----+
> >  | SG2|----------------| SGP1|------ | ASP1|
> >  +----+          +-----|     |       +-----+
> >   PC3            |     +-----+
> >  +----+          |
> >  | SG3|----------+
> >  +----+
> >
> > This diagram is wrong because ASP1 may not be able intstrcut SGP1 to
send
> > a message to a particular SG.
> > [TOLGA]Because IMO it is not correct to see a SGP as part of multiple
> SGs,
> > this one is also not possible.
> >
> > Can SGP1 use PC (which is inside routing lable) to dintinguish SGs?
> > [TOLGA]How can you do this?Are your messages destined only to the PC of
> > SGs?
> >
> > 2)ASP is selecting the correct association to the chosen SGP when SGP
is
> >   a member of single SG.
> >
> >     SG1
> >  +-------+
> >  |+-----+|  ASSOC-1
> >  || SGP1||---------+
> >  |+-----+|         |
> >  |       |         |          +------+
> >  |+-----+| ASSOC-2 +----------|      |
> >  || SGP2||--------------------| ASP1 |
> >  |+-----+|         +----------|      |
> >  |       |         |          +------+
> >  |+-----+|ASSOC-3  |
> >  || SG3 ||---------+
> >  |+-----+|
> >  +-------+
> > [TOLGA]In my opinion, this is, how it should be.
> >
> > Are my interpretations all correct ?
> >
> >  --Samuel
> >
> >   Alcatel USA, Inc.                  Internet: <userid>
> @ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075
> >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> *******
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran





_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 28 08:04:56 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA25633
	for <sigtran-archive@odin.ietf.org>; Fri, 28 Feb 2003 08:04:56 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1SDFQb26717
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 08:15:26 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SDFDp26705;
	Fri, 28 Feb 2003 08:15:13 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SDCap26597
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 08:12:36 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA25566
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 08:01:34 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:8lZuhFXw9+fOXIfBe26GpzwT1OTTelsR@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1SD5UD27523;
	Fri, 28 Feb 2003 06:05:30 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1SD5U007464;
	Fri, 28 Feb 2003 06:05:30 -0700
Date: Fri, 28 Feb 2003 06:05:30 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
Cc: TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
Subject: Re: [Sigtran] [m3ua]-AS-SG
Message-ID: <20030228060530.A7111@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: ksrinivas <ksrini@axes-mach01.usa.alcatel.com>,
	TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
References: <OF56E1405A.0EB79C38-ONC2256CDB.003A1C59@LocalDomain> <3E5F59F1.1C1D4685@axes-mach01.usa.alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E5F59F1.1C1D4685@axes-mach01.usa.alcatel.com>; from ksrini@axes-mach01.usa.alcatel.com on Fri, Feb 28, 2003 at 06:15:37PM +0530
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

ksrinivas,

1.3.2.4 Support for the Management of SCTP Associations between the SGP
        and ASPs.

   ...

   The M3UA layer MAY be instructed by local management to establish an
   SCTP association to a peer M3UA node.  This can be achieved using the
   M-SCTP_ESTABLISH primitives (See Section 1.6.3 for a description of
   management primitives.) to request, indicate and confirm the
   establishment of an SCTP association with a peer M3UA node.  In order
   to avoid redundant SCTP associations between two M3UA peers, one side
   (client) SHOULD be designated to establish the SCTP association, or
   M3UA configuration information maintained to detect redundant
   associations (e.g., via knowledge of the expected local and remote
   SCTP endpoint addresses).

I told you that if you want an SGP to serve 2 SG's it must appear to the
ASPs as 2 different SGPs.  Also that you would need to use a mechanism
such as IP addresses and port numbers to distinguish between these two
views of the SGP.  There is still only one association between an ASP
and an SGP (as viewed by the ASP).

If this is too complicated for you, stick to Figure 5.

--brian

ksrinivas wrote:                                                                 (Fri, 28 Feb 2003 18:15:37)
> Tolga,
> 
> An ASP can be a part of different ASs. Similarly, Can a SGP be a part of
> different SGs?
> 
> As per RFC page 115, the figure has only one Association between any two
> Signalling Processes(SGP/ASP). RFC (or even IG) does not say whether we can
> have a single Association or multiple Association between any to Signalling
> Processes. But Brain mail says we can have multiple associations. What is your
> comment on this?.
> 
> --Srini
> 
> TOLGA ASVEREN wrote:
> 
> > Srini,
> >
> > I don't think, what Brian says is the same with the first diagram of
> > Samuel's mail,  from the viewpoint of an ASP, one SGP belongs to one SG,
> > which is violated by the first diagram of Samuel's mail. Furthermore SGP
> > and SG are not different entities which are communicating with each other.
> > One SGP is part of a SG. That is the reason, why several SGP boxes are in
> > one SG box (whereas in the first figure, we even have SCTP associations
> > between them). I think, the figure Brian has referenced in his message
> > (RFC page 115) is a good one to understand the realtionship between SG and
> > SGP.
> >
> > (And also I don't think it is a good idea to consider interpretations from
> > individuals -regardless of the person (Brian, no offense ;-) )-binding,
> > let's consider just RFC and IG).
> >
> >       Tolga
> >
> > Tolga,
> >
> >     The following mail replied by Brian says, a SGP can be part of multiple
> > SGs. How is this different from Samuel model?.
> >
> > --Srini
> >
> > >Srini,
> >
> > >ksrinivas
> > wrote:
> > (Thu, 30
> > Jan 2003 10:08:41)
> > >> Tolga,
> > >>
> > >> An ASP can be a part of different ASs. Similarly, Can a SGP be a part of
> > >> different SGs?
> >
> > >Sure.  But from the viewpoint of an ASP, the SGP belongs to one SG.  If an
> > >SGP were to perform the function for two SGs for the same AS, the ASP
> > would
> > >have to form two associations with the SGP: one for SG-A, the other for
> > SG-B.
> > >The SGP would also need some way of knowing (e.g. static port allocation)
> > >for which SG the ASP was forming an association.
> >
> > >--brian
> >
> > TOLGA ASVEREN wrote:
> >
> > > Samuel,
> > >
> > > Please find below my interpretations.
> > >
> > >    Tolga
> > >
> > > Hi,
> > >
> > > section 5.4.1.1.1
> > >
> > > ...
> > > - Determine the correct association to the chosen SGP;
> > >
> > > I have interpreted this statement as below.
> > >
> > > 1)ASP is selecting the correct association to the chosen SGP when SGP is
> > >   a member of more than one SG.
> > > [TOLGA]I don't think it is correct from modelling point of view to see
> > one
> > > SGP as a member of multiple SGs. I would think, choosing SCTP association
> > > to a given SGP is just for the case, where you have more than one SCTP
> > > association(for whatever reason) to a SGP from your ASP.
> > >
> > >  +----+   ASSOC-1
> > >  | SG1|----------+
> > >  +----+          |
> > >                  |     +-----+
> > >  +----+  ASSOC-2 +-----|     |------ +-----+
> > >  | SG2|----------------| SGP1|------ | ASP1|
> > >  +----+          +-----|     |------ +-----+
> > >                  |     +-----+
> > >  +----+  ASSOC-3 |
> > >  | SG3|----------+
> > >  +----+
> > >
> > > In this case SGP1 will have 3 IP addresses to have 3 SCTP associations
> > > with ASP1.When ASP1 wants to send a message to SG3 then ASP will select
> > > ASSOC-3.
> > > [TOLGA]As Is aid above, I don't think it is a very good idea to see one
> > SGP
> > > as part of multiple SGs. From ASP point of view, if there are there
> > > different SGs, then there should be three different SGPs:
> > >
> > >   ****SG1********
> > >   * +--------+  *
> > >   * | SGP1   +--*-----+
> > >   * +--------+  *     |
> > >   ***************     |
> > >   ****SG2********     |   +--------+
> > >   * +--------+  *     +---+        |
> > >   * | SGP2   +--*---------+ ASP1   |
> > >   * +--------+  *     +---+        |
> > >   ***************     |   +--------+
> > >   ****SG3********     |
> > >   * +--------+  *     |
> > >   * | SGP3   +--*-----+
> > >   * +--------+  *
> > >   ***************
> > >
> > >  Will the below diagram be possible?
> > > (i think it is not possible)
> > >
> > >   PC1
> > >  +----+
> > >  | SG1|----------+
> > >  +----+          |
> > >   PC2            |     +-----+
> > >  +----+          +-----|     |       +-----+
> > >  | SG2|----------------| SGP1|------ | ASP1|
> > >  +----+          +-----|     |       +-----+
> > >   PC3            |     +-----+
> > >  +----+          |
> > >  | SG3|----------+
> > >  +----+
> > >
> > > This diagram is wrong because ASP1 may not be able intstrcut SGP1 to send
> > > a message to a particular SG.
> > > [TOLGA]Because IMO it is not correct to see a SGP as part of multiple
> > SGs,
> > > this one is also not possible.
> > >
> > > Can SGP1 use PC (which is inside routing lable) to dintinguish SGs?
> > > [TOLGA]How can you do this?Are your messages destined only to the PC of
> > > SGs?
> > >
> > > 2)ASP is selecting the correct association to the chosen SGP when SGP is
> > >   a member of single SG.
> > >
> > >     SG1
> > >  +-------+
> > >  |+-----+|  ASSOC-1
> > >  || SGP1||---------+
> > >  |+-----+|         |
> > >  |       |         |          +------+
> > >  |+-----+| ASSOC-2 +----------|      |
> > >  || SGP2||--------------------| ASP1 |
> > >  |+-----+|         +----------|      |
> > >  |       |         |          +------+
> > >  |+-----+|ASSOC-3  |
> > >  || SG3 ||---------+
> > >  |+-----+|
> > >  +-------+
> > > [TOLGA]In my opinion, this is, how it should be.
> > >
> > > Are my interpretations all correct ?
> > >
> > >  --Samuel
> > >
> > >   Alcatel USA, Inc.                  Internet: <userid>
> > @ssd.usa.alcatel.com
> > >   1000 Coit Road, Plano, Texas 75075
> > >   ******* The opinions expressed are not those of Alcatel USA, Inc.
> > *******
> > >
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 28 09:46:31 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA00197
	for <sigtran-archive@odin.ietf.org>; Fri, 28 Feb 2003 09:46:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1SEv1L01188
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 09:57:01 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SEucp01167;
	Fri, 28 Feb 2003 09:56:38 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SEr4p01004
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 09:53:04 -0500
Received: from mail-gw.radisys.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA00014
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 09:41:59 -0500 (EST)
From: Barry.Nagelberg@radisys.com
Received: by mail-gw.radisys.com (Postfix, from userid 5)
	id 25CAC4E7D0; Fri, 28 Feb 2003 06:41:08 -0800 (PST)
Received: from notes.radisys.com(206.103.52.220)
 via SMTP by mail-gw.radisys.com, id smtpdAAAFoa4UH; Fri Feb 28 06:41:02 2003
To: sigtran@ietf.org
Subject: Re: [Sigtran] [m3ua]-AS-SG
MIME-Version: 1.0
X-Mailer: Lotus Notes Release 5.0.8  June 18, 2001
Message-ID: <OFA705ECB4.7EC8721A-ON85256CDB.004E8905@radisys.com>
Date: Fri, 28 Feb 2003 09:21:51 -0500
X-MIMETrack: Serialize by Router on HQ_ACS_1/Radisys_Corporation/US(Release 5.0.9a |January
 7, 2002) at 02/28/2003 06:38:20 AM,
	Serialize complete at 02/28/2003 06:38:20 AM
Content-Type: text/plain; charset="us-ascii"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Brian,

