From exim@www1.ietf.org  Tue Jul  1 03:03:29 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 DAA16599
	for <sigtran-archive@odin.ietf.org>; Tue, 1 Jul 2003 03:03:29 -0400 (EDT)
Received: (from exim@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h61732C09696
	for sigtran-archive@odin.ietf.org; Tue, 1 Jul 2003 03:03:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XFAK-0002VH-TD; Tue, 01 Jul 2003 03:03:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XF9q-0002Q9-HY
	for sigtran@optimus.ietf.org; Tue, 01 Jul 2003 03:02:30 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA16552;
	Tue, 1 Jul 2003 03:02:26 -0400 (EDT)
From: Albrecht.Schwarz@alcatel.de
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19XF9m-0000Gq-00; Tue, 01 Jul 2003 03:02:26 -0400
Received: from mailrelay3.alcatel.de ([194.113.59.71] helo=mailrelay1.alcatel.de)
	by ietf-mx with esmtp (Exim 4.12)
	id 19XF8n-0000Fl-00; Tue, 01 Jul 2003 03:01:25 -0400
Received: from slsas5.stgl.netd.alcatel.de (destgs0000c.ad1.ad.alcatel.com [149.204.45.57])
	by mailrelay1.alcatel.de (8.9.3/8.9.3) with ESMTP id JAA16853;
	Tue, 1 Jul 2003 09:00:02 +0200 (MET DST)
Subject: Re: [Sigtran] Signaling Gateway Doubts
To: Pablo Salinas <pablopesb@yahoo.com>
Cc: sigtran@ietf.org, sigtran-admin@ietf.org
X-Mailer: Lotus Notes Release 5.0.8  June 18, 2001
Message-ID: <OF6FE0D2A5.E001DF19-ONC1256D56.0025D2F0@stgl.netd.alcatel.de>
Date: Tue, 1 Jul 2003 09:00:20 +0200
X-MIMETrack: Serialize by Router on DEMTA001/DE/ALCATEL(Release 5.0.11  |July 24, 2002) at
 07/01/2003 09:00:23 AM
MIME-Version: 1.0
Content-type: text/plain; charset=iso-8859-1
Content-transfer-encoding: quoted-printable
X-Alcanet-MTA-scanned-and-authorized: yes
X-Alcanet-MTA-scanned-and-authorized: yes
X-Alcanet-MTA-scanned-and-authorized: yes
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable


The motivation is outlined in RFC 2719. See Fig. 2 for instance ... Fur=
ther
you should take into account, that SS7 networks may principally be oper=
ated
in associated, quasi-associated, or non-associated mode; which may lead=
 to
different network approaches in case of VoIP transition ...
./Albrecht




                                                                       =
                                                                       =
     =20
                      Pablo Salinas                                    =
                                                                       =
     =20
                      <pablopesb@yahoo         To:      sigtran@ietf.or=
g                                                                      =
     =20
                      .com>                    cc:                     =
                                                                       =
     =20
                      Sent by:                 Subject: [Sigtran] Signa=
ling Gateway Doubts                                                    =
     =20
                      sigtran-admin@ie                                 =
                                                                       =
     =20
                      tf.org                                           =
                                                                       =
     =20
                                                                       =
                                                                       =
     =20
                                                                       =
                                                                       =
     =20
                      30.06.2003 16:19                                 =
                                                                       =
     =20
                                                                       =
                                                                       =
     =20
                                                                       =
                                                                       =
     =20




Hi there,
            I am having a hard time to understand one thing in a VoIP
Network that uses a softswitch.
Supose, we have the following situation:
There is a softswitch installed in location A. And this softswitch must=

communicate with the PTT in localtion A. Should it have a Signaling Gat=
eway
between the softswitch and the PTT?
If the answer is yes, why use the SCTP, M2UA, M2PA or whatever other
sigtran protocol, if it is possible to connect the softswitch directly =
to
the PTT via the MTP2 protocol.
If the answer is no, why in every network scheme I saw until now, there=
 is
always a SG between the PTT and the softswitch, when the sofswitch
communicates with the media gateways via H.248?

                         thanks,
                                      Pablo Salinas



Do You Yahoo!?

Todo lo que quieres saber de Estados Unidos, Am=E9rica Latina y el rest=
o del
Mundo.
Vis=EDta Yahoo! Noticias.


=



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



From exim@www1.ietf.org  Tue Jul  1 03:15:31 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 DAA01070
	for <sigtran-archive@odin.ietf.org>; Tue, 1 Jul 2003 03:15:31 -0400 (EDT)
Received: (from exim@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h617F4e12467
	for sigtran-archive@odin.ietf.org; Tue, 1 Jul 2003 03:15:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XFLy-0003ER-V5; Tue, 01 Jul 2003 03:15:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XFLw-0003Dr-SG
	for sigtran@optimus.ietf.org; Tue, 01 Jul 2003 03:15:00 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA01057
	for <sigtran@ietf.org>; Tue, 1 Jul 2003 03:14:57 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19XFLn-0000P3-00
	for sigtran@ietf.org; Tue, 01 Jul 2003 03:14:51 -0400
Received: from beamer.mchh.siemens.de ([194.138.158.163])
	by ietf-mx with esmtp (Exim 4.12)
	id 19XFLN-0000On-00
	for sigtran@ietf.org; Tue, 01 Jul 2003 03:14:25 -0400
Received: from blues.mchh.siemens.de ([139.21.204.206])
	by beamer.mchh.siemens.de (8.9.3/8.9.3) with ESMTP id JAA12568;
	Tue, 1 Jul 2003 09:14:00 +0200 (MET DST)
Received: from mchh253e.mchh.siemens.de (mchh253e.mchh.siemens.de [139.21.200.63])
	by blues.mchh.siemens.de (8.9.3/8.9.1) with ESMTP id JAA13022;
	Tue, 1 Jul 2003 09:12:56 +0200 (MET DST)
Received: by mchh253e.mchh.siemens.de with Internet Mail Service (5.5.2653.19)
	id <NTW3ZAQ8>; Tue, 1 Jul 2003 09:13:29 +0200
Message-ID: <D73559B38D51D4119CB800508B9527B505AC1907@mchh266e.mchh.siemens.de>
From: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>
To: "'Pablo Salinas'" <pablopesb@yahoo.com>, sigtran@ietf.org
Subject: RE: [Sigtran] Signaling Gateway Doubts
Date: Tue, 1 Jul 2003 09:13:29 +0200 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C33FA0.458F6AD0"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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.

------_=_NextPart_001_01C33FA0.458F6AD0
Content-Type: text/plain;
	charset="ISO-8859-1"
Content-Transfer-Encoding: quoted-printable

Pablo,
=20
the answer is yes and no!
YES in case your softswitch has continued use of TDM based SS7
NO in case your softswitch has no TDM connections at all. However, also =
in this case there is no absolute necessity to have a separated SG, as =
the SG could be integrated into the STP (when using quasi-associated =
signalling). If you are using associated signalling, the SG might be =
integrated into the MediaGateway.
=20
greetings
=20
HannsJuergen

-----Original Message-----
From: Pablo Salinas [mailto:pablopesb@yahoo.com]
Sent: Monday, June 30, 2003 4:19 PM
To: sigtran@ietf.org
Subject: [Sigtran] Signaling Gateway Doubts


Hi there,
            I am having a hard time to understand one thing in a VoIP =
Network that uses a softswitch.
Supose, we have the following situation:
There is a softswitch installed in location A. And this softswitch must =
communicate with the PTT in localtion A. Should it have a Signaling =
Gateway between the softswitch and the PTT?
If the answer is yes, why use the SCTP, M2UA, M2PA or whatever other =
sigtran protocol, if it is possible to connect the softswitch directly =
to the PTT via the MTP2 protocol.
If the answer is no, why in every network scheme I saw until now, there =
is always a SG between the PTT and the softswitch, when the sofswitch =
communicates with the media gateways via H.248?
=20
                         thanks,
                                      Pablo Salinas




  _____ =20

Do You Yahoo!?
 <http://espanol.yahoo.com/mail_tagline/*http://espanol.news.yahoo.com> =
 Todo lo que quieres saber de Estados Unidos, Am=E9rica Latina y el =
resto del Mundo.
Vis=EDta Yahoo!  =
<http://espanol.yahoo.com/mail_tagline/*http://espanol.news.yahoo.com> =
Noticias.



------_=_NextPart_001_01C33FA0.458F6AD0
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 HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3DISO-8859-1">


<META content=3D"MSHTML 6.00.2800.1141" name=3DGENERATOR></HEAD>
<BODY>
<DIV><SPAN class=3D657550707-01072003><FONT face=3DArial =
color=3D#0000ff=20
size=3D2>Pablo,</FONT></SPAN></DIV>
<DIV><SPAN class=3D657550707-01072003><FONT face=3DArial =
color=3D#0000ff=20
size=3D2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=3D657550707-01072003><FONT face=3DArial =
color=3D#0000ff size=3D2>the=20
answer is yes and no!</FONT></SPAN></DIV>
<DIV><SPAN class=3D657550707-01072003><FONT face=3DArial =
color=3D#0000ff size=3D2>YES in=20
case your softswitch has continued use of TDM based =
SS7</FONT></SPAN></DIV>
<DIV><SPAN class=3D657550707-01072003><FONT face=3DArial =
color=3D#0000ff size=3D2>NO in=20
case your softswitch has no TDM connections at all. However, also in =
this case=20
there is no absolute necessity to have a separated SG, as the SG could =
be=20
integrated into the STP (when using quasi-associated signalling). If =
you are=20
using associated signalling, the SG might be integrated into the=20
MediaGateway.</FONT></SPAN></DIV>
<DIV><SPAN class=3D657550707-01072003><FONT face=3DArial =
color=3D#0000ff=20
size=3D2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=3D657550707-01072003><FONT face=3DArial =
color=3D#0000ff=20
size=3D2>greetings</FONT></SPAN></DIV>
<DIV><SPAN class=3D657550707-01072003><FONT face=3DArial =
color=3D#0000ff=20
size=3D2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=3D657550707-01072003><FONT face=3DArial =
color=3D#0000ff=20
size=3D2>HannsJuergen</FONT></SPAN></DIV>
<BLOCKQUOTE=20
style=3D"PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #0000ff 2px =
solid">
  <DIV class=3DOutlookMessageHeader dir=3Dltr align=3Dleft><FONT =
face=3DTahoma=20
  size=3D2>-----Original Message-----<BR><B>From:</B> Pablo Salinas=20
  [mailto:pablopesb@yahoo.com]<BR><B>Sent:</B> Monday, June 30, 2003 =
4:19=20
  PM<BR><B>To:</B> sigtran@ietf.org<BR><B>Subject:</B> [Sigtran] =
Signaling=20
  Gateway Doubts<BR><BR></FONT></DIV>
  <DIV>Hi there,</DIV>
  =
<DIV>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =
I am=20
  having a hard time to understand one thing in a VoIP Network =
that&nbsp;uses a=20
  softswitch.</DIV>
  <DIV>Supose, we have the following situation:</DIV>
  <DIV>There is a softswitch installed in location A. And this =
softswitch must=20
  communicate with the PTT in localtion A. Should it have a Signaling =
Gateway=20
  between the softswitch and the PTT?</DIV>
  <DIV>If the answer is yes, why use the SCTP, M2UA, M2PA or whatever =
other=20
  sigtran protocol, if it is possible to connect the softswitch =
directly to the=20
  PTT via the MTP2 protocol.</DIV>
  <DIV>If the answer is no, why in every network scheme I saw until =
now, there=20
  is always a SG between the PTT and the softswitch,&nbsp;when the =
sofswitch=20
  communicates with the media gateways via H.248?</DIV>
  <DIV>&nbsp;</DIV>
  =
<DIV>&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
  thanks,</DIV>
  =
<DIV>&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;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;=20
  Pablo Salinas</DIV>
  <P><BR>
  <HR noShade SIZE=3D1>
  <B>Do You Yahoo!?</B><BR><A=20
  =
href=3D"http://espanol.yahoo.com/mail_tagline/*http://espanol.news.yahoo=
.com"><IMG=20
  height=3D26 =
src=3D"http://us.i1.yimg.com/us.yimg.com/i/not/dailynews.gif" =
width=3D30=20
  align=3Dleft border=3D0 NOSEND=3D"1"></A> Todo lo que quieres saber =
de Estados=20
  Unidos, Am=E9rica Latina y el resto del Mundo.<BR>Vis=EDta <A=20
  =
href=3D"http://espanol.yahoo.com/mail_tagline/*http://espanol.news.yahoo=
.com">Yahoo!=20
  Noticias</A>.<BR></BLOCKQUOTE></BODY></HTML>

------_=_NextPart_001_01C33FA0.458F6AD0--

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



From exim@www1.ietf.org  Tue Jul  1 05:40:28 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 FAA04362
	for <sigtran-archive@odin.ietf.org>; Tue, 1 Jul 2003 05:40:28 -0400 (EDT)
Received: (from exim@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h619e3p07595
	for sigtran-archive@odin.ietf.org; Tue, 1 Jul 2003 05:40:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XHcH-0001yF-Gr; Tue, 01 Jul 2003 05:40:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XHcE-0001xb-P9
	for sigtran@optimus.ietf.org; Tue, 01 Jul 2003 05:39:58 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA04332
	for <sigtran@ietf.org>; Tue, 1 Jul 2003 05:39:53 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19XHc9-0001Y6-00
	for sigtran@ietf.org; Tue, 01 Jul 2003 05:39:53 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[qiLbqOmAF7h2g2CNk98Cll2RRBFol9D5])
	by ietf-mx with esmtp (Exim 4.12)
	id 19XHbt-0001Xg-00
	for sigtran@ietf.org; Tue, 01 Jul 2003 05:39:37 -0400
Received: from ns.pigworks.openss7.net (IDENT:iVw0xslyzKxYdhQpXWf/pHgSODBmRJDu@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h619cpX29798;
	Tue, 1 Jul 2003 03:38:51 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h619cpL31889;
	Tue, 1 Jul 2003 03:38:51 -0600
Date: Tue, 1 Jul 2003 03:38:51 -0600
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:n+k issue
Message-ID: <20030701033851.A31866@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	sigtran@ietf.org
References: <20030628041753.B5019@openss7.org> <Pine.SOL.4.10.10306300837510.24913-100000@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.4.10.10306300837510.24913-100000@sun2594.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Mon, Jun 30, 2003 at 09:44:01AM -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>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                      (Mon, 30 Jun 2003 09:44:01)
> 
> Brian,
> 
> On Sat, 28 Jun 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > I should point out that the proposal to require that the SG wait for 'n' to
> > start traffic does not have much merit.
> > 
> > Example: n=256, k=1
> > 
> > 2 ASPs fail, permanently and need to be manually replaced.
> > 
> > A temporary network failure causes SCTP associations to be lost to the
> > remaining 255 ASPs.  They move to the ASP down state and immediately
> > re-establish SCTP associations and move to the ASP-Active state.  Because the
> > number of ASPs is < n, the SG will not restart traffic for some time.
> 
> Your example leads to the following figure.
> 
>          SG1                            AS1
>      
>  SS7 MSG +------+----------------------- ASP (1)
> ---------| SGP1 |----------------------- ASP (2)
>          |      |----------------------- ....
>          +------+----------------------- ASP (256)
>             |
>             | 
>             +---------------------------ASP(k)
> 
> 1. First of all craft will not configure 1 ASP as backup for 256 Active
>    ASPs, if he does, it means to be a poor network configuration.

Well, first of all, craft does not normally choose the number of ASPs used for
backup, engineering does.  Second, k=1 is more than adequate for n=256 where
the ASPs are reasonably reliable and have a short replacement time.

Regardless, I followed with the same example with n=2 and k=1.

> 
> 2. Consider ASP as an MGC, your example leads to 256 MGCs and they are 
>    controlled by a single SGP.A good network will have multiple SGPs to
>    share the POTENTIAL traffic across ASPs.

You seem to misunderstand the concept of redundancy in M3UA: there are not
multiple SGPs or ASPs because of traffic load.  It is easy enough to handle
the traffic of a small to medium sized country through a single processor.

Also, you seem to misunderstand that an SGP and an ASP are not processors.
They are processes.  One host processor could have many (even 256) ASPs or
SGPs (or both for that matter).

> 
> 3. BASIC INTENTION OF 'n+k' is,just to make sure that incoming messages
>    ARE EFFICIENTLY TRANSMITTED TO DESTINATION ASPs WITH THE HELP OF
>    MINIMUM NUMBER OF ASSOCIATIONS.

No.  The intention of 'n+k' is to provide one of the more recently accepted
models for reliability using redundancy (n = m+1, k=0 being the typical system
deployed in the field today, m usually being greater than 2).  Up until
recently, only 2m (or m+m) systems were considered carrier class.  Some
systems were 4m (with 3 out of 4 vote).

You should take some time to look at redundancy and reliability requirements
for carrier grade systems for an understanding of how to calculate and provide
the reliability requirements expected by carriers.  There are good standards
written on the topic.  See the Telcordia NEBS and reliability specs, for
example.  If you have an RBOC RFP handy, take a look at the hardware and
software reliability requirements sections or talk to the engineer in your
company responsible for answering those sections.

> 
>    Say for example
> 
>    Inefficient configuration -just an example
>    ------------------------------------------
>   
>             SLS         DEST
>             =================
>              0           ASP1
>              1           ASP1
>              2           ASP1
>              3           ASP1
>              4           ASP1
>              5           ASP1
>              6           ASP1
>              .            .
>              .            .
>              15          ASP1
>              
>              
>                SG                            AS1
> sls(0,1..15 )+----+  sls(0,1..15)         +------+
> -------------|SGP1|-----------------------| ASP1 |
>              |    |              ASSOC-1  +------+
>              +----+                     
> 
>                       Fig - 1
> 
> In this figure, all incoming messages are transmitted over ASSOC-1 towards
> ASP1.Here there are two potential problems 
> 
>        1. Total incoming traffic is dependent on a single ASP
>        2. ASP1 is always overloaded, messages are processed one by one
>           from the transmission buffer which ultimately introduces time
>           delay(example last arrived msg will be processed at the end).

The above was not in my examples.  My examples had n > 1.

> 
> 
> Efficient configuration
> -----------------------  
> 
>              SLS         DEST
>             =================
>              0           ASP1
>              1           ASP1
>              2           ASP1
>              3           ASP1
>              4           ASP1
>              5           ASP1
>              6           ASP1
>              7           ASP1
>              8           ASP2
>              9           ASP2
>              .            .
>              .            .
>              15          ASP2
> 
>                
>                        
>                        
>               SG                             AS1
> sls(0,1..15 )+----+   sls(0,1,..7)ASSOC-1 +------+
> -------------|SGP1|-----------------------| ASP1 |
>              |    |--------+              +------+
>              +----+        | sls(8,9..15) +------+
>                            +--------------| ASP2 |
>                                  ASSOC-2  +------+
>                         
>                    fig - 2
> 
> In this example incoming traffic is distributed between ASP1 and ASP2.
> 
> IN ORDER TO HANDLE THE HEAVY INCOMING TRAFFIC, SGP1 NEEDS MINIMUM OF
> 2 ACTIVE ASPs BY SETTING n=2.

Do they not comply to SS7 ISUP/SCCP flow and congestion controls?  Are they
running at full capacity in this arrangement?  Are these your ASPs?  Do you
design your systems this way?

Take a look at the reliability engineering.  If either of these ASPs is
handling its maximum rated capacity, it will have to be reliable to .99999
This is because, if one ASP fails, 50% of the system has failed.  A 50% system
failure is deemed a "system" failure rather than a "component" failure in
reliability engineering for carrier grade systems.  Considering proscribed
replacement times on the ASP, this would be difficult to achieve unless both
ASPs were hardware fault tolerant platforms.

Nevertheless, the RFC defines ASPs as a process, not a processor.  There is
nothing restricting the two ASPs in your diagram from being processes on the
same processor.  In that case, when ASP2 is down, ASP1 has double the
processing capacity as it had before.

You certainly cannot make statements as you have made here without
presupposing an implementation architecture outside the scope of the standard.

Besides, the RFC permits you to configure your SGP (and why do you only have
one of them?) so that it waits for 'n' ASPs to become active.  You MAY do
this.  Just don't require everybody else to follow your approach.  (Actually,
if the above is your approach, don't require anyone else to follow it.)

Several other points: ANSI T1.111.4/2000 SLS values are 8-bits.  That means
there are 256 SLS values (gee, I had 256 ASPs in my example).

For another thing, sequenced traffic flows are not only based on SLS.  In an
ITU-T network, each remote signalling point has 16 independent traffic flows
for each AS.  The traffic flows from each remote signalling point is
independent from the traffic flows of other remote signalling points.
Therefore, if there are 400 remote signalling points, there will be 6400
independent traffic flows that the SG can use for distribution of AS traffic
for one AS over as many as 6400 ASPs without risking mis-sequencing of
messages.  For an ANSI T1.111.4/2000 network using 8-bit SLS values with 400
remote signalling points, there would be 102400 independent traffic flows to
the AS that the SG can use to distribute traffic to one AS over as many as
102400 ASPs without risking message mis-sequencing.

> 
> 
> Your example with n=256 is completely a wrong interpretaion of 'n+k'
> redundancy case.

Quite the opposite, you clearly misunderstand redundancy and reliability in
carrier grade systems and obviously have little experience with them.

Here's a little primer for you:

Carrier grade systems were originally designed as hot-standby, fully redundant
systems.  For every processor actively handling traffic, there was a hot
standby processor also actively handling the same traffic.  This typically
required special bus arrangements whereby the I/O bus of one processor was
actively controlling peripheral devices, while the hot standby processor was
issuing I/O bus output that was not terminated to the I/O bus.  The bus also
normally performed a comparison of the outputs from the active side and the
outputs from the inactive side.  Peripheral components, such as switch fabric
were also fully redundant (duplicated).  Switching fabrics are separated into
an A-plane and a B-plane in much the same fashion as the SPCS front-end I/O
bus.

Note that active/standby in M3UA does not address the above model.  2+0
broadcast does.

The problem with such a systems was what we used to call the "kill your
sister" syndrome.  A failure on the inactive side only detectable to the I/O
bus could cause a switch from the active to inactive side, resulting in the
switch running on the erroneous inactive side.  We called it the "kill your
sister" syndrome because it is not so surprising that the psychotic sister
(erred inactive side) kills the sane sister (correct active side).

Primarily to avoid the "kill your sister" syndrome, some switches were
designed with 3 or 4 processors on an cross-checking I/O bus pair.  In that
case, it would take a 2 out of 3 or 3 out of 4 vote to shift from the active
processor.  These systems were far less susceptible to inactive plane errors
that were only detected at the I/O bus.

Note that the only M3UA model that address the above is the 3+0 or 4+0
broadcast model.

When the use of general purpose computing platforms for telecommunications
systems started to become popular, there were several models that emerged.

First was taking two big iron UNIX hosts and running them in active/standby.
In this scenario, one UNIX host would be actively handling traffic and the
other would be warm standby.  By "warm" standby, it means that the device was
powered and was synchronizing state on a checkpoint interval or monitoring the
health of the active hosts with heartbeats.  Shared memory, shared DASD or
message passing are the most common mechanisms of redundancy.  These systems
are considered "highly available" systems rather than the "fault tolerant"
systems used in telecommunications above.  An increased level of reliability
is often achieved by using hardware fault tolerant multiple processor hosts.

In M3UA, the active/standby model addresses these systems.  Although M3UA
provides for a '1+k' model, k is invariably 1.

The major difficulties with these systems are: a) switchover latency between
systems; b) loss of state during switchover; and c) scalability.  Switchover
latency is normally improved by decreasing heartbeat intervals.  State
synchronization is normally improved by decreasing the process checkpoint
interval.  These reductions typically increase cost and they are normally
chosen to minimize a cost/performance/reliability metric.  Scalability was
typically performed by installing larger more powerful hosts with increased
demand.  This was not good, because larger processors tended to have poorer
cost/performance ratios that smaller processors.

Later it was determined that the cost of a complete redundant host that was
wasting CPU cycles while inactive could be avoided while addressing the
deficiencies of the paired host approach above.  To increase state
synchronization, state could be spread over a number of hosts.  If there were
'm' hosts, the failure of one host could at worst affect a proportion, 1/m, of
the state.  Therefore state synchronization would improve as m increased.  To
distribute the state, traffic load would be shared over the m hosts according
to state.  This also served the needs of scalability and provided for closer
to linear cost scalability, because the number of hosts, m, could be increased
while holding the cost/performance ratio of the hosts constant.  The costs of
redundancy could also be reduced over the fully redundant systems described
above, because redundancy could be achieved using 'm+1' processors, where 'm'
processors were required to handle the maximum engineered load.  For systems
as small as 'm=2' hosts with equivalent reliability to that normally
experienced in the '1+1' active/standby scenario above, system reliability
figures could be maintained with less redundancy cost.  It is important to
note, that in these systems, all 'm+1' hosts are active and handling only a
proportion m/(m+1) of their rated capacity.  It is very seldom that the rated
capacity was more that 2/3 of the maximum short-term capacity.

These systems do not solve the heartbeat interval problem of the
active/standby paired hosts system above.  Typically what these systems use is
more sophisticated watchdog approaches rather than passing heartbeat messages
between processes on the active and standby hosts.  Memory sharing, DASD
sharing, message passing, reliable broadcast, voting, and a plethora of
reliable distributed systems techniques have been developed that are useful in
these systems.

Adoption of these latter systems changed the requirement from "duplicated"
systems to systems with no "single point of failure".

Now, the 'n+k' load-sharing and broadcast model was intended to address these
systems.  Where 'n == m' in the description above.  That is why we agreed that
n is the maximum engineered load.  Because that is what 'm' is.  The k is the
1 in the description above, generalized.

First off, you must understand that it is not the intention of the model to
have 'n' ASPs active and 'k' inactive unless 'n=1, k=1'.  This is described in
section 1.4.4.1:

  A "1+1" active/standby redundancy is a subset of this model.

Also no redundancy is possible:

  A simplex "1+0" model is also supported as a subset, with no ASP redundancy.

(The term "simplex" come from the telephone switches described above where the
inactive side is taken off-line.  Such a situation is described as running the
switch "simplex".)

The 'n+k' model is to address the 'm+1' distributed systems described last.
Rereading section 1.4.4.1:

  The failover model supports an "n+k" redundancy model, where "n" APSs is the
  minimum number of redundant [sic] ASPs required to handle traffic and "k"
  ASPs are available to take over for a failed or unavailable ASP.

Note that the state of the "k" ASPs with respect to an AS is not stated.  In
the 'm+1' redundant system described above, the "k" ASPs are active for the
AS.  "k" ASPs active in load sharing mode are "available to take over for a
failed or unavailable ASP".

It is typical to run the "k" ASPs active (and I know you had some questions
about activating the "k" ASPs, but the point is that they are normally already
active), and handling their own share of traffic.  In this case, each ASPs
handles n/(n+k) of their engineered capacity.  Were you actually considering
deploying such a atypical model where the 'k' ASPs are inactive?

"k" ASPs can be taken out of service, upgraded, and brought back into service
without affecting the maximum engineering traffic handling capacity of the
system (in absence of failures during the upgrade).  This assists in graceful
software/hardware upgrades on live systems.  (Keeping them inactive would
serve no purpose and would, decrease maximum traffic handling capability,
increase loss of state on fail-over, and increase fail-over latency: all good 
reasons for not deactivating the 'k' ASPs).

Now, reliability engineering for carrier class systems takes the view that if
50% of the load bearing components in a system fail (stop operating in any
way), then the entire system has failed.  If traffic is stopped above n/2
processes, the system will have failed prematurely.

A requirement to withhold traffic < n will always suffer a prolonged system
failure, because only n/2 are required to make the system available.
Therefore, on startup the reliability of the system is degraded by the
negative exponent n/2, regardless of whether n=256 or n=2; however, increasing
'n' increases the damage done to reliability by waiting for 'n'.

A requirement for the SG to wait for 'n' is unacceptable from the standpoint
of the redundancy model that 'n+k' was meant to address, as well as other
possible models that it can accommodate.

> 
> 
> > 
> > Consider the same example with: n=2, k=1, two ASPs failed.
> > 
> > Any system that introduces hysteresis to aggravate and perpetuate an outage in
> > this fashion is not very useful.
> > 
> > In the examples, what would have been a brief AS-PENDING situation has been
> > turned into a prolonged and complete system failure.  It is certainly not good
> > practice to design redundant systems in that fashion and recommending or
> > requiring such a design would be a bad thing.
> > 
> > To further the example, consider the multiple SG as STP situation.  One SG
> > cannot know the state of ASPs for the its mated SG and therefore cannot
> > distinguish between the restart scenario and the start scenario as far as the
> > AS state is concerned because the SG does not know the state of the AS itself:
> > the AS could be active with other SGs.
> > 
> > This is the same case for SGP.  Unless AS state is synchronized across SGPs,
> > one SGP cannot know the total state of the AS (only its partition).  There is
> > no requirement to synchronize AS state across SGP (on the SGP side, AS-state
> > is per-SGP), and there is certainly no requirement to synchronize AS-state
> > across SGs.
> > 
> > So, take the (n=2, k=0) example where there are two SGs.  Due to a network
> > partitioning at startup, one ASP can communicate only with one SG, and the
> > other ASP can communicate only with the other SG.  Because n=2, neither SG
> > will start traffic and the system suffers 100% failure until the network
> > partitioning ceases.
> > 
> > Consider the same example where there are two SGPs (instead of SGs) and the
> > two ASPs normally only form one association (e.g., ASP1-SGP1, ASP2-SGP2).
> > Note that there is no requirement in the RFC to form full-mesh associations.
> > Because the number of active ASPs with each SGP is < 2, the system will never
> > start.
> > 
> > These are not strange scenarios.  These situations are all permitted by the
> > RFC.  Yet, in each situation it is clear that one cannot recommend that the SG
> > wait.  The result would be a non-functional system.  There are many more
> > examples.
> > 
> > n+k is a model for redundancy (not for congestion and flow control).  A system
> > with an SG that waits for n is, in general, not a fully redundant system at
> > all, but is, rather, prone to these failure syndromes.
> > 
> > When you look at the situation and apply the experience of reliability
> > engineering, the recommendation is that the SG SHOULD start traffic when the
> > first ASP goes active.
> > 
> > Why wait for 'n' anyway?  Because the ASP cannot handle some level of traffic?
> > Any ASP that conforms to MTP-User standards, must have the standardized
> > MTP-User (ISUP/SCCP) congestion and flow controls and must have the SCTP
> > congestion controls and is therefore able to accept any amount of traffic
> > overload.
> > 
> > So there is no real reason for the SG to wait for 'n'.  Nevertheless, the RFC
> > allows this as an OPTION.  Requiring (MUST) or recommending it (SHOULD)
> > regardless of whether the number of active processors is rising or falling,
> > whether start, restart or failure, is unacceptable.
> > 
> > On the other side, at the ASPs, the ASPs have a requirement to share state
> > between them.  They must share circuit state and transaction state in all
> > configurations to meet carrier grade.  This is stated in the RFC.  It is
> > implicit that the ASPs are capable of knowing with which SGs, and which SGPs
> > for which they are active.  They have access to the total state of the AS that
> > they collectively represent, that the SG/SGP does not.  Therefore is it
> > reasonable for the ASPs to coordinate among themselves the number of active
> > ASPs and to ensure that there are sufficient ASPs active to handle the
> > expected load before starting traffic.  Because the ASPs need not respond to
> > user test traffic (UPT/SST), the ASPs are also in control of when remote SS7
> > MTP-User restart traffic toward them.
> > 
> > Therefore, the recommendation that the ASPs SHOULD coordinate among
> > themselves the number of active ASPs and not start traffic until a sufficient
> > number of them are active is sound.  This is not a MUST, permitting ASPs
> > that know that they are capable of handling any overload, to start traffic
> > immediately, or at some earlier point.
> > 
> > Back to the RFC, the RFC sets the normal case where the SG starts traffic on
> > the first active ASP but explicitly permits as an OPTION to wait until it
> > considers that there are sufficient ASPs active to handle the offered load.
> > The RFC also RECOMMENDS that the ASPs coordinate among themselves the number
> > of active ASPs so that there are sufficient ASPs to handle the expected load.
> > 
> > The provisions in the RFC are both sound, where the proposal and section 3.15
> > is clearly not.
> > 
> > Remember when looking at the AS state diagram in the RFC that this is the
> > SGP's view of the AS state and not the total state of the AS, that can only
> > known to the ASPs for all configurations.  'n' as defined has no meaning to
> > the state machine at the SGP because is not required to know, and often cannot
> > know, how many ASPs are active for the AS via other SGPs and SGs, or even via
> > direct connection to the SS7 network.
> > 
> > Some SGs might have special knowledge, or sophisticated algorithms that allow
> > it to determine whether a congestive failure will be the immediate result of
> > starting or restarting traffic.  These SGs are permitted as an OPTION to wait.
> > 
> > The ASPs SHOULD have the knowledge of how many are required to handle the
> > expected load, because they must share circuit/transaction state and they must
> > have congestion and flow controls as required by the standards.
> > 
> > So again, the RFC is adequate in this regard.  The current proposal and
> > section 3.15 are not.
> RFC is inadequate so we need section 3.15 to explain 'n+k' case.

Teaching reliability engineering and the design of redundant systems is
outside the scope of the RFC and the IG.  People, such as yourself, are better
advised to learn the subject matter before attempting to implement or claim a
complete understanding of the protocol.

The current RFC provides the protocol mechanisms and basis for
interoperability for all of the common approaches to redundancy in carrier
class systems, and even provides a general model that can accommodate
non-traditional redundancy architectures.

When confused about how a redundancy architecture works, don't blame the M3UA
RFC, blame those responsible for describing redundant architectures to you.

In the appendix we stated:

   M3UA imposes no restrictions as to the exact layout of the network
   elements, the message distribution algorithms and the distribution of
   signalling processes.  Instead, it provides a framework and a set of
   messages that allow for a flexible and scalable signalling network
   architecture, aiming to provide reliability and performance.

Forcing SGs to wait for 'n' ASPs active before activating is contrary to this.
The existing "MAY" achieves this flexibility.

When reading 1.4.4.1, also remember this statement in the appendix, that used
to be beside 1.4.4.1.  The description of 'n+k' does not dictate the state of
the 'k' or the state of the AS with respect to 'n' to leave the redundancy
model flexible, and accommodate '1+1' and '1+0' in the same model.

Also, please don't presume to tell me to set 'n' to 1, as it is not a
arbitrary value, but the number 'm' in the 'm+1' distributed system in common
use today.  This number is determined as the minimum number of ASPs required
to handle the maximum engineered load, and not, as Tolga would have, the
minimum number of ASPs required to handle ANY load.  If you want your system
to behave as Tolga proposes, just have your SG exercise the "MAY choose not to
direct traffic" option of section 4.3.4.3 of the RFC.

Section 3.15 and Tolga's proposal should be struck because they attempt to
impose a 'n' active 'k' inactive and system cannot restart (3.15) or start
(Tolga) without 'n' view on the redundancy model, where it is clearly the
intent of the RFC to place no such restrictions on the model, but to
accommodate a range of models, that is already achieved by the RFC text.

--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 exim@www1.ietf.org  Tue Jul  1 07:12:27 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 HAA06358
	for <sigtran-archive@odin.ietf.org>; Tue, 1 Jul 2003 07:12:27 -0400 (EDT)
Received: (from exim@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h61BC0924570
	for sigtran-archive@odin.ietf.org; Tue, 1 Jul 2003 07:12:00 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XJ3G-0006O3-7s; Tue, 01 Jul 2003 07:11:58 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XJ3B-0006N7-5b
	for sigtran@optimus.ietf.org; Tue, 01 Jul 2003 07:11:53 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA06327
	for <sigtran@ietf.org>; Tue, 1 Jul 2003 07:11:46 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19XJ34-0002Be-00
	for sigtran@ietf.org; Tue, 01 Jul 2003 07:11:47 -0400
Received: from auds952.usa.alcatel.com ([143.209.238.7])
	by ietf-mx with esmtp (Exim 4.12)
	id 19XJ2u-0002Ac-00
	for sigtran@ietf.org; Tue, 01 Jul 2003 07:11:36 -0400
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h61BAAB3002374;
	Tue, 1 Jul 2003 06:10:11 -0500 (CDT)
Received: from sun2594.ssd.usa.alcatel.com (sun2594.ssd.usa.alcatel.com [143.209.250.252])
	by ssd.usa.alcatel.com (8.12.8/8.12.8) with ESMTP id h61B9Y1C022012;
	Tue, 1 Jul 2003 06:09:35 -0500 (CDT)
Date: Tue, 1 Jul 2003 06:09:34 -0500 (CDT)
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:n+k issue
In-Reply-To: <20030701033851.A31866@openss7.org>
Message-ID: <Pine.SOL.4.10.10307010509100.24913-100000@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,

  Can you bit elaborate your n=256 example (ANSI-SLS) with real-time 
  scenario?(please write to the point)