Please tell us where in the RFC or IG did you find that

 "...if you want an SGP to serve 2 SGs it must appear to the ASPs as 2 
different SGPs".

Barry Nagelberg





"Brian F. G. Bidulock" <bidulock@openss7.org>
Sent by: sigtran-admin@ietf.org
 02/28/2003 08:05 AM
 Please respond to bidulock

 
        To:     ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
        cc:     TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
        Subject:        Re: [Sigtran] [m3ua]-AS-SG


ksrinivas,

1.3.2.4 Support for the Management of SCTP Associations between the SGP
        and ASPs.

   ...

   The M3UA layer MAY be instructed by local management to establish an
   SCTP association to a peer M3UA node.  This can be achieved using the
   M-SCTP_ESTABLISH primitives (See Section 1.6.3 for a description of
   management primitives.) to request, indicate and confirm the
   establishment of an SCTP association with a peer M3UA node.  In order
   to avoid redundant SCTP associations between two M3UA peers, one side
   (client) SHOULD be designated to establish the SCTP association, or
   M3UA configuration information maintained to detect redundant
   associations (e.g., via knowledge of the expected local and remote
   SCTP endpoint addresses).

I told you that if you want an SGP to serve 2 SG's it must appear to the
ASPs as 2 different SGPs.  Also that you would need to use a mechanism
such as IP addresses and port numbers to distinguish between these two
views of the SGP.  There is still only one association between an ASP
and an SGP (as viewed by the ASP).

If this is too complicated for you, stick to Figure 5.

--brian

ksrinivas wrote:       (Fri, 28 Feb 2003 18:15:37)
> Tolga,
> 
> An ASP can be a part of different ASs. Similarly, Can a SGP be a part of
> different SGs?
> 
> As per RFC page 115, the figure has only one Association between any two
> Signalling Processes(SGP/ASP). RFC (or even IG) does not say whether we 
can
> have a single Association or multiple Association between any to 
Signalling
> Processes. But Brain mail says we can have multiple associations. What 
is your
> comment on this?.
> 
> --Srini
> 
> TOLGA ASVEREN wrote:
> 
> > Srini,
> >
> > I don't think, what Brian says is the same with the first diagram of
> > Samuel's mail,  from the viewpoint of an ASP, one SGP belongs to one 
SG,
> > which is violated by the first diagram of Samuel's mail. Furthermore 
SGP
> > and SG are not different entities which are communicating with each 
other.
> > One SGP is part of a SG. That is the reason, why several SGP boxes are 
in
> > one SG box (whereas in the first figure, we even have SCTP 
associations
> > between them). I think, the figure Brian has referenced in his message
> > (RFC page 115) is a good one to understand the realtionship between SG 
and
> > SGP.
> >
> > (And also I don't think it is a good idea to consider interpretations 
from
> > individuals -regardless of the person (Brian, no offense ;-) 
)-binding,
> > let's consider just RFC and IG).
> >
> >       Tolga
> >
> > Tolga,
> >
> >     The following mail replied by Brian says, a SGP can be part of 
multiple
> > SGs. How is this different from Samuel model?.
> >
> > --Srini
> >
> > >Srini,
> >
> > >ksrinivas
> > wrote:
> > (Thu, 30
> > Jan 2003 10:08:41)
> > >> Tolga,
> > >>
> > >> An ASP can be a part of different ASs. Similarly, Can a SGP be a 
part of
> > >> different SGs?
> >
> > >Sure.  But from the viewpoint of an ASP, the SGP belongs to one SG. 
If an
> > >SGP were to perform the function for two SGs for the same AS, the ASP
> > would
> > >have to form two associations with the SGP: one for SG-A, the other 
for
> > SG-B.
> > >The SGP would also need some way of knowing (e.g. static port 
allocation)
> > >for which SG the ASP was forming an association.
> >
> > >--brian
> >
> > TOLGA ASVEREN wrote:
> >
> > > Samuel,
> > >
> > > Please find below my interpretations.
> > >
> > >    Tolga
> > >
> > > Hi,
> > >
> > > section 5.4.1.1.1
> > >
> > > ...
> > > - Determine the correct association to the chosen SGP;
> > >
> > > I have interpreted this statement as below.
> > >
> > > 1)ASP is selecting the correct association to the chosen SGP when 
SGP is
> > >   a member of more than one SG.
> > > [TOLGA]I don't think it is correct from modelling point of view to 
see
> > one
> > > SGP as a member of multiple SGs. I would think, choosing SCTP 
association
> > > to a given SGP is just for the case, where you have more than one 
SCTP
> > > association(for whatever reason) to a SGP from your ASP.
> > >
> > >  +----+   ASSOC-1
> > >  | SG1|----------+
> > >  +----+          |
> > >                  |     +-----+
> > >  +----+  ASSOC-2 +-----|     |------ +-----+
> > >  | SG2|----------------| SGP1|------ | ASP1|
> > >  +----+          +-----|     |------ +-----+
> > >                  |     +-----+
> > >  +----+  ASSOC-3 |
> > >  | SG3|----------+
> > >  +----+
> > >
> > > In this case SGP1 will have 3 IP addresses to have 3 SCTP 
associations
> > > with ASP1.When ASP1 wants to send a message to SG3 then ASP will 
select
> > > ASSOC-3.
> > > [TOLGA]As Is aid above, I don't think it is a very good idea to see 
one
> > SGP
> > > as part of multiple SGs. From ASP point of view, if there are there
> > > different SGs, then there should be three different SGPs:
> > >
> > >   ****SG1********
> > >   * +--------+  *
> > >   * | SGP1   +--*-----+
> > >   * +--------+  *     |
> > >   ***************     |
> > >   ****SG2********     |   +--------+
> > >   * +--------+  *     +---+        |
> > >   * | SGP2   +--*---------+ ASP1   |
> > >   * +--------+  *     +---+        |
> > >   ***************     |   +--------+
> > >   ****SG3********     |
> > >   * +--------+  *     |
> > >   * | SGP3   +--*-----+
> > >   * +--------+  *
> > >   ***************
> > >
> > >  Will the below diagram be possible?
> > > (i think it is not possible)
> > >
> > >   PC1
> > >  +----+
> > >  | SG1|----------+
> > >  +----+          |
> > >   PC2            |     +-----+
> > >  +----+          +-----|     |       +-----+
> > >  | SG2|----------------| SGP1|------ | ASP1|
> > >  +----+          +-----|     |       +-----+
> > >   PC3            |     +-----+
> > >  +----+          |
> > >  | SG3|----------+
> > >  +----+
> > >
> > > This diagram is wrong because ASP1 may not be able intstrcut SGP1 to 
send
> > > a message to a particular SG.
> > > [TOLGA]Because IMO it is not correct to see a SGP as part of 
multiple
> > SGs,
> > > this one is also not possible.
> > >
> > > Can SGP1 use PC (which is inside routing lable) to dintinguish SGs?
> > > [TOLGA]How can you do this?Are your messages destined only to the PC 
of
> > > SGs?
> > >
> > > 2)ASP is selecting the correct association to the chosen SGP when 
SGP is
> > >   a member of single SG.
> > >
> > >     SG1
> > >  +-------+
> > >  |+-----+|  ASSOC-1
> > >  || SGP1||---------+
> > >  |+-----+|         |
> > >  |       |         |          +------+
> > >  |+-----+| ASSOC-2 +----------|      |
> > >  || SGP2||--------------------| ASP1 |
> > >  |+-----+|         +----------|      |
> > >  |       |         |          +------+
> > >  |+-----+|ASSOC-3  |
> > >  || SG3 ||---------+
> > >  |+-----+|
> > >  +-------+
> > > [TOLGA]In my opinion, this is, how it should be.
> > >
> > > Are my interpretations all correct ?
> > >
> > >  --Samuel
> > >



_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 28 10:16:19 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA01161
	for <sigtran-archive@odin.ietf.org>; Fri, 28 Feb 2003 10:16:19 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1SFQo203449
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 10:26:50 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SFQbp03433;
	Fri, 28 Feb 2003 10:26:37 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SFLIp03125
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 10:21:18 -0500
Received: from mail-gw.radisys.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA01026
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 10:10:15 -0500 (EST)
From: Barry.Nagelberg@radisys.com
Received: by mail-gw.radisys.com (Postfix, from userid 5)
	id 3C7804E7D0; Fri, 28 Feb 2003 07:09:23 -0800 (PST)
Received: from notes.radisys.com(206.103.52.220)
 via SMTP by mail-gw.radisys.com, id smtpdAAAlfaGic; Fri Feb 28 07:09:18 2003
To: sigtran@ietf.org
Subject: Re: [Sigtran] [m3ua]-AS-SG
MIME-Version: 1.0
X-Mailer: Lotus Notes Release 5.0.8  June 18, 2001
Message-ID: <OFD043C5B8.D46A7740-ON85256CDB.0050B85B@radisys.com>
Date: Fri, 28 Feb 2003 09:50:05 -0500
X-MIMETrack: Serialize by Router on HQ_ACS_1/Radisys_Corporation/US(Release 5.0.9a |January
 7, 2002) at 02/28/2003 07:06:36 AM,
	Serialize complete at 02/28/2003 07:06:36 AM
Content-Type: text/plain; charset="us-ascii"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Tolga,

Please tell us where in the RFC or IG it is stated that an SGP cannot be 
part of different SGs.

Barry Nagelberg





"TOLGA ASVEREN" <TOLGA.ASVEREN@o2.com.tr>
Sent by: sigtran-admin@ietf.org
 02/28/2003 07:59 AM

 
        To:     sigtran@ietf.org
        cc: 
        Subject:        Re: [Sigtran] [m3ua]-AS-SG



Srini,


Tolga,

An ASP can be a part of different ASs. Similarly, Can a SGP be a part of
different SGs?
[TOLGA]No.

As per RFC page 115, the figure has only one Association between any two
Signalling Processes(SGP/ASP). RFC (or even IG) does not say whether we 
can
have a single Association or multiple Association between any to 
Signalling
Processes. But Brain mail says we can have multiple associations. What is
your
comment on this?.
[TOLGA]I don't think multiple associations between two signalling entities
are forbidden. Just I wouldn't consider it as very usefull. For example,
5.4.1.1.1 (the section Samuel was referring to) talks about determining 
the
correct association to the choosen SGP.

> Srini,
>
> I don't think, what Brian says is the same with the first diagram of
> Samuel's mail,  from the viewpoint of an ASP, one SGP belongs to one SG,
> which is violated by the first diagram of Samuel's mail. Furthermore SGP
> and SG are not different entities which are communicating with each
other.
> One SGP is part of a SG. That is the reason, why several SGP boxes are 
in
> one SG box (whereas in the first figure, we even have SCTP associations
> between them). I think, the figure Brian has referenced in his message
> (RFC page 115) is a good one to understand the realtionship between SG
and
> SGP.
>
> (And also I don't think it is a good idea to consider interpretations
from
> individuals -regardless of the person (Brian, no offense ;-) )-binding,
> let's consider just RFC and IG).
>
>       Tolga
>
> Tolga,
>
>     The following mail replied by Brian says, a SGP can be part of
multiple
> SGs. How is this different from Samuel model?.
>
> --Srini
>
> >Srini,
>
> >ksrinivas
> wrote:
> (Thu, 30
> Jan 2003 10:08:41)
> >> Tolga,
> >>
> >> An ASP can be a part of different ASs. Similarly, Can a SGP be a part
of
> >> different SGs?
>
> >Sure.  But from the viewpoint of an ASP, the SGP belongs to one SG.  If
an
> >SGP were to perform the function for two SGs for the same AS, the ASP
> would
> >have to form two associations with the SGP: one for SG-A, the other for
> SG-B.
> >The SGP would also need some way of knowing (e.g. static port
allocation)
> >for which SG the ASP was forming an association.
>
> >--brian
>
> TOLGA ASVEREN wrote:
>
> > Samuel,
> >
> > Please find below my interpretations.
> >
> >    Tolga
> >
> > Hi,
> >
> > section 5.4.1.1.1
> >
> > ...
> > - Determine the correct association to the chosen SGP;
> >
> > I have interpreted this statement as below.
> >
> > 1)ASP is selecting the correct association to the chosen SGP when SGP
is
> >   a member of more than one SG.
> > [TOLGA]I don't think it is correct from modelling point of view to see
> one
> > SGP as a member of multiple SGs. I would think, choosing SCTP
association
> > to a given SGP is just for the case, where you have more than one SCTP
> > association(for whatever reason) to a SGP from your ASP.
> >
> >  +----+   ASSOC-1
> >  | SG1|----------+
> >  +----+          |
> >                  |     +-----+
> >  +----+  ASSOC-2 +-----|     |------ +-----+
> >  | SG2|----------------| SGP1|------ | ASP1|
> >  +----+          +-----|     |------ +-----+
> >                  |     +-----+
> >  +----+  ASSOC-3 |
> >  | SG3|----------+
> >  +----+
> >
> > In this case SGP1 will have 3 IP addresses to have 3 SCTP associations
> > with ASP1.When ASP1 wants to send a message to SG3 then ASP will 
select
> > ASSOC-3.
> > [TOLGA]As Is aid above, I don't think it is a very good idea to see 
one
> SGP
> > as part of multiple SGs. From ASP point of view, if there are there
> > different SGs, then there should be three different SGPs:
> >
> >   ****SG1********
> >   * +--------+  *
> >   * | SGP1   +--*-----+
> >   * +--------+  *     |
> >   ***************     |
> >   ****SG2********     |   +--------+
> >   * +--------+  *     +---+        |
> >   * | SGP2   +--*---------+ ASP1   |
> >   * +--------+  *     +---+        |
> >   ***************     |   +--------+
> >   ****SG3********     |
> >   * +--------+  *     |
> >   * | SGP3   +--*-----+
> >   * +--------+  *
> >   ***************
> >
> >  Will the below diagram be possible?
> > (i think it is not possible)
> >
> >   PC1
> >  +----+
> >  | SG1|----------+
> >  +----+          |
> >   PC2            |     +-----+
> >  +----+          +-----|     |       +-----+
> >  | SG2|----------------| SGP1|------ | ASP1|
> >  +----+          +-----|     |       +-----+
> >   PC3            |     +-----+
> >  +----+          |
> >  | SG3|----------+
> >  +----+
> >
> > This diagram is wrong because ASP1 may not be able intstrcut SGP1 to
send
> > a message to a particular SG.
> > [TOLGA]Because IMO it is not correct to see a SGP as part of multiple
> SGs,
> > this one is also not possible.
> >
> > Can SGP1 use PC (which is inside routing lable) to dintinguish SGs?
> > [TOLGA]How can you do this?Are your messages destined only to the PC 
of
> > SGs?
> >
> > 2)ASP is selecting the correct association to the chosen SGP when SGP
is
> >   a member of single SG.
> >
> >     SG1
> >  +-------+
> >  |+-----+|  ASSOC-1
> >  || SGP1||---------+
> >  |+-----+|         |
> >  |       |         |          +------+
> >  |+-----+| ASSOC-2 +----------|      |
> >  || SGP2||--------------------| ASP1 |
> >  |+-----+|         +----------|      |
> >  |       |         |          +------+
> >  |+-----+|ASSOC-3  |
> >  || SG3 ||---------+
> >  |+-----+|
> >  +-------+
> > [TOLGA]In my opinion, this is, how it should be.
> >
> > Are my interpretations all correct ?
> >
> >  --Samuel
> >