--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, 1 Jul 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:                      (Mon, 30 Jun 2003 09:44:01)
> > 
> > Brian,
> > 
> > On Sat, 28 Jun 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > I should point out that the proposal to require that the SG wait for 'n' to
> > > start traffic does not have much merit.
> > > 
> > > Example: n=256, k=1
> > > 
> > > 2 ASPs fail, permanently and need to be manually replaced.
> > > 
> > > A temporary network failure causes SCTP associations to be lost to the
> > > remaining 255 ASPs.  They move to the ASP down state and immediately
> > > re-establish SCTP associations and move to the ASP-Active state.  Because the
> > > number of ASPs is < n, the SG will not restart traffic for some time.
> > 
> > Your example leads to the following figure.
> > 
> >          SG1                            AS1
> >      
> >  SS7 MSG +------+----------------------- ASP (1)
> > ---------| SGP1 |----------------------- ASP (2)
> >          |      |----------------------- ....
> >          +------+----------------------- ASP (256)
> >             |
> >             | 
> >             +---------------------------ASP(k)
> > 
> > 1. First of all craft will not configure 1 ASP as backup for 256 Active
> >    ASPs, if he does, it means to be a poor network configuration.
> 
> Well, first of all, craft does not normally choose the number of ASPs used for
> backup, engineering does.  Second, k=1 is more than adequate for n=256 where
> the ASPs are reasonably reliable and have a short replacement time.
> 
> Regardless, I followed with the same example with n=2 and k=1.
> 
> > 
> > 2. Consider ASP as an MGC, your example leads to 256 MGCs and they are 
> >    controlled by a single SGP.A good network will have multiple SGPs to
> >    share the POTENTIAL traffic across ASPs.
> 
> You seem to misunderstand the concept of redundancy in M3UA: there are not
> multiple SGPs or ASPs because of traffic load.  It is easy enough to handle
> the traffic of a small to medium sized country through a single processor.
> 
> Also, you seem to misunderstand that an SGP and an ASP are not processors.
> They are processes.  One host processor could have many (even 256) ASPs or
> SGPs (or both for that matter).
> 
> > 
> > 3. BASIC INTENTION OF 'n+k' is,just to make sure that incoming messages
> >    ARE EFFICIENTLY TRANSMITTED TO DESTINATION ASPs WITH THE HELP OF
> >    MINIMUM NUMBER OF ASSOCIATIONS.
> 
> No.  The intention of 'n+k' is to provide one of the more recently accepted
> models for reliability using redundancy (n = m+1, k=0 being the typical system
> deployed in the field today, m usually being greater than 2).  Up until
> recently, only 2m (or m+m) systems were considered carrier class.  Some
> systems were 4m (with 3 out of 4 vote).
> 
> You should take some time to look at redundancy and reliability requirements
> for carrier grade systems for an understanding of how to calculate and provide
> the reliability requirements expected by carriers.  There are good standards
> written on the topic.  See the Telcordia NEBS and reliability specs, for
> example.  If you have an RBOC RFP handy, take a look at the hardware and
> software reliability requirements sections or talk to the engineer in your
> company responsible for answering those sections.
> 
> > 
> >    Say for example
> > 
> >    Inefficient configuration -just an example
> >    ------------------------------------------
> >   
> >             SLS         DEST
> >             =================
> >              0           ASP1
> >              1           ASP1
> >              2           ASP1
> >              3           ASP1
> >              4           ASP1
> >              5           ASP1
> >              6           ASP1
> >              .            .
> >              .            .
> >              15          ASP1
> >              
> >              
> >                SG                            AS1
> > sls(0,1..15 )+----+  sls(0,1..15)         +------+
> > -------------|SGP1|-----------------------| ASP1 |
> >              |    |              ASSOC-1  +------+
> >              +----+                     
> > 
> >                       Fig - 1
> > 
> > In this figure, all incoming messages are transmitted over ASSOC-1 towards
> > ASP1.Here there are two potential problems 
> > 
> >        1. Total incoming traffic is dependent on a single ASP
> >        2. ASP1 is always overloaded, messages are processed one by one
> >           from the transmission buffer which ultimately introduces time
> >           delay(example last arrived msg will be processed at the end).
> 
> The above was not in my examples.  My examples had n > 1.
> 
> > 
> > 
> > Efficient configuration
> > -----------------------  
> > 
> >              SLS         DEST
> >             =================
> >              0           ASP1
> >              1           ASP1
> >              2           ASP1
> >              3           ASP1
> >              4           ASP1
> >              5           ASP1
> >              6           ASP1
> >              7           ASP1
> >              8           ASP2
> >              9           ASP2
> >              .            .
> >              .            .
> >              15          ASP2
> > 
> >                
> >                        
> >                        
> >               SG                             AS1
> > sls(0,1..15 )+----+   sls(0,1,..7)ASSOC-1 +------+
> > -------------|SGP1|-----------------------| ASP1 |
> >              |    |--------+              +------+
> >              +----+        | sls(8,9..15) +------+
> >                            +--------------| ASP2 |
> >                                  ASSOC-2  +------+
> >                         
> >                    fig - 2
> > 
> > In this example incoming traffic is distributed between ASP1 and ASP2.
> > 
> > IN ORDER TO HANDLE THE HEAVY INCOMING TRAFFIC, SGP1 NEEDS MINIMUM OF
> > 2 ACTIVE ASPs BY SETTING n=2.
> 
> Do they not comply to SS7 ISUP/SCCP flow and congestion controls?  Are they
> running at full capacity in this arrangement?  Are these your ASPs?  Do you
> design your systems this way?
> 
> Take a look at the reliability engineering.  If either of these ASPs is
> handling its maximum rated capacity, it will have to be reliable to .99999
> This is because, if one ASP fails, 50% of the system has failed.  A 50% system
> failure is deemed a "system" failure rather than a "component" failure in
> reliability engineering for carrier grade systems.  Considering proscribed
> replacement times on the ASP, this would be difficult to achieve unless both
> ASPs were hardware fault tolerant platforms.
> 
> Nevertheless, the RFC defines ASPs as a process, not a processor.  There is
> nothing restricting the two ASPs in your diagram from being processes on the
> same processor.  In that case, when ASP2 is down, ASP1 has double the
> processing capacity as it had before.
> 
> You certainly cannot make statements as you have made here without
> presupposing an implementation architecture outside the scope of the standard.
> 
> Besides, the RFC permits you to configure your SGP (and why do you only have
> one of them?) so that it waits for 'n' ASPs to become active.  You MAY do
> this.  Just don't require everybody else to follow your approach.  (Actually,
> if the above is your approach, don't require anyone else to follow it.)
> 
> Several other points: ANSI T1.111.4/2000 SLS values are 8-bits.  That means
> there are 256 SLS values (gee, I had 256 ASPs in my example).
> 
> For another thing, sequenced traffic flows are not only based on SLS.  In an
> ITU-T network, each remote signalling point has 16 independent traffic flows
> for each AS.  The traffic flows from each remote signalling point is
> independent from the traffic flows of other remote signalling points.
> Therefore, if there are 400 remote signalling points, there will be 6400
> independent traffic flows that the SG can use for distribution of AS traffic
> for one AS over as many as 6400 ASPs without risking mis-sequencing of
> messages.  For an ANSI T1.111.4/2000 network using 8-bit SLS values with 400
> remote signalling points, there would be 102400 independent traffic flows to
> the AS that the SG can use to distribute traffic to one AS over as many as
> 102400 ASPs without risking message mis-sequencing.
> 
> > 
> > 
> > Your example with n=256 is completely a wrong interpretaion of 'n+k'
> > redundancy case.
> 
> Quite the opposite, you clearly misunderstand redundancy and reliability in
> carrier grade systems and obviously have little experience with them.
> 
> Here's a little primer for you:
> 
> Carrier grade systems were originally designed as hot-standby, fully redundant
> systems.  For every processor actively handling traffic, there was a hot
> standby processor also actively handling the same traffic.  This typically
> required special bus arrangements whereby the I/O bus of one processor was
> actively controlling peripheral devices, while the hot standby processor was
> issuing I/O bus output that was not terminated to the I/O bus.  The bus also
> normally performed a comparison of the outputs from the active side and the
> outputs from the inactive side.  Peripheral components, such as switch fabric
> were also fully redundant (duplicated).  Switching fabrics are separated into
> an A-plane and a B-plane in much the same fashion as the SPCS front-end I/O
> bus.
> 
> Note that active/standby in M3UA does not address the above model.  2+0
> broadcast does.
> 
> The problem with such a systems was what we used to call the "kill your
> sister" syndrome.  A failure on the inactive side only detectable to the I/O
> bus could cause a switch from the active to inactive side, resulting in the
> switch running on the erroneous inactive side.  We called it the "kill your
> sister" syndrome because it is not so surprising that the psychotic sister
> (erred inactive side) kills the sane sister (correct active side).
> 
> Primarily to avoid the "kill your sister" syndrome, some switches were
> designed with 3 or 4 processors on an cross-checking I/O bus pair.  In that
> case, it would take a 2 out of 3 or 3 out of 4 vote to shift from the active
> processor.  These systems were far less susceptible to inactive plane errors
> that were only detected at the I/O bus.
> 
> Note that the only M3UA model that address the above is the 3+0 or 4+0
> broadcast model.
> 
> When the use of general purpose computing platforms for telecommunications
> systems started to become popular, there were several models that emerged.
> 
> First was taking two big iron UNIX hosts and running them in active/standby.
> In this scenario, one UNIX host would be actively handling traffic and the
> other would be warm standby.  By "warm" standby, it means that the device was
> powered and was synchronizing state on a checkpoint interval or monitoring the
> health of the active hosts with heartbeats.  Shared memory, shared DASD or
> message passing are the most common mechanisms of redundancy.  These systems
> are considered "highly available" systems rather than the "fault tolerant"
> systems used in telecommunications above.  An increased level of reliability
> is often achieved by using hardware fault tolerant multiple processor hosts.
> 
> In M3UA, the active/standby model addresses these systems.  Although M3UA
> provides for a '1+k' model, k is invariably 1.
> 
> The major difficulties with these systems are: a) switchover latency between
> systems; b) loss of state during switchover; and c) scalability.  Switchover
> latency is normally improved by decreasing heartbeat intervals.  State
> synchronization is normally improved by decreasing the process checkpoint
> interval.  These reductions typically increase cost and they are normally
> chosen to minimize a cost/performance/reliability metric.  Scalability was
> typically performed by installing larger more powerful hosts with increased
> demand.  This was not good, because larger processors tended to have poorer
> cost/performance ratios that smaller processors.
> 
> Later it was determined that the cost of a complete redundant host that was
> wasting CPU cycles while inactive could be avoided while addressing the
> deficiencies of the paired host approach above.  To increase state
> synchronization, state could be spread over a number of hosts.  If there were
> 'm' hosts, the failure of one host could at worst affect a proportion, 1/m, of
> the state.  Therefore state synchronization would improve as m increased.  To
> distribute the state, traffic load would be shared over the m hosts according
> to state.  This also served the needs of scalability and provided for closer
> to linear cost scalability, because the number of hosts, m, could be increased
> while holding the cost/performance ratio of the hosts constant.  The costs of
> redundancy could also be reduced over the fully redundant systems described
> above, because redundancy could be achieved using 'm+1' processors, where 'm'
> processors were required to handle the maximum engineered load.  For systems
> as small as 'm=2' hosts with equivalent reliability to that normally
> experienced in the '1+1' active/standby scenario above, system reliability
> figures could be maintained with less redundancy cost.  It is important to
> note, that in these systems, all 'm+1' hosts are active and handling only a
> proportion m/(m+1) of their rated capacity.  It is very seldom that the rated
> capacity was more that 2/3 of the maximum short-term capacity.
> 
> These systems do not solve the heartbeat interval problem of the
> active/standby paired hosts system above.  Typically what these systems use is
> more sophisticated watchdog approaches rather than passing heartbeat messages
> between processes on the active and standby hosts.  Memory sharing, DASD
> sharing, message passing, reliable broadcast, voting, and a plethora of
> reliable distributed systems techniques have been developed that are useful in
> these systems.
> 
> Adoption of these latter systems changed the requirement from "duplicated"
> systems to systems with no "single point of failure".
> 
> Now, the 'n+k' load-sharing and broadcast model was intended to address these
> systems.  Where 'n == m' in the description above.  That is why we agreed that
> n is the maximum engineered load.  Because that is what 'm' is.  The k is the
> 1 in the description above, generalized.
> 
> First off, you must understand that it is not the intention of the model to
> have 'n' ASPs active and 'k' inactive unless 'n=1, k=1'.  This is described in
> section 1.4.4.1:
> 
>   A "1+1" active/standby redundancy is a subset of this model.
> 
> Also no redundancy is possible:
> 
>   A simplex "1+0" model is also supported as a subset, with no ASP redundancy.
> 
> (The term "simplex" come from the telephone switches described above where the
> inactive side is taken off-line.  Such a situation is described as running the
> switch "simplex".)
> 
> The 'n+k' model is to address the 'm+1' distributed systems described last.
> Rereading section 1.4.4.1:
> 
>   The failover model supports an "n+k" redundancy model, where "n" APSs is the
>   minimum number of redundant [sic] ASPs required to handle traffic and "k"
>   ASPs are available to take over for a failed or unavailable ASP.
> 
> Note that the state of the "k" ASPs with respect to an AS is not stated.  In
> the 'm+1' redundant system described above, the "k" ASPs are active for the
> AS.  "k" ASPs active in load sharing mode are "available to take over for a
> failed or unavailable ASP".
> 
> It is typical to run the "k" ASPs active (and I know you had some questions
> about activating the "k" ASPs, but the point is that they are normally already
> active), and handling their own share of traffic.  In this case, each ASPs
> handles n/(n+k) of their engineered capacity.  Were you actually considering
> deploying such a atypical model where the 'k' ASPs are inactive?
> 
> "k" ASPs can be taken out of service, upgraded, and brought back into service
> without affecting the maximum engineering traffic handling capacity of the
> system (in absence of failures during the upgrade).  This assists in graceful
> software/hardware upgrades on live systems.  (Keeping them inactive would
> serve no purpose and would, decrease maximum traffic handling capability,
> increase loss of state on fail-over, and increase fail-over latency: all good 
> reasons for not deactivating the 'k' ASPs).
> 
> Now, reliability engineering for carrier class systems takes the view that if
> 50% of the load bearing components in a system fail (stop operating in any
> way), then the entire system has failed.  If traffic is stopped above n/2
> processes, the system will have failed prematurely.
> 
> A requirement to withhold traffic < n will always suffer a prolonged system
> failure, because only n/2 are required to make the system available.
> Therefore, on startup the reliability of the system is degraded by the
> negative exponent n/2, regardless of whether n=256 or n=2; however, increasing
> 'n' increases the damage done to reliability by waiting for 'n'.
> 
> A requirement for the SG to wait for 'n' is unacceptable from the standpoint
> of the redundancy model that 'n+k' was meant to address, as well as other
> possible models that it can accommodate.
> 
> > 
> > 
> > > 
> > > Consider the same example with: n=2, k=1, two ASPs failed.
> > > 
> > > Any system that introduces hysteresis to aggravate and perpetuate an outage in
> > > this fashion is not very useful.
> > > 
> > > In the examples, what would have been a brief AS-PENDING situation has been
> > > turned into a prolonged and complete system failure.  It is certainly not good
> > > practice to design redundant systems in that fashion and recommending or
> > > requiring such a design would be a bad thing.
> > > 
> > > To further the example, consider the multiple SG as STP situation.  One SG
> > > cannot know the state of ASPs for the its mated SG and therefore cannot
> > > distinguish between the restart scenario and the start scenario as far as the
> > > AS state is concerned because the SG does not know the state of the AS itself:
> > > the AS could be active with other SGs.
> > > 
> > > This is the same case for SGP.  Unless AS state is synchronized across SGPs,
> > > one SGP cannot know the total state of the AS (only its partition).  There is
> > > no requirement to synchronize AS state across SGP (on the SGP side, AS-state
> > > is per-SGP), and there is certainly no requirement to synchronize AS-state
> > > across SGs.
> > > 
> > > So, take the (n=2, k=0) example where there are two SGs.  Due to a network
> > > partitioning at startup, one ASP can communicate only with one SG, and the
> > > other ASP can communicate only with the other SG.  Because n=2, neither SG
> > > will start traffic and the system suffers 100% failure until the network
> > > partitioning ceases.
> > > 
> > > Consider the same example where there are two SGPs (instead of SGs) and the
> > > two ASPs normally only form one association (e.g., ASP1-SGP1, ASP2-SGP2).
> > > Note that there is no requirement in the RFC to form full-mesh associations.
> > > Because the number of active ASPs with each SGP is < 2, the system will never
> > > start.
> > > 
> > > These are not strange scenarios.  These situations are all permitted by the
> > > RFC.  Yet, in each situation it is clear that one cannot recommend that the SG
> > > wait.  The result would be a non-functional system.  There are many more
> > > examples.
> > > 
> > > n+k is a model for redundancy (not for congestion and flow control).  A system
> > > with an SG that waits for n is, in general, not a fully redundant system at
> > > all, but is, rather, prone to these failure syndromes.
> > > 
> > > When you look at the situation and apply the experience of reliability
> > > engineering, the recommendation is that the SG SHOULD start traffic when the
> > > first ASP goes active.
> > > 
> > > Why wait for 'n' anyway?  Because the ASP cannot handle some level of traffic?
> > > Any ASP that conforms to MTP-User standards, must have the standardized
> > > MTP-User (ISUP/SCCP) congestion and flow controls and must have the SCTP
> > > congestion controls and is therefore able to accept any amount of traffic
> > > overload.
> > > 
> > > So there is no real reason for the SG to wait for 'n'.  Nevertheless, the RFC
> > > allows this as an OPTION.  Requiring (MUST) or recommending it (SHOULD)
> > > regardless of whether the number of active processors is rising or falling,
> > > whether start, restart or failure, is unacceptable.
> > > 
> > > On the other side, at the ASPs, the ASPs have a requirement to share state
> > > between them.  They must share circuit state and transaction state in all
> > > configurations to meet carrier grade.  This is stated in the RFC.  It is
> > > implicit that the ASPs are capable of knowing with which SGs, and which SGPs
> > > for which they are active.  They have access to the total state of the AS that
> > > they collectively represent, that the SG/SGP does not.  Therefore is it
> > > reasonable for the ASPs to coordinate among themselves the number of active
> > > ASPs and to ensure that there are sufficient ASPs active to handle the
> > > expected load before starting traffic.  Because the ASPs need not respond to
> > > user test traffic (UPT/SST), the ASPs are also in control of when remote SS7
> > > MTP-User restart traffic toward them.
> > > 
> > > Therefore, the recommendation that the ASPs SHOULD coordinate among
> > > themselves the number of active ASPs and not start traffic until a sufficient
> > > number of them are active is sound.  This is not a MUST, permitting ASPs
> > > that know that they are capable of handling any overload, to start traffic
> > > immediately, or at some earlier point.
> > > 
> > > Back to the RFC, the RFC sets the normal case where the SG starts traffic on
> > > the first active ASP but explicitly permits as an OPTION to wait until it
> > > considers that there are sufficient ASPs active to handle the offered load.
> > > The RFC also RECOMMENDS that the ASPs coordinate among themselves the number
> > > of active ASPs so that there are sufficient ASPs to handle the expected load.
> > > 
> > > The provisions in the RFC are both sound, where the proposal and section 3.15
> > > is clearly not.
> > > 
> > > Remember when looking at the AS state diagram in the RFC that this is the
> > > SGP's view of the AS state and not the total state of the AS, that can only
> > > known to the ASPs for all configurations.  'n' as defined has no meaning to
> > > the state machine at the SGP because is not required to know, and often cannot
> > > know, how many ASPs are active for the AS via other SGPs and SGs, or even via
> > > direct connection to the SS7 network.
> > > 
> > > Some SGs might have special knowledge, or sophisticated algorithms that allow
> > > it to determine whether a congestive failure will be the immediate result of
 > > > starting or restarting traffic.  These SGs are permitted as an
OPTION to wait.
> > > 
> > > The ASPs SHOULD have the knowledge of how many are required to handle the
> > > expected load, because they must share circuit/transaction state and they must
> > > have congestion and flow controls as required by the standards.
> > > 
> > > So again, the RFC is adequate in this regard.  The current proposal and
> > > section 3.15 are not.
> > RFC is inadequate so we need section 3.15 to explain 'n+k' case.
> 
> Teaching reliability engineering and the design of redundant systems is
> outside the scope of the RFC and the IG.  People, such as yourself, are better
> advised to learn the subject matter before attempting to implement or claim a
> complete understanding of the protocol.
> 
> The current RFC provides the protocol mechanisms and basis for
> interoperability for all of the common approaches to redundancy in carrier
> class systems, and even provides a general model that can accommodate
> non-traditional redundancy architectures.
> 
> When confused about how a redundancy architecture works, don't blame the M3UA
> RFC, blame those responsible for describing redundant architectures to you.
> 
> In the appendix we stated:
> 
>    M3UA imposes no restrictions as to the exact layout of the network
>    elements, the message distribution algorithms and the distribution of
>    signalling processes.  Instead, it provides a framework and a set of
>    messages that allow for a flexible and scalable signalling network
>    architecture, aiming to provide reliability and performance.
> 
> Forcing SGs to wait for 'n' ASPs active before activating is contrary to this.
> The existing "MAY" achieves this flexibility.
> 
> When reading 1.4.4.1, also remember this statement in the appendix, that used
> to be beside 1.4.4.1.  The description of 'n+k' does not dictate the state of
> the 'k' or the state of the AS with respect to 'n' to leave the redundancy
> model flexible, and accommodate '1+1' and '1+0' in the same model.
> 
> Also, please don't presume to tell me to set 'n' to 1, as it is not a
> arbitrary value, but the number 'm' in the 'm+1' distributed system in common
> use today.  This number is determined as the minimum number of ASPs required
> to handle the maximum engineered load, and not, as Tolga would have, the
> minimum number of ASPs required to handle ANY load.  If you want your system
> to behave as Tolga proposes, just have your SG exercise the "MAY choose not to
> direct traffic" option of section 4.3.4.3 of the RFC.
> 
> Section 3.15 and Tolga's proposal should be struck because they attempt to
> impose a 'n' active 'k' inactive and system cannot restart (3.15) or start
> (Tolga) without 'n' view on the redundancy model, where it is clearly the
> intent of the RFC to place no such restrictions on the model, but to
> accommodate a range of models, that is already achieved by the RFC text.
> 
> --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 exim@www1.ietf.org  Tue Jul  1 08:33:28 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 IAA15154
	for <sigtran-archive@odin.ietf.org>; Tue, 1 Jul 2003 08:33:28 -0400 (EDT)
Received: (from exim@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h61CX2k11431
	for sigtran-archive@odin.ietf.org; Tue, 1 Jul 2003 08:33:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XKJg-0002y8-Ub; Tue, 01 Jul 2003 08:33:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XKJP-0002wE-OL
	for sigtran@optimus.ietf.org; Tue, 01 Jul 2003 08:32:43 -0400
Received: from hindon.hss.co.in (hindon.hss.co.in [202.54.26.202])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA15067
	for <sigtran@ietf.org>; Tue, 1 Jul 2003 08:32:36 -0400 (EDT)
From: vissharma@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 h61CSEH02989
	for <sigtran@ietf.org>; Tue, 1 Jul 2003 17:58:14 +0530 (IST)
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h61CSAO02977;
	Tue, 1 Jul 2003 17:58:11 +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 ESMTP id h61CVnB15377;
	Tue, 1 Jul 2003 18:01:49 +0530 (IST)
Subject: Re: [Sigtran] Signaling Gateway Doubts
To: Pablo Salinas <pablopesb@yahoo.com>
Cc: sigtran@ietf.org
Date: Tue, 1 Jul 2003 17:56:33 +0530
Message-ID: <OF4FBB1FA7.DD46A7CB-ON65256D56.00443990@hss.hns.com>
X-MIMETrack: Serialize by Router on Sandesh/HSS(Release 6.0|September 26, 2002) at 01/07/2003
 05:55:31 PM
MIME-Version: 1.0
Content-type: text/plain; charset=ISO-8859-1
Content-transfer-encoding: quoted-printable
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable





HI ,
As per my understand
Answer is yes, U may omit use of  SG ,when SS7 links  terminate directl=
y on
the Softswicth .
Above is very much possible in theory but practicably applicable to  sm=
all
networks only (IMO).
On the other side  when you look at the scaling of network and Network
management ,then the answer will be NO. Because there you would like to=

share load
 functionality wise between different logical/physical entities ,say SG=
 and
Softswicth (MGCU). You may not want to load your MGCU process with SS7 =
link
management functionality ,but better provide services to both SS7 netwo=
rk
and IP network subscribers.

Regards
Vishavdeep Sharma
www.hssworld.com
Hughes Software Systems




Pablo Salinas <pablopesb@yahoo.com>@ietf.org on 06/30/2003 07:49:13 PM

Sent by:    sigtran-admin@ietf.org


To:    sigtran@ietf.org
cc:

Subject:    [Sigtran] Signaling Gateway Doubts



Hi there,
=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 I am having a hard time to understand=
 one thing in a VoIP
Network that=A0uses a softswitch.
Supose, we have the following situation:
There is a softswitch installed in location A. And this softswitch must=

communicate with the PTT in localtion A. Should it have a Signaling Gat=
eway
between the softswitch and the PTT?
If the answer is yes, why use the SCTP, M2UA, M2PA or whatever other
sigtran protocol, if it is possible to connect the softswitch directly =
to
the PTT via the MTP2 protocol.
If the answer is no, why in every network scheme I saw until now, there=
 is
always a SG between the PTT and the softswitch,=A0when the sofswitch
communicates with the media gateways via H.248?

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=
 thanks,
=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=
=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 Pablo Salinas



Do You Yahoo!?
[IMAGE]Todo lo que quieres saber de Estados Unidos, Am=E9rica Latina y =
el
resto del Mundo.
Vis=EDta Yahoo! Noticias.

=



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



From exim@www1.ietf.org  Tue Jul  1 09:32:59 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 JAA26375
	for <sigtran-archive@odin.ietf.org>; Tue, 1 Jul 2003 09:32:59 -0400 (EDT)
Received: (from exim@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h61DWV625022
	for sigtran-archive@odin.ietf.org; Tue, 1 Jul 2003 09:32:31 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XLFH-0006VV-Mg
	for sigtran-web-archive@optimus.ietf.org; Tue, 01 Jul 2003 09:32:31 -0400
Date: Tue, 01 Jul 2003 09:32:31 -0400
Message-ID: <20030701133231.18162.89668.Mailman@www1.ietf.org>
Subject: ietf.org mailing list memberships reminder
From: mailman-owner@www1.ietf.org
To: sigtran-web-archive@optimus.ietf.org
X-No-Archive: yes
X-Ack: no
Sender: sip-admin@ietf.org
Errors-To: sip-admin@ietf.org
X-BeenThere: sip@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk

This is a reminder, sent out once a month, about your ietf.org mailing
list memberships.  It includes your subscription info and how to use
it to change it or unsubscribe from a list.

You can visit the URLs to change your membership status or
configuration, including unsubscribing, setting digest-style delivery
or disabling delivery altogether (e.g., for a vacation), and so on.

In addition to the URL interfaces, you can also use email to make such
changes.  For more info, send a message to the '-request' address of
the list (for example, sigtran-request@ietf.org) containing just the
word 'help' in the message body, and an email message will be sent to
you with instructions.

If you have questions, problems, comments, etc, send them to
mailman-owner@www1.ietf.org.  Thanks!

Passwords for sigtran-web-archive@optimus.ietf.org:

List                                     Password // URL
----                                     --------  
sigtran@ietf.org                         emupuc    
https://www1.ietf.org/mailman/options/sigtran/sigtran-web-archive%40optimus.ietf.org



From exim@www1.ietf.org  Tue Jul  1 10:17:30 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 KAA01959
	for <sigtran-archive@odin.ietf.org>; Tue, 1 Jul 2003 10:17:30 -0400 (EDT)
Received: (from exim@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h61EH3S27137
	for sigtran-archive@odin.ietf.org; Tue, 1 Jul 2003 10:17:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XLwL-00073N-IY; Tue, 01 Jul 2003 10:17:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XLwC-00072A-Pl
	for sigtran@optimus.ietf.org; Tue, 01 Jul 2003 10:16:52 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA01862
	for <sigtran@ietf.org>; Tue, 1 Jul 2003 10:16:49 -0400 (EDT)
From: jeffrey.craig@tekelec.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19XLwA-0004xq-00
	for sigtran@ietf.org; Tue, 01 Jul 2003 10:16:50 -0400
Received: from [198.89.32.2] (helo=domino2.nc.tekelec.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19XLw0-0004wK-00
	for sigtran@ietf.org; Tue, 01 Jul 2003 10:16:40 -0400
Received: from domino4.nc.tekelec.com ([198.89.40.9])
          by domino2.nc.tekelec.com (Lotus Domino Release 5.0.12)
          with ESMTP id 2003070110121665:33450 ;
          Tue, 1 Jul 2003 10:12:16 -0400 
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.7  March 21, 2001
Message-ID: <OFAF74D2FB.53416A2F-ON85256D56.004DF08E@nc.tekelec.com>
Date: Tue, 1 Jul 2003 10:13:01 -0400
MIME-Version: 1.0
X-MIMETrack: Serialize by Router on Domino4/Raleigh/TEKELEC(Release 5.0.12  |February 13, 2003) at
 07/01/2003 10:13:02 AM,
	Itemize by SMTP Server on Domino2/Raleigh/TEKELEC(Release 5.0.12  |February
 13, 2003) at 07/01/2003 10:12:17 AM,
	Serialize by Router on Domino2/Raleigh/TEKELEC(Release 5.0.12  |February 13, 2003) at
 07/01/2003 10:15:33 AM,
	Serialize complete at 07/01/2003 10:15:33 AM
Content-type: text/plain; charset=us-ascii
Subject: [Sigtran] (M2PA) Any word on status? Last Call, v9, etc?
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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,

Is there any word on the status of M2PA?
Will there be a version 9?
Is the last call expired and final?

Regards,

Jeff



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



From exim@www1.ietf.org  Tue Jul  1 12:39:27 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 MAA06823
	for <sigtran-archive@odin.ietf.org>; Tue, 1 Jul 2003 12:39:27 -0400 (EDT)
Received: (from exim@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h61Gd2j23738
	for sigtran-archive@odin.ietf.org; Tue, 1 Jul 2003 12:39:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XO9l-0006Ad-2G; Tue, 01 Jul 2003 12:39:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XO9U-0006AP-28
	for sigtran@optimus.ietf.org; Tue, 01 Jul 2003 12:38:44 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA06795
	for <sigtran@ietf.org>; Tue, 1 Jul 2003 12:38:38 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19XO9Q-0006I3-00
	for sigtran@ietf.org; Tue, 01 Jul 2003 12:38:40 -0400
Received: from sjmail.ciena.com ([12.7.169.25] helo=w2ksjexg01.ciena.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19XO9F-0006Hw-00
	for sigtran@ietf.org; Tue, 01 Jul 2003 12:38:30 -0400
Received: by w2ksjexg01.ciena.com with Internet Mail Service (5.5.2653.19)
	id <NS8BKPCD>; Tue, 1 Jul 2003 09:38:02 -0700
Message-ID: <829F074A10F98943A218DC363D09C92A250222@w2ksjexg06.oni.com>
From: "Ong, Lyndon" <LyOng@Ciena.com>
To: "'jeffrey.craig@tekelec.com'" <jeffrey.craig@tekelec.com>,
        sigtran@ietf.org
Subject: RE: [Sigtran] (M2PA) Any word on status? Last Call, v9, etc?
Date: Tue, 1 Jul 2003 09:41:45 -0700 
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 Jeff,

Tom has submitted a new version, we will LC when it comes out.

Cheers,

Lyndon

-----Original Message-----
From: jeffrey.craig@tekelec.com [mailto:jeffrey.craig@tekelec.com]
Sent: Tuesday, July 01, 2003 7:13 AM
To: sigtran@ietf.org
Subject: [Sigtran] (M2PA) Any word on status? Last Call, v9, etc?


Hello All,

Is there any word on the status of M2PA?
Will there be a version 9?
Is the last call expired and final?

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 exim@www1.ietf.org  Tue Jul  1 13:48:14 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 NAA08385
	for <sigtran-archive@odin.ietf.org>; Tue, 1 Jul 2003 13:48:13 -0400 (EDT)
Received: (from exim@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h5RISDh30338
	for sigtran-archive@odin.ietf.org; Fri, 27 Jun 2003 14:28:13 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19VxxF-0007tE-31; Fri, 27 Jun 2003 14:28:13 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Vuhf-0003V4-EI
	for sigtran@optimus.ietf.org; Fri, 27 Jun 2003 10:59:55 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA02709
	for <sigtran@ietf.org>; Fri, 27 Jun 2003 10:59:45 -0400 (EDT)
From: kakaul@hss.hns.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19VuV2-0003Hk-00
	for sigtran@ietf.org; Fri, 27 Jun 2003 10:46:52 -0400
Received: from hindon.hss.co.in ([202.54.26.202])
	by ietf-mx with esmtp (Exim 4.12)
	id 19VuUq-0003H0-00
	for sigtran@ietf.org; Fri, 27 Jun 2003 10:46:41 -0400
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 h5REgjV02814
	for <sigtran@ietf.org>; Fri, 27 Jun 2003 20:12:46 +0530 (IST)
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h5REghO02799
	for <sigtran@ietf.org>; Fri, 27 Jun 2003 20:12:43 +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 ESMTP id h5REkKq19590
	for <sigtran@ietf.org>; Fri, 27 Jun 2003 20:16:20 +0530 (IST)
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.5  September 22, 2000
Message-ID: <OF103D9B85.761870E4-ON65256D52.004F32F2@hss.hns.com>
Date: Fri, 27 Jun 2003 20:11:06 +0530
X-MIMETrack: Serialize by Router on Sandesh/HSS(Release 6.0|September 26, 2002) at 27/06/2003
 08:10:05 PM
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Subject: [Sigtran] M3UA RFC 3332 default treatment of unallocated data
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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,
 Can some one share thoughts about what is the usage in practical sense of
data routed through default AS from SG side.As M3UA RFC 3332 mentions about
default treatment of unallocated data(data for which no routing key is
configured) at SG side. What can ASP do with that data which has been
routed to it by SG for default AS. I am sure while adding default treatment
text in the RFC, usage of that in practical sense must be thought about. I
want to know in what network scenario SG will receive unallocated data and
after routing it to default AS what is the usage of that data at AS. Please
comment.

Regards
Kamesh Kaul
www.hssworld.com
Hughes Software Systems

"DISCLAIMER: This message is proprietary to Hughes Software Systems Limited
(HSS) and is intended solely for the use of the individual to whom it is
addressed. It may contain  privileged or confidential information and
should not be circulated or used for any purpose other than for what it is
intended. If you have received this message in error, please notify the
originator immediately. If you are not the intended recipient, you are
notified that you are strictly prohibited from using, copying, altering, or
disclosing the contents of this message. HSS accepts no responsibility for
loss or damage arising from the use of the information transmitted by this
email including damage from virus."


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



From exim@www1.ietf.org  Tue Jul  1 13:48:33 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 NAA08597
	for <sigtran-archive@odin.ietf.org>; Tue, 1 Jul 2003 13:48:33 -0400 (EDT)
Received: (from exim@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h5RIdCf15991
	for sigtran-archive@odin.ietf.org; Fri, 27 Jun 2003 14:39:12 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Vy7s-00049c-KT; Fri, 27 Jun 2003 14:39:12 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19VwSZ-0001Bp-EI
	for sigtran@optimus.ietf.org; Fri, 27 Jun 2003 12:52:27 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA07225
	for <sigtran@ietf.org>; Fri, 27 Jun 2003 12:52:23 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19VwSX-0004dU-00
	for sigtran@ietf.org; Fri, 27 Jun 2003 12:52:25 -0400
Received: from web12802.mail.yahoo.com ([216.136.174.37])
	by ietf-mx with smtp (Exim 4.12)
	id 19VwSN-0004d8-00
	for sigtran@ietf.org; Fri, 27 Jun 2003 12:52:15 -0400
Message-ID: <20030627165139.56664.qmail@web12802.mail.yahoo.com>
Received: from [208.255.120.2] by web12802.mail.yahoo.com via HTTP; Fri, 27 Jun 2003 09:51:39 PDT
Date: Fri, 27 Jun 2003 09:51:39 -0700 (PDT)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: RE: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
To: sigtran@ietf.org
In-Reply-To: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CA2@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>

IMO it might be good to keep them for both of the
documents, just in case somebody want to use it.
Personally I don't see a benefit of removing CIC/SSN
based loadsharing.

    Tolga
--- "Javier Pastor-Balbas (ECE)"
<javier.pastor-balbas@ece.ericsson.se> wrote:
> Hi Anjali, 
> 
> Thanks for your comment.
> 
> You are right saying that all CIC and SSN references
> were removed from the draft in  M3UA IG v03. Anyhow,
> I don't see any damage in keeping this CIC and SSN
> based loadsharing as another option. In fact it
> seems to me that this change was a bug during the
> process of removing all the SSN and CIC reference
> for the RK. But in any case both documents should be
> aligned. 
> 
> What do you and others think? Do we remove CIC and
> SSN based loadsharing from both documents? or do we
> allow it?
> 
> Thanks // Javier.
> 
> 
> > -----Original Message-----
> > From: agurmukhani@hss.hns.com
> [mailto:agurmukhani@hss.hns.com]
> > Sent: jueves, 26 de junio de 2003 10:16
> > To: sigtran@ietf.org
> > Subject: [Sigtran] M3UA MIB Draft05: Loadsharing
> on CIC ,SSN
> > 
> > 
> > 
> > 
> > 
> > 
> > Hi....
> > 
> > I had a query related to the M3UA MIB draft 05.
> > Both m3uaSpLshMethod OBJECT-TYPE and 
> m3uaStreamLshMethod OBJECT-TYPE
> > mention CIC and SSN as basis of loadsharing
> > In my opinion, there is no point keeping CIC and
> SSN  as 
> > loadsharing basis
> > since this  would require  parsing  the data
> coming from the 
> > M3UA user.
> > Also references to these as loadsharing basis have
> been 
> > removed from RFC as
> > per Implementors guide 03.
> > So IMO these should be removed from the MIB also.
> > 
> > 
> > Thanks
> > Anjali
> > Hughes Software Systems
> > 
> > 
> > 
> > "DISCLAIMER: This message is proprietary to Hughes
> Software 
> > Systems Limited
> > (HSS) and is intended solely for the use of the
> individual to 
> > whom it is
> > addressed. It may contain  privileged or
> confidential information and
> > should not be circulated or used for any purpose
> other than 
> > for what it is
> > intended. If you have received this message in
> error, please 
> > notify the
> > originator immediately. If you are not the
> intended recipient, you are
> > notified that you are strictly prohibited from
> using, 
> > copying, altering, or
> > disclosing the contents of this message. HSS
> accepts no 
> > responsibility for
> > loss or damage arising from the use of the
> information 
> > transmitted by this
> > email including damage from virus."
> > 
> > 
> > _______________________________________________
> > 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!?
SBC Yahoo! DSL - Now only $29.95 per month!
http://sbc.yahoo.com

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



From exim@www1.ietf.org  Tue Jul  1 13:48:41 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 NAA08676
	for <sigtran-archive@odin.ietf.org>; Tue, 1 Jul 2003 13:48:35 -0400 (EDT)
Received: (from exim@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h5RIcUY14465
	for sigtran-archive@odin.ietf.org; Fri, 27 Jun 2003 14:38:30 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Vy77-0003kS-GT; Fri, 27 Jun 2003 14:38:25 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19VwG1-0000jC-Er
	for sigtran@optimus.ietf.org; Fri, 27 Jun 2003 12:39:29 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA06951
	for <sigtran@ietf.org>; Fri, 27 Jun 2003 12:39:10 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19VwFk-0004Y7-00
	for sigtran@ietf.org; Fri, 27 Jun 2003 12:39:12 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19VwFa-0004Xb-00
	for sigtran@ietf.org; Fri, 27 Jun 2003 12:39:02 -0400
Received: from [172.16.1.40] (helo=ccpu00251)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19VwDv-0006ye-6k; Fri, 27 Jun 2003 09:37:19 -0700
Reply-To: <umesh@ccpu.com>
From: "Umesh Vats" <umesh@ccpu.com>
To: "Javier Pastor-Balbas \(ECE\)" <javier.pastor-balbas@ece.ericsson.se>,
        <agurmukhani@hss.hns.com>, <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
Date: Fri, 27 Jun 2003 09:41:54 -0700
Message-ID: <ACEMIJONPBBPDLOKCABHGEPGCAAA.umesh@ccpu.com>
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 IMO, Build 9.0.2416 (9.0.2911.0)
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
In-Reply-To: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CA2@eestqnt105.es.eu.ericsson.se>
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 to believe that CIC and SSN based loadsharing shouldn't be part of
M3UA. It requires parsing of user messages.

thanks
Umesh

-----Original Message-----
From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf Of
Javier Pastor-Balbas (ECE)
Sent: Thursday, June 26, 2003 11:45 PM
To: 'agurmukhani@hss.hns.com'; sigtran@ietf.org
Subject: RE: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN


Hi Anjali,

Thanks for your comment.

You are right saying that all CIC and SSN references were removed from the
draft in  M3UA IG v03. Anyhow, I don't see any damage in keeping this CIC
and SSN based loadsharing as another option. In fact it seems to me that
this change was a bug during the process of removing all the SSN and CIC
reference for the RK. But in any case both documents should be aligned.

What do you and others think? Do we remove CIC and SSN based loadsharing
from both documents? or do we allow it?

Thanks // Javier.


> -----Original Message-----
> From: agurmukhani@hss.hns.com [mailto:agurmukhani@hss.hns.com]
> Sent: jueves, 26 de junio de 2003 10:16
> To: sigtran@ietf.org
> Subject: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
>
>
>
>
>
>
> Hi....
>
> I had a query related to the M3UA MIB draft 05.
> Both m3uaSpLshMethod OBJECT-TYPE and  m3uaStreamLshMethod OBJECT-TYPE
> mention CIC and SSN as basis of loadsharing
> In my opinion, there is no point keeping CIC and SSN  as
> loadsharing basis
> since this  would require  parsing  the data coming from the
> M3UA user.
> Also references to these as loadsharing basis have been
> removed from RFC as
> per Implementors guide 03.
> So IMO these should be removed from the MIB also.
>
>
> Thanks
> Anjali
> Hughes Software Systems
>
>
>
> "DISCLAIMER: This message is proprietary to Hughes Software
> Systems Limited
> (HSS) and is intended solely for the use of the individual to
> whom it is
> addressed. It may contain  privileged or confidential information and
> should not be circulated or used for any purpose other than
> for what it is
> intended. If you have received this message in error, please
> notify the
> originator immediately. If you are not the intended recipient, you are
> notified that you are strictly prohibited from using,
> copying, altering, or
> disclosing the contents of this message. HSS accepts no
> responsibility for
> loss or damage arising from the use of the information
> transmitted by this
> email including damage from virus."
>
>
> _______________________________________________
> 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

---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.493 / Virus Database: 292 - Release Date: 6/25/2003

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.493 / Virus Database: 292 - Release Date: 6/25/2003


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



From exim@www1.ietf.org  Tue Jul  1 13:49:11 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 NAA09072
	for <sigtran-archive@odin.ietf.org>; Tue, 1 Jul 2003 13:49:05 -0400 (EDT)
Received: (from exim@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id h5RIS9I30226
	for sigtran-archive@odin.ietf.org; Fri, 27 Jun 2003 14:28:09 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19VxxB-0007qv-5a; Fri, 27 Jun 2003 14:28:09 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19VuaJ-0002P9-Ah
	for sigtran@optimus.ietf.org; Fri, 27 Jun 2003 10:53:54 -0400
Received: from gw.openss7.com (IDENT:ytIA7E9hkwRQ/mmO/OmRgrAixnMrlKLh@[142.179.199.224])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA20019
	for <sigtran@ietf.org>; Fri, 27 Jun 2003 03:51:41 -0400 (EDT)
Received: from ns.pigworks.openss7.net (IDENT:cGjsfECgTFP5j40uBOCxgMxzmeKcKnJK@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h5R7j7X10304;
	Fri, 27 Jun 2003 01:45:07 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h5R7j6j17676;
	Fri, 27 Jun 2003 01:45:06 -0600
Date: Fri, 27 Jun 2003 01:45:05 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Ong, Lyndon" <LyOng@Ciena.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Re:n+k issue
Message-ID: <20030627014505.A14028@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Ong, Lyndon" <LyOng@Ciena.com>, sigtran@ietf.org
References: <829F074A10F98943A218DC363D09C92A250208@w2ksjexg06.oni.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: <829F074A10F98943A218DC363D09C92A250208@w2ksjexg06.oni.com>; from LyOng@Ciena.com on Thu, Jun 26, 2003 at 06:16:18PM -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>

Lyndon,

I don't think that I have misinterpreted anything as I authored half the text
in question and helped with the other half.  Consider the history of consensus
on this issue:

This began with the thread that I initiated entitled

  "M3UA-10: Example 5.1.4" that extended from Jan 2, 2002 to Jan 7, 2002.

This was a LC issue on M3UA.  The example showed in the "n+k" case the SG
sending NTFY(AS-Active) after this first (yes, the first) ASP activated for
the AS even thought "n=2" in the example.  I asked why the SG did not withhold
NTFY(AS-Active) until after "n=2" ASPs were active for the AS.

Greg proposed the text:

  An SGP or IPSP, upon reception of an ASP Active message for the first ASP in
  a Load-share 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)."

The MAY was intentional to permit the SG to activate on the first ASP.

I proposed that the SG then withhold the NTFY(AS-Active) until after it
directs traffic if the SG follows the MAY and waits for sufficient ASPs.  (The
point of the original issue.)

It was agreed on this thread that some implementations would want to wait for
'n' and some would want to activate immediately, relying on the congestion and
flow controls of the ASP and MTP-User parts to handle any potential traffic
overloads.

Consensus was reached in the LC process and the text was adopted in M3UA-11.

On the thread:

  "n+k redundancy + NTFY  changes" from Jan 30, 2002 to Feb 1, 2002.

Tolga (surprising now) asks John Loughney to echo the changes of these two
passages into SUA and the same procedures are incorporated into SUA by
consensus.

On the thread:

  "m3ua-implementors-guidev00,  OPC list + missing issues" from March 31, 2002
  through April 10, 2002.

The discussion of "n+k" is brought up again by Tolga.  In this thread, Vikas,
Ken Morneault, and I agree that the text is clear.  Vikas and I also defend
the text.  Tolga is the only dissenter.  Umesh asks for clarification in the
IG, but it is agreed that the text in the RFC is clear.

On the thread:

  "m3ua: IG : n+k redundancy" from Jun 21, 2002 to Jun 26, 2002
  and
  "m2ua: IG : n+k redundancy - AS state machine"

Tolga brings up the same arguments again.  Javier's conclusion from previous
discussion initially was that no change was necessary.  Javier and Tolga then
go on to have a discussion between the two of them.  (I would have disagreed
in the middle of this, as consensus was already reached twice, but I was in a
U-haul between Texas and Alberta for the remainder of their exchange).  No one
else commented on the matter.

On the next thread:

  "I-D ACTION:draft-ietf-sigtran-m3ua-implementors-guide-02.txt"

I initiated a number of issued on the new release of the IG including the
section 3.15 which was not included by consensus, but only on the basis of
this brief discussion between Javier and Tolga.   I pointed out to Ken the
previous consensus points on this matter and he agreed to review the mail
threads.  Ken did not comment further.

Next on the thread:

  "M3UA Notify message" from Feb 12, 2003 to Feb 26, 2003
  (rejoined recently and extending from Jun 18, 2003 to Jun 26, 2003)

Srini asks some questions about Notify messages and n+k comes up again.  Srini
agrees that active ASPs < n corresponds to the AS-ACTIVE state as in the RFC
and that the ASP should use congestion and flow control procedures to limit
traffic if it is presented with too much traffic.  Bijoy agrees as well.
Srivats agrees as well with not stopping traffic #ASPs < n.  Samuel also
agrees with not stopping traffic #ASPs < n.  Javier agrees to the MAY.

The thread is rejoined recently.  The only dissenting voice on the thread is
Tolga (and most recently Javier).

Now, it appears to me that Tolga has brought this up on 4 or 5 separate
occasions, on each occasion (except when I was in the U-Haul) consensus was
that the existing RFC text stands and that the SG can activate with the first
ASP active in the AS regardless of 'n' or MAY wait for 'n' and SHOULD withhold
then NTFY(AS-Active) is reiterated.

10 separate individuals have come to consensus 4 times on this issue,
including last call, and the result was always that the RFC stands.  This text
also appears by consensus (as proposed by Tolga!) in SUA.  In subsequent
discussions Tolga was the only dissenting voice.  It appears Tolga wants 'n'
to be required (only) and never preferred.

The existing text allows it to be preferred or required depending on SG
implementation or configuration.  Although Tolga states that "1+k" required is
the same as preferred, the Notify(Insufficient ASPs Active in AS) behavior is
different for "1+k" required than it is for "n+k" preferred.

10 separate individual disagree with Tolga's proposals over the cited threads
(he disagrees with himself on a couple of occasions too).

Section 3.15 was NOT included on the consensus of the WG and stands against
the consensus of the WG.  There is no rough consensus to adopt the proposal as
evidenced by the archive.

As the WG has consistently spoken against these changes over the last 18
months, please do the right thing and strike Section 3.15 from the IG.

Thank you.

--brian


Lyndon Ong wrote:                                                                   (Thu, 26 Jun 2003 18:16:18)
> Hi Folks,
> 
> At the risk of blowing up the issue (although I think we
> are there already), I would like to make the following
> points:
> 
> -- the issue of n+k is in fact confusing in the RFC text
>   - n+k as defined in the introduction is subject to multiple
>     interpretations
>   - n+k has the following procedure defined in 4.3.4.3 -
>   "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."
>     This seems to imply that traffic should only be resumed after
>     n ASPs are active, even if traffic is sent from the SG the ASPs
>     should not reply.
>   - the examples in 5.1.4 and 5.2.3 also appear to reflect this usage.
>   - n+k is not clearly reflected in the current state diagrams
> 
> Given that this appears to be a valid issue with the RFC and
> there seems to be a greater weight of opinion on the side of
> the proposed text I believe that the rough consensus exists
> supporting the proposed text in the M3UA implementor's guide draft.
> 
> However, I understand the concern on the part of some folks
> that have interpreted n+k differently.  As a result I think
> we also should add clarifying text to the n+k definition in
> 1.4.4.1 stating that in this definition, traffic is not sent
> until "n" ASPs are active, and that the "k" ASPs may be either 
> active at the same time as "n" or kept inactive until needed
> due to a failed or unavailable ASP.  Hopefully that will 
> cover other situations acceptably.
> 
> Cheers,
> 
> L. Ong

-- 
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 exim@www1.ietf.org  Wed Jul  2 06:58:49 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA03605
	for <sigtran-archive@odin.ietf.org>; Wed, 2 Jul 2003 06:58:49 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XfJd-0003jA-OJ
	for sigtran-archive@odin.ietf.org; Wed, 02 Jul 2003 06:58:21 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h62AwLUf014329
	for sigtran-archive@odin.ietf.org; Wed, 2 Jul 2003 06:58:21 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XfJN-0003RJ-U5; Wed, 02 Jul 2003 06:58:05 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XfIk-0003GR-H3
	for sigtran@optimus.ietf.org; Wed, 02 Jul 2003 06:57:26 -0400
Received: from CNRI.Reston.VA.US (localhost [127.0.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA03221;
	Wed, 2 Jul 2003 06:57:23 -0400 (EDT)
Message-Id: <200307021057.GAA03221@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: Wed, 02 Jul 2003 06:57:23 -0400
Subject: [Sigtran] I-D ACTION:draft-ietf-sigtran-security-03.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-03.txt
	Pages		: 15
	Date		: 2003-7-1
	
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-03.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-03.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-03.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:	<2003-7-1135014.I-D@ietf.org>

ENCODING mime
FILE /internet-drafts/draft-ietf-sigtran-security-03.txt

--OtherAccess
Content-Type: Message/External-body;
	name="draft-ietf-sigtran-security-03.txt";
	site="ftp.ietf.org";
	access-type="anon-ftp";
	directory="internet-drafts"

Content-Type: text/plain
Content-ID:	<2003-7-1135014.I-D@ietf.org>

--OtherAccess--

--NextPart--



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



From exim@www1.ietf.org  Wed Jul  2 06:58:51 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA03623
	for <sigtran-archive@odin.ietf.org>; Wed, 2 Jul 2003 06:58:50 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XfJf-0003jq-IE
	for sigtran-archive@odin.ietf.org; Wed, 02 Jul 2003 06:58:23 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h62AwNvw014361
	for sigtran-archive@odin.ietf.org; Wed, 2 Jul 2003 06:58:23 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XfJO-0003SF-Ga; Wed, 02 Jul 2003 06:58:06 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XfIr-0003Gl-3R
	for sigtran@optimus.ietf.org; Wed, 02 Jul 2003 06:57:33 -0400
Received: from CNRI.Reston.VA.US (localhost [127.0.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA03244;
	Wed, 2 Jul 2003 06:57:29 -0400 (EDT)
Message-Id: <200307021057.GAA03244@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: Wed, 02 Jul 2003 06:57:29 -0400
Subject: [Sigtran] I-D ACTION:draft-ietf-sigtran-m2pa-09.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		: SS7 MTP2-User Peer-to-Peer Adaptation Layer
	Author(s)	: T. George et al.
	Filename	: draft-ietf-sigtran-m2pa-09.txt
	Pages		: 51
	Date		: 2003-7-1
	
This Internet Draft defines a protocol supporting the transport of
Signaling System Number 7 (SS7) Message Transfer Part (MTP) Level 3
signaling messages over Internet Protocol (IP) using the services of
the Stream Control Transmission Protocol (SCTP).  This protocol would
be used between SS7 Signaling Points using the MTP Level 3
protocol. The SS7 Signaling Points may also use standard SS7 links
using the SS7 MTP Level 2 to provide transport of MTP Level 3
signaling messages. The protocol operates in a manner similar to MTP
Level 2 so as to provide peer-to-peer communication between SS7
endpoints.

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-sigtran-m2pa-09.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-m2pa-09.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-m2pa-09.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:	<2003-7-1135026.I-D@ietf.org>

ENCODING mime
FILE /internet-drafts/draft-ietf-sigtran-m2pa-09.txt

--OtherAccess
Content-Type: Message/External-body;
	name="draft-ietf-sigtran-m2pa-09.txt";
	site="ftp.ietf.org";
	access-type="anon-ftp";
	directory="internet-drafts"

Content-Type: text/plain
Content-ID:	<2003-7-1135026.I-D@ietf.org>

--OtherAccess--

--NextPart--



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



From exim@www1.ietf.org  Wed Jul  2 11:47:12 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA27862
	for <sigtran-archive@odin.ietf.org>; Wed, 2 Jul 2003 11:47:12 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Xjod-0000I7-3n
	for sigtran-archive@odin.ietf.org; Wed, 02 Jul 2003 11:46:45 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h62FkdQY001117
	for sigtran-archive@odin.ietf.org; Wed, 2 Jul 2003 11:46:39 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Xjo6-0008LO-DL; Wed, 02 Jul 2003 11:46:06 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XQVw-00029T-Tx
	for sigtran@optimus.ietf.org; Tue, 01 Jul 2003 15:10:15 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA09830
	for <sigtran@ietf.org>; Tue, 1 Jul 2003 14:18:40 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19XPiE-0006oP-00
	for sigtran@ietf.org; Tue, 01 Jul 2003 14:18:42 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[rjn7IrmWpb8OgJ+7zk/Jn14xWrFyLpe0])
	by ietf-mx with esmtp (Exim 4.12)
	id 19XPiD-0006oG-00
	for sigtran@ietf.org; Tue, 01 Jul 2003 14:18:41 -0400
Received: from ns.pigworks.openss7.net (IDENT:BH88RvlE4QdVo1ZN7up0TSlQp4uj81dF@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h61IHaX06868;
	Tue, 1 Jul 2003 12:17:36 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h61IHaK07066;
	Tue, 1 Jul 2003 12:17:36 -0600
Date: Tue, 1 Jul 2003 12:17:36 -0600
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:n+k issue
Message-ID: <20030701121736.A6925@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	sigtran@ietf.org
References: <20030701033851.A31866@openss7.org> <Pine.SOL.4.10.10307010509100.24913-100000@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.4.10.10307010509100.24913-100000@sun2594.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Tue, Jul 01, 2003 at 06:09:34AM -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>

Samuel,

Perhaps you could answer my questions below first.  Please be
specific.