_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 28 17:11:32 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA15020
	for <sigtran-archive@odin.ietf.org>; Fri, 28 Feb 2003 17:11:32 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1SMKDF00376
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 17:20:13 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SMJtp32766;
	Fri, 28 Feb 2003 17:19:56 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SMI9p32701
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 17:18:09 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA14957
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 17:08:55 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:Jn5u8S1S/qRfTkyb52ZklANkJyZJ5q7e@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1SMArD05816;
	Fri, 28 Feb 2003 15:10:53 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1SMArS20971;
	Fri, 28 Feb 2003 15:10:53 -0700
Date: Fri, 28 Feb 2003 15:10:53 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Barry.Nagelberg@radisys.com
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] [m3ua]-AS-SG
Message-ID: <20030228151053.A20045@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Barry.Nagelberg@radisys.com, sigtran@ietf.org
References: <OFA705ECB4.7EC8721A-ON85256CDB.004E8905@radisys.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OFA705ECB4.7EC8721A-ON85256CDB.004E8905@radisys.com>; from Barry.Nagelberg@radisys.com on Fri, Feb 28, 2003 at 09:21:51AM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Barry.Nagelberg,

Read quote below.

--brian

Barry.Nagelberg@radisys.com wrote:                                         (Fri, 28 Feb 2003 09:21:51)
> Brian,
> 
> Please tell us where in the RFC or IG did you find that
> 
>  "...if you want an SGP to serve 2 SGs it must appear to the ASPs as 2 
> different SGPs".
> 
> Barry Nagelberg
> 
> 
> 
> 
> 
> "Brian F. G. Bidulock" <bidulock@openss7.org>
> Sent by: sigtran-admin@ietf.org
>  02/28/2003 08:05 AM
>  Please respond to bidulock
> 
>  
>         To:     ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
>         cc:     TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
>         Subject:        Re: [Sigtran] [m3ua]-AS-SG
> 
> 
> ksrinivas,
> 
> 1.3.2.4 Support for the Management of SCTP Associations between the SGP
>         and ASPs.
> 
>    ...
> 
>    The M3UA layer MAY be instructed by local management to establish an
>    SCTP association to a peer M3UA node.  This can be achieved using the
>    M-SCTP_ESTABLISH primitives (See Section 1.6.3 for a description of
>    management primitives.) to request, indicate and confirm the
>    establishment of an SCTP association with a peer M3UA node.  In order
>    to avoid redundant SCTP associations between two M3UA peers, one side
>    (client) SHOULD be designated to establish the SCTP association, or
>    M3UA configuration information maintained to detect redundant
>    associations (e.g., via knowledge of the expected local and remote
>    SCTP endpoint addresses).
> 
> I told you that if you want an SGP to serve 2 SG's it must appear to the
> ASPs as 2 different SGPs.  Also that you would need to use a mechanism
> such as IP addresses and port numbers to distinguish between these two
> views of the SGP.  There is still only one association between an ASP
> and an SGP (as viewed by the ASP).
> 
> If this is too complicated for you, stick to Figure 5.
> 
> --brian
> 
> ksrinivas wrote:       (Fri, 28 Feb 2003 18:15:37)
> > Tolga,
> > 
> > An ASP can be a part of different ASs. Similarly, Can a SGP be a part of
> > different SGs?
> > 
> > As per RFC page 115, the figure has only one Association between any two
> > Signalling Processes(SGP/ASP). RFC (or even IG) does not say whether we 
> can
> > have a single Association or multiple Association between any to 
> Signalling
> > Processes. But Brain mail says we can have multiple associations. What 
> is your
> > comment on this?.
> > 
> > --Srini
> > 
> > TOLGA ASVEREN wrote:
> > 
> > > Srini,
> > >
> > > I don't think, what Brian says is the same with the first diagram of
> > > Samuel's mail,  from the viewpoint of an ASP, one SGP belongs to one 
> SG,
> > > which is violated by the first diagram of Samuel's mail. Furthermore 
> SGP
> > > and SG are not different entities which are communicating with each 
> other.
> > > One SGP is part of a SG. That is the reason, why several SGP boxes are 
> in
> > > one SG box (whereas in the first figure, we even have SCTP 
> associations
> > > between them). I think, the figure Brian has referenced in his message
> > > (RFC page 115) is a good one to understand the realtionship between SG 
> and
> > > SGP.
> > >
> > > (And also I don't think it is a good idea to consider interpretations 
> from
> > > individuals -regardless of the person (Brian, no offense ;-) 
> )-binding,
> > > let's consider just RFC and IG).
> > >
> > >       Tolga
> > >
> > > Tolga,
> > >
> > >     The following mail replied by Brian says, a SGP can be part of 
> multiple
> > > SGs. How is this different from Samuel model?.
> > >
> > > --Srini
> > >
> > > >Srini,
> > >
> > > >ksrinivas
> > > wrote:
> > > (Thu, 30
> > > Jan 2003 10:08:41)
> > > >> Tolga,
> > > >>
> > > >> An ASP can be a part of different ASs. Similarly, Can a SGP be a 
> part of
> > > >> different SGs?
> > >
> > > >Sure.  But from the viewpoint of an ASP, the SGP belongs to one SG. 
> If an
> > > >SGP were to perform the function for two SGs for the same AS, the ASP
> > > would
> > > >have to form two associations with the SGP: one for SG-A, the other 
> for
> > > SG-B.
> > > >The SGP would also need some way of knowing (e.g. static port 
> allocation)
> > > >for which SG the ASP was forming an association.
> > >
> > > >--brian
> > >
> > > TOLGA ASVEREN wrote:
> > >
> > > > Samuel,
> > > >
> > > > Please find below my interpretations.
> > > >
> > > >    Tolga
> > > >
> > > > Hi,
> > > >
> > > > section 5.4.1.1.1
> > > >
> > > > ...
> > > > - Determine the correct association to the chosen SGP;
> > > >
> > > > I have interpreted this statement as below.
> > > >
> > > > 1)ASP is selecting the correct association to the chosen SGP when 
> SGP is
> > > >   a member of more than one SG.
> > > > [TOLGA]I don't think it is correct from modelling point of view to 
> see
> > > one
> > > > SGP as a member of multiple SGs. I would think, choosing SCTP 
> association
> > > > to a given SGP is just for the case, where you have more than one 
> SCTP
> > > > association(for whatever reason) to a SGP from your ASP.
> > > >
> > > >  +----+   ASSOC-1
> > > >  | SG1|----------+
> > > >  +----+          |
> > > >                  |     +-----+
> > > >  +----+  ASSOC-2 +-----|     |------ +-----+
> > > >  | SG2|----------------| SGP1|------ | ASP1|
> > > >  +----+          +-----|     |------ +-----+
> > > >                  |     +-----+
> > > >  +----+  ASSOC-3 |
> > > >  | SG3|----------+
> > > >  +----+
> > > >
> > > > In this case SGP1 will have 3 IP addresses to have 3 SCTP 
> associations
> > > > with ASP1.When ASP1 wants to send a message to SG3 then ASP will 
> select
> > > > ASSOC-3.
> > > > [TOLGA]As Is aid above, I don't think it is a very good idea to see 
> one
> > > SGP
> > > > as part of multiple SGs. From ASP point of view, if there are there
> > > > different SGs, then there should be three different SGPs:
> > > >
> > > >   ****SG1********
> > > >   * +--------+  *
> > > >   * | SGP1   +--*-----+
> > > >   * +--------+  *     |
> > > >   ***************     |
> > > >   ****SG2********     |   +--------+
> > > >   * +--------+  *     +---+        |
> > > >   * | SGP2   +--*---------+ ASP1   |
> > > >   * +--------+  *     +---+        |
> > > >   ***************     |   +--------+
> > > >   ****SG3********     |
> > > >   * +--------+  *     |
> > > >   * | SGP3   +--*-----+
> > > >   * +--------+  *
> > > >   ***************
> > > >
> > > >  Will the below diagram be possible?
> > > > (i think it is not possible)
> > > >
> > > >   PC1
> > > >  +----+
> > > >  | SG1|----------+
> > > >  +----+          |
> > > >   PC2            |     +-----+
> > > >  +----+          +-----|     |       +-----+
> > > >  | SG2|----------------| SGP1|------ | ASP1|
> > > >  +----+          +-----|     |       +-----+
> > > >   PC3            |     +-----+
> > > >  +----+          |
> > > >  | SG3|----------+
> > > >  +----+
> > > >
> > > > This diagram is wrong because ASP1 may not be able intstrcut SGP1 to 
> send
> > > > a message to a particular SG.
> > > > [TOLGA]Because IMO it is not correct to see a SGP as part of 
> multiple
> > > SGs,
> > > > this one is also not possible.
> > > >
> > > > Can SGP1 use PC (which is inside routing lable) to dintinguish SGs?
> > > > [TOLGA]How can you do this?Are your messages destined only to the PC 
> of
> > > > SGs?
> > > >
> > > > 2)ASP is selecting the correct association to the chosen SGP when 
> SGP is
> > > >   a member of single SG.
> > > >
> > > >     SG1
> > > >  +-------+
> > > >  |+-----+|  ASSOC-1
> > > >  || SGP1||---------+
> > > >  |+-----+|         |
> > > >  |       |         |          +------+
> > > >  |+-----+| ASSOC-2 +----------|      |
> > > >  || SGP2||--------------------| ASP1 |
> > > >  |+-----+|         +----------|      |
> > > >  |       |         |          +------+
> > > >  |+-----+|ASSOC-3  |
> > > >  || SG3 ||---------+
> > > >  |+-----+|
> > > >  +-------+
> > > > [TOLGA]In my opinion, this is, how it should be.
> > > >
> > > > Are my interpretations all correct ?
> > > >
> > > >  --Samuel
> > > >
> 
> 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 28 17:11:43 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA15033
	for <sigtran-archive@odin.ietf.org>; Fri, 28 Feb 2003 17:11:42 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1SMKOb00394
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 17:20:24 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SMKCp00369;
	Fri, 28 Feb 2003 17:20:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SMIhp32729
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 17:18:43 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA14970
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 17:09:30 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:BkyidMwtIZ/EM2r+khi9dHSCFgf8wMyG@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1SMBSD05829;
	Fri, 28 Feb 2003 15:11:28 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1SMBSS20992;
	Fri, 28 Feb 2003 15:11:28 -0700
Date: Fri, 28 Feb 2003 15:11:28 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Barry.Nagelberg@radisys.com
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] [m3ua]-AS-SG
Message-ID: <20030228151128.B20045@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Barry.Nagelberg@radisys.com, sigtran@ietf.org
References: <OFA705ECB4.7EC8721A-ON85256CDB.004E8905@radisys.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OFA705ECB4.7EC8721A-ON85256CDB.004E8905@radisys.com>; from Barry.Nagelberg@radisys.com on Fri, Feb 28, 2003 at 09:21:51AM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Barry.Nagelberg,

See Figure 5.

--brian

Barry.Nagelberg@radisys.com wrote:                                         (Fri, 28 Feb 2003 09:21:51)
> Brian,
> 
> Please tell us where in the RFC or IG did you find that
> 
>  "...if you want an SGP to serve 2 SGs it must appear to the ASPs as 2 
> different SGPs".
> 
> Barry Nagelberg
> 
> 
> 
> 
> 
> "Brian F. G. Bidulock" <bidulock@openss7.org>
> Sent by: sigtran-admin@ietf.org
>  02/28/2003 08:05 AM
>  Please respond to bidulock
> 
>  
>         To:     ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
>         cc:     TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, sigtran@ietf.org
>         Subject:        Re: [Sigtran] [m3ua]-AS-SG
> 
> 
> ksrinivas,
> 
> 1.3.2.4 Support for the Management of SCTP Associations between the SGP
>         and ASPs.
> 
>    ...
> 
>    The M3UA layer MAY be instructed by local management to establish an
>    SCTP association to a peer M3UA node.  This can be achieved using the
>    M-SCTP_ESTABLISH primitives (See Section 1.6.3 for a description of
>    management primitives.) to request, indicate and confirm the
>    establishment of an SCTP association with a peer M3UA node.  In order
>    to avoid redundant SCTP associations between two M3UA peers, one side
>    (client) SHOULD be designated to establish the SCTP association, or
>    M3UA configuration information maintained to detect redundant
>    associations (e.g., via knowledge of the expected local and remote
>    SCTP endpoint addresses).
> 
> I told you that if you want an SGP to serve 2 SG's it must appear to the
> ASPs as 2 different SGPs.  Also that you would need to use a mechanism
> such as IP addresses and port numbers to distinguish between these two
> views of the SGP.  There is still only one association between an ASP
> and an SGP (as viewed by the ASP).
> 
> If this is too complicated for you, stick to Figure 5.
> 
> --brian
> 
> ksrinivas wrote:       (Fri, 28 Feb 2003 18:15:37)
> > Tolga,
> > 
> > An ASP can be a part of different ASs. Similarly, Can a SGP be a part of
> > different SGs?
> > 
> > As per RFC page 115, the figure has only one Association between any two
> > Signalling Processes(SGP/ASP). RFC (or even IG) does not say whether we 
> can
> > have a single Association or multiple Association between any to 
> Signalling
> > Processes. But Brain mail says we can have multiple associations. What 
> is your
> > comment on this?.
> > 
> > --Srini
> > 
> > TOLGA ASVEREN wrote:
> > 
> > > Srini,
> > >
> > > I don't think, what Brian says is the same with the first diagram of
> > > Samuel's mail,  from the viewpoint of an ASP, one SGP belongs to one 
> SG,
> > > which is violated by the first diagram of Samuel's mail. Furthermore 
> SGP
> > > and SG are not different entities which are communicating with each 
> other.
> > > One SGP is part of a SG. That is the reason, why several SGP boxes are 
> in
> > > one SG box (whereas in the first figure, we even have SCTP 
> associations
> > > between them). I think, the figure Brian has referenced in his message
> > > (RFC page 115) is a good one to understand the realtionship between SG 
> and
> > > SGP.
> > >
> > > (And also I don't think it is a good idea to consider interpretations 
> from
> > > individuals -regardless of the person (Brian, no offense ;-) 
> )-binding,
> > > let's consider just RFC and IG).
> > >
> > >       Tolga
> > >
> > > Tolga,
> > >
> > >     The following mail replied by Brian says, a SGP can be part of 
> multiple
> > > SGs. How is this different from Samuel model?.
> > >
> > > --Srini
> > >
> > > >Srini,
> > >
> > > >ksrinivas
> > > wrote:
> > > (Thu, 30
> > > Jan 2003 10:08:41)
> > > >> Tolga,
> > > >>
> > > >> An ASP can be a part of different ASs. Similarly, Can a SGP be a 
> part of
> > > >> different SGs?
> > >
> > > >Sure.  But from the viewpoint of an ASP, the SGP belongs to one SG. 
> If an
> > > >SGP were to perform the function for two SGs for the same AS, the ASP
> > > would
> > > >have to form two associations with the SGP: one for SG-A, the other 
> for
> > > SG-B.
> > > >The SGP would also need some way of knowing (e.g. static port 
> allocation)
> > > >for which SG the ASP was forming an association.
> > >
> > > >--brian
> > >
> > > TOLGA ASVEREN wrote:
> > >
> > > > Samuel,
> > > >
> > > > Please find below my interpretations.
> > > >
> > > >    Tolga
> > > >
> > > > Hi,
> > > >
> > > > section 5.4.1.1.1
> > > >
> > > > ...
> > > > - Determine the correct association to the chosen SGP;
> > > >
> > > > I have interpreted this statement as below.
> > > >
> > > > 1)ASP is selecting the correct association to the chosen SGP when 
> SGP is
> > > >   a member of more than one SG.
> > > > [TOLGA]I don't think it is correct from modelling point of view to 
> see
> > > one
> > > > SGP as a member of multiple SGs. I would think, choosing SCTP 
> association
> > > > to a given SGP is just for the case, where you have more than one 
> SCTP
> > > > association(for whatever reason) to a SGP from your ASP.
> > > >
> > > >  +----+   ASSOC-1
> > > >  | SG1|----------+
> > > >  +----+          |
> > > >                  |     +-----+
> > > >  +----+  ASSOC-2 +-----|     |------ +-----+
> > > >  | SG2|----------------| SGP1|------ | ASP1|
> > > >  +----+          +-----|     |------ +-----+
> > > >                  |     +-----+
> > > >  +----+  ASSOC-3 |
> > > >  | SG3|----------+
> > > >  +----+
> > > >
> > > > In this case SGP1 will have 3 IP addresses to have 3 SCTP 
> associations
> > > > with ASP1.When ASP1 wants to send a message to SG3 then ASP will 
> select
> > > > ASSOC-3.
> > > > [TOLGA]As Is aid above, I don't think it is a very good idea to see 
> one
> > > SGP
> > > > as part of multiple SGs. From ASP point of view, if there are there
> > > > different SGs, then there should be three different SGPs:
> > > >
> > > >   ****SG1********
> > > >   * +--------+  *
> > > >   * | SGP1   +--*-----+
> > > >   * +--------+  *     |
> > > >   ***************     |
> > > >   ****SG2********     |   +--------+
> > > >   * +--------+  *     +---+        |
> > > >   * | SGP2   +--*---------+ ASP1   |
> > > >   * +--------+  *     +---+        |
> > > >   ***************     |   +--------+
> > > >   ****SG3********     |
> > > >   * +--------+  *     |
> > > >   * | SGP3   +--*-----+
> > > >   * +--------+  *
> > > >   ***************
> > > >
> > > >  Will the below diagram be possible?
> > > > (i think it is not possible)
> > > >
> > > >   PC1
> > > >  +----+
> > > >  | SG1|----------+
> > > >  +----+          |
> > > >   PC2            |     +-----+
> > > >  +----+          +-----|     |       +-----+
> > > >  | SG2|----------------| SGP1|------ | ASP1|
> > > >  +----+          +-----|     |       +-----+
> > > >   PC3            |     +-----+
> > > >  +----+          |
> > > >  | SG3|----------+
> > > >  +----+
> > > >
> > > > This diagram is wrong because ASP1 may not be able intstrcut SGP1 to 
> send
> > > > a message to a particular SG.
> > > > [TOLGA]Because IMO it is not correct to see a SGP as part of 
> multiple
> > > SGs,
> > > > this one is also not possible.
> > > >
> > > > Can SGP1 use PC (which is inside routing lable) to dintinguish SGs?
> > > > [TOLGA]How can you do this?Are your messages destined only to the PC 
> of
> > > > SGs?
> > > >
> > > > 2)ASP is selecting the correct association to the chosen SGP when 
> SGP is
> > > >   a member of single SG.
> > > >
> > > >     SG1
> > > >  +-------+
> > > >  |+-----+|  ASSOC-1
> > > >  || SGP1||---------+
> > > >  |+-----+|         |
> > > >  |       |         |          +------+
> > > >  |+-----+| ASSOC-2 +----------|      |
> > > >  || SGP2||--------------------| ASP1 |
> > > >  |+-----+|         +----------|      |
> > > >  |       |         |          +------+
> > > >  |+-----+|ASSOC-3  |
> > > >  || SG3 ||---------+
> > > >  |+-----+|
> > > >  +-------+
> > > > [TOLGA]In my opinion, this is, how it should be.
> > > >
> > > > Are my interpretations all correct ?
> > > >
> > > >  --Samuel
> > > >
> 
> 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 28 17:26:53 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA15276
	for <sigtran-archive@odin.ietf.org>; Fri, 28 Feb 2003 17:26:53 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1SMZZ800723
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 17:35:35 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SMZMp00707;
	Fri, 28 Feb 2003 17:35:22 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SMYLp00678
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 17:34:21 -0500
Received: from mail-gw.radisys.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA15255
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 17:25:07 -0500 (EST)
From: Barry.Nagelberg@radisys.com
Received: by mail-gw.radisys.com (Postfix, from userid 5)
	id 60C254E7A1; Fri, 28 Feb 2003 14:22:15 -0800 (PST)
Received: from notes.radisys.com(206.103.52.220)
 via SMTP by mail-gw.radisys.com, id smtpdAAAsYaysZ; Fri Feb 28 14:22:09 2003
To: sigtran@ietf.org
Subject: Re: [Sigtran] [m3ua]-AS-SG
MIME-Version: 1.0
X-Mailer: Lotus Notes Release 5.0.8  June 18, 2001
Message-ID: <OFF1C4F3B7.9851890E-ON85256CDB.00788337@radisys.com>
Date: Fri, 28 Feb 2003 17:02:59 -0500
X-MIMETrack: Serialize by Router on HQ_ACS_1/Radisys_Corporation/US(Release 5.0.9a |January
 7, 2002) at 02/28/2003 02:19:29 PM,
	Serialize complete at 02/28/2003 02:19:29 PM
Content-Type: text/plain; charset="us-ascii"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Brian,

As noted in the RFC, Figure 5 is just one example. It does not prohibit 
either multiple ASs for a single ASP, nor multiple SGs for a single SGP.

Barry Nagelberg





"Brian F. G. Bidulock" <bidulock@openss7.org>
 02/28/2003 05:11 PM
 Please respond to bidulock

 
        To:     Barry.Nagelberg@radisys.com
        cc:     sigtran@ietf.org
        Subject:        Re: [Sigtran] [m3ua]-AS-SG


Barry.Nagelberg,

See Figure 5.

--brian

Barry.Nagelberg@radisys.com wrote: (Fri, 28 Feb 2003 09:21:51)
> Brian,
> 
> Please tell us where in the RFC or IG did you find that
> 
>  "...if you want an SGP to serve 2 SGs it must appear to the ASPs as 2 
> different SGPs".
> 
> Barry Nagelberg
> 
> 
> 
> 
> 
> "Brian F. G. Bidulock" <bidulock@openss7.org>
> Sent by: sigtran-admin@ietf.org
>  02/28/2003 08:05 AM
>  Please respond to bidulock
> 
> 
>         To:     ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
>         cc:     TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, 
sigtran@ietf.org
>         Subject:        Re: [Sigtran] [m3ua]-AS-SG
> 
> 
> ksrinivas,
> 
> 1.3.2.4 Support for the Management of SCTP Associations between the SGP
>         and ASPs.
> 
>    ...
> 
>    The M3UA layer MAY be instructed by local management to establish an
>    SCTP association to a peer M3UA node.  This can be achieved using the
>    M-SCTP_ESTABLISH primitives (See Section 1.6.3 for a description of
>    management primitives.) to request, indicate and confirm the
>    establishment of an SCTP association with a peer M3UA node.  In order
>    to avoid redundant SCTP associations between two M3UA peers, one side
>    (client) SHOULD be designated to establish the SCTP association, or
>    M3UA configuration information maintained to detect redundant
>    associations (e.g., via knowledge of the expected local and remote
>    SCTP endpoint addresses).
> 
> I told you that if you want an SGP to serve 2 SG's it must appear to the
> ASPs as 2 different SGPs.  Also that you would need to use a mechanism
> such as IP addresses and port numbers to distinguish between these two
> views of the SGP.  There is still only one association between an ASP
> and an SGP (as viewed by the ASP).
> 
> If this is too complicated for you, stick to Figure 5.
> 
> --brian
> 
> ksrinivas wrote:       (Fri, 28 Feb 2003 18:15:37)
> > Tolga,
> > 
> > An ASP can be a part of different ASs. Similarly, Can a SGP be a part 
of
> > different SGs?
> > 
> > As per RFC page 115, the figure has only one Association between any 
two
> > Signalling Processes(SGP/ASP). RFC (or even IG) does not say whether 
we 
> can
> > have a single Association or multiple Association between any to 
> Signalling
> > Processes. But Brain mail says we can have multiple associations. What 

> is your
> > comment on this?.
> > 
> > --Srini
> > 
> > TOLGA ASVEREN wrote:
> > 
> > > Srini,
> > >
> > > I don't think, what Brian says is the same with the first diagram of
> > > Samuel's mail,  from the viewpoint of an ASP, one SGP belongs to one 

> SG,
> > > which is violated by the first diagram of Samuel's mail. Furthermore 

> SGP
> > > and SG are not different entities which are communicating with each 
> other.
> > > One SGP is part of a SG. That is the reason, why several SGP boxes 
are 
> in
> > > one SG box (whereas in the first figure, we even have SCTP 
> associations
> > > between them). I think, the figure Brian has referenced in his 
message
> > > (RFC page 115) is a good one to understand the realtionship between 
SG 
> and
> > > SGP.
> > >
> > > (And also I don't think it is a good idea to consider 
interpretations 
> from
> > > individuals -regardless of the person (Brian, no offense ;-) 
> )-binding,
> > > let's consider just RFC and IG).
> > >
> > >       Tolga
> > >
> > > Tolga,
> > >
> > >     The following mail replied by Brian says, a SGP can be part of 
> multiple
> > > SGs. How is this different from Samuel model?.
> > >
> > > --Srini
> > >
> > > >Srini,
> > >
> > > >ksrinivas
> > > wrote:
> > > (Thu, 30
> > > Jan 2003 10:08:41)
> > > >> Tolga,
> > > >>
> > > >> An ASP can be a part of different ASs. Similarly, Can a SGP be a 
> part of
> > > >> different SGs?
> > >
> > > >Sure.  But from the viewpoint of an ASP, the SGP belongs to one SG. 

> If an
> > > >SGP were to perform the function for two SGs for the same AS, the 
ASP
> > > would
> > > >have to form two associations with the SGP: one for SG-A, the other 

> for
> > > SG-B.
> > > >The SGP would also need some way of knowing (e.g. static port 
> allocation)
> > > >for which SG the ASP was forming an association.
> > >
> > > >--brian
> > >
> > > TOLGA ASVEREN wrote:
> > >
> > > > Samuel,
> > > >
> > > > Please find below my interpretations.
> > > >
> > > >    Tolga
> > > >
> > > > Hi,
> > > >
> > > > section 5.4.1.1.1
> > > >
> > > > ...
> > > > - Determine the correct association to the chosen SGP;
> > > >
> > > > I have interpreted this statement as below.
> > > >
> > > > 1)ASP is selecting the correct association to the chosen SGP when 
> SGP is
> > > >   a member of more than one SG.
> > > > [TOLGA]I don't think it is correct from modelling point of view to 

> see
> > > one
> > > > SGP as a member of multiple SGs. I would think, choosing SCTP 
> association
> > > > to a given SGP is just for the case, where you have more than one 
> SCTP
> > > > association(for whatever reason) to a SGP from your ASP.
> > > >
> > > >  +----+   ASSOC-1
> > > >  | SG1|----------+
> > > >  +----+          |
> > > >                  |     +-----+
> > > >  +----+  ASSOC-2 +-----|     |------ +-----+
> > > >  | SG2|----------------| SGP1|------ | ASP1|
> > > >  +----+          +-----|     |------ +-----+
> > > >                  |     +-----+
> > > >  +----+  ASSOC-3 |
> > > >  | SG3|----------+
> > > >  +----+
> > > >
> > > > In this case SGP1 will have 3 IP addresses to have 3 SCTP 
> associations
> > > > with ASP1.When ASP1 wants to send a message to SG3 then ASP will 
> select
> > > > ASSOC-3.
> > > > [TOLGA]As Is aid above, I don't think it is a very good idea to 
see 
> one
> > > SGP
> > > > as part of multiple SGs. From ASP point of view, if there are 
there
> > > > different SGs, then there should be three different SGPs:
> > > >
> > > >   ****SG1********
> > > >   * +--------+  *
> > > >   * | SGP1   +--*-----+
> > > >   * +--------+  *     |
> > > >   ***************     |
> > > >   ****SG2********     |   +--------+
> > > >   * +--------+  *     +---+        |
> > > >   * | SGP2   +--*---------+ ASP1   |
> > > >   * +--------+  *     +---+        |
> > > >   ***************     |   +--------+
> > > >   ****SG3********     |
> > > >   * +--------+  *     |
> > > >   * | SGP3   +--*-----+
> > > >   * +--------+  *
> > > >   ***************
> > > >
> > > >  Will the below diagram be possible?
> > > > (i think it is not possible)
> > > >
> > > >   PC1
> > > >  +----+
> > > >  | SG1|----------+
> > > >  +----+          |
> > > >   PC2            |     +-----+
> > > >  +----+          +-----|     |       +-----+
> > > >  | SG2|----------------| SGP1|------ | ASP1|
> > > >  +----+          +-----|     |       +-----+
> > > >   PC3            |     +-----+
> > > >  +----+          |
> > > >  | SG3|----------+
> > > >  +----+
> > > >
> > > > This diagram is wrong because ASP1 may not be able intstrcut SGP1 
to 
> send
> > > > a message to a particular SG.
> > > > [TOLGA]Because IMO it is not correct to see a SGP as part of 
> multiple
> > > SGs,
> > > > this one is also not possible.
> > > >
> > > > Can SGP1 use PC (which is inside routing lable) to dintinguish 
SGs?
> > > > [TOLGA]How can you do this?Are your messages destined only to the 
PC 
> of
> > > > SGs?
> > > >
> > > > 2)ASP is selecting the correct association to the chosen SGP when 
> SGP is
> > > >   a member of single SG.
> > > >
> > > >     SG1
> > > >  +-------+
> > > >  |+-----+|  ASSOC-1
> > > >  || SGP1||---------+
> > > >  |+-----+|         |
> > > >  |       |         |          +------+
> > > >  |+-----+| ASSOC-2 +----------|      |
> > > >  || SGP2||--------------------| ASP1 |
> > > >  |+-----+|         +----------|      |
> > > >  |       |         |          +------+
> > > >  |+-----+|ASSOC-3  |
> > > >  || SG3 ||---------+
> > > >  |+-----+|
> > > >  +-------+
> > > > [TOLGA]In my opinion, this is, how it should be.
> > > >
> > > > Are my interpretations all correct ?
> > > >
> > > >  --Samuel
> > > >
> 
> 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/



_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



From mailnull@www1.ietf.org  Fri Feb 28 17:31:41 2003
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA15376
	for <sigtran-archive@odin.ietf.org>; Fri, 28 Feb 2003 17:31:41 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h1SMeMC01701
	for sigtran-archive@odin.ietf.org; Fri, 28 Feb 2003 17:40:22 -0500
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SMeAp01686;
	Fri, 28 Feb 2003 17:40:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id h1SMdhp01639
	for <sigtran@optimus.ietf.org>; Fri, 28 Feb 2003 17:39:43 -0500
Received: from gw.openss7.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA15364
	for <sigtran@ietf.org>; Fri, 28 Feb 2003 17:30:29 -0500 (EST)
Received: from ns.pigworks.openss7.net (IDENT:0GVtim7c2RBrjVAsWqLbwOXdl3b8t2qj@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h1SMWRD06252;
	Fri, 28 Feb 2003 15:32:27 -0700
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h1SMWRB21568;
	Fri, 28 Feb 2003 15:32:27 -0700
Date: Fri, 28 Feb 2003 15:32:27 -0700
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Barry.Nagelberg@radisys.com
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] [m3ua]-AS-SG
Message-ID: <20030228153227.A21543@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Barry.Nagelberg@radisys.com, sigtran@ietf.org
References: <OFF1C4F3B7.9851890E-ON85256CDB.00788337@radisys.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OFF1C4F3B7.9851890E-ON85256CDB.00788337@radisys.com>; from Barry.Nagelberg@radisys.com on Fri, Feb 28, 2003 at 05:02:59PM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Barry.Nagelberg,

Perhaps you would like to answer Samuel's question then...

--brian

Barry.Nagelberg@radisys.com wrote:                                         (Fri, 28 Feb 2003 17:02:59)
> Brian,
> 
> As noted in the RFC, Figure 5 is just one example. It does not prohibit 
> either multiple ASs for a single ASP, nor multiple SGs for a single SGP.
> 
> Barry Nagelberg
> 
> 
> 
> 
> 
> "Brian F. G. Bidulock" <bidulock@openss7.org>
>  02/28/2003 05:11 PM
>  Please respond to bidulock
> 
>  
>         To:     Barry.Nagelberg@radisys.com
>         cc:     sigtran@ietf.org
>         Subject:        Re: [Sigtran] [m3ua]-AS-SG
> 
> 
> Barry.Nagelberg,
> 
> See Figure 5.
> 
> --brian
> 
> Barry.Nagelberg@radisys.com wrote: (Fri, 28 Feb 2003 09:21:51)
> > Brian,
> > 
> > Please tell us where in the RFC or IG did you find that
> > 
> >  "...if you want an SGP to serve 2 SGs it must appear to the ASPs as 2 
> > different SGPs".
> > 
> > Barry Nagelberg
> > 
> > 
> > 
> > 
> > 
> > "Brian F. G. Bidulock" <bidulock@openss7.org>
> > Sent by: sigtran-admin@ietf.org
> >  02/28/2003 08:05 AM
> >  Please respond to bidulock
> > 
> > 
> >         To:     ksrinivas <ksrini@axes-mach01.usa.alcatel.com>
> >         cc:     TOLGA ASVEREN <TOLGA.ASVEREN@o2.com.tr>, 
> sigtran@ietf.org
> >         Subject:        Re: [Sigtran] [m3ua]-AS-SG
> > 
> > 
> > ksrinivas,
> > 
> > 1.3.2.4 Support for the Management of SCTP Associations between the SGP
> >         and ASPs.
> > 
> >    ...
> > 
> >    The M3UA layer MAY be instructed by local management to establish an
> >    SCTP association to a peer M3UA node.  This can be achieved using the
> >    M-SCTP_ESTABLISH primitives (See Section 1.6.3 for a description of
> >    management primitives.) to request, indicate and confirm the
> >    establishment of an SCTP association with a peer M3UA node.  In order
> >    to avoid redundant SCTP associations between two M3UA peers, one side
> >    (client) SHOULD be designated to establish the SCTP association, or
> >    M3UA configuration information maintained to detect redundant
> >    associations (e.g., via knowledge of the expected local and remote
> >    SCTP endpoint addresses).
> > 
> > I told you that if you want an SGP to serve 2 SG's it must appear to the
> > ASPs as 2 different SGPs.  Also that you would need to use a mechanism
> > such as IP addresses and port numbers to distinguish between these two
> > views of the SGP.  There is still only one association between an ASP
> > and an SGP (as viewed by the ASP).
> > 
> > If this is too complicated for you, stick to Figure 5.
> > 
> > --brian
> > 
> > ksrinivas wrote:       (Fri, 28 Feb 2003 18:15:37)
> > > Tolga,
> > > 
> > > An ASP can be a part of different ASs. Similarly, Can a SGP be a part 
> of
> > > different SGs?
> > > 
> > > As per RFC page 115, the figure has only one Association between any 
> two
> > > Signalling Processes(SGP/ASP). RFC (or even IG) does not say whether 
> we 
> > can
> > > have a single Association or multiple Association between any to 
> > Signalling
> > > Processes. But Brain mail says we can have multiple associations. What 
> 
> > is your
> > > comment on this?.
> > > 
> > > --Srini
> > > 
> > > TOLGA ASVEREN wrote:
> > > 
> > > > Srini,
> > > >
> > > > I don't think, what Brian says is the same with the first diagram of
> > > > Samuel's mail,  from the viewpoint of an ASP, one SGP belongs to one 
> 
> > SG,
> > > > which is violated by the first diagram of Samuel's mail. Furthermore 
> 
> > SGP
> > > > and SG are not different entities which are communicating with each 
> > other.
> > > > One SGP is part of a SG. That is the reason, why several SGP boxes 
> are 
> > in
> > > > one SG box (whereas in the first figure, we even have SCTP 
> > associations
> > > > between them). I think, the figure Brian has referenced in his 
> message
> > > > (RFC page 115) is a good one to understand the realtionship between 
> SG 
> > and
> > > > SGP.
> > > >
> > > > (And also I don't think it is a good idea to consider 
> interpretations 
> > from
> > > > individuals -regardless of the person (Brian, no offense ;-) 
> > )-binding,
> > > > let's consider just RFC and IG).
> > > >
> > > >       Tolga
> > > >
> > > > Tolga,
> > > >
> > > >     The following mail replied by Brian says, a SGP can be part of 
> > multiple
> > > > SGs. How is this different from Samuel model?.
> > > >
> > > > --Srini
> > > >
> > > > >Srini,
> > > >
> > > > >ksrinivas
> > > > wrote:
> > > > (Thu, 30
> > > > Jan 2003 10:08:41)
> > > > >> Tolga,
> > > > >>
> > > > >> An ASP can be a part of different ASs. Similarly, Can a SGP be a 
> > part of
> > > > >> different SGs?
> > > >
> > > > >Sure.  But from the viewpoint of an ASP, the SGP belongs to one SG. 
> 
> > If an
> > > > >SGP were to perform the function for two SGs for the same AS, the 
> ASP
> > > > would
> > > > >have to form two associations with the SGP: one for SG-A, the other 
> 
> > for
> > > > SG-B.
> > > > >The SGP would also need some way of knowing (e.g. static port 
> > allocation)
> > > > >for which SG the ASP was forming an association.
> > > >
> > > > >--brian
> > > >
> > > > TOLGA ASVEREN wrote:
> > > >
> > > > > Samuel,
> > > > >
> > > > > Please find below my interpretations.
> > > > >
> > > > >    Tolga
> > > > >
> > > > > Hi,
> > > > >
> > > > > section 5.4.1.1.1
> > > > >
> > > > > ...
> > > > > - Determine the correct association to the chosen SGP;
> > > > >
> > > > > I have interpreted this statement as below.
> > > > >
> > > > > 1)ASP is selecting the correct association to the chosen SGP when 
> > SGP is
> > > > >   a member of more than one SG.
> > > > > [TOLGA]I don't think it is correct from modelling point of view to 
> 
> > see
> > > > one
> > > > > SGP as a member of multiple SGs. I would think, choosing SCTP 
> > association
> > > > > to a given SGP is just for the case, where you have more than one 
> > SCTP
> > > > > association(for whatever reason) to a SGP from your ASP.
> > > > >
> > > > >  +----+   ASSOC-1
> > > > >  | SG1|----------+
> > > > >  +----+          |
> > > > >                  |     +-----+
> > > > >  +----+  ASSOC-2 +-----|     |------ +-----+
> > > > >  | SG2|----------------| SGP1|------ | ASP1|
> > > > >  +----+          +-----|     |------ +-----+
> > > > >                  |     +-----+
> > > > >  +----+  ASSOC-3 |
> > > > >  | SG3|----------+
> > > > >  +----+
> > > > >
> > > > > In this case SGP1 will have 3 IP addresses to have 3 SCTP 
> > associations
> > > > > with ASP1.When ASP1 wants to send a message to SG3 then ASP will 
> > select
> > > > > ASSOC-3.
> > > > > [TOLGA]As Is aid above, I don't think it is a very good idea to 
> see 
> > one
> > > > SGP
> > > > > as part of multiple SGs. From ASP point of view, if there are 
> there
> > > > > different SGs, then there should be three different SGPs:
> > > > >
> > > > >   ****SG1********
> > > > >   * +--------+  *
> > > > >   * | SGP1   +--*-----+
> > > > >   * +--------+  *     |
> > > > >   ***************     |
> > > > >   ****SG2********     |   +--------+
> > > > >   * +--------+  *     +---+        |
> > > > >   * | SGP2   +--*---------+ ASP1   |
> > > > >   * +--------+  *     +---+        |
> > > > >   ***************     |   +--------+
> > > > >   ****SG3********     |
> > > > >   * +--------+  *     |
> > > > >   * | SGP3   +--*-----+
> > > > >   * +--------+  *
> > > > >   ***************
> > > > >
> > > > >  Will the below diagram be possible?
> > > > > (i think it is not possible)
> > > > >
> > > > >   PC1
> > > > >  +----+
> > > > >  | SG1|----------+
> > > > >  +----+          |
> > > > >   PC2            |     +-----+
> > > > >  +----+          +-----|     |       +-----+
> > > > >  | SG2|----------------| SGP1|------ | ASP1|
> > > > >  +----+          +-----|     |       +-----+
> > > > >   PC3            |     +-----+
> > > > >  +----+          |
> > > > >  | SG3|----------+
> > > > >  +----+
> > > > >
> > > > > This diagram is wrong because ASP1 may not be able intstrcut SGP1 
> to 
> > send
> > > > > a message to a particular SG.
> > > > > [TOLGA]Because IMO it is not correct to see a SGP as part of 
> > multiple
> > > > SGs,
> > > > > this one is also not possible.
> > > > >
> > > > > Can SGP1 use PC (which is inside routing lable) to dintinguish 
> SGs?
> > > > > [TOLGA]How can you do this?Are your messages destined only to the 
> PC 
> > of
> > > > > SGs?
> > > > >
> > > > > 2)ASP is selecting the correct association to the chosen SGP when 
> > SGP is
> > > > >   a member of single SG.
> > > > >
> > > > >     SG1
> > > > >  +-------+
> > > > >  |+-----+|  ASSOC-1
> > > > >  || SGP1||---------+
> > > > >  |+-----+|         |
> > > > >  |       |         |          +------+
> > > > >  |+-----+| ASSOC-2 +----------|      |
> > > > >  || SGP2||--------------------| ASP1 |
> > > > >  |+-----+|         +----------|      |
> > > > >  |       |         |          +------+
> > > > >  |+-----+|ASSOC-3  |
> > > > >  || SG3 ||---------+
> > > > >  |+-----+|
> > > > >  +-------+
> > > > > [TOLGA]In my opinion, this is, how it should be.
> > > > >
> > > > > Are my interpretations all correct ?
> > > > >
> > > > >  --Samuel
> > > > >
> > 
> > 
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> -- 
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
> 
> 
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

-- 
Brian F. G. Bidulock
bidulock@openss7.org
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran