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Tue, 01 Jul 2003 06:09:34)
> 
> Brian,
> 
>   Can you bit elaborate your n=256 example (ANSI-SLS) with real-time 
>   scenario?(please write to the point)
> 
> --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, 1 Jul 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Samuel Dur D. Jeyaseelan wrote:                      (Mon, 30 Jun 2003 09:44:01)
> > > 
> > > Brian,
> > > 
> > > On Sat, 28 Jun 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Samuel,
> > > > 
> > > > I should point out that the proposal to require that the SG wait for 'n' to
> > > > start traffic does not have much merit.
> > > > 
> > > > Example: n=256, k=1
> > > > 
> > > > 2 ASPs fail, permanently and need to be manually replaced.
> > > > 
> > > > A temporary network failure causes SCTP associations to be lost to the
> > > > remaining 255 ASPs.  They move to the ASP down state and immediately
> > > > re-establish SCTP associations and move to the ASP-Active state.  Because the
> > > > number of ASPs is < n, the SG will not restart traffic for some time.
> > > 
> > > Your example leads to the following figure.
> > > 
> > >          SG1                            AS1
> > >      
> > >  SS7 MSG +------+----------------------- ASP (1)
> > > ---------| SGP1 |----------------------- ASP (2)
> > >          |      |----------------------- ....
> > >          +------+----------------------- ASP (256)
> > >             |
> > >             | 
> > >             +---------------------------ASP(k)
> > > 
> > > 1. First of all craft will not configure 1 ASP as backup for 256 Active
> > >    ASPs, if he does, it means to be a poor network configuration.
> > 
> > Well, first of all, craft does not normally choose the number of ASPs used for
> > backup, engineering does.  Second, k=1 is more than adequate for n=256 where
> > the ASPs are reasonably reliable and have a short replacement time.
> > 
> > Regardless, I followed with the same example with n=2 and k=1.
> > 
> > > 
> > > 2. Consider ASP as an MGC, your example leads to 256 MGCs and they are 
> > >    controlled by a single SGP.A good network will have multiple SGPs to
> > >    share the POTENTIAL traffic across ASPs.
> > 
> > You seem to misunderstand the concept of redundancy in M3UA: there are not
> > multiple SGPs or ASPs because of traffic load.  It is easy enough to handle
> > the traffic of a small to medium sized country through a single processor.
> > 
> > Also, you seem to misunderstand that an SGP and an ASP are not processors.
> > They are processes.  One host processor could have many (even 256) ASPs or
> > SGPs (or both for that matter).
> > 
> > > 
> > > 3. BASIC INTENTION OF 'n+k' is,just to make sure that incoming messages
> > >    ARE EFFICIENTLY TRANSMITTED TO DESTINATION ASPs WITH THE HELP OF
> > >    MINIMUM NUMBER OF ASSOCIATIONS.
> > 
> > No.  The intention of 'n+k' is to provide one of the more recently accepted
> > models for reliability using redundancy (n = m+1, k=0 being the typical system
> > deployed in the field today, m usually being greater than 2).  Up until
> > recently, only 2m (or m+m) systems were considered carrier class.  Some
> > systems were 4m (with 3 out of 4 vote).
> > 
> > You should take some time to look at redundancy and reliability requirements
> > for carrier grade systems for an understanding of how to calculate and provide
> > the reliability requirements expected by carriers.  There are good standards
> > written on the topic.  See the Telcordia NEBS and reliability specs, for
> > example.  If you have an RBOC RFP handy, take a look at the hardware and
> > software reliability requirements sections or talk to the engineer in your
> > company responsible for answering those sections.
> > 
> > > 
> > >    Say for example
> > > 
> > >    Inefficient configuration -just an example
> > >    ------------------------------------------
> > >   
> > >             SLS         DEST
> > >             =================
> > >              0           ASP1
> > >              1           ASP1
> > >              2           ASP1
> > >              3           ASP1
> > >              4           ASP1
> > >              5           ASP1
> > >              6           ASP1
> > >              .            .
> > >              .            .
> > >              15          ASP1
> > >              
> > >              
> > >                SG                            AS1
> > > sls(0,1..15 )+----+  sls(0,1..15)         +------+
> > > -------------|SGP1|-----------------------| ASP1 |
> > >              |    |              ASSOC-1  +------+
> > >              +----+                     
> > > 
> > >                       Fig - 1
> > > 
> > > In this figure, all incoming messages are transmitted over ASSOC-1 towards
> > > ASP1.Here there are two potential problems 
> > > 
> > >        1. Total incoming traffic is dependent on a single ASP
> > >        2. ASP1 is always overloaded, messages are processed one by one
> > >           from the transmission buffer which ultimately introduces time
> > >           delay(example last arrived msg will be processed at the end).
> > 
> > The above was not in my examples.  My examples had n > 1.
> > 
> > > 
> > > 
> > > Efficient configuration
> > > -----------------------  
> > > 
> > >              SLS         DEST
> > >             =================
> > >              0           ASP1
> > >              1           ASP1
> > >              2           ASP1
> > >              3           ASP1
> > >              4           ASP1
> > >              5           ASP1
> > >              6           ASP1
> > >              7           ASP1
> > >              8           ASP2
> > >              9           ASP2
> > >              .            .
> > >              .            .
> > >              15          ASP2
> > > 
> > >                
> > >                        
> > >                        
> > >               SG                             AS1
> > > sls(0,1..15 )+----+   sls(0,1,..7)ASSOC-1 +------+
> > > -------------|SGP1|-----------------------| ASP1 |
> > >              |    |--------+              +------+
> > >              +----+        | sls(8,9..15) +------+
> > >                            +--------------| ASP2 |
> > >                                  ASSOC-2  +------+
> > >                         
> > >                    fig - 2
> > > 
> > > In this example incoming traffic is distributed between ASP1 and ASP2.
> > > 
> > > IN ORDER TO HANDLE THE HEAVY INCOMING TRAFFIC, SGP1 NEEDS MINIMUM OF
> > > 2 ACTIVE ASPs BY SETTING n=2.
> > 
> > Do they not comply to SS7 ISUP/SCCP flow and congestion controls?  Are they
> > running at full capacity in this arrangement?  Are these your ASPs?  Do you
> > design your systems this way?
> > 
> > Take a look at the reliability engineering.  If either of these ASPs is
> > handling its maximum rated capacity, it will have to be reliable to .99999
> > This is because, if one ASP fails, 50% of the system has failed.  A 50% system
> > failure is deemed a "system" failure rather than a "component" failure in
> > reliability engineering for carrier grade systems.  Considering proscribed
> > replacement times on the ASP, this would be difficult to achieve unless both
> > ASPs were hardware fault tolerant platforms.
> > 
> > Nevertheless, the RFC defines ASPs as a process, not a processor.  There is
> > nothing restricting the two ASPs in your diagram from being processes on the
> > same processor.  In that case, when ASP2 is down, ASP1 has double the
> > processing capacity as it had before.
> > 
> > You certainly cannot make statements as you have made here without
> > presupposing an implementation architecture outside the scope of the standard.
> > 
> > Besides, the RFC permits you to configure your SGP (and why do you only have
> > one of them?) so that it waits for 'n' ASPs to become active.  You MAY do
> > this.  Just don't require everybody else to follow your approach.  (Actually,
> > if the above is your approach, don't require anyone else to follow it.)
> > 
> > Several other points: ANSI T1.111.4/2000 SLS values are 8-bits.  That means
> > there are 256 SLS values (gee, I had 256 ASPs in my example).
> > 
> > For another thing, sequenced traffic flows are not only based on SLS.  In an
> > ITU-T network, each remote signalling point has 16 independent traffic flows
> > for each AS.  The traffic flows from each remote signalling point is
> > independent from the traffic flows of other remote signalling points.
> > Therefore, if there are 400 remote signalling points, there will be 6400
> > independent traffic flows that the SG can use for distribution of AS traffic
> > for one AS over as many as 6400 ASPs without risking mis-sequencing of
> > messages.  For an ANSI T1.111.4/2000 network using 8-bit SLS values with 400
> > remote signalling points, there would be 102400 independent traffic flows to
> > the AS that the SG can use to distribute traffic to one AS over as many as
> > 102400 ASPs without risking message mis-sequencing.
> > 
> > > 
> > > 
> > > Your example with n=256 is completely a wrong interpretaion of 'n+k'
> > > redundancy case.
> > 
> > Quite the opposite, you clearly misunderstand redundancy and reliability in
> > carrier grade systems and obviously have little experience with them.
> > 
> > Here's a little primer for you:
> > 
> > Carrier grade systems were originally designed as hot-standby, fully redundant
> > systems.  For every processor actively handling traffic, there was a hot
> > standby processor also actively handling the same traffic.  This typically
> > required special bus arrangements whereby the I/O bus of one processor was
> > actively controlling peripheral devices, while the hot standby processor was
> > issuing I/O bus output that was not terminated to the I/O bus.  The bus also
> > normally performed a comparison of the outputs from the active side and the
> > outputs from the inactive side.  Peripheral components, such as switch fabric
> > were also fully redundant (duplicated).  Switching fabrics are separated into
> > an A-plane and a B-plane in much the same fashion as the SPCS front-end I/O
> > bus.
> > 
> > Note that active/standby in M3UA does not address the above model.  2+0
> > broadcast does.
> > 
> > The problem with such a systems was what we used to call the "kill your
> > sister" syndrome.  A failure on the inactive side only detectable to the I/O
> > bus could cause a switch from the active to inactive side, resulting in the
> > switch running on the erroneous inactive side.  We called it the "kill your
> > sister" syndrome because it is not so surprising that the psychotic sister
> > (erred inactive side) kills the sane sister (correct active side).
> > 
> > Primarily to avoid the "kill your sister" syndrome, some switches were
> > designed with 3 or 4 processors on an cross-checking I/O bus pair.  In that
> > case, it would take a 2 out of 3 or 3 out of 4 vote to shift from the active
> > processor.  These systems were far less susceptible to inactive plane errors
> > that were only detected at the I/O bus.
> > 
> > Note that the only M3UA model that address the above is the 3+0 or 4+0
> > broadcast model.
> > 
> > When the use of general purpose computing platforms for telecommunications
> > systems started to become popular, there were several models that emerged.
> > 
> > First was taking two big iron UNIX hosts and running them in active/standby.
> > In this scenario, one UNIX host would be actively handling traffic and the
> > other would be warm standby.  By "warm" standby, it means that the device was
> > powered and was synchronizing state on a checkpoint interval or monitoring the
> > health of the active hosts with heartbeats.  Shared memory, shared DASD or
> > message passing are the most common mechanisms of redundancy.  These systems
> > are considered "highly available" systems rather than the "fault tolerant"
> > systems used in telecommunications above.  An increased level of reliability
> > is often achieved by using hardware fault tolerant multiple processor hosts.
> > 
> > In M3UA, the active/standby model addresses these systems.  Although M3UA
> > provides for a '1+k' model, k is invariably 1.
> > 
> > The major difficulties with these systems are: a) switchover latency between
> > systems; b) loss of state during switchover; and c) scalability.  Switchover
> > latency is normally improved by decreasing heartbeat intervals.  State
> > synchronization is normally improved by decreasing the process checkpoint
> > interval.  These reductions typically increase cost and they are normally
> > chosen to minimize a cost/performance/reliability metric.  Scalability was
> > typically performed by installing larger more powerful hosts with increased
> > demand.  This was not good, because larger processors tended to have poorer
> > cost/performance ratios that smaller processors.
> > 
> > Later it was determined that the cost of a complete redundant host that was
> > wasting CPU cycles while inactive could be avoided while addressing the
> > deficiencies of the paired host approach above.  To increase state
> > synchronization, state could be spread over a number of hosts.  If there were
> > 'm' hosts, the failure of one host could at worst affect a proportion, 1/m, of
> > the state.  Therefore state synchronization would improve as m increased.  To
> > distribute the state, traffic load would be shared over the m hosts according
> > to state.  This also served the needs of scalability and provided for closer
> > to linear cost scalability, because the number of hosts, m, could be increased
> > while holding the cost/performance ratio of the hosts constant.  The costs of
> > redundancy could also be reduced over the fully redundant systems described
> > above, because redundancy could be achieved using 'm+1' processors, where 'm'
> > processors were required to handle the maximum engineered load.  For systems
> > as small as 'm=2' hosts with equivalent reliability to that normally
> > experienced in the '1+1' active/standby scenario above, system reliability
> > figures could be maintained with less redundancy cost.  It is important to
> > note, that in these systems, all 'm+1' hosts are active and handling only a
> > proportion m/(m+1) of their rated capacity.  It is very seldom that the rated
> > capacity was more that 2/3 of the maximum short-term capacity.
> > 
> > These systems do not solve the heartbeat interval problem of the
> > active/standby paired hosts system above.  Typically what these systems use is
> > more sophisticated watchdog approaches rather than passing heartbeat messages
> > between processes on the active and standby hosts.  Memory sharing, DASD
> > sharing, message passing, reliable broadcast, voting, and a plethora of
> > reliable distributed systems techniques have been developed that are useful in
> > these systems.
> > 
> > Adoption of these latter systems changed the requirement from "duplicated"
> > systems to systems with no "single point of failure".
> > 
> > Now, the 'n+k' load-sharing and broadcast model was intended to address these
> > systems.  Where 'n == m' in the description above.  That is why we agreed that
> > n is the maximum engineered load.  Because that is what 'm' is.  The k is the
> > 1 in the description above, generalized.
> > 
> > First off, you must understand that it is not the intention of the model to
> > have 'n' ASPs active and 'k' inactive unless 'n=1, k=1'.  This is described in
> > section 1.4.4.1:
> > 
> >   A "1+1" active/standby redundancy is a subset of this model.
> > 
> > Also no redundancy is possible:
> > 
> >   A simplex "1+0" model is also supported as a subset, with no ASP redundancy.
> > 
> > (The term "simplex" come from the telephone switches described above where the
> > inactive side is taken off-line.  Such a situation is described as running the
> > switch "simplex".)
> > 
> > The 'n+k' model is to address the 'm+1' distributed systems described last.
> > Rereading section 1.4.4.1:
> > 
> >   The failover model supports an "n+k" redundancy model, where "n" APSs is the
> >   minimum number of redundant [sic] ASPs required to handle traffic and "k"
> >   ASPs are available to take over for a failed or unavailable ASP.
> > 
> > Note that the state of the "k" ASPs with respect to an AS is not stated.  In
> > the 'm+1' redundant system described above, the "k" ASPs are active for the
> > AS.  "k" ASPs active in load sharing mode are "available to take over for a
> > failed or unavailable ASP".
> > 
> > It is typical to run the "k" ASPs active (and I know you had some questions
> > about activating the "k" ASPs, but the point is that they are normally already
> > active), and handling their own share of traffic.  In this case, each ASPs
> > handles n/(n+k) of their engineered capacity.  Were you actually considering
> > deploying such a atypical model where the 'k' ASPs are inactive?
> > 
> > "k" ASPs can be taken out of service, upgraded, and brought back into service
> > without affecting the maximum engineering traffic handling capacity of the
> > system (in absence of failures during the upgrade).  This assists in graceful
> > software/hardware upgrades on live systems.  (Keeping them inactive would
> > serve no purpose and would, decrease maximum traffic handling capability,
> > increase loss of state on fail-over, and increase fail-over latency: all good 
> > reasons for not deactivating the 'k' ASPs).
> > 
> > Now, reliability engineering for carrier class systems takes the view that if
> > 50% of the load bearing components in a system fail (stop operating in any
> > way), then the entire system has failed.  If traffic is stopped above n/2
> > processes, the system will have failed prematurely.
> > 
> > A requirement to withhold traffic < n will always suffer a prolonged system
> > failure, because only n/2 are required to make the system available.
> > Therefore, on startup the reliability of the system is degraded by the
> > negative exponent n/2, regardless of whether n=256 or n=2; however, increasing
> > 'n' increases the damage done to reliability by waiting for 'n'.
> > 
> > A requirement for the SG to wait for 'n' is unacceptable from the standpoint
> > of the redundancy model that 'n+k' was meant to address, as well as other
> > possible models that it can accommodate.
> > 
> > > 
> > > 
> > > > 
> > > > Consider the same example with: n=2, k=1, two ASPs failed.
> > > > 
> > > > Any system that introduces hysteresis to aggravate and perpetuate an outage in
> > > > this fashion is not very useful.
> > > > 
> > > > In the examples, what would have been a brief AS-PENDING situation has been
> > > > turned into a prolonged and complete system failure.  It is certainly not good
> > > > practice to design redundant systems in that fashion and recommending or
> > > > requiring such a design would be a bad thing.
> > > > 
> > > > To further the example, consider the multiple SG as STP situation.  One SG
> > > > cannot know the state of ASPs for the its mated SG and therefore cannot
> > > > distinguish between the restart scenario and the start scenario as far as the
> > > > AS state is concerned because the SG does not know the state of the AS itself:
> > > > the AS could be active with other SGs.
> > > > 
> > > > This is the same case for SGP.  Unless AS state is synchronized across SGPs,
> > > > one SGP cannot know the total state of the AS (only its partition).  There is
> > > > no requirement to synchronize AS state across SGP (on the SGP side, AS-state
> > > > is per-SGP), and there is certainly no requirement to synchronize AS-state
> > > > across SGs.
> > > > 
> > > > So, take the (n=2, k=0) example where there are two SGs.  Due to a network
> > > > partitioning at startup, one ASP can communicate only with one SG, and the
> > > > other ASP can communicate only with the other SG.  Because n=2, neither SG
> > > > will start traffic and the system suffers 100% failure until the network
> > > > partitioning ceases.
> > > > 
> > > > Consider the same example where there are two SGPs (instead of SGs) and the
> > > > two ASPs normally only form one association (e.g., ASP1-SGP1, ASP2-SGP2).
> > > > Note that there is no requirement in the RFC to form full-mesh associations.
> > > > Because the number of active ASPs with each SGP is < 2, the system will never
> > > > start.
> > > > 
> > > > These are not strange scenarios.  These situations are all permitted by the
> > > > RFC.  Yet, in each situation it is clear that one cannot recommend that the SG
> > > > wait.  The result would be a non-functional system.  There are many more
> > > > examples.
> > > > 
> > > > n+k is a model for redundancy (not for congestion and flow control).  A system
> > > > with an SG that waits for n is, in general, not a fully redundant system at
> > > > all, but is, rather, prone to these failure syndromes.
> > > > 
> > > > When you look at the situation and apply the experience of reliability
> > > > engineering, the recommendation is that the SG SHOULD start traffic when the
> > > > first ASP goes active.
> > > > 
> > > > Why wait for 'n' anyway?  Because the ASP cannot handle some level of traffic?
> > > > Any ASP that conforms to MTP-User standards, must have the standardized
> > > > MTP-User (ISUP/SCCP) congestion and flow controls and must have the SCTP
> > > > congestion controls and is therefore able to accept any amount of traffic
> > > > overload.
> > > > 
> > > > So there is no real reason for the SG to wait for 'n'.  Nevertheless, the RFC
> > > > allows this as an OPTION.  Requiring (MUST) or recommending it (SHOULD)
> > > > regardless of whether the number of active processors is rising or falling,
> > > > whether start, restart or failure, is unacceptable.
> > > > 
> > > > On the other side, at the ASPs, the ASPs have a requirement to share state
> > > > between them.  They must share circuit state and transaction state in all
> > > > configurations to meet carrier grade.  This is stated in the RFC.  It is
> > > > implicit that the ASPs are capable of knowing with which SGs, and which SGPs
> > > > for which they are active.  They have access to the total state of the AS that
> > > > they collectively represent, that the SG/SGP does not.  Therefore is it
> > > > reasonable for the ASPs to coordinate among themselves the number of active
> > > > ASPs and to ensure that there are sufficient ASPs active to handle the
> > > > expected load before starting traffic.  Because the ASPs need not respond to
> > > > user test traffic (UPT/SST), the ASPs are also in control of when remote SS7
> > > > MTP-User restart traffic toward them.
> > > > 
> > > > Therefore, the recommendation that the ASPs SHOULD coordinate among
> > > > themselves the number of active ASPs and not start traffic until a sufficient
> > > > number of them are active is sound.  This is not a MUST, permitting ASPs
> > > > that know that they are capable of handling any overload, to start traffic
> > > > immediately, or at some earlier point.
> > > > 
> > > > Back to the RFC, the RFC sets the normal case where the SG starts traffic on
> > > > the first active ASP but explicitly permits as an OPTION to wait until it
> > > > considers that there are sufficient ASPs active to handle the offered load.
> > > > The RFC also RECOMMENDS that the ASPs coordinate among themselves the number
> > > > of active ASPs so that there are sufficient ASPs to handle the expected load.
> > > > 
> > > > The provisions in the RFC are both sound, where the proposal and section 3.15
> > > > is clearly not.
> > > > 
> > > > Remember when looking at the AS state diagram in the RFC that this is the
> > > > SGP's view of the AS state and not the total state of the AS, that can only
> > > > known to the ASPs for all configurations.  'n' as defined has no meaning to
> > > > the state machine at the SGP because is not required to know, and often cannot
> > > > know, how many ASPs are active for the AS via other SGPs and SGs, or even via
> > > > direct connection to the SS7 network.
> > > > 
> > > > Some SGs might have special knowledge, or sophisticated algorithms that allow
> > > > it to determine whether a congestive failure will be the immediate result of
>  > > > starting or restarting traffic.  These SGs are permitted as an
> OPTION to wait.
> > > > 
> > > > The ASPs SHOULD have the knowledge of how many are required to handle the
> > > > expected load, because they must share circuit/transaction state and they must
> > > > have congestion and flow controls as required by the standards.
> > > > 
> > > > So again, the RFC is adequate in this regard.  The current proposal and
> > > > section 3.15 are not.
> > > RFC is inadequate so we need section 3.15 to explain 'n+k' case.
> > 
> > Teaching reliability engineering and the design of redundant systems is
> > outside the scope of the RFC and the IG.  People, such as yourself, are better
> > advised to learn the subject matter before attempting to implement or claim a
> > complete understanding of the protocol.
> > 
> > The current RFC provides the protocol mechanisms and basis for
> > interoperability for all of the common approaches to redundancy in carrier
> > class systems, and even provides a general model that can accommodate
> > non-traditional redundancy architectures.
> > 
> > When confused about how a redundancy architecture works, don't blame the M3UA
> > RFC, blame those responsible for describing redundant architectures to you.
> > 
> > In the appendix we stated:
> > 
> >    M3UA imposes no restrictions as to the exact layout of the network
> >    elements, the message distribution algorithms and the distribution of
> >    signalling processes.  Instead, it provides a framework and a set of
> >    messages that allow for a flexible and scalable signalling network
> >    architecture, aiming to provide reliability and performance.
> > 
> > Forcing SGs to wait for 'n' ASPs active before activating is contrary to this.
> > The existing "MAY" achieves this flexibility.
> > 
> > When reading 1.4.4.1, also remember this statement in the appendix, that used
> > to be beside 1.4.4.1.  The description of 'n+k' does not dictate the state of
> > the 'k' or the state of the AS with respect to 'n' to leave the redundancy
> > model flexible, and accommodate '1+1' and '1+0' in the same model.
> > 
> > Also, please don't presume to tell me to set 'n' to 1, as it is not a
> > arbitrary value, but the number 'm' in the 'm+1' distributed system in common
> > use today.  This number is determined as the minimum number of ASPs required
> > to handle the maximum engineered load, and not, as Tolga would have, the
> > minimum number of ASPs required to handle ANY load.  If you want your system
> > to behave as Tolga proposes, just have your SG exercise the "MAY choose not to
> > direct traffic" option of section 4.3.4.3 of the RFC.
> > 
> > Section 3.15 and Tolga's proposal should be struck because they attempt to
> > impose a 'n' active 'k' inactive and system cannot restart (3.15) or start
> > (Tolga) without 'n' view on the redundancy model, where it is clearly the
> > intent of the RFC to place no such restrictions on the model, but to
> > accommodate a range of models, that is already achieved by the RFC text.
> > 
> > --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 exim@www1.ietf.org  Wed Jul  2 13:18:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA02711
	for <sigtran-archive@odin.ietf.org>; Wed, 2 Jul 2003 13:18:33 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XlF6-0000oC-JY
	for sigtran-archive@odin.ietf.org; Wed, 02 Jul 2003 13:18:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h62HI41J003092
	for sigtran-archive@odin.ietf.org; Wed, 2 Jul 2003 13:18:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XlF3-0000mS-T3; Wed, 02 Jul 2003 13:18:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XlEv-0000lk-3r
	for sigtran@optimus.ietf.org; Wed, 02 Jul 2003 13:17:53 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA02691
	for <sigtran@ietf.org>; Wed, 2 Jul 2003 13:17:50 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19XlEt-0004T2-00
	for sigtran@ietf.org; Wed, 02 Jul 2003 13:17:51 -0400
Received: from sjmail.ciena.com ([12.7.169.25] helo=w2ksjexg01.ciena.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19XlEs-0004Sx-00
	for sigtran@ietf.org; Wed, 02 Jul 2003 13:17:50 -0400
Received: by w2ksjexg01.ciena.com with Internet Mail Service (5.5.2653.19)
	id <NS8BLK7D>; Wed, 2 Jul 2003 10:17:47 -0700
Message-ID: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com>
From: "Ong, Lyndon" <LyOng@Ciena.com>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Date: Wed, 2 Jul 2003 10:21:28 -0700 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Sigtran] WG Last Call on M2PA
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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 updated M2PA spec is out, let's do a WG Last Call on the
spec and see if we can approve it out of the WG.

Please have a look at http://ietf.org/internet-drafts/draft-ietf-sigtran-m2pa-09.txt
and bring any comments to the list, with the Subject of the email set to
"M2PA Last Call Comment".

Last Call will run to Wednesday, July 16th.

Cheers,

L. Ong

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



From exim@www1.ietf.org  Thu Jul  3 01:01:35 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA15463
	for <sigtran-archive@odin.ietf.org>; Thu, 3 Jul 2003 01:01:34 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XwDR-0005Tn-BR
	for sigtran-archive@odin.ietf.org; Thu, 03 Jul 2003 01:01:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h63515Rv021042
	for sigtran-archive@odin.ietf.org; Thu, 3 Jul 2003 01:01:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XwDO-0005Sc-Db; Thu, 03 Jul 2003 01:01:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XwCh-0005S2-NF
	for sigtran@optimus.ietf.org; Thu, 03 Jul 2003 01:00:19 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA15402
	for <sigtran@ietf.org>; Thu, 3 Jul 2003 01:00:18 -0400 (EDT)
From: agurmukhani@hss.hns.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19XwCe-0002nQ-00
	for sigtran@ietf.org; Thu, 03 Jul 2003 01:00:16 -0400
Received: from hindon.hss.co.in ([202.54.26.202])
	by ietf-mx with esmtp (Exim 4.12)
	id 19XwCc-0002nA-00
	for sigtran@ietf.org; Thu, 03 Jul 2003 01:00:14 -0400
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 h634ud528027
	for <sigtran@ietf.org>; Thu, 3 Jul 2003 10:26:40 +0530 (IST)
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h634uaO28011;
	Thu, 3 Jul 2003 10:26:36 +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 ESMTP id h6350EO28183;
	Thu, 3 Jul 2003 10:30:15 +0530 (IST)
To: javier.pastor-balbas@ece.ericsson.se
Cc: sigtran@ietf.org
Date: Thu, 3 Jul 2003 10:23:35 +0530
Message-ID: <OFEB796ECC.A92CCF1D-ON65256D58.00195CC7@hss.hns.com>
X-MIMETrack: Serialize by Router on Sandesh/HSS(Release 6.0|September 26, 2002) at 03/07/2003
 10:23:54 AM
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Subject: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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...


In my understanding and  as per the discussion and consensus on the mailing
list , everyone agreed that  CIC and SSN should not be used as loadsharing
basis
be it within streams in a particular SCTP association or for active
signalling processes.

So would these be removed from  the m3uaStreamLshMethod  and
m3uaSpLshMethod OBJECT-TYPE s in M3UA MIB draft.


Thanks and Regards
Anjali
www.hssworld.com
Hughes Software Systems.

"DISCLAIMER: This message is proprietary to Hughes Software Systems Limited
(HSS) and is intended solely for the use of the individual to whom it is
addressed. It may contain  privileged or confidential information and
should not be circulated or used for any purpose other than for what it is
intended. If you have received this message in error, please notify the
originator immediately. If you are not the intended recipient, you are
notified that you are strictly prohibited from using, copying, altering, or
disclosing the contents of this message. HSS accepts no responsibility for
loss or damage arising from the use of the information transmitted by this
email including damage from virus."


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



From exim@www1.ietf.org  Thu Jul  3 05:47:32 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA04151
	for <sigtran-archive@odin.ietf.org>; Thu, 3 Jul 2003 05:47:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Y0gC-0005zz-Db
	for sigtran-archive@odin.ietf.org; Thu, 03 Jul 2003 05:47:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h639l45m023049
	for sigtran-archive@odin.ietf.org; Thu, 3 Jul 2003 05:47:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Y0g9-0005z2-Nc; Thu, 03 Jul 2003 05:47:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Y0fJ-0005yO-Jn
	for sigtran@optimus.ietf.org; Thu, 03 Jul 2003 05:46:09 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA04106
	for <sigtran@ietf.org>; Thu, 3 Jul 2003 05:46:06 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19Y0fG-0004jL-00
	for sigtran@ietf.org; Thu, 03 Jul 2003 05:46:06 -0400
Received: from penguin-ext.wise.edt.ericsson.se ([193.180.251.47] helo=penguin.al.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19Y0fF-0004jI-00
	for sigtran@ietf.org; Thu, 03 Jul 2003 05:46:05 -0400
Received: from esealnt612.al.sw.ericsson.se ([153.88.254.118])
	by penguin.al.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h639k6GJ018187
	for <sigtran@ietf.org>; Thu, 3 Jul 2003 11:46:06 +0200 (MEST)
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 NW44W9KW; Thu, 3 Jul 2003 11:46:06 +0200
Received: by ESEALNT745.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <N94NLZL2>; Thu, 3 Jul 2003 11:45:54 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CC9@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Date: Thu, 3 Jul 2003 11:45:41 +0200 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Sigtran] Re: M3UA MIB Draft05: Loadsharing on CIC ,SSN
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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 Anjali, 

It seems that it makes sense to remove those from the standard MIB. All the people who are interested in having this object can create it within their private MIBs. We may keep SLS since it is a common practise in SS7 unless you have any objection.


Thanks for your responses // Javier.

PS: Samuel: What object do you refer when talking about DPC-based loadsharing within the current MIB? 


> -----Original Message-----
> From: agurmukhani@hss.hns.com [mailto:agurmukhani@hss.hns.com]
> Sent: jueves, 03 de julio de 2003 6:54
> To: J.Javier Pastor (EE/EEM)
> Cc: sigtran@ietf.org
> Subject: M3UA MIB Draft05: Loadsharing on CIC ,SSN
> 
> 
> 
> 
> 
> 
> Hi...
> 
> 
> In my understanding and  as per the discussion and consensus 
> on the mailing
> list , everyone agreed that  CIC and SSN should not be used 
> as loadsharing
> basis
> be it within streams in a particular SCTP association or for active
> signalling processes.
> 
> So would these be removed from  the m3uaStreamLshMethod  and
> m3uaSpLshMethod OBJECT-TYPE s in M3UA MIB draft.
> 
> 
> Thanks and Regards
> Anjali
> www.hssworld.com
> Hughes Software Systems.
> 
> "DISCLAIMER: This message is proprietary to Hughes Software 
> Systems Limited
> (HSS) and is intended solely for the use of the individual to 
> whom it is
> addressed. It may contain  privileged or confidential information and
> should not be circulated or used for any purpose other than 
> for what it is
> intended. If you have received this message in error, please 
> notify the
> originator immediately. If you are not the intended recipient, you are
> notified that you are strictly prohibited from using, 
> copying, altering, or
> disclosing the contents of this message. HSS accepts no 
> responsibility for
> loss or damage arising from the use of the information 
> transmitted by this
> email including damage from virus."
> 

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



From exim@www1.ietf.org  Thu Jul  3 06:34:53 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA04152
	for <sigtran-archive@odin.ietf.org>; Thu, 3 Jul 2003 05:47:32 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Y0gC-000601-Dm
	for sigtran-archive@odin.ietf.org; Thu, 03 Jul 2003 05:47:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h639l4P4023050
	for sigtran-archive@odin.ietf.org; Thu, 3 Jul 2003 05:47:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Y0gA-0005zI-6Z; Thu, 03 Jul 2003 05:47:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Y0fX-0005yT-TP
	for sigtran@optimus.ietf.org; Thu, 03 Jul 2003 05:46:23 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA04119
	for <sigtran@ietf.org>; Thu, 3 Jul 2003 05:46:20 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19Y0fU-0004jX-00
	for sigtran@ietf.org; Thu, 03 Jul 2003 05:46:20 -0400
Received: from penguin-ext.wise.edt.ericsson.se ([193.180.251.47] helo=penguin.al.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19Y0fT-0004jT-00
	for sigtran@ietf.org; Thu, 03 Jul 2003 05:46:19 -0400
Received: from esealnt612.al.sw.ericsson.se ([153.88.254.118])
	by penguin.al.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h639kKGJ018253
	for <sigtran@ietf.org>; Thu, 3 Jul 2003 11:46:20 +0200 (MEST)
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 NW44W9L6; Thu, 3 Jul 2003 11:46:20 +0200
Received: by ESEALNT745.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <N94NLZLP>; Thu, 3 Jul 2003 11:46:08 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CCA@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Date: Thu, 3 Jul 2003 11:45:57 +0200 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Sigtran] M3UA IG v04
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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 guys,

We tried to submit the next IG version before the deadline, but I got confused since I thought it ended at 12:00 instead of 9:00. So the server bounced it. I'll submit the doc again after the IETF meeting. In the meantime you can find it at:

http://standards.ericsson.net/javier/draft-ietf-sigtran-m3ua-implementors-guide-04.txt

We've tried to include all the different opinions in here. Although it seems that the case of waiting till n ASPs is the preferred one, we have included the new possibility of a smooth start for those of you who want to activate the AS when the first ASP gets to ACTIVE. At the end it will be the operator's decission how to proceed during the activation phase. Hopefully everyone agrees on this wording...

Cheers // Javier.



> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: martes, 01 de julio de 2003 20:18
> To: Samuel Dur D. Jeyaseelan
> Cc: sigtran@ietf.org
> Subject: Re: [Sigtran] Re:n+k issue
> 
> 
> Samuel,
> 
> Perhaps you could answer my questions below first.  Please be
> specific.
> 
> --brian
> 
> Samuel Dur D. Jeyaseelan wrote:                               
>                       (Tue, 01 Jul 2003 06:09:34)
> > 
> > Brian,
> > 
> >   Can you bit elaborate your n=256 example (ANSI-SLS) with 
> real-time 
> >   scenario?(please write to the point)
> > 
> > --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, 1 Jul 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                      
> (Mon, 30 Jun 2003 09:44:01)
> > > > 
> > > > Brian,
> > > > 
> > > > On Sat, 28 Jun 2003, Brian F. G. Bidulock wrote:
> > > > 
> > > > > Samuel,
> > > > > 
> > > > > I should point out that the proposal to require that 
> the SG wait for 'n' to
> > > > > start traffic does not have much merit.
> > > > > 
> > > > > Example: n=256, k=1
> > > > > 
> > > > > 2 ASPs fail, permanently and need to be manually replaced.
> > > > > 
> > > > > A temporary network failure causes SCTP associations 
> to be lost to the
> > > > > remaining 255 ASPs.  They move to the ASP down state 
> and immediately
> > > > > re-establish SCTP associations and move to the 
> ASP-Active state.  Because the
> > > > > number of ASPs is < n, the SG will not restart 
> traffic for some time.
> > > > 
> > > > Your example leads to the following figure.
> > > > 
> > > >          SG1                            AS1
> > > >      
> > > >  SS7 MSG +------+----------------------- ASP (1)
> > > > ---------| SGP1 |----------------------- ASP (2)
> > > >          |      |----------------------- ....
> > > >          +------+----------------------- ASP (256)
> > > >             |
> > > >             | 
> > > >             +---------------------------ASP(k)
> > > > 
> > > > 1. First of all craft will not configure 1 ASP as 
> backup for 256 Active
> > > >    ASPs, if he does, it means to be a poor network 
> configuration.
> > > 
> > > Well, first of all, craft does not normally choose the 
> number of ASPs used for
> > > backup, engineering does.  Second, k=1 is more than 
> adequate for n=256 where
> > > the ASPs are reasonably reliable and have a short 
> replacement time.
> > > 
> > > Regardless, I followed with the same example with n=2 and k=1.
> > > 
> > > > 
> > > > 2. Consider ASP as an MGC, your example leads to 256 
> MGCs and they are 
> > > >    controlled by a single SGP.A good network will have 
> multiple SGPs to
> > > >    share the POTENTIAL traffic across ASPs.
> > > 
> > > You seem to misunderstand the concept of redundancy in 
> M3UA: there are not
> > > multiple SGPs or ASPs because of traffic load.  It is 
> easy enough to handle
> > > the traffic of a small to medium sized country through a 
> single processor.
> > > 
> > > Also, you seem to misunderstand that an SGP and an ASP 
> are not processors.
> > > They are processes.  One host processor could have many 
> (even 256) ASPs or
> > > SGPs (or both for that matter).
> > > 
> > > > 
> > > > 3. BASIC INTENTION OF 'n+k' is,just to make sure that 
> incoming messages
> > > >    ARE EFFICIENTLY TRANSMITTED TO DESTINATION ASPs WITH 
> THE HELP OF
> > > >    MINIMUM NUMBER OF ASSOCIATIONS.
> > > 
> > > No.  The intention of 'n+k' is to provide one of the more 
> recently accepted
> > > models for reliability using redundancy (n = m+1, k=0 
> being the typical system
> > > deployed in the field today, m usually being greater than 
> 2).  Up until
> > > recently, only 2m (or m+m) systems were considered 
> carrier class.  Some
> > > systems were 4m (with 3 out of 4 vote).
> > > 
> > > You should take some time to look at redundancy and 
> reliability requirements
> > > for carrier grade systems for an understanding of how to 
> calculate and provide
> > > the reliability requirements expected by carriers.  There 
> are good standards
> > > written on the topic.  See the Telcordia NEBS and 
> reliability specs, for
> > > example.  If you have an RBOC RFP handy, take a look at 
> the hardware and
> > > software reliability requirements sections or talk to the 
> engineer in your
> > > company responsible for answering those sections.
> > > 
> > > > 
> > > >    Say for example
> > > > 
> > > >    Inefficient configuration -just an example
> > > >    ------------------------------------------
> > > >   
> > > >             SLS         DEST
> > > >             =================
> > > >              0           ASP1
> > > >              1           ASP1
> > > >              2           ASP1
> > > >              3           ASP1
> > > >              4           ASP1
> > > >              5           ASP1
> > > >              6           ASP1
> > > >              .            .
> > > >              .            .
> > > >              15          ASP1
> > > >              
> > > >              
> > > >                SG                            AS1
> > > > sls(0,1..15 )+----+  sls(0,1..15)         +------+
> > > > -------------|SGP1|-----------------------| ASP1 |
> > > >              |    |              ASSOC-1  +------+
> > > >              +----+                     
> > > > 
> > > >                       Fig - 1
> > > > 
> > > > In this figure, all incoming messages are transmitted 
> over ASSOC-1 towards
> > > > ASP1.Here there are two potential problems 
> > > > 
> > > >        1. Total incoming traffic is dependent on a single ASP
> > > >        2. ASP1 is always overloaded, messages are 
> processed one by one
> > > >           from the transmission buffer which ultimately 
> introduces time
> > > >           delay(example last arrived msg will be 
> processed at the end).
> > > 
> > > The above was not in my examples.  My examples had n > 1.
> > > 
> > > > 
> > > > 
> > > > Efficient configuration
> > > > -----------------------  
> > > > 
> > > >              SLS         DEST
> > > >             =================
> > > >              0           ASP1
> > > >              1           ASP1
> > > >              2           ASP1
> > > >              3           ASP1
> > > >              4           ASP1
> > > >              5           ASP1
> > > >              6           ASP1
> > > >              7           ASP1
> > > >              8           ASP2
> > > >              9           ASP2
> > > >              .            .
> > > >              .            .
> > > >              15          ASP2
> > > > 
> > > >                
> > > >                        
> > > >                        
> > > >               SG                             AS1
> > > > sls(0,1..15 )+----+   sls(0,1,..7)ASSOC-1 +------+
> > > > -------------|SGP1|-----------------------| ASP1 |
> > > >              |    |--------+              +------+
> > > >              +----+        | sls(8,9..15) +------+
> > > >                            +--------------| ASP2 |
> > > >                                  ASSOC-2  +------+
> > > >                         
> > > >                    fig - 2
> > > > 
> > > > In this example incoming traffic is distributed between 
> ASP1 and ASP2.
> > > > 
> > > > IN ORDER TO HANDLE THE HEAVY INCOMING TRAFFIC, SGP1 
> NEEDS MINIMUM OF
> > > > 2 ACTIVE ASPs BY SETTING n=2.
> > > 
> > > Do they not comply to SS7 ISUP/SCCP flow and congestion 
> controls?  Are they
> > > running at full capacity in this arrangement?  Are these 
> your ASPs?  Do you
> > > design your systems this way?
> > > 
> > > Take a look at the reliability engineering.  If either of 
> these ASPs is
> > > handling its maximum rated capacity, it will have to be 
> reliable to .99999
> > > This is because, if one ASP fails, 50% of the system has 
> failed.  A 50% system
> > > failure is deemed a "system" failure rather than a 
> "component" failure in
> > > reliability engineering for carrier grade systems.  
> Considering proscribed
> > > replacement times on the ASP, this would be difficult to 
> achieve unless both
> > > ASPs were hardware fault tolerant platforms.
> > > 
> > > Nevertheless, the RFC defines ASPs as a process, not a 
> processor.  There is
> > > nothing restricting the two ASPs in your diagram from 
> being processes on the
> > > same processor.  In that case, when ASP2 is down, ASP1 
> has double the
> > > processing capacity as it had before.
> > > 
> > > You certainly cannot make statements as you have made here without
> > > presupposing an implementation architecture outside the 
> scope of the standard.
> > > 
> > > Besides, the RFC permits you to configure your SGP (and 
> why do you only have
> > > one of them?) so that it waits for 'n' ASPs to become 
> active.  You MAY do
> > > this.  Just don't require everybody else to follow your 
> approach.  (Actually,
> > > if the above is your approach, don't require anyone else 
> to follow it.)
> > > 
> > > Several other points: ANSI T1.111.4/2000 SLS values are 
> 8-bits.  That means
> > > there are 256 SLS values (gee, I had 256 ASPs in my example).
> > > 
> > > For another thing, sequenced traffic flows are not only 
> based on SLS.  In an
> > > ITU-T network, each remote signalling point has 16 
> independent traffic flows
> > > for each AS.  The traffic flows from each remote 
> signalling point is
> > > independent from the traffic flows of other remote 
> signalling points.
> > > Therefore, if there are 400 remote signalling points, 
> there will be 6400
> > > independent traffic flows that the SG can use for 
> distribution of AS traffic
> > > for one AS over as many as 6400 ASPs without risking 
> mis-sequencing of
> > > messages.  For an ANSI T1.111.4/2000 network using 8-bit 
> SLS values with 400
> > > remote signalling points, there would be 102400 
> independent traffic flows to
> > > the AS that the SG can use to distribute traffic to one 
> AS over as many as
> > > 102400 ASPs without risking message mis-sequencing.
> > > 
> > > > 
> > > > 
> > > > Your example with n=256 is completely a wrong 
> interpretaion of 'n+k'
> > > > redundancy case.
> > > 
> > > Quite the opposite, you clearly misunderstand redundancy 
> and reliability in
> > > carrier grade systems and obviously have little 
> experience with them.
> > > 
> > > Here's a little primer for you:
> > > 
> > > Carrier grade systems were originally designed as 
> hot-standby, fully redundant
> > > systems.  For every processor actively handling traffic, 
> there was a hot
> > > standby processor also actively handling the same 
> traffic.  This typically
> > > required special bus arrangements whereby the I/O bus of 
> one processor was
> > > actively controlling peripheral devices, while the hot 
> standby processor was
> > > issuing I/O bus output that was not terminated to the I/O 
> bus.  The bus also
> > > normally performed a comparison of the outputs from the 
> active side and the
> > > outputs from the inactive side.  Peripheral components, 
> such as switch fabric
> > > were also fully redundant (duplicated).  Switching 
> fabrics are separated into
> > > an A-plane and a B-plane in much the same fashion as the 
> SPCS front-end I/O
> > > bus.
> > > 
> > > Note that active/standby in M3UA does not address the 
> above model.  2+0
> > > broadcast does.
> > > 
> > > The problem with such a systems was what we used to call 
> the "kill your
> > > sister" syndrome.  A failure on the inactive side only 
> detectable to the I/O
> > > bus could cause a switch from the active to inactive 
> side, resulting in the
> > > switch running on the erroneous inactive side.  We called 
> it the "kill your
> > > sister" syndrome because it is not so surprising that the 
> psychotic sister
> > > (erred inactive side) kills the sane sister (correct active side).
> > > 
> > > Primarily to avoid the "kill your sister" syndrome, some 
> switches were
> > > designed with 3 or 4 processors on an cross-checking I/O 
> bus pair.  In that
> > > case, it would take a 2 out of 3 or 3 out of 4 vote to 
> shift from the active
> > > processor.  These systems were far less susceptible to 
> inactive plane errors
> > > that were only detected at the I/O bus.
> > > 
> > > Note that the only M3UA model that address the above is 
> the 3+0 or 4+0
> > > broadcast model.
> > > 
> > > When the use of general purpose computing platforms for 
> telecommunications
> > > systems started to become popular, there were several 
> models that emerged.
> > > 
> > > First was taking two big iron UNIX hosts and running them 
> in active/standby.
> > > In this scenario, one UNIX host would be actively 
> handling traffic and the
> > > other would be warm standby.  By "warm" standby, it means 
> that the device was
> > > powered and was synchronizing state on a checkpoint 
> interval or monitoring the
> > > health of the active hosts with heartbeats.  Shared 
> memory, shared DASD or
> > > message passing are the most common mechanisms of 
> redundancy.  These systems
> > > are considered "highly available" systems rather than the 
> "fault tolerant"
> > > systems used in telecommunications above.  An increased 
> level of reliability
> > > is often achieved by using hardware fault tolerant 
> multiple processor hosts.
> > > 
> > > In M3UA, the active/standby model addresses these 
> systems.  Although M3UA
> > > provides for a '1+k' model, k is invariably 1.
> > > 
> > > The major difficulties with these systems are: a) 
> switchover latency between
> > > systems; b) loss of state during switchover; and c) 
> scalability.  Switchover
> > > latency is normally improved by decreasing heartbeat 
> intervals.  State
> > > synchronization is normally improved by decreasing the 
> process checkpoint
> > > interval.  These reductions typically increase cost and 
> they are normally
> > > chosen to minimize a cost/performance/reliability metric. 
>  Scalability was
> > > typically performed by installing larger more powerful 
> hosts with increased
> > > demand.  This was not good, because larger processors 
> tended to have poorer
> > > cost/performance ratios that smaller processors.
> > > 
> > > Later it was determined that the cost of a complete 
> redundant host that was
> > > wasting CPU cycles while inactive could be avoided while 
> addressing the
> > > deficiencies of the paired host approach above.  To increase state
> > > synchronization, state could be spread over a number of 
> hosts.  If there were
> > > 'm' hosts, the failure of one host could at worst affect 
> a proportion, 1/m, of
> > > the state.  Therefore state synchronization would improve 
> as m increased.  To
> > > distribute the state, traffic load would be shared over 
> the m hosts according
> > > to state.  This also served the needs of scalability and 
> provided for closer
> > > to linear cost scalability, because the number of hosts, 
> m, could be increased
> > > while holding the cost/performance ratio of the hosts 
> constant.  The costs of
> > > redundancy could also be reduced over the fully redundant 
> systems described
> > > above, because redundancy could be achieved using 'm+1' 
> processors, where 'm'
> > > processors were required to handle the maximum engineered 
> load.  For systems
> > > as small as 'm=2' hosts with equivalent reliability to 
> that normally
> > > experienced in the '1+1' active/standby scenario above, 
> system reliability
> > > figures could be maintained with less redundancy cost.  
> It is important to
> > > note, that in these systems, all 'm+1' hosts are active 
> and handling only a
> > > proportion m/(m+1) of their rated capacity.  It is very 
> seldom that the rated
> > > capacity was more that 2/3 of the maximum short-term capacity.
> > > 
> > > These systems do not solve the heartbeat interval problem of the
> > > active/standby paired hosts system above.  Typically what 
> these systems use is
> > > more sophisticated watchdog approaches rather than 
> passing heartbeat messages
> > > between processes on the active and standby hosts.  
> Memory sharing, DASD
> > > sharing, message passing, reliable broadcast, voting, and 
> a plethora of
> > > reliable distributed systems techniques have been 
> developed that are useful in
> > > these systems.
> > > 
> > > Adoption of these latter systems changed the requirement 
> from "duplicated"
> > > systems to systems with no "single point of failure".
> > > 
> > > Now, the 'n+k' load-sharing and broadcast model was 
> intended to address these
> > > systems.  Where 'n == m' in the description above.  That 
> is why we agreed that
> > > n is the maximum engineered load.  Because that is what 
> 'm' is.  The k is the
> > > 1 in the description above, generalized.
> > > 
> > > First off, you must understand that it is not the 
> intention of the model to
> > > have 'n' ASPs active and 'k' inactive unless 'n=1, k=1'.  
> This is described in
> > > section 1.4.4.1:
> > > 
> > >   A "1+1" active/standby redundancy is a subset of this model.
> > > 
> > > Also no redundancy is possible:
> > > 
> > >   A simplex "1+0" model is also supported as a subset, 
> with no ASP redundancy.
> > > 
> > > (The term "simplex" come from the telephone switches 
> described above where the
> > > inactive side is taken off-line.  Such a situation is 
> described as running the
> > > switch "simplex".)
> > > 
> > > The 'n+k' model is to address the 'm+1' distributed 
> systems described last.
> > > Rereading section 1.4.4.1:
> > > 
> > >   The failover model supports an "n+k" redundancy model, 
> where "n" APSs is the
> > >   minimum number of redundant [sic] ASPs required to 
> handle traffic and "k"
> > >   ASPs are available to take over for a failed or unavailable ASP.
> > > 
> > > Note that the state of the "k" ASPs with respect to an AS 
> is not stated.  In
> > > the 'm+1' redundant system described above, the "k" ASPs 
> are active for the
> > > AS.  "k" ASPs active in load sharing mode are "available 
> to take over for a
> > > failed or unavailable ASP".
> > > 
> > > It is typical to run the "k" ASPs active (and I know you 
> had some questions
> > > about activating the "k" ASPs, but the point is that they 
> are normally already
> > > active), and handling their own share of traffic.  In 
> this case, each ASPs
> > > handles n/(n+k) of their engineered capacity.  Were you 
> actually considering
> > > deploying such a atypical model where the 'k' ASPs are inactive?
> > > 
> > > "k" ASPs can be taken out of service, upgraded, and 
> brought back into service
> > > without affecting the maximum engineering traffic 
> handling capacity of the
> > > system (in absence of failures during the upgrade).  This 
> assists in graceful
> > > software/hardware upgrades on live systems.  (Keeping 
> them inactive would
> > > serve no purpose and would, decrease maximum traffic 
> handling capability,
> > > increase loss of state on fail-over, and increase 
> fail-over latency: all good 
> > > reasons for not deactivating the 'k' ASPs).
> > > 
> > > Now, reliability engineering for carrier class systems 
> takes the view that if
> > > 50% of the load bearing components in a system fail (stop 
> operating in any
> > > way), then the entire system has failed.  If traffic is 
> stopped above n/2
> > > processes, the system will have failed prematurely.
> > > 
> > > A requirement to withhold traffic < n will always suffer 
> a prolonged system
> > > failure, because only n/2 are required to make the system 
> available.
> > > Therefore, on startup the reliability of the system is 
> degraded by the
> > > negative exponent n/2, regardless of whether n=256 or 
> n=2; however, increasing
> > > 'n' increases the damage done to reliability by waiting for 'n'.
> > > 
> > > A requirement for the SG to wait for 'n' is unacceptable 
> from the standpoint
> > > of the redundancy model that 'n+k' was meant to address, 
> as well as other
> > > possible models that it can accommodate.
> > > 
> > > > 
> > > > 
> > > > > 
> > > > > Consider the same example with: n=2, k=1, two ASPs failed.
> > > > > 
> > > > > Any system that introduces hysteresis to aggravate 
> and perpetuate an outage in
> > > > > this fashion is not very useful.
> > > > > 
> > > > > In the examples, what would have been a brief 
> AS-PENDING situation has been
> > > > > turned into a prolonged and complete system failure.  
> It is certainly not good
> > > > > practice to design redundant systems in that fashion 
> and recommending or
> > > > > requiring such a design would be a bad thing.
> > > > > 
> > > > > To further the example, consider the multiple SG as 
> STP situation.  One SG
> > > > > cannot know the state of ASPs for the its mated SG 
> and therefore cannot
> > > > > distinguish between the restart scenario and the 
> start scenario as far as the
> > > > > AS state is concerned because the SG does not know 
> the state of the AS itself:
> > > > > the AS could be active with other SGs.
> > > > > 
> > > > > This is the same case for SGP.  Unless AS state is 
> synchronized across SGPs,
> > > > > one SGP cannot know the total state of the AS (only 
> its partition).  There is
> > > > > no requirement to synchronize AS state across SGP (on 
> the SGP side, AS-state
> > > > > is per-SGP), and there is certainly no requirement to 
> synchronize AS-state
> > > > > across SGs.
> > > > > 
> > > > > So, take the (n=2, k=0) example where there are two 
> SGs.  Due to a network
> > > > > partitioning at startup, one ASP can communicate only 
> with one SG, and the
> > > > > other ASP can communicate only with the other SG.  
> Because n=2, neither SG
> > > > > will start traffic and the system suffers 100% 
> failure until the network
> > > > > partitioning ceases.
> > > > > 
> > > > > Consider the same example where there are two SGPs 
> (instead of SGs) and the
> > > > > two ASPs normally only form one association (e.g., 
> ASP1-SGP1, ASP2-SGP2).
> > > > > Note that there is no requirement in the RFC to form 
> full-mesh associations.
> > > > > Because the number of active ASPs with each SGP is < 
> 2, the system will never
> > > > > start.
> > > > > 
> > > > > These are not strange scenarios.  These situations 
> are all permitted by the
> > > > > RFC.  Yet, in each situation it is clear that one 
> cannot recommend that the SG
> > > > > wait.  The result would be a non-functional system.  
> There are many more
> > > > > examples.
> > > > > 
> > > > > n+k is a model for redundancy (not for congestion and 
> flow control).  A system
> > > > > with an SG that waits for n is, in general, not a 
> fully redundant system at
> > > > > all, but is, rather, prone to these failure syndromes.
> > > > > 
> > > > > When you look at the situation and apply the 
> experience of reliability
> > > > > engineering, the recommendation is that the SG SHOULD 
> start traffic when the
> > > > > first ASP goes active.
> > > > > 
> > > > > Why wait for 'n' anyway?  Because the ASP cannot 
> handle some level of traffic?
> > > > > Any ASP that conforms to MTP-User standards, must 
> have the standardized
> > > > > MTP-User (ISUP/SCCP) congestion and flow controls and 
> must have the SCTP
> > > > > congestion controls and is therefore able to accept 
> any amount of traffic
> > > > > overload.
> > > > > 
> > > > > So there is no real reason for the SG to wait for 
> 'n'.  Nevertheless, the RFC
> > > > > allows this as an OPTION.  Requiring (MUST) or 
> recommending it (SHOULD)
> > > > > regardless of whether the number of active processors 
> is rising or falling,
> > > > > whether start, restart or failure, is unacceptable.
> > > > > 
> > > > > On the other side, at the ASPs, the ASPs have a 
> requirement to share state
> > > > > between them.  They must share circuit state and 
> transaction state in all
> > > > > configurations to meet carrier grade.  This is stated 
> in the RFC.  It is
> > > > > implicit that the ASPs are capable of knowing with 
> which SGs, and which SGPs
> > > > > for which they are active.  They have access to the 
> total state of the AS that
> > > > > they collectively represent, that the SG/SGP does 
> not.  Therefore is it
> > > > > reasonable for the ASPs to coordinate among 
> themselves the number of active
> > > > > ASPs and to ensure that there are sufficient ASPs 
> active to handle the
> > > > > expected load before starting traffic.  Because the 
> ASPs need not respond to
> > > > > user test traffic (UPT/SST), the ASPs are also in 
> control of when remote SS7
> > > > > MTP-User restart traffic toward them.
> > > > > 
> > > > > Therefore, the recommendation that the ASPs SHOULD 
> coordinate among
> > > > > themselves the number of active ASPs and not start 
> traffic until a sufficient
> > > > > number of them are active is sound.  This is not a 
> MUST, permitting ASPs
> > > > > that know that they are capable of handling any 
> overload, to start traffic
> > > > > immediately, or at some earlier point.
> > > > > 
> > > > > Back to the RFC, the RFC sets the normal case where 
> the SG starts traffic on
> > > > > the first active ASP but explicitly permits as an 
> OPTION to wait until it
> > > > > considers that there are sufficient ASPs active to 
> handle the offered load.
> > > > > The RFC also RECOMMENDS that the ASPs coordinate 
> among themselves the number
> > > > > of active ASPs so that there are sufficient ASPs to 
> handle the expected load.
> > > > > 
> > > > > The provisions in the RFC are both sound, where the 
> proposal and section 3.15
> > > > > is clearly not.
> > > > > 
> > > > > Remember when looking at the AS state diagram in the 
> RFC that this is the
> > > > > SGP's view of the AS state and not the total state of 
> the AS, that can only
> > > > > known to the ASPs for all configurations.  'n' as 
> defined has no meaning to
> > > > > the state machine at the SGP because is not required 
> to know, and often cannot
> > > > > know, how many ASPs are active for the AS via other 
> SGPs and SGs, or even via
> > > > > direct connection to the SS7 network.
> > > > > 
> > > > > Some SGs might have special knowledge, or 
> sophisticated algorithms that allow
> > > > > it to determine whether a congestive failure will be 
> the immediate result of
> >  > > > starting or restarting traffic.  These SGs are 
> permitted as an
> > OPTION to wait.
> > > > > 
> > > > > The ASPs SHOULD have the knowledge of how many are 
> required to handle the
> > > > > expected load, because they must share 
> circuit/transaction state and they must
> > > > > have congestion and flow controls as required by the 
> standards.
> > > > > 
> > > > > So again, the RFC is adequate in this regard.  The 
> current proposal and
> > > > > section 3.15 are not.
> > > > RFC is inadequate so we need section 3.15 to explain 'n+k' case.
> > > 
> > > Teaching reliability engineering and the design of 
> redundant systems is
> > > outside the scope of the RFC and the IG.  People, such as 
> yourself, are better
> > > advised to learn the subject matter before attempting to 
> implement or claim a
> > > complete understanding of the protocol.
> > > 
> > > The current RFC provides the protocol mechanisms and basis for
> > > interoperability for all of the common approaches to 
> redundancy in carrier
> > > class systems, and even provides a general model that can 
> accommodate
> > > non-traditional redundancy architectures.
> > > 
> > > When confused about how a redundancy architecture works, 
> don't blame the M3UA
> > > RFC, blame those responsible for describing redundant 
> architectures to you.
> > > 
> > > In the appendix we stated:
> > > 
> > >    M3UA imposes no restrictions as to the exact layout of 
> the network
> > >    elements, the message distribution algorithms and the 
> distribution of
> > >    signalling processes.  Instead, it provides a 
> framework and a set of
> > >    messages that allow for a flexible and scalable 
> signalling network
> > >    architecture, aiming to provide reliability and performance.
> > > 
> > > Forcing SGs to wait for 'n' ASPs active before activating 
> is contrary to this.
> > > The existing "MAY" achieves this flexibility.
> > > 
> > > When reading 1.4.4.1, also remember this statement in the 
> appendix, that used
> > > to be beside 1.4.4.1.  The description of 'n+k' does not 
> dictate the state of
> > > the 'k' or the state of the AS with respect to 'n' to 
> leave the redundancy
> > > model flexible, and accommodate '1+1' and '1+0' in the same model.
> > > 
> > > Also, please don't presume to tell me to set 'n' to 1, as 
> it is not a
> > > arbitrary value, but the number 'm' in the 'm+1' 
> distributed system in common
> > > use today.  This number is determined as the minimum 
> number of ASPs required
> > > to handle the maximum engineered load, and not, as Tolga 
> would have, the
> > > minimum number of ASPs required to handle ANY load.  If 
> you want your system
> > > to behave as Tolga proposes, just have your SG exercise 
> the "MAY choose not to
> > > direct traffic" option of section 4.3.4.3 of the RFC.
> > > 
> > > Section 3.15 and Tolga's proposal should be struck 
> because they attempt to
> > > impose a 'n' active 'k' inactive and system cannot 
> restart (3.15) or start
> > > (Tolga) without 'n' view on the redundancy model, where 
> it is clearly the
> > > intent of the RFC to place no such restrictions on the 
> model, but to
> > > accommodate a range of models, that is already achieved 
> by the RFC text.
> > > 
> > > --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 exim@www1.ietf.org  Thu Jul  3 07:24:35 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA06588
	for <sigtran-archive@odin.ietf.org>; Thu, 3 Jul 2003 07:24:35 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Y2C5-0008Cd-HS
	for sigtran-archive@odin.ietf.org; Thu, 03 Jul 2003 07:24:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h63BO5WX031525
	for sigtran-archive@odin.ietf.org; Thu, 3 Jul 2003 07:24:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Y2C1-0008Bk-Bx; Thu, 03 Jul 2003 07:24:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Y2BT-0008BS-Sx
	for sigtran@optimus.ietf.org; Thu, 03 Jul 2003 07:23:27 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA06478
	for <sigtran@ietf.org>; Thu, 3 Jul 2003 07:23:26 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19Y2BS-0005IU-00
	for sigtran@ietf.org; Thu, 03 Jul 2003 07:23:26 -0400
Received: from auds953.usa.alcatel.com ([143.209.238.6])
	by ietf-mx with esmtp (Exim 4.12)
	id 19Y2BR-0005Hh-00
	for sigtran@ietf.org; Thu, 03 Jul 2003 07:23:25 -0400
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h63BMqLX002948;
	Thu, 3 Jul 2003 06:22:52 -0500 (CDT)
Received: from sun2594.ssd.usa.alcatel.com (sun2594.ssd.usa.alcatel.com [143.209.250.252])
	by ssd.usa.alcatel.com (8.12.8/8.12.8) with ESMTP id h63BMb1C003106;
	Thu, 3 Jul 2003 06:22:37 -0500 (CDT)
Date: Thu, 3 Jul 2003 06:22:37 -0500 (CDT)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>
cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] Re: M3UA MIB Draft05: Loadsharing on CIC ,SSN
In-Reply-To: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CC9@eestqnt105.es.eu.ericsson.se>
Message-ID: <Pine.SOL.4.10.10307030614420.2696-100000@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>


Pastor,

  Going back to old emails ...on 11-Apr-2003 i asked
  'What is loadsharing based on DPC?'

  Nobody gave proper reply to this email.
  Still i dont undrstand it.If anybody is there to explain what is it?
  otherwise we will delete this from MIB-04.

Following is the email which i sent  on 11-Apr-2003.

----------------------------------------------
Date: Fri, 11 Apr 2003 06:09:28 -0500 (CDT)
From: Samuel Dur D. Jeyaseelan <sjeyasee@ssd.usa.alcatel.com>
To: sigtran@ietf.org
Subject: [m3ua]-ldshar basing on dpc
---------------------------------------------------


--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, 3 Jul 2003, Javier Pastor-Balbas (ECE) wrote:

> 
> Hi Anjali, 
> 
> It seems that it makes sense to remove those from the standard MIB. All the people who are interested in having this object can create it within their private MIBs. We may keep SLS since it is a common practise in SS7 unless you have any objection.
> 
> 
> Thanks for your responses // Javier.
> 
> PS: Samuel: What object do you refer when talking about DPC-based loadsharing within the current MIB? 
> 
> 
> > -----Original Message-----
> > From: agurmukhani@hss.hns.com [mailto:agurmukhani@hss.hns.com]
> > Sent: jueves, 03 de julio de 2003 6:54
> > To: J.Javier Pastor (EE/EEM)
> > Cc: sigtran@ietf.org
> > Subject: M3UA MIB Draft05: Loadsharing on CIC ,SSN
> > 
> > 
> > 
> > 
> > 
> > 
> > Hi...
> > 
> > 
> > In my understanding and  as per the discussion and consensus 
> > on the mailing
> > list , everyone agreed that  CIC and SSN should not be used 
> > as loadsharing
> > basis
> > be it within streams in a particular SCTP association or for active
> > signalling processes.
> > 
> > So would these be removed from  the m3uaStreamLshMethod  and
> > m3uaSpLshMethod OBJECT-TYPE s in M3UA MIB draft.
> > 
> > 
> > Thanks and Regards
> > Anjali
> > www.hssworld.com
> > Hughes Software Systems.
> > 
> > "DISCLAIMER: This message is proprietary to Hughes Software 
> > Systems Limited
> > (HSS) and is intended solely for the use of the individual to 
> > whom it is
> > addressed. It may contain  privileged or confidential information and
> > should not be circulated or used for any purpose other than 
> > for what it is
> > intended. If you have received this message in error, please 
> > notify the
> > originator immediately. If you are not the intended recipient, you are
> > notified that you are strictly prohibited from using, 
> > copying, altering, or
> > disclosing the contents of this message. HSS accepts no 
> > responsibility for
> > loss or damage arising from the use of the information 
> > transmitted by this
> > email including damage from virus."
> > 
> 
> _______________________________________________
> 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 exim@www1.ietf.org  Thu Jul  3 09:20:34 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA09922
	for <sigtran-archive@odin.ietf.org>; Thu, 3 Jul 2003 09:20:34 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Y40L-0003nw-MH
	for sigtran-archive@odin.ietf.org; Thu, 03 Jul 2003 09:20:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h63DK5E7014625
	for sigtran-archive@odin.ietf.org; Thu, 3 Jul 2003 09:20:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Y40I-0003nW-Fq; Thu, 03 Jul 2003 09:20:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Y3zW-0003mg-7A
	for sigtran@optimus.ietf.org; Thu, 03 Jul 2003 09:19:14 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA09904
	for <sigtran@ietf.org>; Thu, 3 Jul 2003 09:19:12 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19Y3zU-0006CL-00
	for sigtran@ietf.org; Thu, 03 Jul 2003 09:19:12 -0400
Received: from penguin-ext.wise.edt.ericsson.se ([193.180.251.47] helo=penguin.al.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19Y3zT-0006CI-00
	for sigtran@ietf.org; Thu, 03 Jul 2003 09:19:11 -0400
Received: from esealnt612.al.sw.ericsson.se ([153.88.254.118])
	by penguin.al.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h63DJAGJ014176;
	Thu, 3 Jul 2003 15:19:10 +0200 (MEST)
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 NW44YH4D; Thu, 3 Jul 2003 15:19:10 +0200
Received: by ESEALNT745.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <N94NL7QP>; Thu, 3 Jul 2003 15:18:58 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CD1@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>
To: "'Samuel Dur D. Jeyaseelan'" <sjeyasee@ssd.usa.alcatel.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: RE: [Sigtran] Re: M3UA MIB Draft05: Loadsharing on CIC ,SSN
Date: Thu, 3 Jul 2003 15:18:44 +0200 
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>

Samuel, 

Are you referring to any particular object within MIB-04? I mean, what Anjali was referring to were two particular objects that not include that kind of loadsharing...


Thanks // Javier.


> -----Original Message-----
> From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> Sent: jueves, 03 de julio de 2003 13:23
> To: J.Javier Pastor (EE/EEM)
> Cc: 'sigtran@ietf.org'
> Subject: Re: [Sigtran] Re: M3UA MIB Draft05: Loadsharing on CIC ,SSN
> 
> 
> 
> Pastor,
> 
>   Going back to old emails ...on 11-Apr-2003 i asked
>   'What is loadsharing based on DPC?'
> 
>   Nobody gave proper reply to this email.
>   Still i dont undrstand it.If anybody is there to explain what is it?
>   otherwise we will delete this from MIB-04.
> 
> Following is the email which i sent  on 11-Apr-2003.
> 
> ----------------------------------------------
> Date: Fri, 11 Apr 2003 06:09:28 -0500 (CDT)
> From: Samuel Dur D. Jeyaseelan <sjeyasee@ssd.usa.alcatel.com>
> To: sigtran@ietf.org
> Subject: [m3ua]-ldshar basing on dpc
> ---------------------------------------------------
> 
> 
> --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, 3 Jul 2003, Javier Pastor-Balbas (ECE) wrote:
> 
> > 
> > Hi Anjali, 
> > 
> > It seems that it makes sense to remove those from the 
> standard MIB. All the people who are interested in having 
> this object can create it within their private MIBs. We may 
> keep SLS since it is a common practise in SS7 unless you have 
> any objection.
> > 
> > 
> > Thanks for your responses // Javier.
> > 
> > PS: Samuel: What object do you refer when talking about 
> DPC-based loadsharing within the current MIB? 
> > 
> > 
> > > -----Original Message-----
> > > From: agurmukhani@hss.hns.com [mailto:agurmukhani@hss.hns.com]
> > > Sent: jueves, 03 de julio de 2003 6:54
> > > To: J.Javier Pastor (EE/EEM)
> > > Cc: sigtran@ietf.org
> > > Subject: M3UA MIB Draft05: Loadsharing on CIC ,SSN
> > > 
> > > 
> > > 
> > > 
> > > 
> > > 
> > > Hi...
> > > 
> > > 
> > > In my understanding and  as per the discussion and consensus 
> > > on the mailing
> > > list , everyone agreed that  CIC and SSN should not be used 
> > > as loadsharing
> > > basis
> > > be it within streams in a particular SCTP association or 
> for active
> > > signalling processes.
> > > 
> > > So would these be removed from  the m3uaStreamLshMethod  and
> > > m3uaSpLshMethod OBJECT-TYPE s in M3UA MIB draft.
> > > 
> > > 
> > > Thanks and Regards
> > > Anjali
> > > www.hssworld.com
> > > Hughes Software Systems.
> > > 
> > > "DISCLAIMER: This message is proprietary to Hughes Software 
> > > Systems Limited
> > > (HSS) and is intended solely for the use of the individual to 
> > > whom it is
> > > addressed. It may contain  privileged or confidential 
> information and
> > > should not be circulated or used for any purpose other than 
> > > for what it is
> > > intended. If you have received this message in error, please 
> > > notify the
> > > originator immediately. If you are not the intended 
> recipient, you are
> > > notified that you are strictly prohibited from using, 
> > > copying, altering, or
> > > disclosing the contents of this message. HSS accepts no 
> > > responsibility for
> > > loss or damage arising from the use of the information 
> > > transmitted by this
> > > email including damage from virus."
> > > 
> > 
> > _______________________________________________
> > 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 exim@www1.ietf.org  Thu Jul  3 11:33:50 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA17262
	for <sigtran-archive@odin.ietf.org>; Thu, 3 Jul 2003 11:33:50 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Y65J-0002Fi-Or
	for sigtran-archive@odin.ietf.org; Thu, 03 Jul 2003 11:33:21 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h63FXLXZ008650
	for sigtran-archive@odin.ietf.org; Thu, 3 Jul 2003 11:33:21 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Y656-0002C8-57; Thu, 03 Jul 2003 11:33:08 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Y64j-00021a-4N
	for sigtran@optimus.ietf.org; Thu, 03 Jul 2003 11:32:45 -0400
Received: from CNRI.Reston.VA.US (localhost [127.0.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA16910;
	Thu, 3 Jul 2003 11:32:42 -0400 (EDT)
Message-Id: <200307031532.LAA16910@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: Thu, 03 Jul 2003 11:32:42 -0400
Subject: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.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		: Signalling Connection Control Part User Adaptation 
                          Layer (SUA)
	Author(s)	: J. Loughney, G. Sidebottom
	Filename	: draft-ietf-sigtran-sua-15.txt
	Pages		: 118
	Date		: 2003-7-3
	
This Internet Draft defines a protocol for the transport of any         Signalling Connection Control Part-User signalling over IP using the   Stream Control Transport Protocol. The protocol is designed to be 
modular and symmetric, to allow it to work in diverse architectures, 
such as a Signalling Gateway to IP Signalling Endpoint architecture 
as well as a peer-to-peer IP Signalling Endpoint architecture.

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-sigtran-sua-15.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-sua-15.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-sua-15.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:	<2003-7-3112356.I-D@ietf.org>

ENCODING mime
FILE /internet-drafts/draft-ietf-sigtran-sua-15.txt

--OtherAccess
Content-Type: Message/External-body;
	name="draft-ietf-sigtran-sua-15.txt";
	site="ftp.ietf.org";
	access-type="anon-ftp";
	directory="internet-drafts"

Content-Type: text/plain
Content-ID:	<2003-7-3112356.I-D@ietf.org>

--OtherAccess--

--NextPart--



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



From exim@www1.ietf.org  Thu Jul  3 12:19:56 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA17263
	for <sigtran-archive@odin.ietf.org>; Thu, 3 Jul 2003 11:33:50 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Y65J-0002Fk-Ow
	for sigtran-archive@odin.ietf.org; Thu, 03 Jul 2003 11:33:21 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h63FXL21008649
	for sigtran-archive@odin.ietf.org; Thu, 3 Jul 2003 11:33:21 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Y655-0002Bh-CR; Thu, 03 Jul 2003 11:33:07 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Y64e-00020O-54
	for sigtran@optimus.ietf.org; Thu, 03 Jul 2003 11:32:40 -0400
Received: from CNRI.Reston.VA.US (localhost [127.0.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA16887;
	Thu, 3 Jul 2003 11:32:37 -0400 (EDT)
Message-Id: <200307031532.LAA16887@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: Thu, 03 Jul 2003 11:32:37 -0400
Subject: [Sigtran] I-D ACTION:draft-ietf-sigtran-m3ua-implementors-guide-04.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
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id LAA16888
Content-Transfer-Encoding: quoted-printable

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

	Title		: M3UA Implementor=C6s Guide
	Author(s)	: J. Pastor, K. Morneault
	Filename	: draft-ietf-sigtran-m3ua-implementors-guide-04.txt
	Pages		: 58
	Date		: 2003-7-3
=09
This document contains a compilation of all defects found up until
the publication date for M3UA [RFC3332]. These defects may be of an
editorial or technical nature. This document may be thought of as a
companion document to be used in the implementation of M3UA to
clarify errors in the original M3UA document. This document updates
RFC3332 and text within this document supersedes the text found in
RFC3332.

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-sigtran-m3ua-implementors-=
guide-04.txt

To remove yourself from the IETF Announcement list, send a message to=20
ietf-announce-request with the word unsubscribe in the body of the messag=
e.

Internet-Drafts are also available by anonymous FTP. Login with the usern=
ame
"anonymous" and a password of your e-mail address. After logging in,
type "cd internet-drafts" and then
	"get draft-ietf-sigtran-m3ua-implementors-guide-04.txt".

A list of Internet-Drafts directories can be found in
http://www.ietf.org/shadow.html=20
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-m3ua-implementors-guide-04.txt=
".
=09
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.
	=09
	=09
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:	<2003-7-3112345.I-D@ietf.org>

ENCODING mime
FILE /internet-drafts/draft-ietf-sigtran-m3ua-implementors-guide-04.txt

--OtherAccess
Content-Type: Message/External-body;
	name="draft-ietf-sigtran-m3ua-implementors-guide-04.txt";
	site="ftp.ietf.org";
	access-type="anon-ftp";
	directory="internet-drafts"

Content-Type: text/plain
Content-ID:	<2003-7-3112345.I-D@ietf.org>

--OtherAccess--

--NextPart--



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



From exim@www1.ietf.org  Thu Jul  3 12:30:37 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA22190
	for <sigtran-archive@odin.ietf.org>; Thu, 3 Jul 2003 12:30:37 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Y6yH-0006aV-Cp
	for sigtran-archive@odin.ietf.org; Thu, 03 Jul 2003 12:30:10 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h63GU9bx025317
	for sigtran-archive@odin.ietf.org; Thu, 3 Jul 2003 12:30:09 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Y6y9-0006a7-E7; Thu, 03 Jul 2003 12:30:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Y6xf-0006WU-2n
	for sigtran@optimus.ietf.org; Thu, 03 Jul 2003 12:29:31 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA22058
	for <sigtran@ietf.org>; Thu, 3 Jul 2003 12:29:28 -0400 (EDT)
From: john.loughney@nokia.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19Y6xd-0002BX-00
	for sigtran@ietf.org; Thu, 03 Jul 2003 12:29:29 -0400
Received: from mgw-x4.nokia.com ([131.228.20.27])
	by ietf-mx with esmtp (Exim 4.12)
	id 19Y6xc-0002BP-00
	for sigtran@ietf.org; Thu, 03 Jul 2003 12:29:28 -0400
Received: from esvir03nok.nokia.com (esvir03nokt.ntc.nokia.com [172.21.143.35])
	by mgw-x4.nokia.com (Switch-2.2.6/Switch-2.2.6) with ESMTP id h63GTQ928925
	for <sigtran@ietf.org>; Thu, 3 Jul 2003 19:29:26 +0300 (EET DST)
Received: from esebh001.NOE.Nokia.com (unverified) by esvir03nok.nokia.com
 (Content Technologies SMTPRS 4.2.5) with ESMTP id <T63360cbb26ac158f23077@esvir03nok.nokia.com> for <sigtran@ietf.org>;
 Thu, 3 Jul 2003 19:29:26 +0300
Received: from esebe010.NOE.Nokia.com ([172.21.138.49]) by esebh001.NOE.Nokia.com with Microsoft SMTPSVC(5.0.2195.6139);
	 Thu, 3 Jul 2003 19:29:25 +0300
Received: from esebe023.NOE.Nokia.com ([172.21.138.115]) by esebe010.NOE.Nokia.com with Microsoft SMTPSVC(5.0.2195.6139);
	 Thu, 3 Jul 2003 19:29:24 +0300
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"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Date: Thu, 3 Jul 2003 19:29:23 +0300
Message-ID: <DADF50F5EC506B41A0F375ABEB32063658F061@esebe023.ntc.nokia.com>
Thread-Topic: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Thread-Index: AcNBeNCN+OPV66NiRz2QUXHtMtAv4wAB2wyQ
To: <sigtran@ietf.org>
X-OriginalArrivalTime: 03 Jul 2003 16:29:24.0917 (UTC) FILETIME=[4402F650:01C34180]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Hi all,

I have updated SUA, I have tried to incorporate some of the major issues
from the M3UA Implementation Guide.  All interested parties, please
give the document a quick read, because I would like to send this to the
IESG for final approval.

thanks,
John

> -----Original Message-----
> From: ext Internet-Drafts@ietf.org [mailto:Internet-Drafts@ietf.org]
> Sent: 03 July, 2003 18:33
> Cc: sigtran@ietf.org
> Subject: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
>=20
>=20
> A New Internet-Draft is available from the on-line=20
> Internet-Drafts directories.
> This draft is a work item of the Signaling Transport Working=20
> Group of the IETF.
>=20
> 	Title		: Signalling Connection Control Part=20
> User Adaptation=20
>                           Layer (SUA)
> 	Author(s)	: J. Loughney, G. Sidebottom
> 	Filename	: draft-ietf-sigtran-sua-15.txt
> 	Pages		: 118
> 	Date		: 2003-7-3
> =09
> This Internet Draft defines a protocol for the transport of=20
> any         Signalling Connection Control Part-User=20
> signalling over IP using the   Stream Control Transport=20
> Protocol. The protocol is designed to be=20
> modular and symmetric, to allow it to work in diverse architectures,=20
> such as a Signalling Gateway to IP Signalling Endpoint architecture=20
> as well as a peer-to-peer IP Signalling Endpoint architecture.
>=20
> A URL for this Internet-Draft is:
> http://www.ietf.org/internet-drafts/draft-ietf-sigtran-sua-15.txt
>=20
> To remove yourself from the IETF Announcement list, send a message to=20
> ietf-announce-request with the word unsubscribe in the body=20
> of the message.
>=20
> Internet-Drafts are also available by anonymous FTP. Login=20
> 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-sua-15.txt".
>=20
> A list of Internet-Drafts directories can be found in
> http://www.ietf.org/shadow.html=20
> or ftp://ftp.ietf.org/ietf/1shadow-sites.txt
>=20
>=20
> Internet-Drafts can also be obtained by e-mail.
>=20
> Send a message to:
> 	mailserv@ietf.org.
> In the body type:
> 	"FILE /internet-drafts/draft-ietf-sigtran-sua-15.txt".
> =09
> 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=20
> 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.
> 	=09
> 	=09
> Below is the data which will enable a MIME compliant mail reader
> implementation to automatically retrieve the ASCII version of the
> Internet-Draft.
>=20

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



From exim@www1.ietf.org  Thu Jul  3 18:52:34 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA17193
	for <sigtran-archive@odin.ietf.org>; Thu, 3 Jul 2003 18:49:07 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YCjr-0006S8-OG
	for sigtran-archive@odin.ietf.org; Thu, 03 Jul 2003 18:39:39 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h63Mdd11024790
	for sigtran-archive@odin.ietf.org; Thu, 3 Jul 2003 18:39:39 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YCjr-0006Re-IM
	for sigtran-web-archive@optimus.ietf.org; Thu, 03 Jul 2003 18:39:39 -0400
Date: Thu, 03 Jul 2003 18:39:39 -0400
Message-ID: <20030703223939.4508.3579.Mailman@www1.ietf.org>
Subject: ietf.org mailing list memberships reminder
From: mailman-owner@www1.ietf.org
To: sigtran-web-archive@optimus.ietf.org
X-No-Archive: yes
X-Ack: no
Sender: mailman-admin@ietf.org
Errors-To: mailman-admin@ietf.org
X-BeenThere: mailman@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk

There was an error in the last monthly reminder, in that the NOTE WELL
statement (below) was not included. Therefore, the reminder is being
sent out again.

This is a reminder, sent out once a month, about your ietf.org mailing
list memberships.  It includes your subscription info and how to use
it to change it or unsubscribe from a list.

You can visit the URLs to change your membership status or
configuration, including unsubscribing, setting digest-style delivery
or disabling delivery altogether (e.g., for a vacation), and so on.

In addition to the URL interfaces, you can also use email to make such
changes.  For more info, send a message to the '-request' address of
the list (for example, sigtran-request@ietf.org) containing just the
word 'help' in the message body, and an email message will be sent to
you with instructions.

***************************************************************************


                              Note Well

All statements related to the activities of the IETF and addressed to
the IETF are subject to all provisions of Section 10 of RFC 2026,
which grants to the IETF and its participants certain licenses and
rights in such statements. Such statements include verbal statements
in IETF meetings, as well as written and electronic communications
made at any time or place, which are addressed to

        * the IETF plenary session,
        * any IETF working group or portion thereof,
        * the IESG, or any member thereof on behalf of the IESG,
        * the IAB or any member thereof on behalf of the IAB,
        * any IETF mailing list, including the IETF list itself, any
working
            group or design team list, or any other list functioning
under IETF
            auspices,
        * the RFC Editor or the Internet-Drafts function

Statements made outside of an IETF meeting, mailing list or other
function, that are clearly not intended to be input to an IETF
activity, group or function, are not subject to these provisions.

   
***************************************************************************


If you have questions, problems, comments, etc, send them to
mailman-owner@www1.ietf.org.  Thanks!

Passwords for sigtran-web-archive@optimus.ietf.org:

List                                     Password // URL
----                                     --------  
sigtran@ietf.org                         emupuc    
https://www1.ietf.org/mailman/options/sigtran/sigtran-web-archive%40optimus.ietf.org



From exim@www1.ietf.org  Thu Jul  3 23:13:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA01454
	for <sigtran-archive@odin.ietf.org>; Thu, 3 Jul 2003 23:13:33 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YH0T-0002nk-5L
	for sigtran-archive@odin.ietf.org; Thu, 03 Jul 2003 23:13:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h643D5em010767
	for sigtran-archive@odin.ietf.org; Thu, 3 Jul 2003 23:13:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YH0O-0002nQ-Ko; Thu, 03 Jul 2003 23:13:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YH04-0002n0-A0
	for sigtran@optimus.ietf.org; Thu, 03 Jul 2003 23:12:40 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA01447
	for <sigtran@ietf.org>; Thu, 3 Jul 2003 23:12:37 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YH01-0000wb-00
	for sigtran@ietf.org; Thu, 03 Jul 2003 23:12:37 -0400
Received: from auds953.usa.alcatel.com ([143.209.238.6])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YH00-0000wJ-00
	for sigtran@ietf.org; Thu, 03 Jul 2003 23:12:36 -0400
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h643C4LX015072
	for <sigtran@ietf.org>; Thu, 3 Jul 2003 22:12:04 -0500 (CDT)
Received: from sun2594.ssd.usa.alcatel.com (sun2594.ssd.usa.alcatel.com [143.209.250.252])
	by ssd.usa.alcatel.com (8.12.8/8.12.8) with ESMTP id h643Bx1C004888
	for <sigtran@ietf.org>; Thu, 3 Jul 2003 22:11:59 -0500 (CDT)
Date: Thu, 3 Jul 2003 22:11:59 -0500 (CDT)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: RE: [Sigtran] Re: M3UA MIB Draft05: Loadsharing on CIC ,SSN
In-Reply-To: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CD1@eestqnt105.es.eu.ericsson.se>
Message-ID: <Pine.SOL.4.10.10307032203570.2696-100000@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>


Pastor,

 In MIB-04 section "4.  Definitions"
 
       m3uaSpLshMethod OBJECT-TYPE
          SYNTAX      INTEGER {
                          none(0),    -- Random
                          sls(1),     -- Basing on SLS
                          cic(2),     -- Basing on CIC
                          ssn(3),     -- Basing on SSN
                          dpg(4),     -- Basing on DPC
                          other(5)    -- Other method
                      }
          MAX-ACCESS  read-write
          STATUS      current
          DESCRIPTION


Here i think CIC and SSN are agreed to be removed.My question is 
related to "dpg(4),     -- Basing on DPC".
How does a node loadshare its traffice based on DPC?
I think this concept also should be removed from MIB-04.

--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, 3 Jul 2003, Javier Pastor-Balbas (ECE) wrote:

> Samuel, 
> 
> Are you referring to any particular object within MIB-04? I mean, what Anjali was referring to were two particular objects that not include that kind of loadsharing...
> 
> 
> Thanks // Javier.
> 
> 
> > -----Original Message-----
> > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > Sent: jueves, 03 de julio de 2003 13:23
> > To: J.Javier Pastor (EE/EEM)
> > Cc: 'sigtran@ietf.org'
> > Subject: Re: [Sigtran] Re: M3UA MIB Draft05: Loadsharing on CIC ,SSN
> > 
> > 
> > 
> > Pastor,
> > 
> >   Going back to old emails ...on 11-Apr-2003 i asked
> >   'What is loadsharing based on DPC?'
> > 
> >   Nobody gave proper reply to this email.
> >   Still i dont undrstand it.If anybody is there to explain what is it?
> >   otherwise we will delete this from MIB-04.
> > 
> > Following is the email which i sent  on 11-Apr-2003.
> > 
> > ----------------------------------------------
> > Date: Fri, 11 Apr 2003 06:09:28 -0500 (CDT)
> > From: Samuel Dur D. Jeyaseelan <sjeyasee@ssd.usa.alcatel.com>
> > To: sigtran@ietf.org
> > Subject: [m3ua]-ldshar basing on dpc
> > ---------------------------------------------------
> > 
> > 
> > --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, 3 Jul 2003, Javier Pastor-Balbas (ECE) wrote:
> > 
> > > 
> > > Hi Anjali, 
> > > 
> > > It seems that it makes sense to remove those from the 
> > standard MIB. All the people who are interested in having 
> > this object can create it within their private MIBs. We may 
> > keep SLS since it is a common practise in SS7 unless you have 
> > any objection.
> > > 
> > > 
> > > Thanks for your responses // Javier.
> > > 
> > > PS: Samuel: What object do you refer when talking about 
> > DPC-based loadsharing within the current MIB? 
> > > 
> > > 
> > > > -----Original Message-----
> > > > From: agurmukhani@hss.hns.com [mailto:agurmukhani@hss.hns.com]
> > > > Sent: jueves, 03 de julio de 2003 6:54
> > > > To: J.Javier Pastor (EE/EEM)
> > > > Cc: sigtran@ietf.org
> > > > Subject: M3UA MIB Draft05: Loadsharing on CIC ,SSN
> > > > 
> > > > 
> > > > 
> > > > 
> > > > 
> > > > 
> > > > Hi...
> > > > 
> > > > 
> > > > In my understanding and  as per the discussion and consensus 
> > > > on the mailing
> > > > list , everyone agreed that  CIC and SSN should not be used 
> > > > as loadsharing
> > > > basis
> > > > be it within streams in a particular SCTP association or 
> > for active
> > > > signalling processes.
> > > > 
> > > > So would these be removed from  the m3uaStreamLshMethod  and
> > > > m3uaSpLshMethod OBJECT-TYPE s in M3UA MIB draft.
> > > > 
> > > > 
> > > > Thanks and Regards
> > > > Anjali
> > > > www.hssworld.com
> > > > Hughes Software Systems.
> > > > 
> > > > "DISCLAIMER: This message is proprietary to Hughes Software 
> > > > Systems Limited
> > > > (HSS) and is intended solely for the use of the individual to 
> > > > whom it is
> > > > addressed. It may contain  privileged or confidential 
> > information and
> > > > should not be circulated or used for any purpose other than 
> > > > for what it is
> > > > intended. If you have received this message in error, please 
> > > > notify the
> > > > originator immediately. If you are not the intended 
> > recipient, you are
> > > > notified that you are strictly prohibited from using, 
> > > > copying, altering, or
> > > > disclosing the contents of this message. HSS accepts no 
> > > > responsibility for
> > > > loss or damage arising from the use of the information 
> > > > transmitted by this
> > > > email including damage from virus."
> > > > 
> > > 
> > > _______________________________________________
> > > 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 exim@www1.ietf.org  Thu Jul  3 23:29:32 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA01841
	for <sigtran-archive@odin.ietf.org>; Thu, 3 Jul 2003 23:29:32 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YHFv-0003hT-RP
	for sigtran-archive@odin.ietf.org; Thu, 03 Jul 2003 23:29:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h643T3ta014221
	for sigtran-archive@odin.ietf.org; Thu, 3 Jul 2003 23:29:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YHFt-0003h8-CE; Thu, 03 Jul 2003 23:29:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YHFH-0003eU-Nj
	for sigtran@optimus.ietf.org; Thu, 03 Jul 2003 23:28:23 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA01819
	for <sigtran@ietf.org>; Thu, 3 Jul 2003 23:28:21 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YHFF-00017u-00
	for sigtran@ietf.org; Thu, 03 Jul 2003 23:28:21 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[KXanRFMmz/IPFzUDReVRjZx6NO60THaE])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YHFE-00017k-00
	for sigtran@ietf.org; Thu, 03 Jul 2003 23:28:20 -0400
Received: from ns.pigworks.openss7.net (IDENT:OeCw/ney/KDI846TMu9dHG99M3Swiyf/@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h643SEX21489;
	Thu, 3 Jul 2003 21:28:14 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h643SDF17940;
	Thu, 3 Jul 2003 21:28:13 -0600
Date: Thu, 3 Jul 2003 21:28:13 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: agurmukhani@hss.hns.com
Cc: javier.pastor-balbas@ece.ericsson.se, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
Message-ID: <20030703212813.F17069@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: agurmukhani@hss.hns.com,
	javier.pastor-balbas@ece.ericsson.se, sigtran@ietf.org
References: <OFEB796ECC.A92CCF1D-ON65256D58.00195CC7@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: <OFEB796ECC.A92CCF1D-ON65256D58.00195CC7@hss.hns.com>; from agurmukhani@hss.hns.com on Thu, Jul 03, 2003 at 10:23:35AM +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>

agurmukhani,

I think that they should be removed from the MIB, but we did NOT agree
that CIC or SSN should not be used for loadsharing.  Any criteria that
the SG/ASP desires to use for loadsharing being congnisant of the need
for ordering of user part messages is allowed for loadsharing.  We DID
agree to remove CIC, SSN from the Routing Key, because a rational and
interoperable basis for splitting CICs and SSNs across AS could not be
decided.

The meanse oadsharing, on the other hand, is not specified in M3UA.
Some examples are given, but no requirements.  Some things that could
clearly be used are SLS, DPC/SLS, OPC/DPC/SLS, CIC, DPC/CIC, OPC/DPC/CIC,
OPC/DPC/CIC/SLS, SSN, ....  and other permutations of these and even
higher layer parameters such as TID.

My argument for removing them from the MIB is not that they cannot be
used, but that they have not been standardized.

--brian

agurmukhani@hss.hns.com wrote:                                                 (Thu, 03 Jul 2003 10:23:35)
> 
> 
> 
> 
> Hi...
> 
> 
> In my understanding and  as per the discussion and consensus on the mailing
> list , everyone agreed that  CIC and SSN should not be used as loadsharing
> basis
> be it within streams in a particular SCTP association or for active
> signalling processes.
> 
> So would these be removed from  the m3uaStreamLshMethod  and
> m3uaSpLshMethod OBJECT-TYPE s in M3UA MIB draft.
> 
> 
> Thanks and Regards
> Anjali
> www.hssworld.com
> Hughes Software Systems.
> 
> "DISCLAIMER: This message is proprietary to Hughes Software Systems Limited
> (HSS) and is intended solely for the use of the individual to whom it is
> addressed. It may contain  privileged or confidential information and
> should not be circulated or used for any purpose other than for what it is
> intended. If you have received this message in error, please notify the
> originator immediately. If you are not the intended recipient, you are
> notified that you are strictly prohibited from using, copying, altering, or
> disclosing the contents of this message. HSS accepts no responsibility for
> loss or damage arising from the use of the information transmitted by this
> email including damage from virus."
> 
> 
> _______________________________________________
> 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 exim@www1.ietf.org  Fri Jul  4 00:01:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA02424
	for <sigtran-archive@odin.ietf.org>; Fri, 4 Jul 2003 00:01:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YHkt-00067b-MJ
	for sigtran-archive@odin.ietf.org; Fri, 04 Jul 2003 00:01:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h64413dL023511
	for sigtran-archive@odin.ietf.org; Fri, 4 Jul 2003 00:01:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YHkr-00066Z-Mo; Fri, 04 Jul 2003 00:01:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YHkU-00065u-To
	for sigtran@optimus.ietf.org; Fri, 04 Jul 2003 00:00:38 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA02392
	for <sigtran@ietf.org>; Fri, 4 Jul 2003 00:00:36 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YHkS-0001QC-00
	for sigtran@ietf.org; Fri, 04 Jul 2003 00:00:36 -0400
Received: from auds953.usa.alcatel.com ([143.209.238.6])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YHkR-0001Pa-00
	for sigtran@ietf.org; Fri, 04 Jul 2003 00:00:35 -0400
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h64405LX017247
	for <sigtran@ietf.org>; Thu, 3 Jul 2003 23:00:05 -0500 (CDT)
Received: from sun2594.ssd.usa.alcatel.com (sun2594.ssd.usa.alcatel.com [143.209.250.252])
	by ssd.usa.alcatel.com (8.12.8/8.12.8) with ESMTP id h643xp1C010467
	for <sigtran@ietf.org>; Thu, 3 Jul 2003 22:59:51 -0500 (CDT)
Date: Thu, 3 Jul 2003 22:59:51 -0500 (CDT)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
In-Reply-To: <20030703212813.F17069@openss7.org>
Message-ID: <Pine.SOL.4.10.10307032245400.2696-100000@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>


 this is just an info regarding CIC and SSN - load-sharing.

 A node can load-share its traffic based on CIC or SSN with the
 help of static configuration but this is outside of the scope of M3UA
 rfc.When an ASP vendor and SGP vendor agrees to do load-sharing based on
 CIC or SSN with static confiuration they are welcome to do this,
 since this has not been standardised, taking this to interop will
 lead to problems.So almost everybody agrees to delete CIC and SSN
 loadsharing concept from MIB-04.


--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, 3 Jul 2003, Brian F. G. Bidulock wrote:

> agurmukhani,
> 
> I think that they should be removed from the MIB, but we did NOT agree
> that CIC or SSN should not be used for loadsharing.  Any criteria that
> the SG/ASP desires to use for loadsharing being congnisant of the need
> for ordering of user part messages is allowed for loadsharing.  We DID
> agree to remove CIC, SSN from the Routing Key, because a rational and
> interoperable basis for splitting CICs and SSNs across AS could not be
> decided.
> 
> The meanse oadsharing, on the other hand, is not specified in M3UA.
> Some examples are given, but no requirements.  Some things that could
> clearly be used are SLS, DPC/SLS, OPC/DPC/SLS, CIC, DPC/CIC, OPC/DPC/CIC,
> OPC/DPC/CIC/SLS, SSN, ....  and other permutations of these and even
> higher layer parameters such as TID.
> 
> My argument for removing them from the MIB is not that they cannot be
> used, but that they have not been standardized.
> 
> --brian
> 
> agurmukhani@hss.hns.com wrote:                                                 (Thu, 03 Jul 2003 10:23:35)
> > 
> > 
> > 
> > 
> > Hi...
> > 
> > 
> > In my understanding and  as per the discussion and consensus on the mailing
> > list , everyone agreed that  CIC and SSN should not be used as loadsharing
> > basis
> > be it within streams in a particular SCTP association or for active
> > signalling processes.
> > 
> > So would these be removed from  the m3uaStreamLshMethod  and
> > m3uaSpLshMethod OBJECT-TYPE s in M3UA MIB draft.
> > 
> > 
> > Thanks and Regards
> > Anjali
> > www.hssworld.com
> > Hughes Software Systems.
> > 
> > "DISCLAIMER: This message is proprietary to Hughes Software Systems Limited
> > (HSS) and is intended solely for the use of the individual to whom it is
> > addressed. It may contain  privileged or confidential information and
> > should not be circulated or used for any purpose other than for what it is
> > intended. If you have received this message in error, please notify the
> > originator immediately. If you are not the intended recipient, you are
> > notified that you are strictly prohibited from using, copying, altering, or
> > disclosing the contents of this message. HSS accepts no responsibility for
> > loss or damage arising from the use of the information transmitted by this
> > email including damage from virus."
> > 
> > 
> > _______________________________________________
> > 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 exim@www1.ietf.org  Fri Jul  4 00:33:36 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA03421
	for <sigtran-archive@odin.ietf.org>; Fri, 4 Jul 2003 00:33:36 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YIFw-0008Ac-IQ
	for sigtran-archive@odin.ietf.org; Fri, 04 Jul 2003 00:33:08 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h644X87r031402
	for sigtran-archive@odin.ietf.org; Fri, 4 Jul 2003 00:33:08 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YIFp-0008AA-81; Fri, 04 Jul 2003 00:33:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YIFF-00085I-Dv
	for sigtran@optimus.ietf.org; Fri, 04 Jul 2003 00:32:25 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA03399
	for <sigtran@ietf.org>; Fri, 4 Jul 2003 00:32:22 -0400 (EDT)
From: kakaul@hss.hns.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YIFC-0001nQ-00
	for sigtran@ietf.org; Fri, 04 Jul 2003 00:32:22 -0400
Received: from hindon.hss.co.in ([202.54.26.202])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YIFA-0001my-00
	for sigtran@ietf.org; Fri, 04 Jul 2003 00:32:21 -0400
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 h644Sx929470
	for <sigtran@ietf.org>; Fri, 4 Jul 2003 09:58:59 +0530 (IST)
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h644SvO29459;
	Fri, 4 Jul 2003 09:58:58 +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 ESMTP id h644WZT14398;
	Fri, 4 Jul 2003 10:02:36 +0530 (IST)
Subject: Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.5  September 22, 2000
Message-ID: <OF1BE595EA.64A3471E-ON65256D59.0017A7D6@hss.hns.com>
Date: Fri, 4 Jul 2003 09:57:16 +0530
X-MIMETrack: Serialize by Router on Sandesh/HSS(Release 6.0|September 26, 2002) at 04/07/2003
 09:56:13 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>





Hi Samule,
  Does MTP3 use CIC or SSN for load-distribution among the available
links.I guess only SLS parameter in MTP3 is used for load-distribution
among the available link.I want to know how an MTP3  user (ISUP or SCCP)
selects SLS for effective load distribution at MTP3 layer among the links
available to reach the  a given DPC.

Regards
Kamesh Kaul
www.hssworld.com
Hughes Software Systems




"Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
07/04/2003 09:29:51 AM

Sent by:    sigtran-admin@ietf.org


To:    sigtran@ietf.org
cc:

Subject:    Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN



 this is just an info regarding CIC and SSN - load-sharing.

 A node can load-share its traffic based on CIC or SSN with the
 help of static configuration but this is outside of the scope of M3UA
 rfc.When an ASP vendor and SGP vendor agrees to do load-sharing based on
 CIC or SSN with static confiuration they are welcome to do this,
 since this has not been standardised, taking this to interop will
 lead to problems.So almost everybody agrees to delete CIC and SSN
 loadsharing concept from MIB-04.


--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, 3 Jul 2003, Brian F. G. Bidulock wrote:

> agurmukhani,
>
> I think that they should be removed from the MIB, but we did NOT agree
> that CIC or SSN should not be used for loadsharing.  Any criteria that
> the SG/ASP desires to use for loadsharing being congnisant of the need
> for ordering of user part messages is allowed for loadsharing.  We DID
> agree to remove CIC, SSN from the Routing Key, because a rational and
> interoperable basis for splitting CICs and SSNs across AS could not be
> decided.
>
> The meanse oadsharing, on the other hand, is not specified in M3UA.
> Some examples are given, but no requirements.  Some things that could
> clearly be used are SLS, DPC/SLS, OPC/DPC/SLS, CIC, DPC/CIC, OPC/DPC/CIC,
> OPC/DPC/CIC/SLS, SSN, ....  and other permutations of these and even
> higher layer parameters such as TID.
>
> My argument for removing them from the MIB is not that they cannot be
> used, but that they have not been standardized.
>
> --brian
>
> agurmukhani@hss.hns.com wrote:
(Thu, 03 Jul 2003 10:23:35)
> >
> >
> >
> >
> > Hi...
> >
> >
> > In my understanding and  as per the discussion and consensus on the
mailing
> > list , everyone agreed that  CIC and SSN should not be used as
loadsharing
> > basis
> > be it within streams in a particular SCTP association or for active
> > signalling processes.
> >
> > So would these be removed from  the m3uaStreamLshMethod  and
> > m3uaSpLshMethod OBJECT-TYPE s in M3UA MIB draft.
> >
> >
> > Thanks and Regards
> > Anjali
> > www.hssworld.com
> > Hughes Software Systems.
> >
> > "DISCLAIMER: This message is proprietary to Hughes Software Systems
Limited
> > (HSS) and is intended solely for the use of the individual to whom it
is
> > addressed. It may contain  privileged or confidential information and
> > should not be circulated or used for any purpose other than for what it
is
> > intended. If you have received this message in error, please notify the
> > originator immediately. If you are not the intended recipient, you are
> > notified that you are strictly prohibited from using, copying,
altering, or
> > disclosing the contents of this message. HSS accepts no responsibility
for
> > loss or damage arising from the use of the information transmitted by
this
> > email including damage from virus."
> >
> >
> > _______________________________________________
> > 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 exim@www1.ietf.org  Fri Jul  4 00:49:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA02425
	for <sigtran-archive@odin.ietf.org>; Fri, 4 Jul 2003 00:01:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YHkt-00067X-MQ
	for sigtran-archive@odin.ietf.org; Fri, 04 Jul 2003 00:01:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h64413T0023510
	for sigtran-archive@odin.ietf.org; Fri, 4 Jul 2003 00:01:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YHks-00066h-3w; Fri, 04 Jul 2003 00:01:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YHkl-000664-Tv
	for sigtran@optimus.ietf.org; Fri, 04 Jul 2003 00:00:55 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA02399
	for <sigtran@ietf.org>; Fri, 4 Jul 2003 00:00:53 -0400 (EDT)
From: kakaul@hss.hns.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YHkj-0001QS-00
	for sigtran@ietf.org; Fri, 04 Jul 2003 00:00:53 -0400
Received: from hindon.hss.co.in ([202.54.26.202])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YHkh-0001Q9-00
	for sigtran@ietf.org; Fri, 04 Jul 2003 00:00:52 -0400
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 h643va523801
	for <sigtran@ietf.org>; Fri, 4 Jul 2003 09:27:37 +0530 (IST)
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h643vXO23788;
	Fri, 4 Jul 2003 09:27:33 +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 ESMTP id h6441BV11519;
	Fri, 4 Jul 2003 09:31:11 +0530 (IST)
Subject: RE: [Sigtran] Re: M3UA MIB Draft05: Loadsharing on CIC ,SSN
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.5  September 22, 2000
Message-ID: <OFC0C51E96.0DE6B8CD-ON65256D59.00152E89@hss.hns.com>
Date: Fri, 4 Jul 2003 09:25:51 +0530
X-MIMETrack: Serialize by Router on Sandesh/HSS(Release 6.0|September 26, 2002) at 04/07/2003
 09:24:48 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>





Hi Samule,
  dpg(4),     -- Basing on DPC     was present in MIB 04 but has been
removed from MIB05 as no body on the list proposed a fruitful loadsharing
scheme based on DPC.

Regards
Kamesh Kaul
www.hssworld.com
Hughes Software Systems




"Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
07/04/2003 08:41:59 AM

Sent by:    sigtran-admin@ietf.org


To:    "'sigtran@ietf.org'" <sigtran@ietf.org>
cc:

Subject:    RE: [Sigtran] Re: M3UA MIB Draft05: Loadsharing on CIC ,SSN



Pastor,

 In MIB-04 section "4.  Definitions"

       m3uaSpLshMethod OBJECT-TYPE
          SYNTAX      INTEGER {
                          none(0),    -- Random
                          sls(1),     -- Basing on SLS
                          cic(2),     -- Basing on CIC
                          ssn(3),     -- Basing on SSN
                          dpg(4),     -- Basing on DPC
                          other(5)    -- Other method
                      }
          MAX-ACCESS  read-write
          STATUS      current
          DESCRIPTION


Here i think CIC and SSN are agreed to be removed.My question is
related to "dpg(4),     -- Basing on DPC".
How does a node loadshare its traffice based on DPC?
I think this concept also should be removed from MIB-04.

--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, 3 Jul 2003, Javier Pastor-Balbas (ECE) wrote:

> Samuel,
>
> Are you referring to any particular object within MIB-04? I mean, what
Anjali was referring to were two particular objects that not include that
kind of loadsharing...
>
>
> Thanks // Javier.
>
>
> > -----Original Message-----
> > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > Sent: jueves, 03 de julio de 2003 13:23
> > To: J.Javier Pastor (EE/EEM)
> > Cc: 'sigtran@ietf.org'
> > Subject: Re: [Sigtran] Re: M3UA MIB Draft05: Loadsharing on CIC ,SSN
> >
> >
> >
> > Pastor,
> >
> >   Going back to old emails ...on 11-Apr-2003 i asked
> >   'What is loadsharing based on DPC?'
> >
> >   Nobody gave proper reply to this email.
> >   Still i dont undrstand it.If anybody is there to explain what is it?
> >   otherwise we will delete this from MIB-04.
> >
> > Following is the email which i sent  on 11-Apr-2003.
> >
> > ----------------------------------------------
> > Date: Fri, 11 Apr 2003 06:09:28 -0500 (CDT)
> > From: Samuel Dur D. Jeyaseelan <sjeyasee@ssd.usa.alcatel.com>
> > To: sigtran@ietf.org
> > Subject: [m3ua]-ldshar basing on dpc
> > ---------------------------------------------------
> >
> >
> > --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, 3 Jul 2003, Javier Pastor-Balbas (ECE) wrote:
> >
> > >
> > > Hi Anjali,
> > >
> > > It seems that it makes sense to remove those from the
> > standard MIB. All the people who are interested in having
> > this object can create it within their private MIBs. We may
> > keep SLS since it is a common practise in SS7 unless you have
> > any objection.
> > >
> > >
> > > Thanks for your responses // Javier.
> > >
> > > PS: Samuel: What object do you refer when talking about
> > DPC-based loadsharing within the current MIB?
> > >
> > >
> > > > -----Original Message-----
> > > > From: agurmukhani@hss.hns.com [mailto:agurmukhani@hss.hns.com]
> > > > Sent: jueves, 03 de julio de 2003 6:54
> > > > To: J.Javier Pastor (EE/EEM)
> > > > Cc: sigtran@ietf.org
> > > > Subject: M3UA MIB Draft05: Loadsharing on CIC ,SSN
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > > Hi...
> > > >
> > > >
> > > > In my understanding and  as per the discussion and consensus
> > > > on the mailing
> > > > list , everyone agreed that  CIC and SSN should not be used
> > > > as loadsharing
> > > > basis
> > > > be it within streams in a particular SCTP association or
> > for active
> > > > signalling processes.
> > > >
> > > > So would these be removed from  the m3uaStreamLshMethod  and
> > > > m3uaSpLshMethod OBJECT-TYPE s in M3UA MIB draft.
> > > >
> > > >
> > > > Thanks and Regards
> > > > Anjali
> > > > www.hssworld.com
> > > > Hughes Software Systems.
> > > >
> > > > "DISCLAIMER: This message is proprietary to Hughes Software
> > > > Systems Limited
> > > > (HSS) and is intended solely for the use of the individual to
> > > > whom it is
> > > > addressed. It may contain  privileged or confidential
> > information and
> > > > should not be circulated or used for any purpose other than
> > > > for what it is
> > > > intended. If you have received this message in error, please
> > > > notify the
> > > > originator immediately. If you are not the intended
> > recipient, you are
> > > > notified that you are strictly prohibited from using,
> > > > copying, altering, or
> > > > disclosing the contents of this message. HSS accepts no
> > > > responsibility for
> > > > loss or damage arising from the use of the information
> > > > transmitted by this
> > > > email including damage from virus."
> > > >
> > >
> > > _______________________________________________
> > > 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 exim@www1.ietf.org  Fri Jul  4 01:30:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA04492
	for <sigtran-archive@odin.ietf.org>; Fri, 4 Jul 2003 01:30:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YJ90-0003qj-So
	for sigtran-archive@odin.ietf.org; Fri, 04 Jul 2003 01:30:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h645U2TB014791
	for sigtran-archive@odin.ietf.org; Fri, 4 Jul 2003 01:30:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YJ8z-0003qK-Qr; Fri, 04 Jul 2003 01:30:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YJ8d-0003pd-QI
	for sigtran@optimus.ietf.org; Fri, 04 Jul 2003 01:29:39 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA04489
	for <sigtran@ietf.org>; Fri, 4 Jul 2003 01:29:37 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YJ8a-0002GB-00
	for sigtran@ietf.org; Fri, 04 Jul 2003 01:29:36 -0400
Received: from auds951.usa.alcatel.com ([143.209.238.80])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YJ8Z-0002G1-00
	for sigtran@ietf.org; Fri, 04 Jul 2003 01:29:35 -0400
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h645T6wA000688
	for <sigtran@ietf.org>; Fri, 4 Jul 2003 00:29:06 -0500 (CDT)
Received: from sun2594.ssd.usa.alcatel.com (sun2594.ssd.usa.alcatel.com [143.209.250.252])
	by ssd.usa.alcatel.com (8.12.8/8.12.8) with ESMTP id h645SE1C001556
	for <sigtran@ietf.org>; Fri, 4 Jul 2003 00:28:14 -0500 (CDT)
Date: Fri, 4 Jul 2003 00:28:14 -0500 (CDT)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
In-Reply-To: <OF1BE595EA.64A3471E-ON65256D59.0017A7D6@hss.hns.com>
Message-ID: <Pine.SOL.4.10.10307040006200.2696-100000@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>


Pastor,

 MIB-05 does not have 'load-shar on dpc',i was referring mib-04 which lead
 me into confusion.
 
 Thanks kakaul for our mail.

Kakaul,

 As you said MTP3 does not route messages based on CIC or SSN because
 there are no standards available, i agree with it.

 Have a look, suppose SG receives an MSU with routing indicator="Route on
 PointCode/SSN", if your SG supports static configuration then it is easy
 to route these messages to the corresponding subSystems based on
 statically configured SSN-ASP table the same way CIC-ASP table can be
 used to route ISUP messages to the corresponding MGCs.Basically CIC was
 removed from RFC to avoid complex problems say a msg comes with CIC
 range,this CIC range can be split across MGCs here the problem is
 potential.

 These are the reasons caused me to say that with static configuration
 CIC or/and SSN can be supported if an SG vendor and ASP vendor has
 mutual understanding.(but this is outside the scope of M3UA RFC)

--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, 4 Jul 2003 kakaul@hss.hns.com wrote:

> 
> 
> 
> 
> Hi Samule,
>   Does MTP3 use CIC or SSN for load-distribution among the available
> links.I guess only SLS parameter in MTP3 is used for load-distribution
> among the available link.I want to know how an MTP3  user (ISUP or SCCP)
> selects SLS for effective load distribution at MTP3 layer among the links
> available to reach the  a given DPC.
> 
> Regards
> Kamesh Kaul
> www.hssworld.com
> Hughes Software Systems
> 
> 
> 
> 
> "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
> 07/04/2003 09:29:51 AM
> 
> Sent by:    sigtran-admin@ietf.org
> 
> 
> To:    sigtran@ietf.org
> cc:
> 
> Subject:    Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
> 
> 
> 
>  this is just an info regarding CIC and SSN - load-sharing.
> 
>  A node can load-share its traffic based on CIC or SSN with the
>  help of static configuration but this is outside of the scope of M3UA
>  rfc.When an ASP vendor and SGP vendor agrees to do load-sharing based on
>  CIC or SSN with static confiuration they are welcome to do this,
>  since this has not been standardised, taking this to interop will
>  lead to problems.So almost everybody agrees to delete CIC and SSN
>  loadsharing concept from MIB-04.
> 
> 
> --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, 3 Jul 2003, Brian F. G. Bidulock wrote:
> 
> > agurmukhani,
> >
> > I think that they should be removed from the MIB, but we did NOT agree
> > that CIC or SSN should not be used for loadsharing.  Any criteria that
> > the SG/ASP desires to use for loadsharing being congnisant of the need
> > for ordering of user part messages is allowed for loadsharing.  We DID
> > agree to remove CIC, SSN from the Routing Key, because a rational and
> > interoperable basis for splitting CICs and SSNs across AS could not be
> > decided.
> >
> > The meanse oadsharing, on the other hand, is not specified in M3UA.
> > Some examples are given, but no requirements.  Some things that could
> > clearly be used are SLS, DPC/SLS, OPC/DPC/SLS, CIC, DPC/CIC, OPC/DPC/CIC,
> > OPC/DPC/CIC/SLS, SSN, ....  and other permutations of these and even
> > higher layer parameters such as TID.
> >
> > My argument for removing them from the MIB is not that they cannot be
> > used, but that they have not been standardized.
> >
> > --brian
> >
> > agurmukhani@hss.hns.com wrote:
> (Thu, 03 Jul 2003 10:23:35)
> > >
> > >
> > >
> > >
> > > Hi...
> > >
> > >
> > > In my understanding and  as per the discussion and consensus on the
> mailing
> > > list , everyone agreed that  CIC and SSN should not be used as
> loadsharing
> > > basis
> > > be it within streams in a particular SCTP association or for active
> > > signalling processes.
> > >
> > > So would these be removed from  the m3uaStreamLshMethod  and
> > > m3uaSpLshMethod OBJECT-TYPE s in M3UA MIB draft.
> > >
> > >
> > > Thanks and Regards
> > > Anjali
> > > www.hssworld.com
> > > Hughes Software Systems.
> > >
> > > "DISCLAIMER: This message is proprietary to Hughes Software Systems
> Limited
> > > (HSS) and is intended solely for the use of the individual to whom it
> is
> > > addressed. It may contain  privileged or confidential information and
> > > should not be circulated or used for any purpose other than for what it
> is
> > > intended. If you have received this message in error, please notify the
> > > originator immediately. If you are not the intended recipient, you are
> > > notified that you are strictly prohibited from using, copying,
> altering, or
> > > disclosing the contents of this message. HSS accepts no responsibility
> for
> > > loss or damage arising from the use of the information transmitted by
> this
> > > email including damage from virus."
> > >
> > >
> > > _______________________________________________
> > > 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 exim@www1.ietf.org  Fri Jul  4 02:45:32 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA17998
	for <sigtran-archive@odin.ietf.org>; Fri, 4 Jul 2003 02:45:32 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YKJc-00018p-KJ
	for sigtran-archive@odin.ietf.org; Fri, 04 Jul 2003 02:45:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h646j4Dr004388
	for sigtran-archive@odin.ietf.org; Fri, 4 Jul 2003 02:45:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YKJZ-00018S-TY; Fri, 04 Jul 2003 02:45:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YKIj-00017k-NS
	for sigtran@optimus.ietf.org; Fri, 04 Jul 2003 02:44:09 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA17927
	for <sigtran@ietf.org>; Fri, 4 Jul 2003 02:44:06 -0400 (EDT)
From: kakaul@hss.hns.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YKIg-0002gM-00
	for sigtran@ietf.org; Fri, 04 Jul 2003 02:44:06 -0400
Received: from hindon.hss.co.in ([202.54.26.202])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YKId-0002g5-00
	for sigtran@ietf.org; Fri, 04 Jul 2003 02:44:04 -0400
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 h646ees27615
	for <sigtran@ietf.org>; Fri, 4 Jul 2003 12:10:41 +0530 (IST)
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h646eWO27587;
	Fri, 4 Jul 2003 12:10:35 +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 ESMTP id h646i5e28122;
	Fri, 4 Jul 2003 12:14:07 +0530 (IST)
Subject: Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.5  September 22, 2000
Message-ID: <OF523A56B2.94AAE275-ON65256D59.00245419@hss.hns.com>
Date: Fri, 4 Jul 2003 12:08:48 +0530
X-MIMETrack: Serialize by Router on Sandesh/HSS(Release 6.0|September 26, 2002) at 04/07/2003
 12:07:50 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>





Hi Samule,
  You have not replied for my second question :
I want to know how an MTP3  user (ISUP or SCCP) selects SLS for effective
load distribution at MTP3 layer among the links available to reach a given
DPC. Is it that SLS is chosen per DPC for a logical session or globally per
each logical session irrespective of DPC for which this session is
intended.I know it is implementation dependent option but what do most of
the implementers opt for. By logical session i mean logically related data.

Regards
Kamesh Kaul
www.hssworld.com
Hughes Software Systems





"Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
07/04/2003 10:58:14 AM

Sent by:    sigtran-admin@ietf.org


To:    sigtran@ietf.org
cc:

Subject:    Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN



Pastor,

 MIB-05 does not have 'load-shar on dpc',i was referring mib-04 which lead
 me into confusion.

 Thanks kakaul for our mail.

Kakaul,

 As you said MTP3 does not route messages based on CIC or SSN because
 there are no standards available, i agree with it.

 Have a look, suppose SG receives an MSU with routing indicator="Route on
 PointCode/SSN", if your SG supports static configuration then it is easy
 to route these messages to the corresponding subSystems based on
 statically configured SSN-ASP table the same way CIC-ASP table can be
 used to route ISUP messages to the corresponding MGCs.Basically CIC was
 removed from RFC to avoid complex problems say a msg comes with CIC
 range,this CIC range can be split across MGCs here the problem is
 potential.

 These are the reasons caused me to say that with static configuration
 CIC or/and SSN can be supported if an SG vendor and ASP vendor has
 mutual understanding.(but this is outside the scope of M3UA RFC)

--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, 4 Jul 2003 kakaul@hss.hns.com wrote:

>
>
>
>
> Hi Samule,
>   Does MTP3 use CIC or SSN for load-distribution among the available
> links.I guess only SLS parameter in MTP3 is used for load-distribution
> among the available link.I want to know how an MTP3  user (ISUP or SCCP)
> selects SLS for effective load distribution at MTP3 layer among the links
> available to reach the  a given DPC.
>
> Regards
> Kamesh Kaul
> www.hssworld.com
> Hughes Software Systems
>
>
>
>
> "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
> 07/04/2003 09:29:51 AM
>
> Sent by:    sigtran-admin@ietf.org
>
>
> To:    sigtran@ietf.org
> cc:
>
> Subject:    Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
>
>
>
>  this is just an info regarding CIC and SSN - load-sharing.
>
>  A node can load-share its traffic based on CIC or SSN with the
>  help of static configuration but this is outside of the scope of M3UA
>  rfc.When an ASP vendor and SGP vendor agrees to do load-sharing based on
>  CIC or SSN with static confiuration they are welcome to do this,
>  since this has not been standardised, taking this to interop will
>  lead to problems.So almost everybody agrees to delete CIC and SSN
>  loadsharing concept from MIB-04.
>
>
> --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, 3 Jul 2003, Brian F. G. Bidulock wrote:
>
> > agurmukhani,
> >
> > I think that they should be removed from the MIB, but we did NOT agree
> > that CIC or SSN should not be used for loadsharing.  Any criteria that
> > the SG/ASP desires to use for loadsharing being congnisant of the need
> > for ordering of user part messages is allowed for loadsharing.  We DID
> > agree to remove CIC, SSN from the Routing Key, because a rational and
> > interoperable basis for splitting CICs and SSNs across AS could not be
> > decided.
> >
> > The meanse oadsharing, on the other hand, is not specified in M3UA.
> > Some examples are given, but no requirements.  Some things that could
> > clearly be used are SLS, DPC/SLS, OPC/DPC/SLS, CIC, DPC/CIC,
OPC/DPC/CIC,
> > OPC/DPC/CIC/SLS, SSN, ....  and other permutations of these and even
> > higher layer parameters such as TID.
> >
> > My argument for removing them from the MIB is not that they cannot be
> > used, but that they have not been standardized.
> >
> > --brian
> >
> > agurmukhani@hss.hns.com wrote:
> (Thu, 03 Jul 2003 10:23:35)
> > >
> > >
> > >
> > >
> > > Hi...
> > >
> > >
> > > In my understanding and  as per the discussion and consensus on the
> mailing
> > > list , everyone agreed that  CIC and SSN should not be used as
> loadsharing
> > > basis
> > > be it within streams in a particular SCTP association or for active
> > > signalling processes.
> > >
> > > So would these be removed from  the m3uaStreamLshMethod  and
> > > m3uaSpLshMethod OBJECT-TYPE s in M3UA MIB draft.
> > >
> > >
> > > Thanks and Regards
> > > Anjali
> > > www.hssworld.com
> > > Hughes Software Systems.
> > >
> > > "DISCLAIMER: This message is proprietary to Hughes Software Systems
> Limited
> > > (HSS) and is intended solely for the use of the individual to whom it
> is
> > > addressed. It may contain  privileged or confidential information and
> > > should not be circulated or used for any purpose other than for what
it
> is
> > > intended. If you have received this message in error, please notify
the
> > > originator immediately. If you are not the intended recipient, you
are
> > > notified that you are strictly prohibited from using, copying,
> altering, or
> > > disclosing the contents of this message. HSS accepts no
responsibility
> for
> > > loss or damage arising from the use of the information transmitted by
> this
> > > email including damage from virus."
> > >
> > >
> > > _______________________________________________
> > > 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



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



From exim@www1.ietf.org  Fri Jul  4 03:38:34 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA19203
	for <sigtran-archive@odin.ietf.org>; Fri, 4 Jul 2003 03:38:34 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YL8w-0005Gw-49
	for sigtran-archive@odin.ietf.org; Fri, 04 Jul 2003 03:38:06 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h647c6W1020265
	for sigtran-archive@odin.ietf.org; Fri, 4 Jul 2003 03:38:06 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YL8s-0005GY-8X; Fri, 04 Jul 2003 03:38:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YL8W-0005BA-P8
	for sigtran@optimus.ietf.org; Fri, 04 Jul 2003 03:37:40 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA19197
	for <sigtran@ietf.org>; Fri, 4 Jul 2003 03:37:39 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YL8U-000344-00
	for sigtran@ietf.org; Fri, 04 Jul 2003 03:37:38 -0400
Received: from auds953.usa.alcatel.com ([143.209.238.6])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YL8T-000340-00
	for sigtran@ietf.org; Fri, 04 Jul 2003 03:37:37 -0400
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h647b7LX028615
	for <sigtran@ietf.org>; Fri, 4 Jul 2003 02:37:08 -0500 (CDT)
Received: from sun2594.ssd.usa.alcatel.com (sun2594.ssd.usa.alcatel.com [143.209.250.252])
	by ssd.usa.alcatel.com (8.12.8/8.12.8) with ESMTP id h647ar1C017889
	for <sigtran@ietf.org>; Fri, 4 Jul 2003 02:36:53 -0500 (CDT)
Date: Fri, 4 Jul 2003 02:36:53 -0500 (CDT)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
In-Reply-To: <OF523A56B2.94AAE275-ON65256D59.00245419@hss.hns.com>
Message-ID: <Pine.SOL.4.10.10307040204090.8257-100000@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>


Kamesh,

  MTP3 layer uses SLS to route the messages(pertaining to the same
  logically  related data) to the destination node.Users(ISUP or SCCP)
  can not instruct MTP3 to do any work related with loadsharing, it is
  upto the decision of MTP3 layer to decide the field(like SLS) on which
  it is going to loadshare its traffic.

  To the point, say for example you have an IAM which has LI > 272, ISUP
  layer would split this IAM into 2 IAMs(just an example) and send it
  to STP(SG), remeber both IAMs must have the same SLS. Now STP'S MTP3
  layer finds an outgoing linkset based on the DPC received on the MSU.

  Once linkset has been selected ,SLS is used to select a particular link
  within the linkset, so the above 2 IAMs will be sent in a sequence to
  the destination, here there is no way of mis-sequencing of MSUs.

  Is this what you expect?

--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, 4 Jul 2003 kakaul@hss.hns.com wrote:

> 
> 
> 
> 
> Hi Samule,
>   You have not replied for my second question :
> I want to know how an MTP3  user (ISUP or SCCP) selects SLS for effective
> load distribution at MTP3 layer among the links available to reach a given
> DPC. Is it that SLS is chosen per DPC for a logical session or globally per
> each logical session irrespective of DPC for which this session is
> intended.I know it is implementation dependent option but what do most of
> the implementers opt for. By logical session i mean logically related data.
> 
> Regards
> Kamesh Kaul
> www.hssworld.com
> Hughes Software Systems
> 
> 
> 
> 
> 
> "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
> 07/04/2003 10:58:14 AM
> 
> Sent by:    sigtran-admin@ietf.org
> 
> 
> To:    sigtran@ietf.org
> cc:
> 
> Subject:    Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
> 
> 
> 
> Pastor,
> 
>  MIB-05 does not have 'load-shar on dpc',i was referring mib-04 which lead
>  me into confusion.
> 
>  Thanks kakaul for our mail.
> 
> Kakaul,
> 
>  As you said MTP3 does not route messages based on CIC or SSN because
>  there are no standards available, i agree with it.
> 
>  Have a look, suppose SG receives an MSU with routing indicator="Route on
>  PointCode/SSN", if your SG supports static configuration then it is easy
>  to route these messages to the corresponding subSystems based on
>  statically configured SSN-ASP table the same way CIC-ASP table can be
>  used to route ISUP messages to the corresponding MGCs.Basically CIC was
>  removed from RFC to avoid complex problems say a msg comes with CIC
>  range,this CIC range can be split across MGCs here the problem is
>  potential.
> 
>  These are the reasons caused me to say that with static configuration
>  CIC or/and SSN can be supported if an SG vendor and ASP vendor has
>  mutual understanding.(but this is outside the scope of M3UA RFC)
> 
> --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, 4 Jul 2003 kakaul@hss.hns.com wrote:
> 
> >
> >
> >
> >
> > Hi Samule,
> >   Does MTP3 use CIC or SSN for load-distribution among the available
> > links.I guess only SLS parameter in MTP3 is used for load-distribution
> > among the available link.I want to know how an MTP3  user (ISUP or SCCP)
> > selects SLS for effective load distribution at MTP3 layer among the links
> > available to reach the  a given DPC.
> >
> > Regards
> > Kamesh Kaul
> > www.hssworld.com
> > Hughes Software Systems
> >
> >
> >
> >
> > "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
> > 07/04/2003 09:29:51 AM
> >
> > Sent by:    sigtran-admin@ietf.org
> >
> >
> > To:    sigtran@ietf.org
> > cc:
> >
> > Subject:    Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
> >
> >
> >
> >  this is just an info regarding CIC and SSN - load-sharing.
> >
> >  A node can load-share its traffic based on CIC or SSN with the
> >  help of static configuration but this is outside of the scope of M3UA
> >  rfc.When an ASP vendor and SGP vendor agrees to do load-sharing based on
> >  CIC or SSN with static confiuration they are welcome to do this,
> >  since this has not been standardised, taking this to interop will
> >  lead to problems.So almost everybody agrees to delete CIC and SSN
> >  loadsharing concept from MIB-04.
> >
> >
> > --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, 3 Jul 2003, Brian F. G. Bidulock wrote:
> >
> > > agurmukhani,
> > >
> > > I think that they should be removed from the MIB, but we did NOT agree
> > > that CIC or SSN should not be used for loadsharing.  Any criteria that
> > > the SG/ASP desires to use for loadsharing being congnisant of the need
> > > for ordering of user part messages is allowed for loadsharing.  We DID
> > > agree to remove CIC, SSN from the Routing Key, because a rational and
> > > interoperable basis for splitting CICs and SSNs across AS could not be
> > > decided.
> > >
> > > The meanse oadsharing, on the other hand, is not specified in M3UA.
> > > Some examples are given, but no requirements.  Some things that could
> > > clearly be used are SLS, DPC/SLS, OPC/DPC/SLS, CIC, DPC/CIC,
> OPC/DPC/CIC,
> > > OPC/DPC/CIC/SLS, SSN, ....  and other permutations of these and even
> > > higher layer parameters such as TID.
> > >
> > > My argument for removing them from the MIB is not that they cannot be
> > > used, but that they have not been standardized.
> > >
> > > --brian
> > >
> > > agurmukhani@hss.hns.com wrote:
> > (Thu, 03 Jul 2003 10:23:35)
> > > >
> > > >
> > > >
> > > >
> > > > Hi...
> > > >
> > > >
> > > > In my understanding and  as per the discussion and consensus on the
> > mailing
> > > > list , everyone agreed that  CIC and SSN should not be used as
> > loadsharing
> > > > basis
> > > > be it within streams in a particular SCTP association or for active
> > > > signalling processes.
> > > >
> > > > So would these be removed from  the m3uaStreamLshMethod  and
> > > > m3uaSpLshMethod OBJECT-TYPE s in M3UA MIB draft.
> > > >
> > > >
> > > > Thanks and Regards
> > > > Anjali
> > > > www.hssworld.com
> > > > Hughes Software Systems.
> > > >
> > > > "DISCLAIMER: This message is proprietary to Hughes Software Systems
> > Limited
> > > > (HSS) and is intended solely for the use of the individual to whom it
> > is
> > > > addressed. It may contain  privileged or confidential information and
> > > > should not be circulated or used for any purpose other than for what
> it
> > is
> > > > intended. If you have received this message in error, please notify
> the
> > > > originator immediately. If you are not the intended recipient, you
> are
> > > > notified that you are strictly prohibited from using, copying,
> > altering, or
> > > > disclosing the contents of this message. HSS accepts no
> responsibility
> > for
> > > > loss or damage arising from the use of the information transmitted by
> > this
> > > > email including damage from virus."
> > > >
> > > >
> > > > _______________________________________________
> > > > 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
> 
> 
> 
> _______________________________________________
> 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 exim@www1.ietf.org  Fri Jul  4 06:31:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA22596
	for <sigtran-archive@odin.ietf.org>; Fri, 4 Jul 2003 06:31:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YNqL-0007ka-Fx
	for sigtran-archive@odin.ietf.org; Fri, 04 Jul 2003 06:31:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h64AV5dm029789
	for sigtran-archive@odin.ietf.org; Fri, 4 Jul 2003 06:31:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YNqI-0007kB-IY; Fri, 04 Jul 2003 06:31:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YNps-0007jg-L8
	for sigtran@optimus.ietf.org; Fri, 04 Jul 2003 06:30:36 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA22567
	for <sigtran@ietf.org>; Fri, 4 Jul 2003 06:30:32 -0400 (EDT)
From: ygahlaut@hss.hns.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YNpo-00042m-00
	for sigtran@ietf.org; Fri, 04 Jul 2003 06:30:32 -0400
Received: from hindon.hss.co.in ([202.54.26.202])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YNpn-00042O-00
	for sigtran@ietf.org; Fri, 04 Jul 2003 06:30:31 -0400
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 h64AQ4p12592
	for <sigtran@ietf.org>; Fri, 4 Jul 2003 15:56:04 +0530 (IST)
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h64AQ2O12579;
	Fri, 4 Jul 2003 15:56:02 +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 ESMTP id h64ATdY19877;
	Fri, 4 Jul 2003 15:59:39 +0530 (IST)
Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
To: john.loughney@nokia.com
Cc: sigtran@ietf.org
Date: Fri, 4 Jul 2003 15:54:23 +0530
Message-ID: <OFC2563063.F33E00B2-ON65256D59.002CE761@hss.hns.com>
X-MIMETrack: Serialize by Router on Sandesh/HSS(Release 6.0|September 26, 2002) at 04/07/2003
 03:53:18 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>






 Hi  John,
 Some comments and queries :-

 >>  Section 3.2.1  CLDT
                Sequence Control is set to 0 for class 0 messages.
 what is the intention behind setting this value to 0 ?

 >>  Section 3.4.4 SCON   says
   "Note 1:    When the SSN is included it must be equal to 1. If the
              SSN is included the SCON corresponds to the N-PCSTATE
              primitive used to convey the Restricted Importance Level
              (RIL) to the SCCP user. When the SSN is not included, the
              SCON message corresponds to the SCCP N-PCSTATE primitive
              reporting signalling point or network congestion status.   "

but the section 3.10.24  (congestion level parameter  description) still talks about the N-State primitive.
So Which is correct out of the two ? Please clarify

>>  All SSNM messages have a note stating "Affected Point Code Field can contain
 one point code when SSN is present in the message".  What about if SSNM
contains more than one affected point codes and SSN as well, then how should that
SSNM be handled ?
Is it a valid message ? if yes then how should it be handled ?
Please clarify on this point if I am missing something.

>>   In Section 3.10.24  size of congestion level field is written 8 bits but in figure it is shown as 32 bits
 Which one is correct?


>>   In Section 3.10.23  segmentation TLV, Length is 32 it should be 8.

>>  Section 4.3 and 4.3.1  are using M3UA term instead of SUA.

>>  Section 4.3.1  SCTP CDI and SCTP RI are described twice.

>>  Section 4.3.1 refers "Figure 4 ASP/ IPSP States per AS"   as Figure 3.

 Thanks and Regards
 Yogesh
 www.hssworld.com
 Hughes Software Systems


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



From exim@www1.ietf.org  Fri Jul  4 11:44:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA29979
	for <sigtran-archive@odin.ietf.org>; Fri, 4 Jul 2003 11:44:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YSjC-00062r-M4
	for sigtran-archive@odin.ietf.org; Fri, 04 Jul 2003 11:44:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h64Fi28J023233
	for sigtran-archive@odin.ietf.org; Fri, 4 Jul 2003 11:44:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YSjA-00062T-Mq; Fri, 04 Jul 2003 11:44:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YSiQ-000629-Nx
	for sigtran@optimus.ietf.org; Fri, 04 Jul 2003 11:43:14 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA29960
	for <sigtran@ietf.org>; Fri, 4 Jul 2003 11:43:12 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YSiP-0006wP-00
	for sigtran@ietf.org; Fri, 04 Jul 2003 11:43:13 -0400
Received: from web41812.mail.yahoo.com ([66.218.93.146])
	by ietf-mx with smtp (Exim 4.12)
	id 19YSiO-0006wB-00
	for sigtran@ietf.org; Fri, 04 Jul 2003 11:43:12 -0400
Message-ID: <20030704154241.44959.qmail@web41812.mail.yahoo.com>
Received: from [65.198.40.51] by web41812.mail.yahoo.com via HTTP; Fri, 04 Jul 2003 10:42:41 CDT
Date: Fri, 4 Jul 2003 10:42:41 -0500 (CDT)
From: =?iso-8859-1?q?Pablo=20Salinas?= <pablopesb@yahoo.com>
To: sigtran@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="0-1538456865-1057333361=:44505"
Subject: [Sigtran] M-SCTP-ESTABLISH request
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

--0-1538456865-1057333361=:44505
Content-Type: text/plain; charset=iso-8859-1
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id LAA29961
Content-Transfer-Encoding: quoted-printable

Hi there,
             First of all, in page 12 of FRC 3331 says:
"M-SCTP_ESTABLISH request
Direction: LM -> M2UA
Purpose: LM requests ASP to establish an SCTP asssociation with an SGP."
=20
Is ti rigth to say "...LM requests ASP to establish...". Shouldn't it be =
the ASP the one that request the M2UA Layer to establish an sctp associat=
ion?
=20
Second of all, I don't understand how the M2UA layer in the MGC maps ASP'=
s with it's local IP addresses.
In the RSG, this information is pre-configured or configured on demand (w=
ith dynamic registration). And it maps ASP's with SCTP associations.
How is this mapping done in th MGC?
=20
                        thanks,
                                     Pablo Salinas



---------------------------------
Do You Yahoo!?
Todo lo que quieres saber de Estados Unidos, Am=E9rica Latina y el resto =
del Mundo.
Vis=EDta Yahoo! Noticias.

--0-1538456865-1057333361=:44505
Content-Type: text/html; charset=iso-8859-1
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id LAA29961
Content-Transfer-Encoding: quoted-printable

<DIV>Hi there,</DIV>
<DIV>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=
bsp; First of all, in page 12 of FRC 3331 says:</DIV>
<DIV>"M-SCTP_ESTABLISH request</DIV>
<DIV>Direction: LM -&gt; M2UA</DIV>
<DIV>Purpose: LM requests ASP to establish an SCTP asssociation with an S=
GP."</DIV>
<DIV>&nbsp;</DIV>
<DIV>Is ti rigth to say "...LM requests ASP to establish...". Shouldn't i=
t be the ASP the one that request the M2UA Layer to establish an sctp ass=
ociation?</DIV>
<DIV>&nbsp;</DIV>
<DIV>Second of all, I don't understand how the M2UA layer in the MGC&nbsp=
;maps ASP's&nbsp;with it's local IP addresses.</DIV>
<DIV>In the RSG, this information is pre-configured or configured on dema=
nd (with dynamic registration). And it maps ASP's with SCTP associations.=
</DIV>
<DIV>How is this mapping done in th MGC?</DIV>
<DIV>&nbsp;</DIV>
<DIV>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=
bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; th=
anks,</DIV>
<DIV>&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; Pablo Salinas</DIV><p><br><hr size=3D1 noshade><b>Do You Yahoo!?</b><b=
r>
<a href=3Dhttp://espanol.yahoo.com/mail_tagline/*http://espanol.news.yaho=
o.com><img src=3D"http://us.i1.yimg.com/us.yimg.com/i/not/dailynews.gif" =
width=3D30 height=3D26 border=3D0 align=3Dleft></a>
Todo lo que quieres saber de Estados Unidos, Am=E9rica Latina y el resto =
del Mundo.<br>
Vis=EDta <a href=3Dhttp://espanol.yahoo.com/mail_tagline/*http://espanol.=
news.yahoo.com>Yahoo! Noticias</a>.<br>
--0-1538456865-1057333361=:44505--

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



From exim@www1.ietf.org  Sat Jul  5 00:16:35 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA17230
	for <sigtran-archive@odin.ietf.org>; Sat, 5 Jul 2003 00:16:35 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YeT1-0005HO-SP
	for sigtran-archive@odin.ietf.org; Sat, 05 Jul 2003 00:16:08 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h654G7oh020288
	for sigtran-archive@odin.ietf.org; Sat, 5 Jul 2003 00:16:07 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YeSu-0005Gm-OK; Sat, 05 Jul 2003 00:16:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YeSs-0005GV-VJ
	for sigtran@optimus.ietf.org; Sat, 05 Jul 2003 00:15:59 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA17215
	for <sigtran@ietf.org>; Sat, 5 Jul 2003 00:15:55 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YeSq-00068a-00
	for sigtran@ietf.org; Sat, 05 Jul 2003 00:15:56 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[NyqC1mFeHmvwQR+Q2iA9tHOQtKJchy7J])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YeSp-00068W-00
	for sigtran@ietf.org; Sat, 05 Jul 2003 00:15:55 -0400
Received: from ns.pigworks.openss7.net (IDENT:UzFiDjZ2ANtVoQT/ahtp7Jgp6Ngts1hb@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h654FrX08589;
	Fri, 4 Jul 2003 22:15:53 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h654FqB01980;
	Fri, 4 Jul 2003 22:15:52 -0600
Date: Fri, 4 Jul 2003 22:15:52 -0600
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 MIB Draft05: Loadsharing on CIC ,SSN
Message-ID: <20030704221552.D1541@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	sigtran@ietf.org
References: <OF523A56B2.94AAE275-ON65256D59.00245419@hss.hns.com> <Pine.SOL.4.10.10307040204090.8257-100000@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.4.10.10307040204090.8257-100000@sun2594.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Fri, Jul 04, 2003 at 02:36:53AM -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>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 04 Jul 2003 02:36:53)
> 
> Kamesh,
> 
>   MTP3 layer uses SLS to route the messages(pertaining to the same
>   logically  related data) to the destination node.Users(ISUP or SCCP)
>   can not instruct MTP3 to do any work related with loadsharing, it is
>   upto the decision of MTP3 layer to decide the field(like SLS) on which
>   it is going to loadshare its traffic.

Incorrect.  It is the MTP-User's responsibility to select SLS value to
ensure load sharing of user messages with consideration for sequence
requirements for interrelated messages.  See Q.704, Q.714, Q.764.

--brian

> 
>   To the point, say for example you have an IAM which has LI > 272, ISUP
>   layer would split this IAM into 2 IAMs(just an example) and send it
>   to STP(SG), remeber both IAMs must have the same SLS. Now STP'S MTP3
>   layer finds an outgoing linkset based on the DPC received on the MSU.
> 
>   Once linkset has been selected ,SLS is used to select a particular link
>   within the linkset, so the above 2 IAMs will be sent in a sequence to
>   the destination, here there is no way of mis-sequencing of MSUs.
> 
>   Is this what you expect?
> 
> --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, 4 Jul 2003 kakaul@hss.hns.com wrote:
> 
> > 
> > 
> > 
> > 
> > Hi Samule,
> >   You have not replied for my second question :
> > I want to know how an MTP3  user (ISUP or SCCP) selects SLS for effective
> > load distribution at MTP3 layer among the links available to reach a given
> > DPC. Is it that SLS is chosen per DPC for a logical session or globally per
> > each logical session irrespective of DPC for which this session is
> > intended.I know it is implementation dependent option but what do most of
> > the implementers opt for. By logical session i mean logically related data.
> > 
> > Regards
> > Kamesh Kaul
> > www.hssworld.com
> > Hughes Software Systems
> > 
> > 
> > 
> > 
> > 
> > "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
> > 07/04/2003 10:58:14 AM
> > 
> > Sent by:    sigtran-admin@ietf.org
> > 
> > 
> > To:    sigtran@ietf.org
> > cc:
> > 
> > Subject:    Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
> > 
> > 
> > 
> > Pastor,
> > 
> >  MIB-05 does not have 'load-shar on dpc',i was referring mib-04 which lead
> >  me into confusion.
> > 
> >  Thanks kakaul for our mail.
> > 
> > Kakaul,
> > 
> >  As you said MTP3 does not route messages based on CIC or SSN because
> >  there are no standards available, i agree with it.
> > 
> >  Have a look, suppose SG receives an MSU with routing indicator="Route on
> >  PointCode/SSN", if your SG supports static configuration then it is easy
> >  to route these messages to the corresponding subSystems based on
> >  statically configured SSN-ASP table the same way CIC-ASP table can be
> >  used to route ISUP messages to the corresponding MGCs.Basically CIC was
> >  removed from RFC to avoid complex problems say a msg comes with CIC
> >  range,this CIC range can be split across MGCs here the problem is
> >  potential.
> > 
> >  These are the reasons caused me to say that with static configuration
> >  CIC or/and SSN can be supported if an SG vendor and ASP vendor has
> >  mutual understanding.(but this is outside the scope of M3UA RFC)
> > 
> > --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, 4 Jul 2003 kakaul@hss.hns.com wrote:
> > 
> > >
> > >
> > >
> > >
> > > Hi Samule,
> > >   Does MTP3 use CIC or SSN for load-distribution among the available
> > > links.I guess only SLS parameter in MTP3 is used for load-distribution
> > > among the available link.I want to know how an MTP3  user (ISUP or SCCP)
> > > selects SLS for effective load distribution at MTP3 layer among the links
> > > available to reach the  a given DPC.
> > >
> > > Regards
> > > Kamesh Kaul
> > > www.hssworld.com
> > > Hughes Software Systems
> > >
> > >
> > >
> > >
> > > "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
> > > 07/04/2003 09:29:51 AM
> > >
> > > Sent by:    sigtran-admin@ietf.org
> > >
> > >
> > > To:    sigtran@ietf.org
> > > cc:
> > >
> > > Subject:    Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
> > >
> > >
> > >
> > >  this is just an info regarding CIC and SSN - load-sharing.
> > >
> > >  A node can load-share its traffic based on CIC or SSN with the
> > >  help of static configuration but this is outside of the scope of M3UA
> > >  rfc.When an ASP vendor and SGP vendor agrees to do load-sharing based on
> > >  CIC or SSN with static confiuration they are welcome to do this,
> > >  since this has not been standardised, taking this to interop will
> > >  lead to problems.So almost everybody agrees to delete CIC and SSN
> > >  loadsharing concept from MIB-04.
> > >
> > >
> > > --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, 3 Jul 2003, Brian F. G. Bidulock wrote:
> > >
> > > > agurmukhani,
> > > >
> > > > I think that they should be removed from the MIB, but we did NOT agree
> > > > that CIC or SSN should not be used for loadsharing.  Any criteria that
> > > > the SG/ASP desires to use for loadsharing being congnisant of the need
> > > > for ordering of user part messages is allowed for loadsharing.  We DID
> > > > agree to remove CIC, SSN from the Routing Key, because a rational and
> > > > interoperable basis for splitting CICs and SSNs across AS could not be
> > > > decided.
> > > >
> > > > The meanse oadsharing, on the other hand, is not specified in M3UA.
> > > > Some examples are given, but no requirements.  Some things that could
> > > > clearly be used are SLS, DPC/SLS, OPC/DPC/SLS, CIC, DPC/CIC,
> > OPC/DPC/CIC,
> > > > OPC/DPC/CIC/SLS, SSN, ....  and other permutations of these and even
> > > > higher layer parameters such as TID.
> > > >
> > > > My argument for removing them from the MIB is not that they cannot be
> > > > used, but that they have not been standardized.
> > > >
> > > > --brian
> > > >
> > > > agurmukhani@hss.hns.com wrote:
> > > (Thu, 03 Jul 2003 10:23:35)
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > Hi...
> > > > >
> > > > >
> > > > > In my understanding and  as per the discussion and consensus on the
> > > mailing
> > > > > list , everyone agreed that  CIC and SSN should not be used as
> > > loadsharing
> > > > > basis
> > > > > be it within streams in a particular SCTP association or for active
> > > > > signalling processes.
> > > > >
> > > > > So would these be removed from  the m3uaStreamLshMethod  and
> > > > > m3uaSpLshMethod OBJECT-TYPE s in M3UA MIB draft.
> > > > >
> > > > >
> > > > > Thanks and Regards
> > > > > Anjali
> > > > > www.hssworld.com
> > > > > Hughes Software Systems.
> > > > >
> > > > > "DISCLAIMER: This message is proprietary to Hughes Software Systems
> > > Limited
> > > > > (HSS) and is intended solely for the use of the individual to whom it
> > > is
> > > > > addressed. It may contain  privileged or confidential information and
> > > > > should not be circulated or used for any purpose other than for what
> > it
> > > is
> > > > > intended. If you have received this message in error, please notify
> > the
> > > > > originator immediately. If you are not the intended recipient, you
> > are
> > > > > notified that you are strictly prohibited from using, copying,
> > > altering, or
> > > > > disclosing the contents of this message. HSS accepts no
> > responsibility
> > > for
> > > > > loss or damage arising from the use of the information transmitted by
> > > this
> > > > > email including damage from virus."
> > > > >
> > > > >
> > > > > _______________________________________________
> > > > > 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
> > 
> > 
> > 
> > _______________________________________________
> > 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 exim@www1.ietf.org  Sat Jul  5 02:18:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA00547
	for <sigtran-archive@odin.ietf.org>; Sat, 5 Jul 2003 02:18:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YgN2-0004uP-GR
	for sigtran-archive@odin.ietf.org; Sat, 05 Jul 2003 02:18:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h656I4sJ018870
	for sigtran-archive@odin.ietf.org; Sat, 5 Jul 2003 02:18:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YgMz-0004u7-Aa; Sat, 05 Jul 2003 02:18:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YgM4-0004hJ-Q5
	for sigtran@optimus.ietf.org; Sat, 05 Jul 2003 02:17:04 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA00415
	for <sigtran@ietf.org>; Sat, 5 Jul 2003 02:17:01 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YgM1-0006Xe-00
	for sigtran@ietf.org; Sat, 05 Jul 2003 02:17:01 -0400
Received: from auds952.usa.alcatel.com ([143.209.238.7])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YgLz-0006Xb-00
	for sigtran@ietf.org; Sat, 05 Jul 2003 02:17:00 -0400
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h656GOB3026940
	for <sigtran@ietf.org>; Sat, 5 Jul 2003 01:16:24 -0500 (CDT)
Received: from sun2594.ssd.usa.alcatel.com (sun2594.ssd.usa.alcatel.com [143.209.250.252])
	by ssd.usa.alcatel.com (8.12.8/8.12.8) with ESMTP id h656FQ1C009533
	for <sigtran@ietf.org>; Sat, 5 Jul 2003 01:15:26 -0500 (CDT)
Date: Sat, 5 Jul 2003 01:15:26 -0500 (CDT)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
In-Reply-To: <20030704221552.D1541@openss7.org>
Message-ID: <Pine.SOL.4.10.10307050111150.25952-100000@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,

See the underlined sentences.

  Alcatel USA, 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, 4 Jul 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 04 Jul 2003 02:36:53)
> > 
> > Kamesh,
> > 
> >   MTP3 layer uses SLS to route the messages(pertaining to the same
> >   logically  related data) to the destination node.Users(ISUP or SCCP)
> >   can not instruct MTP3 to do any work related with loadsharing, it is
> >   upto the decision of MTP3 layer to decide the field(like SLS) on which
> >   it is going to loadshare its traffic.
> 
> Incorrect.  It is the MTP-User's responsibility to select SLS value to
> ensure load sharing of user messages with consideration for sequence
> requirements for interrelated messages.  See Q.704, Q.714, Q.764.
> 
> --brian
> 
> > 

> >   To the point, say for example you have an IAM which has LI > 272, ISUP
                                                                        --
> >   layer would split this IAM into 2 IAMs(just an example) and send it
      ------------------------------------------------------------------
> >   to STP(SG), remeber both IAMs must have the same SLS. Now STP'S MTP3
      ------------------------------------------------------
> >   layer finds an outgoing linkset based on the DPC received on the MSU.
> > 
> >   Once linkset has been selected ,SLS is used to select a particular link
> >   within the linkset, so the above 2 IAMs will be sent in a sequence to
> >   the destination, here there is no way of mis-sequencing of MSUs.
> > 
> >   Is this what you expect?
> > 
> > --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, 4 Jul 2003 kakaul@hss.hns.com wrote:
> > 
> > > 
> > > 
> > > 
> > > 
> > > Hi Samule,
> > >   You have not replied for my second question :
> > > I want to know how an MTP3  user (ISUP or SCCP) selects SLS for effective
> > > load distribution at MTP3 layer among the links available to reach a given
> > > DPC. Is it that SLS is chosen per DPC for a logical session or globally per
> > > each logical session irrespective of DPC for which this session is
> > > intended.I know it is implementation dependent option but what do most of
> > > the implementers opt for. By logical session i mean logically related data.
> > > 
> > > Regards
> > > Kamesh Kaul
> > > www.hssworld.com
> > > Hughes Software Systems
> > > 
> > > 
> > > 
> > > 
> > > 
> > > "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
> > > 07/04/2003 10:58:14 AM
> > > 
> > > Sent by:    sigtran-admin@ietf.org
> > > 
> > > 
> > > To:    sigtran@ietf.org
> > > cc:
> > > 
> > > Subject:    Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
> > > 
> > > 
> > > 
> > > Pastor,
> > > 
> > >  MIB-05 does not have 'load-shar on dpc',i was referring mib-04 which lead
> > >  me into confusion.
> > > 
> > >  Thanks kakaul for our mail.
> > > 
> > > Kakaul,
> > > 
> > >  As you said MTP3 does not route messages based on CIC or SSN because
> > >  there are no standards available, i agree with it.
> > > 
> > >  Have a look, suppose SG receives an MSU with routing indicator="Route on
> > >  PointCode/SSN", if your SG supports static configuration then it is easy
> > >  to route these messages to the corresponding subSystems based on
> > >  statically configured SSN-ASP table the same way CIC-ASP table can be
> > >  used to route ISUP messages to the corresponding MGCs.Basically CIC was
> > >  removed from RFC to avoid complex problems say a msg comes with CIC
> > >  range,this CIC range can be split across MGCs here the problem is
> > >  potential.
> > > 
> > >  These are the reasons caused me to say that with static configuration
> > >  CIC or/and SSN can be supported if an SG vendor and ASP vendor has
> > >  mutual understanding.(but this is outside the scope of M3UA RFC)
> > > 
> > > --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, 4 Jul 2003 kakaul@hss.hns.com wrote:
> > > 
> > > >
> > > >
> > > >
> > > >
> > > > Hi Samule,
> > > >   Does MTP3 use CIC or SSN for load-distribution among the available
> > > > links.I guess only SLS parameter in MTP3 is used for load-distribution
> > > > among the available link.I want to know how an MTP3  user (ISUP or SCCP)
> > > > selects SLS for effective load distribution at MTP3 layer among the links
> > > > available to reach the  a given DPC.
> > > >
> > > > Regards
> > > > Kamesh Kaul
> > > > www.hssworld.com
> > > > Hughes Software Systems
> > > >
> > > >
> > > >
> > > >
> > > > "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
> > > > 07/04/2003 09:29:51 AM
> > > >
> > > > Sent by:    sigtran-admin@ietf.org
> > > >
> > > >
> > > > To:    sigtran@ietf.org
> > > > cc:
> > > >
> > > > Subject:    Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
> > > >
> > > >
> > > >
> > > >  this is just an info regarding CIC and SSN - load-sharing.
> > > >
> > > >  A node can load-share its traffic based on CIC or SSN with the
> > > >  help of static configuration but this is outside of the scope of M3UA
> > > >  rfc.When an ASP vendor and SGP vendor agrees to do load-sharing based on
> > > >  CIC or SSN with static confiuration they are welcome to do this,
> > > >  since this has not been standardised, taking this to interop will
> > > >  lead to problems.So almost everybody agrees to delete CIC and SSN
> > > >  loadsharing concept from MIB-04.
> > > >
> > > >
> > > > --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, 3 Jul 2003, Brian F. G. Bidulock wrote:
> > > >
> > > > > agurmukhani,
> > > > >
> > > > > I think that they should be removed from the MIB, but we did NOT agree
> > > > > that CIC or SSN should not be used for loadsharing.  Any criteria that
> > > > > the SG/ASP desires to use for loadsharing being congnisant of the need
> > > > > for ordering of user part messages is allowed for loadsharing.  We DID
> > > > > agree to remove CIC, SSN from the Routing Key, because a rational and
> > > > > interoperable basis for splitting CICs and SSNs across AS could not be
> > > > > decided.
> > > > >
> > > > > The meanse oadsharing, on the other hand, is not specified in M3UA.
> > > > > Some examples are given, but no requirements.  Some things that could
> > > > > clearly be used are SLS, DPC/SLS, OPC/DPC/SLS, CIC, DPC/CIC,
> > > OPC/DPC/CIC,
> > > > > OPC/DPC/CIC/SLS, SSN, ....  and other permutations of these and even
> > > > > higher layer parameters such as TID.
> > > > >
> > > > > My argument for removing them from the MIB is not that they cannot be
> > > > > used, but that they have not been standardized.
> > > > >
> > > > > --brian
> > > > >
> > > > > agurmukhani@hss.hns.com wrote:
> > > > (Thu, 03 Jul 2003 10:23:35)
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > Hi...
> > > > > >
> > > > > >
> > > > > > In my understanding and  as per the discussion and consensus on the
> > > > mailing
> > > > > > list , everyone agreed that  CIC and SSN should not be used as
> > > > loadsharing
> > > > > > basis
> > > > > > be it within streams in a particular SCTP association or for active
> > > > > > signalling processes.
> > > > > >
> > > > > > So would these be removed from  the m3uaStreamLshMethod  and
> > > > > > m3uaSpLshMethod OBJECT-TYPE s in M3UA MIB draft.
> > > > > >
> > > > > >
> > > > > > Thanks and Regards
> > > > > > Anjali
> > > > > > www.hssworld.com
> > > > > > Hughes Software Systems.
> > > > > >
> > > > > > "DISCLAIMER: This message is proprietary to Hughes Software Systems
> > > > Limited
> > > > > > (HSS) and is intended solely for the use of the individual to whom it
> > > > is
> > > > > > addressed. It may contain  privileged or confidential information and
> > > > > > should not be circulated or used for any purpose other than for what
> > > it
> > > > is
> > > > > > intended. If you have received this message in error, please notify
> > > the
> > > > > > originator immediately. If you are not the intended recipient, you
> > > are
> > > > > > notified that you are strictly prohibited from using, copying,
> > > > altering, or
> > > > > > disclosing the contents of this message. HSS accepts no
> > > responsibility
> > > > for
> > > > > > loss or damage arising from the use of the information transmitted by
> > > > this
> > > > > > email including damage from virus."
> > > > > >
> > > > > >
> > > > > > _______________________________________________
> > > > > > 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
> > > 
> > > 
> > > 
> > > _______________________________________________
> > > 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 exim@www1.ietf.org  Sat Jul  5 02:23:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA00868
	for <sigtran-archive@odin.ietf.org>; Sat, 5 Jul 2003 02:23:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YgRs-0005CI-Lw
	for sigtran-archive@odin.ietf.org; Sat, 05 Jul 2003 02:23:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h656N42I019975
	for sigtran-archive@odin.ietf.org; Sat, 5 Jul 2003 02:23:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YgRr-0005Bw-RG; Sat, 05 Jul 2003 02:23:03 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YgQt-0005BK-MC
	for sigtran@optimus.ietf.org; Sat, 05 Jul 2003 02:22:03 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA00846
	for <sigtran@ietf.org>; Sat, 5 Jul 2003 02:21:59 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YgQp-0006YY-00
	for sigtran@ietf.org; Sat, 05 Jul 2003 02:21:59 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[tpyR1sX3TOlh1A5kEOgqckrB+YAtIXws])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YgQo-0006YV-00
	for sigtran@ietf.org; Sat, 05 Jul 2003 02:21:58 -0400
Received: from ns.pigworks.openss7.net (IDENT:cn8yhuaEf8gkruqoJO7/FFTQmu7u9CPd@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h656LwX10028;
	Sat, 5 Jul 2003 00:21:58 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h656LwQ03346;
	Sat, 5 Jul 2003 00:21:58 -0600
Date: Sat, 5 Jul 2003 00:21:58 -0600
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 MIB Draft05: Loadsharing on CIC ,SSN
Message-ID: <20030705002158.A3334@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	sigtran@ietf.org
References: <20030704221552.D1541@openss7.org> <Pine.SOL.4.10.10307050111150.25952-100000@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.4.10.10307050111150.25952-100000@sun2594.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Sat, Jul 05, 2003 at 01:15:26AM -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>

Samuel,

See the underlined sentences that are incorrect:

Samuel Dur D. Jeyaseelan wrote:                                                     (Sat, 05 Jul 2003 01:15:26)
> 
> Brian,
> 
> See the underlined sentences.
> 
>   Alcatel USA, 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, 4 Jul 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 04 Jul 2003 02:36:53)
> > > 
> > > Kamesh,
> > > 
> > >   MTP3 layer uses SLS to route the messages(pertaining to the same
> > >   logically  related data) to the destination node.Users(ISUP or SCCP)
> > >   can not instruct MTP3 to do any work related with loadsharing, it is
                                                                       -----
> > >   upto the decision of MTP3 layer to decide the field(like SLS) on which
        ----------------------------------------------------------------------
> > >   it is going to loadshare its traffic.
        -------------------------------------
> > 
> > Incorrect.  It is the MTP-User's responsibility to select SLS value to
> > ensure load sharing of user messages with consideration for sequence
> > requirements for interrelated messages.  See Q.704, Q.714, Q.764.
> > 
> > --brian
> > 
> > > 
> 
> > >   To the point, say for example you have an IAM which has LI > 272, ISUP
>                                                                         --
> > >   layer would split this IAM into 2 IAMs(just an example) and send it
>       ------------------------------------------------------------------
> > >   to STP(SG), remeber both IAMs must have the same SLS. Now STP'S MTP3
>       ------------------------------------------------------
> > >   layer finds an outgoing linkset based on the DPC received on the MSU.
> > > 
> > >   Once linkset has been selected ,SLS is used to select a particular link
> > >   within the linkset, so the above 2 IAMs will be sent in a sequence to
> > >   the destination, here there is no way of mis-sequencing of MSUs.
> > > 
> > >   Is this what you expect?
> > > 
> > > --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, 4 Jul 2003 kakaul@hss.hns.com wrote:
> > > 
> > > > 
> > > > 
> > > > 
> > > > 
> > > > Hi Samule,
> > > >   You have not replied for my second question :
> > > > I want to know how an MTP3  user (ISUP or SCCP) selects SLS for effective
> > > > load distribution at MTP3 layer among the links available to reach a given
> > > > DPC. Is it that SLS is chosen per DPC for a logical session or globally per
> > > > each logical session irrespective of DPC for which this session is
> > > > intended.I know it is implementation dependent option but what do most of
> > > > the implementers opt for. By logical session i mean logically related data.
> > > > 
> > > > Regards
> > > > Kamesh Kaul
> > > > www.hssworld.com
> > > > Hughes Software Systems
> > > > 
> > > > 
> > > > 
> > > > 
> > > > 
> > > > "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
> > > > 07/04/2003 10:58:14 AM
> > > > 
> > > > Sent by:    sigtran-admin@ietf.org
> > > > 
> > > > 
> > > > To:    sigtran@ietf.org
> > > > cc:
> > > > 
> > > > Subject:    Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
> > > > 
> > > > 
> > > > 
> > > > Pastor,
> > > > 
> > > >  MIB-05 does not have 'load-shar on dpc',i was referring mib-04 which lead
> > > >  me into confusion.
> > > > 
> > > >  Thanks kakaul for our mail.
> > > > 
> > > > Kakaul,
> > > > 
> > > >  As you said MTP3 does not route messages based on CIC or SSN because
> > > >  there are no standards available, i agree with it.
> > > > 
> > > >  Have a look, suppose SG receives an MSU with routing indicator="Route on
> > > >  PointCode/SSN", if your SG supports static configuration then it is easy
> > > >  to route these messages to the corresponding subSystems based on
> > > >  statically configured SSN-ASP table the same way CIC-ASP table can be
> > > >  used to route ISUP messages to the corresponding MGCs.Basically CIC was
> > > >  removed from RFC to avoid complex problems say a msg comes with CIC
> > > >  range,this CIC range can be split across MGCs here the problem is
> > > >  potential.
> > > > 
> > > >  These are the reasons caused me to say that with static configuration
> > > >  CIC or/and SSN can be supported if an SG vendor and ASP vendor has
> > > >  mutual understanding.(but this is outside the scope of M3UA RFC)
> > > > 
> > > > --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, 4 Jul 2003 kakaul@hss.hns.com wrote:
> > > > 
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > Hi Samule,
> > > > >   Does MTP3 use CIC or SSN for load-distribution among the available
> > > > > links.I guess only SLS parameter in MTP3 is used for load-distribution
> > > > > among the available link.I want to know how an MTP3  user (ISUP or SCCP)
> > > > > selects SLS for effective load distribution at MTP3 layer among the links
> > > > > available to reach the  a given DPC.
> > > > >
> > > > > Regards
> > > > > Kamesh Kaul
> > > > > www.hssworld.com
> > > > > Hughes Software Systems
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
> > > > > 07/04/2003 09:29:51 AM
> > > > >
> > > > > Sent by:    sigtran-admin@ietf.org
> > > > >
> > > > >
> > > > > To:    sigtran@ietf.org
> > > > > cc:
> > > > >
> > > > > Subject:    Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
> > > > >
> > > > >
> > > > >
> > > > >  this is just an info regarding CIC and SSN - load-sharing.
> > > > >
> > > > >  A node can load-share its traffic based on CIC or SSN with the
> > > > >  help of static configuration but this is outside of the scope of M3UA
> > > > >  rfc.When an ASP vendor and SGP vendor agrees to do load-sharing based on
> > > > >  CIC or SSN with static confiuration they are welcome to do this,
> > > > >  since this has not been standardised, taking this to interop will
> > > > >  lead to problems.So almost everybody agrees to delete CIC and SSN
> > > > >  loadsharing concept from MIB-04.
> > > > >
> > > > >
> > > > > --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, 3 Jul 2003, Brian F. G. Bidulock wrote:
> > > > >
> > > > > > agurmukhani,
> > > > > >
> > > > > > I think that they should be removed from the MIB, but we did NOT agree
> > > > > > that CIC or SSN should not be used for loadsharing.  Any criteria that
> > > > > > the SG/ASP desires to use for loadsharing being congnisant of the need
> > > > > > for ordering of user part messages is allowed for loadsharing.  We DID
> > > > > > agree to remove CIC, SSN from the Routing Key, because a rational and
> > > > > > interoperable basis for splitting CICs and SSNs across AS could not be
> > > > > > decided.
> > > > > >
> > > > > > The meanse oadsharing, on the other hand, is not specified in M3UA.
> > > > > > Some examples are given, but no requirements.  Some things that could
> > > > > > clearly be used are SLS, DPC/SLS, OPC/DPC/SLS, CIC, DPC/CIC,
> > > > OPC/DPC/CIC,
> > > > > > OPC/DPC/CIC/SLS, SSN, ....  and other permutations of these and even
> > > > > > higher layer parameters such as TID.
> > > > > >
> > > > > > My argument for removing them from the MIB is not that they cannot be
> > > > > > used, but that they have not been standardized.
> > > > > >
> > > > > > --brian
> > > > > >
> > > > > > agurmukhani@hss.hns.com wrote:
> > > > > (Thu, 03 Jul 2003 10:23:35)
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > Hi...
> > > > > > >
> > > > > > >
> > > > > > > In my understanding and  as per the discussion and consensus on the
> > > > > mailing
> > > > > > > list , everyone agreed that  CIC and SSN should not be used as
> > > > > loadsharing
> > > > > > > basis
> > > > > > > be it within streams in a particular SCTP association or for active
> > > > > > > signalling processes.
> > > > > > >
> > > > > > > So would these be removed from  the m3uaStreamLshMethod  and
> > > > > > > m3uaSpLshMethod OBJECT-TYPE s in M3UA MIB draft.
> > > > > > >
> > > > > > >
> > > > > > > Thanks and Regards
> > > > > > > Anjali
> > > > > > > www.hssworld.com
> > > > > > > Hughes Software Systems.
> > > > > > >
> > > > > > > "DISCLAIMER: This message is proprietary to Hughes Software Systems
> > > > > Limited
> > > > > > > (HSS) and is intended solely for the use of the individual to whom it
> > > > > is
> > > > > > > addressed. It may contain  privileged or confidential information and
> > > > > > > should not be circulated or used for any purpose other than for what
> > > > it
> > > > > is
> > > > > > > intended. If you have received this message in error, please notify
> > > > the
> > > > > > > originator immediately. If you are not the intended recipient, you
> > > > are
> > > > > > > notified that you are strictly prohibited from using, copying,
> > > > > altering, or
> > > > > > > disclosing the contents of this message. HSS accepts no
> > > > responsibility
> > > > > for
> > > > > > > loss or damage arising from the use of the information transmitted by
> > > > > this
> > > > > > > email including damage from virus."
> > > > > > >
> > > > > > >
> > > > > > > _______________________________________________
> > > > > > > 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
> > > > 
> > > > 
> > > > 
> > > > _______________________________________________
> > > > 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 exim@www1.ietf.org  Sat Jul  5 02:39:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA01174
	for <sigtran-archive@odin.ietf.org>; Sat, 5 Jul 2003 02:39:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YghL-0006H8-9D
	for sigtran-archive@odin.ietf.org; Sat, 05 Jul 2003 02:39:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h656d3Gw024122
	for sigtran-archive@odin.ietf.org; Sat, 5 Jul 2003 02:39:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YghJ-0006Gh-QR; Sat, 05 Jul 2003 02:39:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YghG-0006GW-RU
	for sigtran@optimus.ietf.org; Sat, 05 Jul 2003 02:38:58 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA01166
	for <sigtran@ietf.org>; Sat, 5 Jul 2003 02:38:55 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YghD-0006h6-00
	for sigtran@ietf.org; Sat, 05 Jul 2003 02:38:55 -0400
Received: from auds952.usa.alcatel.com ([143.209.238.7])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YghC-0006gw-00
	for sigtran@ietf.org; Sat, 05 Jul 2003 02:38:54 -0400
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h656cOB3027397
	for <sigtran@ietf.org>; Sat, 5 Jul 2003 01:38:24 -0500 (CDT)
Received: from sun2594.ssd.usa.alcatel.com (sun2594.ssd.usa.alcatel.com [143.209.250.252])
	by ssd.usa.alcatel.com (8.12.8/8.12.8) with ESMTP id h656bx1C012113
	for <sigtran@ietf.org>; Sat, 5 Jul 2003 01:37:59 -0500 (CDT)
Date: Sat, 5 Jul 2003 01:37:59 -0500 (CDT)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
In-Reply-To: <20030705002158.A3334@openss7.org>
Message-ID: <Pine.SOL.4.10.10307050132040.25952-100000@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,
 
 I dont know the intention of your email.In this i two paragraphs 
 2nd para says ISUP layer at MGC constructs an IAM (WITH SLS) and sends
 it to STP(SG), now STP load-shares its traffic based on received SLS.

 
--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, 5 Jul 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> See the underlined sentences that are incorrect:
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Sat, 05 Jul 2003 01:15:26)
> > 
> > Brian,
> > 
> > See the underlined sentences.
> > 
> >   Alcatel USA, 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, 4 Jul 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 04 Jul 2003 02:36:53)
> > > > 
> > > > Kamesh,
> > > > 
> > > >   MTP3 layer uses SLS to route the messages(pertaining to the same
> > > >   logically  related data) to the destination node.Users(ISUP or SCCP)
> > > >   can not instruct MTP3 to do any work related with loadsharing, it is
>                                                                        -----
> > > >   upto the decision of MTP3 layer to decide the field(like SLS) on which
>         ----------------------------------------------------------------------
> > > >   it is going to loadshare its traffic.
>         -------------------------------------
> > > 
> > > Incorrect.  It is the MTP-User's responsibility to select SLS value to
> > > ensure load sharing of user messages with consideration for sequence
> > > requirements for interrelated messages.  See Q.704, Q.714, Q.764.
> > > 
> > > --brian
> > > 
> > > > 
> > 
> > > >   To the point, say for example you have an IAM which has LI > 272, ISUP
> >                                                                         --
> > > >   layer would split this IAM into 2 IAMs(just an example) and send it
> >       ------------------------------------------------------------------
> > > >   to STP(SG), remeber both IAMs must have the same SLS. Now STP'S MTP3
> >       ------------------------------------------------------
> > > >   layer finds an outgoing linkset based on the DPC received on the MSU.
> > > > 
> > > >   Once linkset has been selected ,SLS is used to select a particular link
> > > >   within the linkset, so the above 2 IAMs will be sent in a sequence to
> > > >   the destination, here there is no way of mis-sequencing of MSUs.
> > > > 
> > > >   Is this what you expect?
> > > > 
> > > > --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, 4 Jul 2003 kakaul@hss.hns.com wrote:
> > > > 
> > > > > 
> > > > > 
> > > > > 
> > > > > 
> > > > > Hi Samule,
> > > > >   You have not replied for my second question :
> > > > > I want to know how an MTP3  user (ISUP or SCCP) selects SLS for effective
> > > > > load distribution at MTP3 layer among the links available to reach a given
> > > > > DPC. Is it that SLS is chosen per DPC for a logical session or globally per
> > > > > each logical session irrespective of DPC for which this session is
> > > > > intended.I know it is implementation dependent option but what do most of
> > > > > the implementers opt for. By logical session i mean logically related data.
> > > > > 
> > > > > Regards
> > > > > Kamesh Kaul
> > > > > www.hssworld.com
> > > > > Hughes Software Systems
> > > > > 
> > > > > 
> > > > > 
> > > > > 
> > > > > 
> > > > > "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
> > > > > 07/04/2003 10:58:14 AM
> > > > > 
> > > > > Sent by:    sigtran-admin@ietf.org
> > > > > 
> > > > > 
> > > > > To:    sigtran@ietf.org
> > > > > cc:
> > > > > 
> > > > > Subject:    Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
> > > > > 
> > > > > 
> > > > > 
> > > > > Pastor,
> > > > > 
> > > > >  MIB-05 does not have 'load-shar on dpc',i was referring mib-04 which lead
> > > > >  me into confusion.
> > > > > 
> > > > >  Thanks kakaul for our mail.
> > > > > 
> > > > > Kakaul,
> > > > > 
> > > > >  As you said MTP3 does not route messages based on CIC or SSN because
> > > > >  there are no standards available, i agree with it.
> > > > > 
> > > > >  Have a look, suppose SG receives an MSU with routing indicator="Route on
> > > > >  PointCode/SSN", if your SG supports static configuration then it is easy
> > > > >  to route these messages to the corresponding subSystems based on
> > > > >  statically configured SSN-ASP table the same way CIC-ASP table can be
> > > > >  used to route ISUP messages to the corresponding MGCs.Basically CIC was
> > > > >  removed from RFC to avoid complex problems say a msg comes with CIC
> > > > >  range,this CIC range can be split across MGCs here the problem is
> > > > >  potential.
> > > > > 
> > > > >  These are the reasons caused me to say that with static configuration
> > > > >  CIC or/and SSN can be supported if an SG vendor and ASP vendor has
> > > > >  mutual understanding.(but this is outside the scope of M3UA RFC)
> > > > > 
> > > > > --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, 4 Jul 2003 kakaul@hss.hns.com wrote:
> > > > > 
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > Hi Samule,
> > > > > >   Does MTP3 use CIC or SSN for load-distribution among the available
> > > > > > links.I guess only SLS parameter in MTP3 is used for load-distribution
> > > > > > among the available link.I want to know how an MTP3  user (ISUP or SCCP)
> > > > > > selects SLS for effective load distribution at MTP3 layer among the links
> > > > > > available to reach the  a given DPC.
> > > > > >
> > > > > > Regards
> > > > > > Kamesh Kaul
> > > > > > www.hssworld.com
> > > > > > Hughes Software Systems
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
> > > > > > 07/04/2003 09:29:51 AM
> > > > > >
> > > > > > Sent by:    sigtran-admin@ietf.org
> > > > > >
> > > > > >
> > > > > > To:    sigtran@ietf.org
> > > > > > cc:
> > > > > >
> > > > > > Subject:    Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
> > > > > >
> > > > > >
> > > > > >
> > > > > >  this is just an info regarding CIC and SSN - load-sharing.
> > > > > >
> > > > > >  A node can load-share its traffic based on CIC or SSN with the
> > > > > >  help of static configuration but this is outside of the scope of M3UA
> > > > > >  rfc.When an ASP vendor and SGP vendor agrees to do load-sharing based on
> > > > > >  CIC or SSN with static confiuration they are welcome to do this,
> > > > > >  since this has not been standardised, taking this to interop will
> > > > > >  lead to problems.So almost everybody agrees to delete CIC and SSN
> > > > > >  loadsharing concept from MIB-04.
> > > > > >
> > > > > >
> > > > > > --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, 3 Jul 2003, Brian F. G. Bidulock wrote:
> > > > > >
> > > > > > > agurmukhani,
> > > > > > >
> > > > > > > I think that they should be removed from the MIB, but we did NOT agree
> > > > > > > that CIC or SSN should not be used for loadsharing.  Any criteria that
> > > > > > > the SG/ASP desires to use for loadsharing being congnisant of the need
> > > > > > > for ordering of user part messages is allowed for loadsharing.  We DID
> > > > > > > agree to remove CIC, SSN from the Routing Key, because a rational and
> > > > > > > interoperable basis for splitting CICs and SSNs across AS could not be
> > > > > > > decided.
> > > > > > >
> > > > > > > The meanse oadsharing, on the other hand, is not specified in M3UA.
> > > > > > > Some examples are given, but no requirements.  Some things that could
> > > > > > > clearly be used are SLS, DPC/SLS, OPC/DPC/SLS, CIC, DPC/CIC,
> > > > > OPC/DPC/CIC,
> > > > > > > OPC/DPC/CIC/SLS, SSN, ....  and other permutations of these and even
> > > > > > > higher layer parameters such as TID.
> > > > > > >
> > > > > > > My argument for removing them from the MIB is not that they cannot be
> > > > > > > used, but that they have not been standardized.
> > > > > > >
> > > > > > > --brian
> > > > > > >
> > > > > > > agurmukhani@hss.hns.com wrote:
> > > > > > (Thu, 03 Jul 2003 10:23:35)
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > > Hi...
> > > > > > > >
> > > > > > > >
> > > > > > > > In my understanding and  as per the discussion and consensus on the
> > > > > > mailing
> > > > > > > > list , everyone agreed that  CIC and SSN should not be used as
> > > > > > loadsharing
> > > > > > > > basis
> > > > > > > > be it within streams in a particular SCTP association or for active
> > > > > > > > signalling processes.
> > > > > > > >
> > > > > > > > So would these be removed from  the m3uaStreamLshMethod  and
> > > > > > > > m3uaSpLshMethod OBJECT-TYPE s in M3UA MIB draft.
> > > > > > > >
> > > > > > > >
> > > > > > > > Thanks and Regards
> > > > > > > > Anjali
> > > > > > > > www.hssworld.com
> > > > > > > > Hughes Software Systems.
> > > > > > > >
> > > > > > > > "DISCLAIMER: This message is proprietary to Hughes Software Systems
> > > > > > Limited
> > > > > > > > (HSS) and is intended solely for the use of the individual to whom it
> > > > > > is
> > > > > > > > addressed. It may contain  privileged or confidential information and
> > > > > > > > should not be circulated or used for any purpose other than for what
> > > > > it
> > > > > > is
> > > > > > > > intended. If you have received this message in error, please notify
> > > > > the
> > > > > > > > originator immediately. If you are not the intended recipient, you
> > > > > are
> > > > > > > > notified that you are strictly prohibited from using, copying,
> > > > > > altering, or
> > > > > > > > disclosing the contents of this message. HSS accepts no
> > > > > responsibility
> > > > > > for
> > > > > > > > loss or damage arising from the use of the information transmitted by
> > > > > > this
> > > > > > > > email including damage from virus."
> > > > > > > >
> > > > > > > >
> > > > > > > > _______________________________________________
> > > > > > > > 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
> > > > > 
> > > > > 
> > > > > 
> > > > > _______________________________________________
> > > > > 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 exim@www1.ietf.org  Sat Jul  5 02:42:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA01242
	for <sigtran-archive@odin.ietf.org>; Sat, 5 Jul 2003 02:42:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YgkD-0006bJ-8c
	for sigtran-archive@odin.ietf.org; Sat, 05 Jul 2003 02:42:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h656g13C025363
	for sigtran-archive@odin.ietf.org; Sat, 5 Jul 2003 02:42:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YgkD-0006az-3H; Sat, 05 Jul 2003 02:42:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YgjO-0006Pv-Pk
	for sigtran@optimus.ietf.org; Sat, 05 Jul 2003 02:41:10 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA01219
	for <sigtran@ietf.org>; Sat, 5 Jul 2003 02:41:07 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YgjL-0006hj-00
	for sigtran@ietf.org; Sat, 05 Jul 2003 02:41:07 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[AtDmyDFRzlErOqgdjgJd4DPV3ze+5AOq])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YgjK-0006hf-00
	for sigtran@ietf.org; Sat, 05 Jul 2003 02:41:06 -0400
Received: from ns.pigworks.openss7.net (IDENT:8LyLmxcbdRh9BzF1wS0q9Yrezz2Z+m+t@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h656f6X10271;
	Sat, 5 Jul 2003 00:41:06 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h656f6b03541;
	Sat, 5 Jul 2003 00:41:06 -0600
Date: Sat, 5 Jul 2003 00:41:06 -0600
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 MIB Draft05: Loadsharing on CIC ,SSN
Message-ID: <20030705004106.A3494@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	sigtran@ietf.org
References: <20030705002158.A3334@openss7.org> <Pine.SOL.4.10.10307050132040.25952-100000@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.4.10.10307050132040.25952-100000@sun2594.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Sat, Jul 05, 2003 at 01:37:59AM -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>

Samuel,

Yes, the passage I underlined was incorrect.

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Sat, 05 Jul 2003 01:37:59)
> 
> Brian,
>  
>  I dont know the intention of your email.In this i two paragraphs 
>  2nd para says ISUP layer at MGC constructs an IAM (WITH SLS) and sends
>  it to STP(SG), now STP load-shares its traffic based on received SLS.
> 
>  
> --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, 5 Jul 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > See the underlined sentences that are incorrect:
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Sat, 05 Jul 2003 01:15:26)
> > > 
> > > Brian,
> > > 
> > > See the underlined sentences.
> > > 
> > >   Alcatel USA, 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, 4 Jul 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Samuel,
> > > > 
> > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 04 Jul 2003 02:36:53)
> > > > > 
> > > > > Kamesh,
> > > > > 
> > > > >   MTP3 layer uses SLS to route the messages(pertaining to the same
> > > > >   logically  related data) to the destination node.Users(ISUP or SCCP)
> > > > >   can not instruct MTP3 to do any work related with loadsharing, it is
> >                                                                        -----
> > > > >   upto the decision of MTP3 layer to decide the field(like SLS) on which
> >         ----------------------------------------------------------------------
> > > > >   it is going to loadshare its traffic.
> >         -------------------------------------
> > > > 
> > > > Incorrect.  It is the MTP-User's responsibility to select SLS value to
> > > > ensure load sharing of user messages with consideration for sequence
> > > > requirements for interrelated messages.  See Q.704, Q.714, Q.764.
> > > > 
> > > > --brian
> > > > 
> > > > > 
> > > 
> > > > >   To the point, say for example you have an IAM which has LI > 272, ISUP
> > >                                                                         --
> > > > >   layer would split this IAM into 2 IAMs(just an example) and send it
> > >       ------------------------------------------------------------------
> > > > >   to STP(SG), remeber both IAMs must have the same SLS. Now STP'S MTP3
> > >       ------------------------------------------------------
> > > > >   layer finds an outgoing linkset based on the DPC received on the MSU.
> > > > > 
> > > > >   Once linkset has been selected ,SLS is used to select a particular link
> > > > >   within the linkset, so the above 2 IAMs will be sent in a sequence to
> > > > >   the destination, here there is no way of mis-sequencing of MSUs.
> > > > > 
> > > > >   Is this what you expect?
> > > > > 
> > > > > --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, 4 Jul 2003 kakaul@hss.hns.com wrote:
> > > > > 
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > > Hi Samule,
> > > > > >   You have not replied for my second question :
> > > > > > I want to know how an MTP3  user (ISUP or SCCP) selects SLS for effective
> > > > > > load distribution at MTP3 layer among the links available to reach a given
> > > > > > DPC. Is it that SLS is chosen per DPC for a logical session or globally per
> > > > > > each logical session irrespective of DPC for which this session is
> > > > > > intended.I know it is implementation dependent option but what do most of
> > > > > > the implementers opt for. By logical session i mean logically related data.
> > > > > > 
> > > > > > Regards
> > > > > > Kamesh Kaul
> > > > > > www.hssworld.com
> > > > > > Hughes Software Systems
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > > "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
> > > > > > 07/04/2003 10:58:14 AM
> > > > > > 
> > > > > > Sent by:    sigtran-admin@ietf.org
> > > > > > 
> > > > > > 
> > > > > > To:    sigtran@ietf.org
> > > > > > cc:
> > > > > > 
> > > > > > Subject:    Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > > Pastor,
> > > > > > 
> > > > > >  MIB-05 does not have 'load-shar on dpc',i was referring mib-04 which lead
> > > > > >  me into confusion.
> > > > > > 
> > > > > >  Thanks kakaul for our mail.
> > > > > > 
> > > > > > Kakaul,
> > > > > > 
> > > > > >  As you said MTP3 does not route messages based on CIC or SSN because
> > > > > >  there are no standards available, i agree with it.
> > > > > > 
> > > > > >  Have a look, suppose SG receives an MSU with routing indicator="Route on
> > > > > >  PointCode/SSN", if your SG supports static configuration then it is easy
> > > > > >  to route these messages to the corresponding subSystems based on
> > > > > >  statically configured SSN-ASP table the same way CIC-ASP table can be
> > > > > >  used to route ISUP messages to the corresponding MGCs.Basically CIC was
> > > > > >  removed from RFC to avoid complex problems say a msg comes with CIC
> > > > > >  range,this CIC range can be split across MGCs here the problem is
> > > > > >  potential.
> > > > > > 
> > > > > >  These are the reasons caused me to say that with static configuration
> > > > > >  CIC or/and SSN can be supported if an SG vendor and ASP vendor has
> > > > > >  mutual understanding.(but this is outside the scope of M3UA RFC)
> > > > > > 
> > > > > > --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, 4 Jul 2003 kakaul@hss.hns.com wrote:
> > > > > > 
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > Hi Samule,
> > > > > > >   Does MTP3 use CIC or SSN for load-distribution among the available
> > > > > > > links.I guess only SLS parameter in MTP3 is used for load-distribution
> > > > > > > among the available link.I want to know how an MTP3  user (ISUP or SCCP)
> > > > > > > selects SLS for effective load distribution at MTP3 layer among the links
> > > > > > > available to reach the  a given DPC.
> > > > > > >
> > > > > > > Regards
> > > > > > > Kamesh Kaul
> > > > > > > www.hssworld.com
> > > > > > > Hughes Software Systems
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
> > > > > > > 07/04/2003 09:29:51 AM
> > > > > > >
> > > > > > > Sent by:    sigtran-admin@ietf.org
> > > > > > >
> > > > > > >
> > > > > > > To:    sigtran@ietf.org
> > > > > > > cc:
> > > > > > >
> > > > > > > Subject:    Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >  this is just an info regarding CIC and SSN - load-sharing.
> > > > > > >
> > > > > > >  A node can load-share its traffic based on CIC or SSN with the
> > > > > > >  help of static configuration but this is outside of the scope of M3UA
> > > > > > >  rfc.When an ASP vendor and SGP vendor agrees to do load-sharing based on
> > > > > > >  CIC or SSN with static confiuration they are welcome to do this,
> > > > > > >  since this has not been standardised, taking this to interop will
> > > > > > >  lead to problems.So almost everybody agrees to delete CIC and SSN
> > > > > > >  loadsharing concept from MIB-04.
> > > > > > >
> > > > > > >
> > > > > > > --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, 3 Jul 2003, Brian F. G. Bidulock wrote:
> > > > > > >
> > > > > > > > agurmukhani,
> > > > > > > >
> > > > > > > > I think that they should be removed from the MIB, but we did NOT agree
> > > > > > > > that CIC or SSN should not be used for loadsharing.  Any criteria that
> > > > > > > > the SG/ASP desires to use for loadsharing being congnisant of the need
> > > > > > > > for ordering of user part messages is allowed for loadsharing.  We DID
> > > > > > > > agree to remove CIC, SSN from the Routing Key, because a rational and
> > > > > > > > interoperable basis for splitting CICs and SSNs across AS could not be
> > > > > > > > decided.
> > > > > > > >
> > > > > > > > The meanse oadsharing, on the other hand, is not specified in M3UA.
> > > > > > > > Some examples are given, but no requirements.  Some things that could
> > > > > > > > clearly be used are SLS, DPC/SLS, OPC/DPC/SLS, CIC, DPC/CIC,
> > > > > > OPC/DPC/CIC,
> > > > > > > > OPC/DPC/CIC/SLS, SSN, ....  and other permutations of these and even
> > > > > > > > higher layer parameters such as TID.
> > > > > > > >
> > > > > > > > My argument for removing them from the MIB is not that they cannot be
> > > > > > > > used, but that they have not been standardized.
> > > > > > > >
> > > > > > > > --brian
> > > > > > > >
> > > > > > > > agurmukhani@hss.hns.com wrote:
> > > > > > > (Thu, 03 Jul 2003 10:23:35)
> > > > > > > > >
> > > > > > > > >
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > Hi...
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > In my understanding and  as per the discussion and consensus on the
> > > > > > > mailing
> > > > > > > > > list , everyone agreed that  CIC and SSN should not be used as
> > > > > > > loadsharing
> > > > > > > > > basis
> > > > > > > > > be it within streams in a particular SCTP association or for active
> > > > > > > > > signalling processes.
> > > > > > > > >
> > > > > > > > > So would these be removed from  the m3uaStreamLshMethod  and
> > > > > > > > > m3uaSpLshMethod OBJECT-TYPE s in M3UA MIB draft.
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > Thanks and Regards
> > > > > > > > > Anjali
> > > > > > > > > www.hssworld.com
> > > > > > > > > Hughes Software Systems.
> > > > > > > > >
> > > > > > > > > "DISCLAIMER: This message is proprietary to Hughes Software Systems
> > > > > > > Limited
> > > > > > > > > (HSS) and is intended solely for the use of the individual to whom it
> > > > > > > is
> > > > > > > > > addressed. It may contain  privileged or confidential information and
> > > > > > > > > should not be circulated or used for any purpose other than for what
> > > > > > it
> > > > > > > is
> > > > > > > > > intended. If you have received this message in error, please notify
> > > > > > the
> > > > > > > > > originator immediately. If you are not the intended recipient, you
> > > > > > are
> > > > > > > > > notified that you are strictly prohibited from using, copying,
> > > > > > > altering, or
> > > > > > > > > disclosing the contents of this message. HSS accepts no
> > > > > > responsibility
> > > > > > > for
> > > > > > > > > loss or damage arising from the use of the information transmitted by
> > > > > > > this
> > > > > > > > > email including damage from virus."
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > _______________________________________________
> > > > > > > > > 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
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > > _______________________________________________
> > > > > > 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

-- 
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 exim@www1.ietf.org  Sat Jul  5 02:56:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA01454
	for <sigtran-archive@odin.ietf.org>; Sat, 5 Jul 2003 02:56:33 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Ygxq-0007PK-PF
	for sigtran-archive@odin.ietf.org; Sat, 05 Jul 2003 02:56:06 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h656u6Te028475
	for sigtran-archive@odin.ietf.org; Sat, 5 Jul 2003 02:56:06 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Ygxk-0007P1-Lk; Sat, 05 Jul 2003 02:56:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Ygxj-0007Oh-1L
	for sigtran@optimus.ietf.org; Sat, 05 Jul 2003 02:55:59 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA01442
	for <sigtran@ietf.org>; Sat, 5 Jul 2003 02:55:55 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19Ygxf-0006mZ-00
	for sigtran@ietf.org; Sat, 05 Jul 2003 02:55:55 -0400
Received: from auds951.usa.alcatel.com ([143.209.238.80])
	by ietf-mx with esmtp (Exim 4.12)
	id 19Ygxe-0006mW-00
	for sigtran@ietf.org; Sat, 05 Jul 2003 02:55:54 -0400
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h656tPwA010932;
	Sat, 5 Jul 2003 01:55:25 -0500 (CDT)
Received: from sun2594.ssd.usa.alcatel.com (sun2594.ssd.usa.alcatel.com [143.209.250.252])
	by ssd.usa.alcatel.com (8.12.8/8.12.8) with ESMTP id h656ss1C014042;
	Sat, 5 Jul 2003 01:54:54 -0500 (CDT)
Date: Sat, 5 Jul 2003 01:54:54 -0500 (CDT)
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 MIB Draft05: Loadsharing on CIC ,SSN
In-Reply-To: <20030705004106.A3494@openss7.org>
Message-ID: <Pine.SOL.4.10.10307050153190.25952-100000@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,

 If an MSU has both CIC and SLS, based on what field load-sharing 
 takesplace?

--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, 5 Jul 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Yes, the passage I underlined was incorrect.
> 
> --brian
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Sat, 05 Jul 2003 01:37:59)
> > 
> > Brian,
> >  
> >  I dont know the intention of your email.In this i two paragraphs 
> >  2nd para says ISUP layer at MGC constructs an IAM (WITH SLS) and sends
> >  it to STP(SG), now STP load-shares its traffic based on received SLS.
> > 
> >  
> > --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, 5 Jul 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > See the underlined sentences that are incorrect:
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                                                     (Sat, 05 Jul 2003 01:15:26)
> > > > 
> > > > Brian,
> > > > 
> > > > See the underlined sentences.
> > > > 
> > > >   Alcatel USA, 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, 4 Jul 2003, Brian F. G. Bidulock wrote:
> > > > 
> > > > > Samuel,
> > > > > 
> > > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 04 Jul 2003 02:36:53)
> > > > > > 
> > > > > > Kamesh,
> > > > > > 
> > > > > >   MTP3 layer uses SLS to route the messages(pertaining to the same
> > > > > >   logically  related data) to the destination node.Users(ISUP or SCCP)
> > > > > >   can not instruct MTP3 to do any work related with loadsharing, it is
> > >                                                                        -----
> > > > > >   upto the decision of MTP3 layer to decide the field(like SLS) on which
> > >         ----------------------------------------------------------------------
> > > > > >   it is going to loadshare its traffic.
> > >         -------------------------------------
> > > > > 
> > > > > Incorrect.  It is the MTP-User's responsibility to select SLS value to
> > > > > ensure load sharing of user messages with consideration for sequence
> > > > > requirements for interrelated messages.  See Q.704, Q.714, Q.764.
> > > > > 
> > > > > --brian
> > > > > 
> > > > > > 
> > > > 
> > > > > >   To the point, say for example you have an IAM which has LI > 272, ISUP
> > > >                                                                         --
> > > > > >   layer would split this IAM into 2 IAMs(just an example) and send it
> > > >       ------------------------------------------------------------------
> > > > > >   to STP(SG), remeber both IAMs must have the same SLS. Now STP'S MTP3
> > > >       ------------------------------------------------------
> > > > > >   layer finds an outgoing linkset based on the DPC received on the MSU.
> > > > > > 
> > > > > >   Once linkset has been selected ,SLS is used to select a particular link
> > > > > >   within the linkset, so the above 2 IAMs will be sent in a sequence to
> > > > > >   the destination, here there is no way of mis-sequencing of MSUs.
> > > > > > 
> > > > > >   Is this what you expect?
> > > > > > 
> > > > > > --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, 4 Jul 2003 kakaul@hss.hns.com wrote:
> > > > > > 
> > > > > > > 
> > > > > > > 
> > > > > > > 
> > > > > > > 
> > > > > > > Hi Samule,
> > > > > > >   You have not replied for my second question :
> > > > > > > I want to know how an MTP3  user (ISUP or SCCP) selects SLS for effective
> > > > > > > load distribution at MTP3 layer among the links available to reach a given
> > > > > > > DPC. Is it that SLS is chosen per DPC for a logical session or globally per
> > > > > > > each logical session irrespective of DPC for which this session is
> > > > > > > intended.I know it is implementation dependent option but what do most of
> > > > > > > the implementers opt for. By logical session i mean logically related data.
> > > > > > > 
> > > > > > > Regards
> > > > > > > Kamesh Kaul
> > > > > > > www.hssworld.com
> > > > > > > Hughes Software Systems
> > > > > > > 
> > > > > > > 
> > > > > > > 
> > > > > > > 
> > > > > > > 
> > > > > > > "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
> > > > > > > 07/04/2003 10:58:14 AM
> > > > > > > 
> > > > > > > Sent by:    sigtran-admin@ietf.org
> > > > > > > 
> > > > > > > 
> > > > > > > To:    sigtran@ietf.org
> > > > > > > cc:
> > > > > > > 
> > > > > > > Subject:    Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
> > > > > > > 
> > > > > > > 
> > > > > > > 
> > > > > > > Pastor,
> > > > > > > 
> > > > > > >  MIB-05 does not have 'load-shar on dpc',i was referring mib-04 which lead
> > > > > > >  me into confusion.
> > > > > > > 
> > > > > > >  Thanks kakaul for our mail.
> > > > > > > 
> > > > > > > Kakaul,
> > > > > > > 
> > > > > > >  As you said MTP3 does not route messages based on CIC or SSN because
> > > > > > >  there are no standards available, i agree with it.
> > > > > > > 
> > > > > > >  Have a look, suppose SG receives an MSU with routing indicator="Route on
> > > > > > >  PointCode/SSN", if your SG supports static configuration then it is easy
> > > > > > >  to route these messages to the corresponding subSystems based on
> > > > > > >  statically configured SSN-ASP table the same way CIC-ASP table can be
> > > > > > >  used to route ISUP messages to the corresponding MGCs.Basically CIC was
> > > > > > >  removed from RFC to avoid complex problems say a msg comes with CIC
> > > > > > >  range,this CIC range can be split across MGCs here the problem is
> > > > > > >  potential.
> > > > > > > 
> > > > > > >  These are the reasons caused me to say that with static configuration
> > > > > > >  CIC or/and SSN can be supported if an SG vendor and ASP vendor has
> > > > > > >  mutual understanding.(but this is outside the scope of M3UA RFC)
> > > > > > > 
> > > > > > > --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, 4 Jul 2003 kakaul@hss.hns.com wrote:
> > > > > > > 
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > > Hi Samule,
> > > > > > > >   Does MTP3 use CIC or SSN for load-distribution among the available
> > > > > > > > links.I guess only SLS parameter in MTP3 is used for load-distribution
> > > > > > > > among the available link.I want to know how an MTP3  user (ISUP or SCCP)
> > > > > > > > selects SLS for effective load distribution at MTP3 layer among the links
> > > > > > > > available to reach the  a given DPC.
> > > > > > > >
> > > > > > > > Regards
> > > > > > > > Kamesh Kaul
> > > > > > > > www.hssworld.com
> > > > > > > > Hughes Software Systems
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > > "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
> > > > > > > > 07/04/2003 09:29:51 AM
> > > > > > > >
> > > > > > > > Sent by:    sigtran-admin@ietf.org
> > > > > > > >
> > > > > > > >
> > > > > > > > To:    sigtran@ietf.org
> > > > > > > > cc:
> > > > > > > >
> > > > > > > > Subject:    Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > >  this is just an info regarding CIC and SSN - load-sharing.
> > > > > > > >
> > > > > > > >  A node can load-share its traffic based on CIC or SSN with the
> > > > > > > >  help of static configuration but this is outside of the scope of M3UA
> > > > > > > >  rfc.When an ASP vendor and SGP vendor agrees to do load-sharing based on
> > > > > > > >  CIC or SSN with static confiuration they are welcome to do this,
> > > > > > > >  since this has not been standardised, taking this to interop will
> > > > > > > >  lead to problems.So almost everybody agrees to delete CIC and SSN
> > > > > > > >  loadsharing concept from MIB-04.
> > > > > > > >
> > > > > > > >
> > > > > > > > --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, 3 Jul 2003, Brian F. G. Bidulock wrote:
> > > > > > > >
> > > > > > > > > agurmukhani,
> > > > > > > > >
> > > > > > > > > I think that they should be removed from the MIB, but we did NOT agree
> > > > > > > > > that CIC or SSN should not be used for loadsharing.  Any criteria that
> > > > > > > > > the SG/ASP desires to use for loadsharing being congnisant of the need
> > > > > > > > > for ordering of user part messages is allowed for loadsharing.  We DID
> > > > > > > > > agree to remove CIC, SSN from the Routing Key, because a rational and
> > > > > > > > > interoperable basis for splitting CICs and SSNs across AS could not be
> > > > > > > > > decided.
> > > > > > > > >
> > > > > > > > > The meanse oadsharing, on the other hand, is not specified in M3UA.
> > > > > > > > > Some examples are given, but no requirements.  Some things that could
> > > > > > > > > clearly be used are SLS, DPC/SLS, OPC/DPC/SLS, CIC, DPC/CIC,
> > > > > > > OPC/DPC/CIC,
> > > > > > > > > OPC/DPC/CIC/SLS, SSN, ....  and other permutations of these and even
> > > > > > > > > higher layer parameters such as TID.
> > > > > > > > >
> > > > > > > > > My argument for removing them from the MIB is not that they cannot be
> > > > > > > > > used, but that they have not been standardized.
> > > > > > > > >
> > > > > > > > > --brian
> > > > > > > > >
> > > > > > > > > agurmukhani@hss.hns.com wrote:
> > > > > > > > (Thu, 03 Jul 2003 10:23:35)
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > Hi...
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > In my understanding and  as per the discussion and consensus on the
> > > > > > > > mailing
> > > > > > > > > > list , everyone agreed that  CIC and SSN should not be used as
> > > > > > > > loadsharing
> > > > > > > > > > basis
> > > > > > > > > > be it within streams in a particular SCTP association or for active
> > > > > > > > > > signalling processes.
> > > > > > > > > >
> > > > > > > > > > So would these be removed from  the m3uaStreamLshMethod  and
> > > > > > > > > > m3uaSpLshMethod OBJECT-TYPE s in M3UA MIB draft.
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > Thanks and Regards
> > > > > > > > > > Anjali
> > > > > > > > > > www.hssworld.com
> > > > > > > > > > Hughes Software Systems.
> > > > > > > > > >
> > > > > > > > > > "DISCLAIMER: This message is proprietary to Hughes Software Systems
> > > > > > > > Limited
> > > > > > > > > > (HSS) and is intended solely for the use of the individual to whom it
> > > > > > > > is
> > > > > > > > > > addressed. It may contain  privileged or confidential information and
> > > > > > > > > > should not be circulated or used for any purpose other than for what
> > > > > > > it
> > > > > > > > is
> > > > > > > > > > intended. If you have received this message in error, please notify
> > > > > > > the
> > > > > > > > > > originator immediately. If you are not the intended recipient, you
> > > > > > > are
> > > > > > > > > > notified that you are strictly prohibited from using, copying,
> > > > > > > > altering, or
> > > > > > > > > > disclosing the contents of this message. HSS accepts no
> > > > > > > responsibility
> > > > > > > > for
> > > > > > > > > > loss or damage arising from the use of the information transmitted by
> > > > > > > > this
> > > > > > > > > > email including damage from virus."
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > _______________________________________________
> > > > > > > > > > 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
> > > > > > > 
> > > > > > > 
> > > > > > > 
> > > > > > > _______________________________________________
> > > > > > > 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
> 
> -- 
> 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 exim@www1.ietf.org  Sat Jul  5 03:02:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA01565
	for <sigtran-archive@odin.ietf.org>; Sat, 5 Jul 2003 03:02:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Yh3a-0007tX-Bt
	for sigtran-archive@odin.ietf.org; Sat, 05 Jul 2003 03:02:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h65722T2030340
	for sigtran-archive@odin.ietf.org; Sat, 5 Jul 2003 03:02:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Yh3Z-0007t7-Lx; Sat, 05 Jul 2003 03:02:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Yh2v-0007mn-BD
	for sigtran@optimus.ietf.org; Sat, 05 Jul 2003 03:01:21 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA01556
	for <sigtran@ietf.org>; Sat, 5 Jul 2003 03:01:17 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19Yh2r-0006pr-00
	for sigtran@ietf.org; Sat, 05 Jul 2003 03:01:17 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[v/g/O4wBc5to4QEroNGl+9CguPYHXIY4])
	by ietf-mx with esmtp (Exim 4.12)
	id 19Yh2q-0006po-00
	for sigtran@ietf.org; Sat, 05 Jul 2003 03:01:16 -0400
Received: from ns.pigworks.openss7.net (IDENT:ujKWTgek2T9nr2Y/mjIQclHvsYafEFJk@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6571HX10540;
	Sat, 5 Jul 2003 01:01:17 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6571HL03934;
	Sat, 5 Jul 2003 01:01:17 -0600
Date: Sat, 5 Jul 2003 01:01:16 -0600
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 MIB Draft05: Loadsharing on CIC ,SSN
Message-ID: <20030705010116.A3701@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	sigtran@ietf.org
References: <20030705004106.A3494@openss7.org> <Pine.SOL.4.10.10307050153190.25952-100000@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.4.10.10307050153190.25952-100000@sun2594.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Sat, Jul 05, 2003 at 01:54:54AM -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>

Samuel,

See Note 1 Table 1/Q.701 and Section 2.2 and 2.3/Q.704.  Load sharing in MTP3
Signalling Routing Function takes place on SLS only.  For ITU-T and ETSI Q.764
CIC and SLS are statically related.  For ANSI T1.113 they are related only for
the duration of a call.

--brian

Samuel Dur D. Jeyaseelan wrote:                       (Sat, 05 Jul 2003 01:54:54)
> 
> Brian,
> 
>  If an MSU has both CIC and SLS, based on what field load-sharing 
>  takesplace?
> 
> --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, 5 Jul 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Yes, the passage I underlined was incorrect.
> > 
> > --brian
> > 
> > Samuel Dur D. Jeyaseelan wrote:                   (Sat, 05 Jul 2003 01:37:59)
> > > 
> > > Brian,
> > >  
> > >  I dont know the intention of your email.In this i two paragraphs 
> > >  2nd para says ISUP layer at MGC constructs an IAM (WITH SLS) and sends
> > >  it to STP(SG), now STP load-shares its traffic based on received SLS.
> > > 
> > >  
> > > --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, 5 Jul 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Samuel,
> > > > 
> > > > See the underlined sentences that are incorrect:
> > > > 
> > > > Samuel Dur D. Jeyaseelan wrote:               (Sat, 05 Jul 2003 01:15:26)
> > > > > 
> > > > > Brian,
> > > > > 
> > > > > See the underlined sentences.
> > > > > 
> > > > >   Alcatel USA, 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, 4 Jul 2003, Brian F. G. Bidulock wrote:
> > > > > 
> > > > > > Samuel,
> > > > > > 
> > > > > > Samuel Dur D. Jeyaseelan wrote:           (Fri, 04 Jul 2003 02:36:53)
> > > > > > > 
> > > > > > > Kamesh,
> > > > > > > 
> > > > > > >   MTP3 layer uses SLS to route the messages(pertaining to the same
> > > > > > >   logically  related data) to the destination node.Users(ISUP or SCCP)
> > > > > > >   can not instruct MTP3 to do any work related with loadsharing, it is
> > > >                                                                        -----
> > > > > > >   upto the decision of MTP3 layer to decide the field(like SLS) on which
> > > >         ----------------------------------------------------------------------
> > > > > > >   it is going to loadshare its traffic.
> > > >         -------------------------------------
> > > > > > 
> > > > > > Incorrect.  It is the MTP-User's responsibility to select SLS value to
> > > > > > ensure load sharing of user messages with consideration for sequence
> > > > > > requirements for interrelated messages.  See Q.704, Q.714, Q.764.
> > > > > > 
> > > > > > --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 exim@www1.ietf.org  Sat Jul  5 03:13:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA01738
	for <sigtran-archive@odin.ietf.org>; Sat, 5 Jul 2003 03:13:33 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YhEH-0000Bh-Qr
	for sigtran-archive@odin.ietf.org; Sat, 05 Jul 2003 03:13:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h657D5nN000721
	for sigtran-archive@odin.ietf.org; Sat, 5 Jul 2003 03:13:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YhEB-0000A4-Ty; Sat, 05 Jul 2003 03:12:59 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YhE9-00009s-9j
	for sigtran@optimus.ietf.org; Sat, 05 Jul 2003 03:12:57 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA01728
	for <sigtran@ietf.org>; Sat, 5 Jul 2003 03:12:54 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YhE7-0006xM-00
	for sigtran@ietf.org; Sat, 05 Jul 2003 03:12:55 -0400
Received: from auds951.usa.alcatel.com ([143.209.238.80])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YhE6-0006xI-00
	for sigtran@ietf.org; Sat, 05 Jul 2003 03:12:54 -0400
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h657CPwA011601
	for <sigtran@ietf.org>; Sat, 5 Jul 2003 02:12:25 -0500 (CDT)
Received: from sun2594.ssd.usa.alcatel.com (sun2594.ssd.usa.alcatel.com [143.209.250.252])
	by ssd.usa.alcatel.com (8.12.8/8.12.8) with ESMTP id h657Bf1C016061
	for <sigtran@ietf.org>; Sat, 5 Jul 2003 02:11:41 -0500 (CDT)
Date: Sat, 5 Jul 2003 02:11:41 -0500 (CDT)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
In-Reply-To: <20030705010116.A3701@openss7.org>
Message-ID: <Pine.SOL.4.10.10307050207520.25952-100000@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,

You are not able to answer my email to the point.Please now you read my
email(2 paragraphs) from top to bottom it can easily be understood.
If you read somewhere middle of the line then you will be sending one more
emial to me.

Thanks
--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, 5 Jul 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> See Note 1 Table 1/Q.701 and Section 2.2 and 2.3/Q.704.  Load sharing in MTP3
> Signalling Routing Function takes place on SLS only.  For ITU-T and ETSI Q.764
> CIC and SLS are statically related.  For ANSI T1.113 they are related only for
> the duration of a call.
> 
> --brian
> 
> Samuel Dur D. Jeyaseelan wrote:                       (Sat, 05 Jul 2003 01:54:54)
> > 
> > Brian,
> > 
> >  If an MSU has both CIC and SLS, based on what field load-sharing 
> >  takesplace?
> > 
> > --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, 5 Jul 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > Yes, the passage I underlined was incorrect.
> > > 
> > > --brian
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                   (Sat, 05 Jul 2003 01:37:59)
> > > > 
> > > > Brian,
> > > >  
> > > >  I dont know the intention of your email.In this i two paragraphs 
> > > >  2nd para says ISUP layer at MGC constructs an IAM (WITH SLS) and sends
> > > >  it to STP(SG), now STP load-shares its traffic based on received SLS.
> > > > 
> > > >  
> > > > --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, 5 Jul 2003, Brian F. G. Bidulock wrote:
> > > > 
> > > > > Samuel,
> > > > > 
> > > > > See the underlined sentences that are incorrect:
> > > > > 
> > > > > Samuel Dur D. Jeyaseelan wrote:               (Sat, 05 Jul 2003 01:15:26)
> > > > > > 
> > > > > > Brian,
> > > > > > 
> > > > > > See the underlined sentences.
> > > > > > 
> > > > > >   Alcatel USA, 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, 4 Jul 2003, Brian F. G. Bidulock wrote:
> > > > > > 
> > > > > > > Samuel,
> > > > > > > 
> > > > > > > Samuel Dur D. Jeyaseelan wrote:           (Fri, 04 Jul 2003 02:36:53)
> > > > > > > > 
> > > > > > > > Kamesh,
> > > > > > > > 
> > > > > > > >   MTP3 layer uses SLS to route the messages(pertaining to the same
> > > > > > > >   logically  related data) to the destination node.Users(ISUP or SCCP)
> > > > > > > >   can not instruct MTP3 to do any work related with loadsharing, it is
> > > > >                                                                        -----
> > > > > > > >   upto the decision of MTP3 layer to decide the field(like SLS) on which
> > > > >         ----------------------------------------------------------------------
> > > > > > > >   it is going to loadshare its traffic.
> > > > >         -------------------------------------
> > > > > > > 
> > > > > > > Incorrect.  It is the MTP-User's responsibility to select SLS value to
> > > > > > > ensure load sharing of user messages with consideration for sequence
> > > > > > > requirements for interrelated messages.  See Q.704, Q.714, Q.764.
> > > > > > > 
> > > > > > > --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 exim@www1.ietf.org  Sat Jul  5 03:27:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA02040
	for <sigtran-archive@odin.ietf.org>; Sat, 5 Jul 2003 03:27:33 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YhRp-0000yc-LK
	for sigtran-archive@odin.ietf.org; Sat, 05 Jul 2003 03:27:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h657R5Vj003753
	for sigtran-archive@odin.ietf.org; Sat, 5 Jul 2003 03:27:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YhRl-0000yI-NL; Sat, 05 Jul 2003 03:27:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YhRN-0000tA-Ls
	for sigtran@optimus.ietf.org; Sat, 05 Jul 2003 03:26:37 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA02026
	for <sigtran@ietf.org>; Sat, 5 Jul 2003 03:26:35 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YhRL-00075O-00
	for sigtran@ietf.org; Sat, 05 Jul 2003 03:26:35 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[gBCWLnqGw7vXB3E2VYbOSTJFvv+GdQFH])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YhRK-00075K-00
	for sigtran@ietf.org; Sat, 05 Jul 2003 03:26:34 -0400
Received: from ns.pigworks.openss7.net (IDENT:wAvNs957GfHDBllGdD1gid2fvrq0hWiT@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h657QZX10794;
	Sat, 5 Jul 2003 01:26:35 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h657QY904116;
	Sat, 5 Jul 2003 01:26:34 -0600
Date: Sat, 5 Jul 2003 01:26:34 -0600
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 MIB Draft05: Loadsharing on CIC ,SSN
Message-ID: <20030705012634.B3701@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	sigtran@ietf.org
References: <20030705010116.A3701@openss7.org> <Pine.SOL.4.10.10307050207520.25952-100000@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.4.10.10307050207520.25952-100000@sun2594.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Sat, Jul 05, 2003 at 02:11:41AM -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>

Samuel,

This first paragraph (underlined portions) is incorrect per the
standards and sections quoted.

The second paragraph is innane because all the messages relating to
the same ISUP call have the same SLS value, regardless of whether they
are segmented or not.

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Sat, 05 Jul 2003 02:11:41)
> 
> Brian,
> 
> You are not able to answer my email to the point.Please now you read my
> email(2 paragraphs) from top to bottom it can easily be understood.
> If you read somewhere middle of the line then you will be sending one more
> emial to me.
> 
> Thanks
> --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, 5 Jul 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > See Note 1 Table 1/Q.701 and Section 2.2 and 2.3/Q.704.  Load sharing in MTP3
> > Signalling Routing Function takes place on SLS only.  For ITU-T and ETSI Q.764
> > CIC and SLS are statically related.  For ANSI T1.113 they are related only for
> > the duration of a call.
> > 
> > --brian
> > 
> > Samuel Dur D. Jeyaseelan wrote:                       (Sat, 05 Jul 2003 01:54:54)
> > > 
> > > Brian,
> > > 
> > >  If an MSU has both CIC and SLS, based on what field load-sharing 
> > >  takesplace?
> > > 
> > > --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, 5 Jul 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Samuel,
> > > > 
> > > > Yes, the passage I underlined was incorrect.
> > > > 
> > > > --brian
> > > > 
> > > > Samuel Dur D. Jeyaseelan wrote:                   (Sat, 05 Jul 2003 01:37:59)
> > > > > 
> > > > > Brian,
> > > > >  
> > > > >  I dont know the intention of your email.In this i two paragraphs 
> > > > >  2nd para says ISUP layer at MGC constructs an IAM (WITH SLS) and sends
> > > > >  it to STP(SG), now STP load-shares its traffic based on received SLS.
> > > > > 
> > > > >  
> > > > > --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, 5 Jul 2003, Brian F. G. Bidulock wrote:
> > > > > 
> > > > > > Samuel,
> > > > > > 
> > > > > > See the underlined sentences that are incorrect:
> > > > > > 
> > > > > > Samuel Dur D. Jeyaseelan wrote:               (Sat, 05 Jul 2003 01:15:26)
> > > > > > > 
> > > > > > > Brian,
> > > > > > > 
> > > > > > > See the underlined sentences.
> > > > > > > 
> > > > > > >   Alcatel USA, 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, 4 Jul 2003, Brian F. G. Bidulock wrote:
> > > > > > > 
> > > > > > > > Samuel,
> > > > > > > > 
> > > > > > > > Samuel Dur D. Jeyaseelan wrote:           (Fri, 04 Jul 2003 02:36:53)
> > > > > > > > > 
> > > > > > > > > Kamesh,
> > > > > > > > > 
> > > > > > > > >   MTP3 layer uses SLS to route the messages(pertaining to the same
> > > > > > > > >   logically  related data) to the destination node.Users(ISUP or SCCP)
> > > > > > > > >   can not instruct MTP3 to do any work related with loadsharing, it is
> > > > > >                                                                        -----
> > > > > > > > >   upto the decision of MTP3 layer to decide the field(like SLS) on which
> > > > > >         ----------------------------------------------------------------------
> > > > > > > > >   it is going to loadshare its traffic.
> > > > > >         -------------------------------------
> > > > > > > > 
> > > > > > > > Incorrect.  It is the MTP-User's responsibility to select SLS value to
> > > > > > > > ensure load sharing of user messages with consideration for sequence
> > > > > > > > requirements for interrelated messages.  See Q.704, Q.714, Q.764.
> > > > > > > > 
> > > > > > > > --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 exim@www1.ietf.org  Mon Jul  7 10:33:35 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA27587
	for <sigtran-archive@odin.ietf.org>; Mon, 7 Jul 2003 10:33:35 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZX3F-0003z2-6q
	for sigtran-archive@odin.ietf.org; Mon, 07 Jul 2003 10:33:09 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h67EX9KC015294
	for sigtran-archive@odin.ietf.org; Mon, 7 Jul 2003 10:33:09 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZX37-0003y7-OQ; Mon, 07 Jul 2003 10:33:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZX2O-0003uY-0M
	for sigtran@optimus.ietf.org; Mon, 07 Jul 2003 10:32:16 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA27544
	for <sigtran@ietf.org>; Mon, 7 Jul 2003 10:32:12 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ZX2L-0004oq-00
	for sigtran@ietf.org; Mon, 07 Jul 2003 10:32:13 -0400
Received: from sj-iport-1-in.cisco.com ([171.71.176.70] helo=sj-iport-1.cisco.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19ZX2L-0004om-00
	for sigtran@ietf.org; Mon, 07 Jul 2003 10:32:13 -0400
Received: from fruitpie.cisco.com (IDENT:mirapoint@fruitpie.cisco.com [64.102.16.27])
	by sj-core-2.cisco.com (8.12.9/8.12.6) with ESMTP id h67EVeMr001983
	for <sigtran@ietf.org>; Mon, 7 Jul 2003 07:31:40 -0700 (PDT)
Received: from kmorneau-w2k01.cisco.com (rtp-vpn2-434.cisco.com [10.82.241.178])
	by fruitpie.cisco.com (Mirapoint Messaging Server MOS 3.3.3-GR)
	with ESMTP id AGE91778;
	Mon, 7 Jul 2003 07:31:39 -0700 (PDT)
Message-Id: <4.3.2.7.2.20030707102133.04cc0528@cia.cisco.com>
X-Sender: kmorneau@cia.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Mon, 07 Jul 2003 10:31:36 -0400
To: sigtran@ietf.org
From: Ken Morneault <kmorneau@cisco.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Subject: [Sigtran] Sigtran Plugtest at ETSI 9/03
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>



All:

Which version of M2PA should be tested at the upcoming
Sigtran plugtest in 9/03?  Currently, the ETSI Plugtest site
indicates version 8, but version 9 was just released.  Version
6 was tested at the last Plugtest.

Please email your preference.

Regards,

Ken


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



From exim@www1.ietf.org  Mon Jul  7 11:01:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA28735
	for <sigtran-archive@odin.ietf.org>; Mon, 7 Jul 2003 11:01:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZXUF-0005TH-7B
	for sigtran-archive@odin.ietf.org; Mon, 07 Jul 2003 11:01:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h67F13Cb021023
	for sigtran-archive@odin.ietf.org; Mon, 7 Jul 2003 11:01:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZXUD-0005St-OU; Mon, 07 Jul 2003 11:01:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZXU2-0005S6-Hw
	for sigtran@optimus.ietf.org; Mon, 07 Jul 2003 11:00:50 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA28693
	for <sigtran@ietf.org>; Mon, 7 Jul 2003 11:00:46 -0400 (EDT)
From: vissharma@hss.hns.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ZXU0-0005AM-00
	for sigtran@ietf.org; Mon, 07 Jul 2003 11:00:48 -0400
Received: from hindon.hss.co.in ([202.54.26.202])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ZXTy-0005AE-00
	for sigtran@ietf.org; Mon, 07 Jul 2003 11:00:46 -0400
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 h67EvPm10357
	for <sigtran@ietf.org>; Mon, 7 Jul 2003 20:27:25 +0530 (IST)
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h67EvOV10351
	for <sigtran@ietf.org>; Mon, 7 Jul 2003 20:27:24 +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 ESMTP id h67F12o19611
	for <sigtran@ietf.org>; Mon, 7 Jul 2003 20:31:02 +0530 (IST)
To: sigtran@ietf.org
Date: Mon, 7 Jul 2003 20:25:44 +0530
Message-ID: <OFB18495D4.2465602B-ON65256D5C.0050957D@hss.hns.com>
X-MIMETrack: Serialize by Router on Sandesh/HSS(Release 6.0|September 26, 2002) at 07/07/2003
 08:24:37 PM
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Subject: [Sigtran] m3ua -ASP receives an ASP Inactive Ack out of blow scenario -Please comment
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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 ,

Refer to RFC 3332 Section 4.3.4.4  "ASP Inactive Procedures"
RFC s describes handling of ASP Inactive Ack out of blow scenario ie.
without ASP have sent ASP Inactive in following text

If the ASP receives an ASP Inactive Ack without having sent an ASP Inactive message, the ASP should now consider itself as in the ASP-INACTIVE state.
If the ASP was previously in the ASP-ACTIVE state, the ASP should then initiate procedures to return itself to its previous state.

Please comment on the following :-

1. What are the situations /Network scenarios when an ASP will get such ASP Inactive Acks.
2. When an ASP receive such a message in active state what are that parameters which will govern when the procedures to return to
  previous state shall be started by ASP .If there are  different thought of scenarios in which purposefully ASP may receive such Ack messages
  then i think the parameters may be different depending on case to case .


Thanks and Regards
Vishavdeep Sharma
www.hssworld.com


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



From exim@www1.ietf.org  Mon Jul  7 17:12:39 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA16200
	for <sigtran-archive@odin.ietf.org>; Mon, 7 Jul 2003 17:12:39 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZdHP-0006Xq-CN
	for sigtran-archive@odin.ietf.org; Mon, 07 Jul 2003 17:12:12 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h67LCBcp025121
	for sigtran-archive@odin.ietf.org; Mon, 7 Jul 2003 17:12:11 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZdHH-0006WK-Fj; Mon, 07 Jul 2003 17:12:03 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19YQLW-0002Ie-2C
	for sigtran@optimus.ietf.org; Fri, 04 Jul 2003 09:11:26 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA25329
	for <sigtran@ietf.org>; Fri, 4 Jul 2003 09:11:23 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YQLU-0005Cj-00
	for sigtran@ietf.org; Fri, 04 Jul 2003 09:11:24 -0400
Received: from zdemail04.zdem.compaq.com ([161.114.112.28])
	by ietf-mx with esmtp (Exim 4.12)
	id 19YQLT-0005CP-00
	for sigtran@ietf.org; Fri, 04 Jul 2003 09:11:23 -0400
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail04.zdem.compaq.com (Postfix) with ESMTP
	id 635C8E2A; Fri,  4 Jul 2003 15:10:47 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Fri, 4 Jul 2003 15:10:47 +0200
Content-Class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
X-MimeOLE: Produced By Microsoft Exchange V6.0.6375.0
Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Date: Fri, 4 Jul 2003 15:10:45 +0200
Message-ID: <BE462396B4A022449E0EB2DD4753D054022578E8@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Thread-Index: AcNCF4oY+tc9NDhuTUuIn8281QARkgAFMD9g
From: "Desiderio, Didier" <didier.desiderio@hp.com>
To: <john.loughney@nokia.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 04 Jul 2003 13:10:47.0094 (UTC) FILETIME=[AED93D60:01C3422D]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable



Some comments...



section 3.2.2 Connectionless Data Response (CLDR)=20



If we refer to

ANSI T1.112 chap 4 & 4.11=20

     T1.112.3-26 Table 11A/T1.112.3

ITU-T Q713 Chap 3.16, 3.20 and 4.11



Data parameter seems to be mandatory in UDTS, XUDTS and LUDTS.

Isn't it possible to have the DATA parameter of the CLDR message=20

mandatory or semi optional (at least mandatory in case of ASP-SGP=20

communication) and not only optional.



Regards



Didier

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



From exim@www1.ietf.org  Mon Jul  7 17:12:41 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA16215
	for <sigtran-archive@odin.ietf.org>; Mon, 7 Jul 2003 17:12:41 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZdHP-0006Xs-CH
	for sigtran-archive@odin.ietf.org; Mon, 07 Jul 2003 17:12:14 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h67LCBvu025122
	for sigtran-archive@odin.ietf.org; Mon, 7 Jul 2003 17:12:11 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZdHG-0006WC-Vv; Mon, 07 Jul 2003 17:12:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XzwQ-0004fw-0F
	for sigtran@optimus.ietf.org; Thu, 03 Jul 2003 04:59:47 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA02912
	for <sigtran@ietf.org>; Thu, 3 Jul 2003 04:59:43 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19XzwM-0004P3-00
	for sigtran@ietf.org; Thu, 03 Jul 2003 04:59:42 -0400
Received: from penguin-ext.wise.edt.ericsson.se ([193.180.251.47] helo=penguin.al.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19XzwL-0004P0-00
	for sigtran@ietf.org; Thu, 03 Jul 2003 04:59:41 -0400
Received: from esealnt612.al.sw.ericsson.se ([153.88.254.118])
	by penguin.al.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h638xfGJ005448;
	Thu, 3 Jul 2003 10:59:41 +0200 (MEST)
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 NW44W4PA; Thu, 3 Jul 2003 10:59:41 +0200
Received: by ESEALNT744.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <N94M1V8H>; Thu, 3 Jul 2003 10:59:42 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CC4@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>
To: sigtran@ietf.org
Cc: "'bidulock@openss7.org'" <bidulock@openss7.org>,
        "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Date: Thu, 3 Jul 2003 10:59:17 +0200 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Sigtran] M3UA IG v04
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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 guys,

We tried to submit the next IG version before the deadline, but I got confused since I thought it ended at 12:00 instead of 9:00. So the server bounced it. I'll submit the doc again after the IETF meeting. In the meantime you can find it at:

http://standards.ericsson.net/javier/draft-ietf-sigtran-m3ua-implementors-guide-04.txt

We've tried to include all the different opinions in here. Although it seems that the case of waiting till n ASPs is the preferred one, we have included the new possibility of a smooth start for those of you who want to activate the AS when the first ASP gets to ACTIVE. At the end it will be the operator's decission how to proceed during the activation phase. Hopefully everyone agrees on this wording...

Cheers // Javier.



> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: martes, 01 de julio de 2003 20:18
> To: Samuel Dur D. Jeyaseelan
> Cc: sigtran@ietf.org
> Subject: Re: [Sigtran] Re:n+k issue
> 
> 
> Samuel,
> 
> Perhaps you could answer my questions below first.  Please be
> specific.
> 
> --brian
> 
> Samuel Dur D. Jeyaseelan wrote:                               
>                       (Tue, 01 Jul 2003 06:09:34)
> > 
> > Brian,
> > 
> >   Can you bit elaborate your n=256 example (ANSI-SLS) with 
> real-time 
> >   scenario?(please write to the point)
> > 
> > --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, 1 Jul 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                      
> (Mon, 30 Jun 2003 09:44:01)
> > > > 
> > > > Brian,
> > > > 
> > > > On Sat, 28 Jun 2003, Brian F. G. Bidulock wrote:
> > > > 
> > > > > Samuel,
> > > > > 
> > > > > I should point out that the proposal to require that 
> the SG wait for 'n' to
> > > > > start traffic does not have much merit.
> > > > > 
> > > > > Example: n=256, k=1
> > > > > 
> > > > > 2 ASPs fail, permanently and need to be manually replaced.
> > > > > 
> > > > > A temporary network failure causes SCTP associations 
> to be lost to the
> > > > > remaining 255 ASPs.  They move to the ASP down state 
> and immediately
> > > > > re-establish SCTP associations and move to the 
> ASP-Active state.  Because the
> > > > > number of ASPs is < n, the SG will not restart 
> traffic for some time.
> > > > 
> > > > Your example leads to the following figure.
> > > > 
> > > >          SG1                            AS1
> > > >      
> > > >  SS7 MSG +------+----------------------- ASP (1)
> > > > ---------| SGP1 |----------------------- ASP (2)
> > > >          |      |----------------------- ....
> > > >          +------+----------------------- ASP (256)
> > > >             |
> > > >             | 
> > > >             +---------------------------ASP(k)
> > > > 
> > > > 1. First of all craft will not configure 1 ASP as 
> backup for 256 Active
> > > >    ASPs, if he does, it means to be a poor network 
> configuration.
> > > 
> > > Well, first of all, craft does not normally choose the 
> number of ASPs used for
> > > backup, engineering does.  Second, k=1 is more than 
> adequate for n=256 where
> > > the ASPs are reasonably reliable and have a short 
> replacement time.
> > > 
> > > Regardless, I followed with the same example with n=2 and k=1.
> > > 
> > > > 
> > > > 2. Consider ASP as an MGC, your example leads to 256 
> MGCs and they are 
> > > >    controlled by a single SGP.A good network will have 
> multiple SGPs to
> > > >    share the POTENTIAL traffic across ASPs.
> > > 
> > > You seem to misunderstand the concept of redundancy in 
> M3UA: there are not
> > > multiple SGPs or ASPs because of traffic load.  It is 
> easy enough to handle
> > > the traffic of a small to medium sized country through a 
> single processor.
> > > 
> > > Also, you seem to misunderstand that an SGP and an ASP 
> are not processors.
> > > They are processes.  One host processor could have many 
> (even 256) ASPs or
> > > SGPs (or both for that matter).
> > > 
> > > > 
> > > > 3. BASIC INTENTION OF 'n+k' is,just to make sure that 
> incoming messages
> > > >    ARE EFFICIENTLY TRANSMITTED TO DESTINATION ASPs WITH 
> THE HELP OF
> > > >    MINIMUM NUMBER OF ASSOCIATIONS.
> > > 
> > > No.  The intention of 'n+k' is to provide one of the more 
> recently accepted
> > > models for reliability using redundancy (n = m+1, k=0 
> being the typical system
> > > deployed in the field today, m usually being greater than 
> 2).  Up until
> > > recently, only 2m (or m+m) systems were considered 
> carrier class.  Some
> > > systems were 4m (with 3 out of 4 vote).
> > > 
> > > You should take some time to look at redundancy and 
> reliability requirements
> > > for carrier grade systems for an understanding of how to 
> calculate and provide
> > > the reliability requirements expected by carriers.  There 
> are good standards
> > > written on the topic.  See the Telcordia NEBS and 
> reliability specs, for
> > > example.  If you have an RBOC RFP handy, take a look at 
> the hardware and
> > > software reliability requirements sections or talk to the 
> engineer in your
> > > company responsible for answering those sections.
> > > 
> > > > 
> > > >    Say for example
> > > > 
> > > >    Inefficient configuration -just an example
> > > >    ------------------------------------------
> > > >   
> > > >             SLS         DEST
> > > >             =================
> > > >              0           ASP1
> > > >              1           ASP1
> > > >              2           ASP1
> > > >              3           ASP1
> > > >              4           ASP1
> > > >              5           ASP1
> > > >              6           ASP1
> > > >              .            .
> > > >              .            .
> > > >              15          ASP1
> > > >              
> > > >              
> > > >                SG                            AS1
> > > > sls(0,1..15 )+----+  sls(0,1..15)         +------+
> > > > -------------|SGP1|-----------------------| ASP1 |
> > > >              |    |              ASSOC-1  +------+
> > > >              +----+                     
> > > > 
> > > >                       Fig - 1
> > > > 
> > > > In this figure, all incoming messages are transmitted 
> over ASSOC-1 towards
> > > > ASP1.Here there are two potential problems 
> > > > 
> > > >        1. Total incoming traffic is dependent on a single ASP
> > > >        2. ASP1 is always overloaded, messages are 
> processed one by one
> > > >           from the transmission buffer which ultimately 
> introduces time
> > > >           delay(example last arrived msg will be 
> processed at the end).
> > > 
> > > The above was not in my examples.  My examples had n > 1.
> > > 
> > > > 
> > > > 
> > > > Efficient configuration
> > > > -----------------------  
> > > > 
> > > >              SLS         DEST
> > > >             =================
> > > >              0           ASP1
> > > >              1           ASP1
> > > >              2           ASP1
> > > >              3           ASP1
> > > >              4           ASP1
> > > >              5           ASP1
> > > >              6           ASP1
> > > >              7           ASP1
> > > >              8           ASP2
> > > >              9           ASP2
> > > >              .            .
> > > >              .            .
> > > >              15          ASP2
> > > > 
> > > >                
> > > >                        
> > > >                        
> > > >               SG                             AS1
> > > > sls(0,1..15 )+----+   sls(0,1,..7)ASSOC-1 +------+
> > > > -------------|SGP1|-----------------------| ASP1 |
> > > >              |    |--------+              +------+
> > > >              +----+        | sls(8,9..15) +------+
> > > >                            +--------------| ASP2 |
> > > >                                  ASSOC-2  +------+
> > > >                         
> > > >                    fig - 2
> > > > 
> > > > In this example incoming traffic is distributed between 
> ASP1 and ASP2.
> > > > 
> > > > IN ORDER TO HANDLE THE HEAVY INCOMING TRAFFIC, SGP1 
> NEEDS MINIMUM OF
> > > > 2 ACTIVE ASPs BY SETTING n=2.
> > > 
> > > Do they not comply to SS7 ISUP/SCCP flow and congestion 
> controls?  Are they
> > > running at full capacity in this arrangement?  Are these 
> your ASPs?  Do you
> > > design your systems this way?
> > > 
> > > Take a look at the reliability engineering.  If either of 
> these ASPs is
> > > handling its maximum rated capacity, it will have to be 
> reliable to .99999
> > > This is because, if one ASP fails, 50% of the system has 
> failed.  A 50% system
> > > failure is deemed a "system" failure rather than a 
> "component" failure in
> > > reliability engineering for carrier grade systems.  
> Considering proscribed
> > > replacement times on the ASP, this would be difficult to 
> achieve unless both
> > > ASPs were hardware fault tolerant platforms.
> > > 
> > > Nevertheless, the RFC defines ASPs as a process, not a 
> processor.  There is
> > > nothing restricting the two ASPs in your diagram from 
> being processes on the
> > > same processor.  In that case, when ASP2 is down, ASP1 
> has double the
> > > processing capacity as it had before.
> > > 
> > > You certainly cannot make statements as you have made here without
> > > presupposing an implementation architecture outside the 
> scope of the standard.
> > > 
> > > Besides, the RFC permits you to configure your SGP (and 
> why do you only have
> > > one of them?) so that it waits for 'n' ASPs to become 
> active.  You MAY do
> > > this.  Just don't require everybody else to follow your 
> approach.  (Actually,
> > > if the above is your approach, don't require anyone else 
> to follow it.)
> > > 
> > > Several other points: ANSI T1.111.4/2000 SLS values are 
> 8-bits.  That means
> > > there are 256 SLS values (gee, I had 256 ASPs in my example).
> > > 
> > > For another thing, sequenced traffic flows are not only 
> based on SLS.  In an
> > > ITU-T network, each remote signalling point has 16 
> independent traffic flows
> > > for each AS.  The traffic flows from each remote 
> signalling point is
> > > independent from the traffic flows of other remote 
> signalling points.
> > > Therefore, if there are 400 remote signalling points, 
> there will be 6400
> > > independent traffic flows that the SG can use for 
> distribution of AS traffic
> > > for one AS over as many as 6400 ASPs without risking 
> mis-sequencing of
> > > messages.  For an ANSI T1.111.4/2000 network using 8-bit 
> SLS values with 400
> > > remote signalling points, there would be 102400 
> independent traffic flows to
> > > the AS that the SG can use to distribute traffic to one 
> AS over as many as
> > > 102400 ASPs without risking message mis-sequencing.
> > > 
> > > > 
> > > > 
> > > > Your example with n=256 is completely a wrong 
> interpretaion of 'n+k'
> > > > redundancy case.
> > > 
> > > Quite the opposite, you clearly misunderstand redundancy 
> and reliability in
> > > carrier grade systems and obviously have little 
> experience with them.
> > > 
> > > Here's a little primer for you:
> > > 
> > > Carrier grade systems were originally designed as 
> hot-standby, fully redundant
> > > systems.  For every processor actively handling traffic, 
> there was a hot
> > > standby processor also actively handling the same 
> traffic.  This typically
> > > required special bus arrangements whereby the I/O bus of 
> one processor was
> > > actively controlling peripheral devices, while the hot 
> standby processor was
> > > issuing I/O bus output that was not terminated to the I/O 
> bus.  The bus also
> > > normally performed a comparison of the outputs from the 
> active side and the
> > > outputs from the inactive side.  Peripheral components, 
> such as switch fabric
> > > were also fully redundant (duplicated).  Switching 
> fabrics are separated into
> > > an A-plane and a B-plane in much the same fashion as the 
> SPCS front-end I/O
> > > bus.
> > > 
> > > Note that active/standby in M3UA does not address the 
> above model.  2+0
> > > broadcast does.
> > > 
> > > The problem with such a systems was what we used to call 
> the "kill your
> > > sister" syndrome.  A failure on the inactive side only 
> detectable to the I/O
> > > bus could cause a switch from the active to inactive 
> side, resulting in the
> > > switch running on the erroneous inactive side.  We called 
> it the "kill your
> > > sister" syndrome because it is not so surprising that the 
> psychotic sister
> > > (erred inactive side) kills the sane sister (correct active side).
> > > 
> > > Primarily to avoid the "kill your sister" syndrome, some 
> switches were
> > > designed with 3 or 4 processors on an cross-checking I/O 
> bus pair.  In that
> > > case, it would take a 2 out of 3 or 3 out of 4 vote to 
> shift from the active
> > > processor.  These systems were far less susceptible to 
> inactive plane errors
> > > that were only detected at the I/O bus.
> > > 
> > > Note that the only M3UA model that address the above is 
> the 3+0 or 4+0
> > > broadcast model.
> > > 
> > > When the use of general purpose computing platforms for 
> telecommunications
> > > systems started to become popular, there were several 
> models that emerged.
> > > 
> > > First was taking two big iron UNIX hosts and running them 
> in active/standby.
> > > In this scenario, one UNIX host would be actively 
> handling traffic and the
> > > other would be warm standby.  By "warm" standby, it means 
> that the device was
> > > powered and was synchronizing state on a checkpoint 
> interval or monitoring the
> > > health of the active hosts with heartbeats.  Shared 
> memory, shared DASD or
> > > message passing are the most common mechanisms of 
> redundancy.  These systems
> > > are considered "highly available" systems rather than the 
> "fault tolerant"
> > > systems used in telecommunications above.  An increased 
> level of reliability
> > > is often achieved by using hardware fault tolerant 
> multiple processor hosts.
> > > 
> > > In M3UA, the active/standby model addresses these 
> systems.  Although M3UA
> > > provides for a '1+k' model, k is invariably 1.
> > > 
> > > The major difficulties with these systems are: a) 
> switchover latency between
> > > systems; b) loss of state during switchover; and c) 
> scalability.  Switchover
> > > latency is normally improved by decreasing heartbeat 
> intervals.  State
> > > synchronization is normally improved by decreasing the 
> process checkpoint
> > > interval.  These reductions typically increase cost and 
> they are normally
> > > chosen to minimize a cost/performance/reliability metric. 
>  Scalability was
> > > typically performed by installing larger more powerful 
> hosts with increased
> > > demand.  This was not good, because larger processors 
> tended to have poorer
> > > cost/performance ratios that smaller processors.
> > > 
> > > Later it was determined that the cost of a complete 
> redundant host that was
> > > wasting CPU cycles while inactive could be avoided while 
> addressing the
> > > deficiencies of the paired host approach above.  To increase state
> > > synchronization, state could be spread over a number of 
> hosts.  If there were
> > > 'm' hosts, the failure of one host could at worst affect 
> a proportion, 1/m, of
> > > the state.  Therefore state synchronization would improve 
> as m increased.  To
> > > distribute the state, traffic load would be shared over 
> the m hosts according
> > > to state.  This also served the needs of scalability and 
> provided for closer
> > > to linear cost scalability, because the number of hosts, 
> m, could be increased
> > > while holding the cost/performance ratio of the hosts 
> constant.  The costs of
> > > redundancy could also be reduced over the fully redundant 
> systems described
> > > above, because redundancy could be achieved using 'm+1' 
> processors, where 'm'
> > > processors were required to handle the maximum engineered 
> load.  For systems
> > > as small as 'm=2' hosts with equivalent reliability to 
> that normally
> > > experienced in the '1+1' active/standby scenario above, 
> system reliability
> > > figures could be maintained with less redundancy cost.  
> It is important to
> > > note, that in these systems, all 'm+1' hosts are active 
> and handling only a
> > > proportion m/(m+1) of their rated capacity.  It is very 
> seldom that the rated
> > > capacity was more that 2/3 of the maximum short-term capacity.
> > > 
> > > These systems do not solve the heartbeat interval problem of the
> > > active/standby paired hosts system above.  Typically what 
> these systems use is
> > > more sophisticated watchdog approaches rather than 
> passing heartbeat messages
> > > between processes on the active and standby hosts.  
> Memory sharing, DASD
> > > sharing, message passing, reliable broadcast, voting, and 
> a plethora of
> > > reliable distributed systems techniques have been 
> developed that are useful in
> > > these systems.
> > > 
> > > Adoption of these latter systems changed the requirement 
> from "duplicated"
> > > systems to systems with no "single point of failure".
> > > 
> > > Now, the 'n+k' load-sharing and broadcast model was 
> intended to address these
> > > systems.  Where 'n == m' in the description above.  That 
> is why we agreed that
> > > n is the maximum engineered load.  Because that is what 
> 'm' is.  The k is the
> > > 1 in the description above, generalized.
> > > 
> > > First off, you must understand that it is not the 
> intention of the model to
> > > have 'n' ASPs active and 'k' inactive unless 'n=1, k=1'.  
> This is described in
> > > section 1.4.4.1:
> > > 
> > >   A "1+1" active/standby redundancy is a subset of this model.
> > > 
> > > Also no redundancy is possible:
> > > 
> > >   A simplex "1+0" model is also supported as a subset, 
> with no ASP redundancy.
> > > 
> > > (The term "simplex" come from the telephone switches 
> described above where the
> > > inactive side is taken off-line.  Such a situation is 
> described as running the
> > > switch "simplex".)
> > > 
> > > The 'n+k' model is to address the 'm+1' distributed 
> systems described last.
> > > Rereading section 1.4.4.1:
> > > 
> > >   The failover model supports an "n+k" redundancy model, 
> where "n" APSs is the
> > >   minimum number of redundant [sic] ASPs required to 
> handle traffic and "k"
> > >   ASPs are available to take over for a failed or unavailable ASP.
> > > 
> > > Note that the state of the "k" ASPs with respect to an AS 
> is not stated.  In
> > > the 'm+1' redundant system described above, the "k" ASPs 
> are active for the
> > > AS.  "k" ASPs active in load sharing mode are "available 
> to take over for a
> > > failed or unavailable ASP".
> > > 
> > > It is typical to run the "k" ASPs active (and I know you 
> had some questions
> > > about activating the "k" ASPs, but the point is that they 
> are normally already
> > > active), and handling their own share of traffic.  In 
> this case, each ASPs
> > > handles n/(n+k) of their engineered capacity.  Were you 
> actually considering
> > > deploying such a atypical model where the 'k' ASPs are inactive?
> > > 
> > > "k" ASPs can be taken out of service, upgraded, and 
> brought back into service
> > > without affecting the maximum engineering traffic 
> handling capacity of the
> > > system (in absence of failures during the upgrade).  This 
> assists in graceful
> > > software/hardware upgrades on live systems.  (Keeping 
> them inactive would
> > > serve no purpose and would, decrease maximum traffic 
> handling capability,
> > > increase loss of state on fail-over, and increase 
> fail-over latency: all good 
> > > reasons for not deactivating the 'k' ASPs).
> > > 
> > > Now, reliability engineering for carrier class systems 
> takes the view that if
> > > 50% of the load bearing components in a system fail (stop 
> operating in any
> > > way), then the entire system has failed.  If traffic is 
> stopped above n/2
> > > processes, the system will have failed prematurely.
> > > 
> > > A requirement to withhold traffic < n will always suffer 
> a prolonged system
> > > failure, because only n/2 are required to make the system 
> available.
> > > Therefore, on startup the reliability of the system is 
> degraded by the
> > > negative exponent n/2, regardless of whether n=256 or 
> n=2; however, increasing
> > > 'n' increases the damage done to reliability by waiting for 'n'.
> > > 
> > > A requirement for the SG to wait for 'n' is unacceptable 
> from the standpoint
> > > of the redundancy model that 'n+k' was meant to address, 
> as well as other
> > > possible models that it can accommodate.
> > > 
> > > > 
> > > > 
> > > > > 
> > > > > Consider the same example with: n=2, k=1, two ASPs failed.
> > > > > 
> > > > > Any system that introduces hysteresis to aggravate 
> and perpetuate an outage in
> > > > > this fashion is not very useful.
> > > > > 
> > > > > In the examples, what would have been a brief 
> AS-PENDING situation has been
> > > > > turned into a prolonged and complete system failure.  
> It is certainly not good
> > > > > practice to design redundant systems in that fashion 
> and recommending or
> > > > > requiring such a design would be a bad thing.
> > > > > 
> > > > > To further the example, consider the multiple SG as 
> STP situation.  One SG
> > > > > cannot know the state of ASPs for the its mated SG 
> and therefore cannot
> > > > > distinguish between the restart scenario and the 
> start scenario as far as the
> > > > > AS state is concerned because the SG does not know 
> the state of the AS itself:
> > > > > the AS could be active with other SGs.
> > > > > 
> > > > > This is the same case for SGP.  Unless AS state is 
> synchronized across SGPs,
> > > > > one SGP cannot know the total state of the AS (only 
> its partition).  There is
> > > > > no requirement to synchronize AS state across SGP (on 
> the SGP side, AS-state
> > > > > is per-SGP), and there is certainly no requirement to 
> synchronize AS-state
> > > > > across SGs.
> > > > > 
> > > > > So, take the (n=2, k=0) example where there are two 
> SGs.  Due to a network
> > > > > partitioning at startup, one ASP can communicate only 
> with one SG, and the
> > > > > other ASP can communicate only with the other SG.  
> Because n=2, neither SG
> > > > > will start traffic and the system suffers 100% 
> failure until the network
> > > > > partitioning ceases.
> > > > > 
> > > > > Consider the same example where there are two SGPs 
> (instead of SGs) and the
> > > > > two ASPs normally only form one association (e.g., 
> ASP1-SGP1, ASP2-SGP2).
> > > > > Note that there is no requirement in the RFC to form 
> full-mesh associations.
> > > > > Because the number of active ASPs with each SGP is < 
> 2, the system will never
> > > > > start.
> > > > > 
> > > > > These are not strange scenarios.  These situations 
> are all permitted by the
> > > > > RFC.  Yet, in each situation it is clear that one 
> cannot recommend that the SG
> > > > > wait.  The result would be a non-functional system.  
> There are many more
> > > > > examples.
> > > > > 
> > > > > n+k is a model for redundancy (not for congestion and 
> flow control).  A system
> > > > > with an SG that waits for n is, in general, not a 
> fully redundant system at
> > > > > all, but is, rather, prone to these failure syndromes.
> > > > > 
> > > > > When you look at the situation and apply the 
> experience of reliability
> > > > > engineering, the recommendation is that the SG SHOULD 
> start traffic when the
> > > > > first ASP goes active.
> > > > > 
> > > > > Why wait for 'n' anyway?  Because the ASP cannot 
> handle some level of traffic?
> > > > > Any ASP that conforms to MTP-User standards, must 
> have the standardized
> > > > > MTP-User (ISUP/SCCP) congestion and flow controls and 
> must have the SCTP
> > > > > congestion controls and is therefore able to accept 
> any amount of traffic
> > > > > overload.
> > > > > 
> > > > > So there is no real reason for the SG to wait for 
> 'n'.  Nevertheless, the RFC
> > > > > allows this as an OPTION.  Requiring (MUST) or 
> recommending it (SHOULD)
> > > > > regardless of whether the number of active processors 
> is rising or falling,
> > > > > whether start, restart or failure, is unacceptable.
> > > > > 
> > > > > On the other side, at the ASPs, the ASPs have a 
> requirement to share state
> > > > > between them.  They must share circuit state and 
> transaction state in all
> > > > > configurations to meet carrier grade.  This is stated 
> in the RFC.  It is
> > > > > implicit that the ASPs are capable of knowing with 
> which SGs, and which SGPs
> > > > > for which they are active.  They have access to the 
> total state of the AS that
> > > > > they collectively represent, that the SG/SGP does 
> not.  Therefore is it
> > > > > reasonable for the ASPs to coordinate among 
> themselves the number of active
> > > > > ASPs and to ensure that there are sufficient ASPs 
> active to handle the
> > > > > expected load before starting traffic.  Because the 
> ASPs need not respond to
> > > > > user test traffic (UPT/SST), the ASPs are also in 
> control of when remote SS7
> > > > > MTP-User restart traffic toward them.
> > > > > 
> > > > > Therefore, the recommendation that the ASPs SHOULD 
> coordinate among
> > > > > themselves the number of active ASPs and not start 
> traffic until a sufficient
> > > > > number of them are active is sound.  This is not a 
> MUST, permitting ASPs
> > > > > that know that they are capable of handling any 
> overload, to start traffic
> > > > > immediately, or at some earlier point.
> > > > > 
> > > > > Back to the RFC, the RFC sets the normal case where 
> the SG starts traffic on
> > > > > the first active ASP but explicitly permits as an 
> OPTION to wait until it
> > > > > considers that there are sufficient ASPs active to 
> handle the offered load.
> > > > > The RFC also RECOMMENDS that the ASPs coordinate 
> among themselves the number
> > > > > of active ASPs so that there are sufficient ASPs to 
> handle the expected load.
> > > > > 
> > > > > The provisions in the RFC are both sound, where the 
> proposal and section 3.15
> > > > > is clearly not.
> > > > > 
> > > > > Remember when looking at the AS state diagram in the 
> RFC that this is the
> > > > > SGP's view of the AS state and not the total state of 
> the AS, that can only
> > > > > known to the ASPs for all configurations.  'n' as 
> defined has no meaning to
> > > > > the state machine at the SGP because is not required 
> to know, and often cannot
> > > > > know, how many ASPs are active for the AS via other 
> SGPs and SGs, or even via
> > > > > direct connection to the SS7 network.
> > > > > 
> > > > > Some SGs might have special knowledge, or 
> sophisticated algorithms that allow
> > > > > it to determine whether a congestive failure will be 
> the immediate result of
> >  > > > starting or restarting traffic.  These SGs are 
> permitted as an
> > OPTION to wait.
> > > > > 
> > > > > The ASPs SHOULD have the knowledge of how many are 
> required to handle the
> > > > > expected load, because they must share 
> circuit/transaction state and they must
> > > > > have congestion and flow controls as required by the 
> standards.
> > > > > 
> > > > > So again, the RFC is adequate in this regard.  The 
> current proposal and
> > > > > section 3.15 are not.
> > > > RFC is inadequate so we need section 3.15 to explain 'n+k' case.
> > > 
> > > Teaching reliability engineering and the design of 
> redundant systems is
> > > outside the scope of the RFC and the IG.  People, such as 
> yourself, are better
> > > advised to learn the subject matter before attempting to 
> implement or claim a
> > > complete understanding of the protocol.
> > > 
> > > The current RFC provides the protocol mechanisms and basis for
> > > interoperability for all of the common approaches to 
> redundancy in carrier
> > > class systems, and even provides a general model that can 
> accommodate
> > > non-traditional redundancy architectures.
> > > 
> > > When confused about how a redundancy architecture works, 
> don't blame the M3UA
> > > RFC, blame those responsible for describing redundant 
> architectures to you.
> > > 
> > > In the appendix we stated:
> > > 
> > >    M3UA imposes no restrictions as to the exact layout of 
> the network
> > >    elements, the message distribution algorithms and the 
> distribution of
> > >    signalling processes.  Instead, it provides a 
> framework and a set of
> > >    messages that allow for a flexible and scalable 
> signalling network
> > >    architecture, aiming to provide reliability and performance.
> > > 
> > > Forcing SGs to wait for 'n' ASPs active before activating 
> is contrary to this.
> > > The existing "MAY" achieves this flexibility.
> > > 
> > > When reading 1.4.4.1, also remember this statement in the 
> appendix, that used
> > > to be beside 1.4.4.1.  The description of 'n+k' does not 
> dictate the state of
> > > the 'k' or the state of the AS with respect to 'n' to 
> leave the redundancy
> > > model flexible, and accommodate '1+1' and '1+0' in the same model.
> > > 
> > > Also, please don't presume to tell me to set 'n' to 1, as 
> it is not a
> > > arbitrary value, but the number 'm' in the 'm+1' 
> distributed system in common
> > > use today.  This number is determined as the minimum 
> number of ASPs required
> > > to handle the maximum engineered load, and not, as Tolga 
> would have, the
> > > minimum number of ASPs required to handle ANY load.  If 
> you want your system
> > > to behave as Tolga proposes, just have your SG exercise 
> the "MAY choose not to
> > > direct traffic" option of section 4.3.4.3 of the RFC.
> > > 
> > > Section 3.15 and Tolga's proposal should be struck 
> because they attempt to
> > > impose a 'n' active 'k' inactive and system cannot 
> restart (3.15) or start
> > > (Tolga) without 'n' view on the redundancy model, where 
> it is clearly the
> > > intent of the RFC to place no such restrictions on the 
> model, but to
> > > accommodate a range of models, that is already achieved 
> by the RFC text.
> > > 
> > > --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 exim@www1.ietf.org  Mon Jul  7 18:00:09 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA16218
	for <sigtran-archive@odin.ietf.org>; Mon, 7 Jul 2003 17:12:41 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZdHP-0006XS-Az
	for sigtran-archive@odin.ietf.org; Mon, 07 Jul 2003 17:12:14 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h67LCB4k025120
	for sigtran-archive@odin.ietf.org; Mon, 7 Jul 2003 17:12:11 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZdHG-0006W4-AO; Mon, 07 Jul 2003 17:12:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19XzwF-0004fr-IB
	for sigtran@optimus.ietf.org; Thu, 03 Jul 2003 04:59:35 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA02908
	for <sigtran@ietf.org>; Thu, 3 Jul 2003 04:59:33 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19XzwC-0004Ox-00
	for sigtran@ietf.org; Thu, 03 Jul 2003 04:59:32 -0400
Received: from albatross-ext.wise.edt.ericsson.se ([193.180.251.49] helo=albatross.tn.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19XzwB-0004Ou-00
	for sigtran@ietf.org; Thu, 03 Jul 2003 04:59:31 -0400
Received: from esealnt610.al.sw.ericsson.se ([153.88.254.120])
	by albatross.tn.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h638xNIi009500;
	Thu, 3 Jul 2003 10:59:23 +0200 (MEST)
Received: from ESEALNT744.al.sw.ericsson.se ([153.88.251.4]) by esealnt610.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55)
	id NW4P9WH1; Thu, 3 Jul 2003 10:59:51 +0200
Received: by ESEALNT744.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <N94M1V81>; Thu, 3 Jul 2003 10:59:24 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CC2@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>
To: "'agurmukhani@hss.hns.com'" <agurmukhani@hss.hns.com>
Cc: sigtran@ietf.org
Date: Thu, 3 Jul 2003 10:59:00 +0200 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Sigtran] RE: M3UA MIB Draft05: Loadsharing on CIC ,SSN
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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 Anjali, 

It seems that it makes sense to remove those from the standard MIB. All the people who are interested in having this object can create it within their private MIBs. We may keep SLS since it is a common practise in SS7 unless you have any objection.


Thanks for your responses // Javier.

PS: Samuel: What object do you refer when talking about DPC-based loadsharing within the current MIB? 


> -----Original Message-----
> From: agurmukhani@hss.hns.com [mailto:agurmukhani@hss.hns.com]
> Sent: jueves, 03 de julio de 2003 6:54
> To: J.Javier Pastor (EE/EEM)
> Cc: sigtran@ietf.org
> Subject: M3UA MIB Draft05: Loadsharing on CIC ,SSN
> 
> 
> 
> 
> 
> 
> Hi...
> 
> 
> In my understanding and  as per the discussion and consensus 
> on the mailing
> list , everyone agreed that  CIC and SSN should not be used 
> as loadsharing
> basis
> be it within streams in a particular SCTP association or for active
> signalling processes.
> 
> So would these be removed from  the m3uaStreamLshMethod  and
> m3uaSpLshMethod OBJECT-TYPE s in M3UA MIB draft.
> 
> 
> Thanks and Regards
> Anjali
> www.hssworld.com
> Hughes Software Systems.
> 
> "DISCLAIMER: This message is proprietary to Hughes Software 
> Systems Limited
> (HSS) and is intended solely for the use of the individual to 
> whom it is
> addressed. It may contain  privileged or confidential information and
> should not be circulated or used for any purpose other than 
> for what it is
> intended. If you have received this message in error, please 
> notify the
> originator immediately. If you are not the intended recipient, you are
> notified that you are strictly prohibited from using, 
> copying, altering, or
> disclosing the contents of this message. HSS accepts no 
> responsibility for
> loss or damage arising from the use of the information 
> transmitted by this
> email including damage from virus."
> 

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



From exim@www1.ietf.org  Tue Jul  8 00:14:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA26144
	for <sigtran-archive@odin.ietf.org>; Tue, 8 Jul 2003 00:14:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Zjre-0007uZ-Sc
	for sigtran-archive@odin.ietf.org; Tue, 08 Jul 2003 00:14:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h684E2n8030407
	for sigtran-archive@odin.ietf.org; Tue, 8 Jul 2003 00:14:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Zjrd-0007uC-07; Tue, 08 Jul 2003 00:14:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZjrN-0007tw-4F
	for sigtran@optimus.ietf.org; Tue, 08 Jul 2003 00:13:45 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA26133
	for <sigtran@ietf.org>; Tue, 8 Jul 2003 00:13:41 -0400 (EDT)
From: kakaul@hss.hns.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ZjrK-00060q-00
	for sigtran@ietf.org; Tue, 08 Jul 2003 00:13:42 -0400
Received: from hindon.hss.co.in ([202.54.26.202])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ZjrJ-00060X-00
	for sigtran@ietf.org; Tue, 08 Jul 2003 00:13:41 -0400
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 h6849Sn11608
	for <sigtran@ietf.org>; Tue, 8 Jul 2003 09:39:29 +0530 (IST)
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h6849QV11588
	for <sigtran@ietf.org>; Tue, 8 Jul 2003 09:39:26 +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 ESMTP id h684CjJ13779
	for <sigtran@ietf.org>; Tue, 8 Jul 2003 09:42:45 +0530 (IST)
Subject: Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.5  September 22, 2000
Message-ID: <OF293C2F2C.5A18F04C-ON65256D5D.0015A546@hss.hns.com>
Date: Tue, 8 Jul 2003 09:37:15 +0530
X-MIMETrack: Serialize by Router on Sandesh/HSS(Release 6.0|September 26, 2002) at 08/07/2003
 09:36:20 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>






Hi Brian and Samule,
  I have not got to the point answer to my question :
I want to know how an MTP3  user (ISUP or SCCP) selects SLS for effective
load distribution at MTP3 layer among the links available to reach a given
DPC. Is it that SLS is chosen per DPC for a logical session or globally per
each logical session irrespective of DPC for which this session is
intended.I know it is implementation dependent option but i want to know
how do most of the MTP3 user implementations choose SLS. By logical session
i mean logically related data.

Regards
Kamesh Kaul
www.hssworld.com
Hughes Software Systems





"Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
07/04/2003 10:58:14 AM

Sent by:    sigtran-admin@ietf.org


To:    sigtran@ietf.org
cc:

Subject:    Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN



Pastor,

 MIB-05 does not have 'load-shar on dpc',i was referring mib-04 which lead
 me into confusion.

 Thanks kakaul for our mail.

Kakaul,

 As you said MTP3 does not route messages based on CIC or SSN because
 there are no standards available, i agree with it.

 Have a look, suppose SG receives an MSU with routing indicator="Route on
 PointCode/SSN", if your SG supports static configuration then it is easy
 to route these messages to the corresponding subSystems based on
 statically configured SSN-ASP table the same way CIC-ASP table can be
 used to route ISUP messages to the corresponding MGCs.Basically CIC was
 removed from RFC to avoid complex problems say a msg comes with CIC
 range,this CIC range can be split across MGCs here the problem is
 potential.

 These are the reasons caused me to say that with static configuration
 CIC or/and SSN can be supported if an SG vendor and ASP vendor has
 mutual understanding.(but this is outside the scope of M3UA RFC)

--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, 4 Jul 2003 kakaul@hss.hns.com wrote:

>
>
>
>
> Hi Samule,
>   Does MTP3 use CIC or SSN for load-distribution among the available
> links.I guess only SLS parameter in MTP3 is used for load-distribution
> among the available link.I want to know how an MTP3  user (ISUP or SCCP)
> selects SLS for effective load distribution at MTP3 layer among the links
> available to reach the  a given DPC.
>
> Regards
> Kamesh Kaul
> www.hssworld.com
> Hughes Software Systems
>
>
>
>
> "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>@ietf.org on
> 07/04/2003 09:29:51 AM
>
> Sent by:    sigtran-admin@ietf.org
>
>
> To:    sigtran@ietf.org
> cc:
>
> Subject:    Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
>
>
>
>  this is just an info regarding CIC and SSN - load-sharing.
>
>  A node can load-share its traffic based on CIC or SSN with the
>  help of static configuration but this is outside of the scope of M3UA
>  rfc.When an ASP vendor and SGP vendor agrees to do load-sharing based on
>  CIC or SSN with static confiuration they are welcome to do this,
>  since this has not been standardised, taking this to interop will
>  lead to problems.So almost everybody agrees to delete CIC and SSN
>  loadsharing concept from MIB-04.
>
>
> --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, 3 Jul 2003, Brian F. G. Bidulock wrote:
>
> > agurmukhani,
> >
> > I think that they should be removed from the MIB, but we did NOT agree
> > that CIC or SSN should not be used for loadsharing.  Any criteria that
> > the SG/ASP desires to use for loadsharing being congnisant of the need
> > for ordering of user part messages is allowed for loadsharing.  We DID
> > agree to remove CIC, SSN from the Routing Key, because a rational and
> > interoperable basis for splitting CICs and SSNs across AS could not be
> > decided.
> >
> > The meanse oadsharing, on the other hand, is not specified in M3UA.
> > Some examples are given, but no requirements.  Some things that could
> > clearly be used are SLS, DPC/SLS, OPC/DPC/SLS, CIC, DPC/CIC,
OPC/DPC/CIC,
> > OPC/DPC/CIC/SLS, SSN, ....  and other permutations of these and even
> > higher layer parameters such as TID.
> >
> > My argument for removing them from the MIB is not that they cannot be
> > used, but that they have not been standardized.
> >
> > --brian
> >
> > agurmukhani@hss.hns.com wrote:
> (Thu, 03 Jul 2003 10:23:35)
> > >
> > >
> > >
> > >
> > > Hi...
> > >
> > >
> > > In my understanding and  as per the discussion and consensus on the
> mailing
> > > list , everyone agreed that  CIC and SSN should not be used as
> loadsharing
> > > basis
> > > be it within streams in a particular SCTP association or for active
> > > signalling processes.
> > >
> > > So would these be removed from  the m3uaStreamLshMethod  and
> > > m3uaSpLshMethod OBJECT-TYPE s in M3UA MIB draft.
> > >
> > >
> > > Thanks and Regards
> > > Anjali
> > > www.hssworld.com
> > > Hughes Software Systems.
> > >
> > > "DISCLAIMER: This message is proprietary to Hughes Software Systems
> Limited
> > > (HSS) and is intended solely for the use of the individual to whom it
> is
> > > addressed. It may contain  privileged or confidential information and
> > > should not be circulated or used for any purpose other than for what
it
> is
> > > intended. If you have received this message in error, please notify
the
> > > originator immediately. If you are not the intended recipient, you
are
> > > notified that you are strictly prohibited from using, copying,
> altering, or
> > > disclosing the contents of this message. HSS accepts no
responsibility
> for
> > > loss or damage arising from the use of the information transmitted by
> this
> > > email including damage from virus."
> > >
> > >
> > > _______________________________________________
> > > 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



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


"DISCLAIMER: This message is proprietary to Hughes Software Systems Limited
(HSS) and is intended solely for the use of the individual to whom it is
addressed. It may contain  privileged or confidential information and
should not be circulated or used for any purpose other than for what it is
intended. If you have received this message in error, please notify the
originator immediately. If you are not the intended recipient, you are
notified that you are strictly prohibited from using, copying, altering, or
disclosing the contents of this message. HSS accepts no responsibility for
loss or damage arising from the use of the information transmitted by this
email including damage from virus."


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



From exim@www1.ietf.org  Tue Jul  8 00:33:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA26487
	for <sigtran-archive@odin.ietf.org>; Tue, 8 Jul 2003 00:33:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZkA2-0008MP-4x
	for sigtran-archive@odin.ietf.org; Tue, 08 Jul 2003 00:33:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h684X2wc032136
	for sigtran-archive@odin.ietf.org; Tue, 8 Jul 2003 00:33:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZkA1-0008M5-1T; Tue, 08 Jul 2003 00:33:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Zk9L-0008J6-7G
	for sigtran@optimus.ietf.org; Tue, 08 Jul 2003 00:32:19 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA26457
	for <sigtran@ietf.org>; Tue, 8 Jul 2003 00:32:15 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19Zk9I-00066a-00
	for sigtran@ietf.org; Tue, 08 Jul 2003 00:32:16 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[EWXbJwV4fyPhL3s0kctQ9BEtIYtdZIYs])
	by ietf-mx with esmtp (Exim 4.12)
	id 19Zk9H-00066X-00
	for sigtran@ietf.org; Tue, 08 Jul 2003 00:32:16 -0400
Received: from ns.pigworks.openss7.net (IDENT:56mGd/ctJm/I9/32wPvK+8uhIfTOX0b1@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h684WAX08913;
	Mon, 7 Jul 2003 22:32:10 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h684WAJ29897;
	Mon, 7 Jul 2003 22:32:10 -0600
Date: Mon, 7 Jul 2003 22:32:10 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: kakaul@hss.hns.com
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA MIB Draft05: Loadsharing on CIC ,SSN
Message-ID: <20030707223210.A29733@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: kakaul@hss.hns.com, sigtran@ietf.org
References: <OF293C2F2C.5A18F04C-ON65256D5D.0015A546@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: <OF293C2F2C.5A18F04C-ON65256D5D.0015A546@hss.hns.com>; from kakaul@hss.hns.com on Tue, Jul 08, 2003 at 09:37:15AM +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>

kakaul,

I'm sorry that we did not answer your question, but it is rather outside the
scope of SIGTRAN and is in the domain of SS7 standards and implementation.

Is there not anyone within your own organization that understands SS7 well
enough to answer this question for you?

If not, you are probably better to ask SS7 questions on the forums at
http://www.c7.com/ or on the OpenSS7 mailing list at http://www.openss7.org/

ITU-T/ETSI ISUP, for example, associates a fixed SLS value with each CIC of an
ISUP trunk group and use that fixed SLS value for each ISUP message pertaining
to the CIC.

ANSI T1.113 ISUP does not associate an SLS value with a CIC.  Rather, the ISUP
user selects an SLS for each call.  This performs better SLS load distribution
that is achieved by the ITU-T/ETSI approach above.

The most direct approach is to simply use global variables and increment them
with each call, each SCCP connection, or each SCCP unit data.

This is not done on a per-destination basis for SCCP and ANSI ISUP.  For
ITU-T/ETSI ISUP it just happens that the fixed SLS assignment per CIC happens
to be on a per-destination basis, but this does not achieve optimal load
sharing.

If one wants to load share on anything less than a global basis, it should be
done on a per-outgoing-route basis rather than a per-route-set basis.  The
difference is that SLS is used to load share between signalling links
belonging to a combined link set (routes of equal priority) and it is on this
basis that it needs to be best distributed to achieve load sharing.

--brian


kakaul@hss.hns.com wrote:                     (Tue, 08 Jul 2003 09:37:15)
> 
> Hi Brian and Samule,
>   I have not got to the point answer to my question :
> I want to know how an MTP3  user (ISUP or SCCP) selects SLS for effective
> load distribution at MTP3 layer among the links available to reach a given
> DPC. Is it that SLS is chosen per DPC for a logical session or globally per
> each logical session irrespective of DPC for which this session is
> intended.I know it is implementation dependent option but i want to know
> how do most of the MTP3 user implementations choose SLS. By logical session
> i mean logically related data.
> 
> Regards
> Kamesh Kaul
> www.hssworld.com
> Hughes Software Systems
> 

-- 
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 exim@www1.ietf.org  Tue Jul  8 03:49:32 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA13143
	for <sigtran-archive@odin.ietf.org>; Tue, 8 Jul 2003 03:49:32 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZnDk-0002v5-2A
	for sigtran-archive@odin.ietf.org; Tue, 08 Jul 2003 03:49:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h687n4ZV011206
	for sigtran-archive@odin.ietf.org; Tue, 8 Jul 2003 03:49:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZnDh-0002uR-PX; Tue, 08 Jul 2003 03:49:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZnD2-0002rg-Ue
	for sigtran@optimus.ietf.org; Tue, 08 Jul 2003 03:48:20 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA13132
	for <sigtran@ietf.org>; Tue, 8 Jul 2003 03:48:18 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ZnD0-0007HV-00
	for sigtran@ietf.org; Tue, 08 Jul 2003 03:48:18 -0400
Received: from mail-n.franken.de ([193.175.24.27] ident=postfix)
	by ietf-mx with esmtp (Exim 4.12)
	id 19ZnCz-0007HS-00
	for sigtran@ietf.org; Tue, 08 Jul 2003 03:48:18 -0400
Received: from lurchi.franken.de (fb02np06.FH-Muenster.DE [194.95.73.131])
	by mail-n.franken.de (Postfix) with ESMTP
	id 0C419247DB; Tue,  8 Jul 2003 09:46:50 +0200 (CEST)
Date: Tue, 8 Jul 2003 09:45:20 +0200
Subject: Re: [Sigtran] Sigtran Plugtest at ETSI 9/03
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v552)
Cc: sigtran@ietf.org
To: Ken Morneault <kmorneau@cisco.com>
From: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>
In-Reply-To: <4.3.2.7.2.20030707102133.04cc0528@cia.cisco.com>
Message-Id: <1FB94CFA-B118-11D7-A807-0030654C1AB6@lurchi.franken.de>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.552)
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

Dear all,

we have some time until the Plugtest, so I would prefer to use
that latest version, which is 09.

Best regards
Michael

On Monday, Jul 7, 2003, at 16:31 Europe/Berlin, Ken Morneault wrote:

>
>
> All:
>
> Which version of M2PA should be tested at the upcoming
> Sigtran plugtest in 9/03?  Currently, the ETSI Plugtest site
> indicates version 8, but version 9 was just released.  Version
> 6 was tested at the last Plugtest.
>
> Please email your preference.
>
> Regards,
>
> Ken
>
>
> _______________________________________________
> 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 exim@www1.ietf.org  Tue Jul  8 04:51:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA14534
	for <sigtran-archive@odin.ietf.org>; Tue, 8 Jul 2003 04:51:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZoBj-0005ag-MS
	for sigtran-archive@odin.ietf.org; Tue, 08 Jul 2003 04:51:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h688p3TV021491
	for sigtran-archive@odin.ietf.org; Tue, 8 Jul 2003 04:51:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZoBg-0005aN-Mv; Tue, 08 Jul 2003 04:51:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZoB3-0005ZD-9n
	for sigtran@optimus.ietf.org; Tue, 08 Jul 2003 04:50:21 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA14512
	for <sigtran@ietf.org>; Tue, 8 Jul 2003 04:50:17 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ZoB0-0007f4-00
	for sigtran@ietf.org; Tue, 08 Jul 2003 04:50:18 -0400
Received: from penguin-ext.wise.edt.ericsson.se ([193.180.251.47] helo=penguin.al.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19ZoAz-0007f1-00
	for sigtran@ietf.org; Tue, 08 Jul 2003 04:50:17 -0400
Received: from esealnt613.al.sw.ericsson.se ([153.88.254.125])
	by penguin.al.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h688oIGJ020897
	for <sigtran@ietf.org>; Tue, 8 Jul 2003 10:50:18 +0200 (MEST)
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 N8X2X0S6; Tue, 8 Jul 2003 10:50:18 +0200
Received: by ESEALNT745.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <N94NMZ8L>; Tue, 8 Jul 2003 10:49:59 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CDD@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
        sigtran@ietf.org
Subject: RE: [Sigtran] M3UA IG v04
Date: Tue, 8 Jul 2003 10:49:53 +0200 
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>


It seems that finally it went through.... (Better late than never) :-D

http://www.ietf.org/internet-drafts/draft-ietf-sigtran-m3ua-implementors-guide-04.txt

Cheers // Javier.

> -----Original Message-----
> From: J.Javier Pastor (EE/EEM) 
> Sent: jueves, 03 de julio de 2003 10:59
> To: sigtran@ietf.org
> Cc: 'bidulock@openss7.org'; Samuel Dur D. Jeyaseelan
> Subject: [Sigtran] M3UA IG v04
> 
> 
> 
> Hi guys,
> 
> We tried to submit the next IG version before the deadline, 
> but I got confused since I thought it ended at 12:00 instead 
> of 9:00. So the server bounced it. I'll submit the doc again 
> after the IETF meeting. In the meantime you can find it at:
> 
> http://standards.ericsson.net/javier/draft-ietf-sigtran-m3ua-i
> mplementors-guide-04.txt
> 
> We've tried to include all the different opinions in here. 
> Although it seems that the case of waiting till n ASPs is the 
> preferred one, we have included the new possibility of a 
> smooth start for those of you who want to activate the AS 
> when the first ASP gets to ACTIVE. At the end it will be the 
> operator's decission how to proceed during the activation 
> phase. Hopefully everyone agrees on this wording...
> 
> Cheers // Javier.
> 
> 
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: martes, 01 de julio de 2003 20:18
> > To: Samuel Dur D. Jeyaseelan
> > Cc: sigtran@ietf.org
> > Subject: Re: [Sigtran] Re:n+k issue
> > 
> > 
> > Samuel,
> > 
> > Perhaps you could answer my questions below first.  Please be
> > specific.
> > 
> > --brian
> > 
> > Samuel Dur D. Jeyaseelan wrote:                               
> >                       (Tue, 01 Jul 2003 06:09:34)
> > > 
> > > Brian,
> > > 
> > >   Can you bit elaborate your n=256 example (ANSI-SLS) with 
> > real-time 
> > >   scenario?(please write to the point)
> > > 
> > > --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, 1 Jul 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Samuel,
> > > > 
> > > > Samuel Dur D. Jeyaseelan wrote:                      
> > (Mon, 30 Jun 2003 09:44:01)
> > > > > 
> > > > > Brian,
> > > > > 
> > > > > On Sat, 28 Jun 2003, Brian F. G. Bidulock wrote:
> > > > > 
> > > > > > Samuel,
> > > > > > 
> > > > > > I should point out that the proposal to require that 
> > the SG wait for 'n' to
> > > > > > start traffic does not have much merit.
> > > > > > 
> > > > > > Example: n=256, k=1
> > > > > > 
> > > > > > 2 ASPs fail, permanently and need to be manually replaced.
> > > > > > 
> > > > > > A temporary network failure causes SCTP associations 
> > to be lost to the
> > > > > > remaining 255 ASPs.  They move to the ASP down state 
> > and immediately
> > > > > > re-establish SCTP associations and move to the 
> > ASP-Active state.  Because the
> > > > > > number of ASPs is < n, the SG will not restart 
> > traffic for some time.
> > > > > 
> > > > > Your example leads to the following figure.
> > > > > 
> > > > >          SG1                            AS1
> > > > >      
> > > > >  SS7 MSG +------+----------------------- ASP (1)
> > > > > ---------| SGP1 |----------------------- ASP (2)
> > > > >          |      |----------------------- ....
> > > > >          +------+----------------------- ASP (256)
> > > > >             |
> > > > >             | 
> > > > >             +---------------------------ASP(k)
> > > > > 
> > > > > 1. First of all craft will not configure 1 ASP as 
> > backup for 256 Active
> > > > >    ASPs, if he does, it means to be a poor network 
> > configuration.
> > > > 
> > > > Well, first of all, craft does not normally choose the 
> > number of ASPs used for
> > > > backup, engineering does.  Second, k=1 is more than 
> > adequate for n=256 where
> > > > the ASPs are reasonably reliable and have a short 
> > replacement time.
> > > > 
> > > > Regardless, I followed with the same example with n=2 and k=1.
> > > > 
> > > > > 
> > > > > 2. Consider ASP as an MGC, your example leads to 256 
> > MGCs and they are 
> > > > >    controlled by a single SGP.A good network will have 
> > multiple SGPs to
> > > > >    share the POTENTIAL traffic across ASPs.
> > > > 
> > > > You seem to misunderstand the concept of redundancy in 
> > M3UA: there are not
> > > > multiple SGPs or ASPs because of traffic load.  It is 
> > easy enough to handle
> > > > the traffic of a small to medium sized country through a 
> > single processor.
> > > > 
> > > > Also, you seem to misunderstand that an SGP and an ASP 
> > are not processors.
> > > > They are processes.  One host processor could have many 
> > (even 256) ASPs or
> > > > SGPs (or both for that matter).
> > > > 
> > > > > 
> > > > > 3. BASIC INTENTION OF 'n+k' is,just to make sure that 
> > incoming messages
> > > > >    ARE EFFICIENTLY TRANSMITTED TO DESTINATION ASPs WITH 
> > THE HELP OF
> > > > >    MINIMUM NUMBER OF ASSOCIATIONS.
> > > > 
> > > > No.  The intention of 'n+k' is to provide one of the more 
> > recently accepted
> > > > models for reliability using redundancy (n = m+1, k=0 
> > being the typical system
> > > > deployed in the field today, m usually being greater than 
> > 2).  Up until
> > > > recently, only 2m (or m+m) systems were considered 
> > carrier class.  Some
> > > > systems were 4m (with 3 out of 4 vote).
> > > > 
> > > > You should take some time to look at redundancy and 
> > reliability requirements
> > > > for carrier grade systems for an understanding of how to 
> > calculate and provide
> > > > the reliability requirements expected by carriers.  There 
> > are good standards
> > > > written on the topic.  See the Telcordia NEBS and 
> > reliability specs, for
> > > > example.  If you have an RBOC RFP handy, take a look at 
> > the hardware and
> > > > software reliability requirements sections or talk to the 
> > engineer in your
> > > > company responsible for answering those sections.
> > > > 
> > > > > 
> > > > >    Say for example
> > > > > 
> > > > >    Inefficient configuration -just an example
> > > > >    ------------------------------------------
> > > > >   
> > > > >             SLS         DEST
> > > > >             =================
> > > > >              0           ASP1
> > > > >              1           ASP1
> > > > >              2           ASP1
> > > > >              3           ASP1
> > > > >              4           ASP1
> > > > >              5           ASP1
> > > > >              6           ASP1
> > > > >              .            .
> > > > >              .            .
> > > > >              15          ASP1
> > > > >              
> > > > >              
> > > > >                SG                            AS1
> > > > > sls(0,1..15 )+----+  sls(0,1..15)         +------+
> > > > > -------------|SGP1|-----------------------| ASP1 |
> > > > >              |    |              ASSOC-1  +------+
> > > > >              +----+                     
> > > > > 
> > > > >                       Fig - 1
> > > > > 
> > > > > In this figure, all incoming messages are transmitted 
> > over ASSOC-1 towards
> > > > > ASP1.Here there are two potential problems 
> > > > > 
> > > > >        1. Total incoming traffic is dependent on a single ASP
> > > > >        2. ASP1 is always overloaded, messages are 
> > processed one by one
> > > > >           from the transmission buffer which ultimately 
> > introduces time
> > > > >           delay(example last arrived msg will be 
> > processed at the end).
> > > > 
> > > > The above was not in my examples.  My examples had n > 1.
> > > > 
> > > > > 
> > > > > 
> > > > > Efficient configuration
> > > > > -----------------------  
> > > > > 
> > > > >              SLS         DEST
> > > > >             =================
> > > > >              0           ASP1
> > > > >              1           ASP1
> > > > >              2           ASP1
> > > > >              3           ASP1
> > > > >              4           ASP1
> > > > >              5           ASP1
> > > > >              6           ASP1
> > > > >              7           ASP1
> > > > >              8           ASP2
> > > > >              9           ASP2
> > > > >              .            .
> > > > >              .            .
> > > > >              15          ASP2
> > > > > 
> > > > >                
> > > > >                        
> > > > >                        
> > > > >               SG                             AS1
> > > > > sls(0,1..15 )+----+   sls(0,1,..7)ASSOC-1 +------+
> > > > > -------------|SGP1|-----------------------| ASP1 |
> > > > >              |    |--------+              +------+
> > > > >              +----+        | sls(8,9..15) +------+
> > > > >                            +--------------| ASP2 |
> > > > >                                  ASSOC-2  +------+
> > > > >                         
> > > > >                    fig - 2
> > > > > 
> > > > > In this example incoming traffic is distributed between 
> > ASP1 and ASP2.
> > > > > 
> > > > > IN ORDER TO HANDLE THE HEAVY INCOMING TRAFFIC, SGP1 
> > NEEDS MINIMUM OF
> > > > > 2 ACTIVE ASPs BY SETTING n=2.
> > > > 
> > > > Do they not comply to SS7 ISUP/SCCP flow and congestion 
> > controls?  Are they
> > > > running at full capacity in this arrangement?  Are these 
> > your ASPs?  Do you
> > > > design your systems this way?
> > > > 
> > > > Take a look at the reliability engineering.  If either of 
> > these ASPs is
> > > > handling its maximum rated capacity, it will have to be 
> > reliable to .99999
> > > > This is because, if one ASP fails, 50% of the system has 
> > failed.  A 50% system
> > > > failure is deemed a "system" failure rather than a 
> > "component" failure in
> > > > reliability engineering for carrier grade systems.  
> > Considering proscribed
> > > > replacement times on the ASP, this would be difficult to 
> > achieve unless both
> > > > ASPs were hardware fault tolerant platforms.
> > > > 
> > > > Nevertheless, the RFC defines ASPs as a process, not a 
> > processor.  There is
> > > > nothing restricting the two ASPs in your diagram from 
> > being processes on the
> > > > same processor.  In that case, when ASP2 is down, ASP1 
> > has double the
> > > > processing capacity as it had before.
> > > > 
> > > > You certainly cannot make statements as you have made 
> here without
> > > > presupposing an implementation architecture outside the 
> > scope of the standard.
> > > > 
> > > > Besides, the RFC permits you to configure your SGP (and 
> > why do you only have
> > > > one of them?) so that it waits for 'n' ASPs to become 
> > active.  You MAY do
> > > > this.  Just don't require everybody else to follow your 
> > approach.  (Actually,
> > > > if the above is your approach, don't require anyone else 
> > to follow it.)
> > > > 
> > > > Several other points: ANSI T1.111.4/2000 SLS values are 
> > 8-bits.  That means
> > > > there are 256 SLS values (gee, I had 256 ASPs in my example).
> > > > 
> > > > For another thing, sequenced traffic flows are not only 
> > based on SLS.  In an
> > > > ITU-T network, each remote signalling point has 16 
> > independent traffic flows
> > > > for each AS.  The traffic flows from each remote 
> > signalling point is
> > > > independent from the traffic flows of other remote 
> > signalling points.
> > > > Therefore, if there are 400 remote signalling points, 
> > there will be 6400
> > > > independent traffic flows that the SG can use for 
> > distribution of AS traffic
> > > > for one AS over as many as 6400 ASPs without risking 
> > mis-sequencing of
> > > > messages.  For an ANSI T1.111.4/2000 network using 8-bit 
> > SLS values with 400
> > > > remote signalling points, there would be 102400 
> > independent traffic flows to
> > > > the AS that the SG can use to distribute traffic to one 
> > AS over as many as
> > > > 102400 ASPs without risking message mis-sequencing.
> > > > 
> > > > > 
> > > > > 
> > > > > Your example with n=256 is completely a wrong 
> > interpretaion of 'n+k'
> > > > > redundancy case.
> > > > 
> > > > Quite the opposite, you clearly misunderstand redundancy 
> > and reliability in
> > > > carrier grade systems and obviously have little 
> > experience with them.
> > > > 
> > > > Here's a little primer for you:
> > > > 
> > > > Carrier grade systems were originally designed as 
> > hot-standby, fully redundant
> > > > systems.  For every processor actively handling traffic, 
> > there was a hot
> > > > standby processor also actively handling the same 
> > traffic.  This typically
> > > > required special bus arrangements whereby the I/O bus of 
> > one processor was
> > > > actively controlling peripheral devices, while the hot 
> > standby processor was
> > > > issuing I/O bus output that was not terminated to the I/O 
> > bus.  The bus also
> > > > normally performed a comparison of the outputs from the 
> > active side and the
> > > > outputs from the inactive side.  Peripheral components, 
> > such as switch fabric
> > > > were also fully redundant (duplicated).  Switching 
> > fabrics are separated into
> > > > an A-plane and a B-plane in much the same fashion as the 
> > SPCS front-end I/O
> > > > bus.
> > > > 
> > > > Note that active/standby in M3UA does not address the 
> > above model.  2+0
> > > > broadcast does.
> > > > 
> > > > The problem with such a systems was what we used to call 
> > the "kill your
> > > > sister" syndrome.  A failure on the inactive side only 
> > detectable to the I/O
> > > > bus could cause a switch from the active to inactive 
> > side, resulting in the
> > > > switch running on the erroneous inactive side.  We called 
> > it the "kill your
> > > > sister" syndrome because it is not so surprising that the 
> > psychotic sister
> > > > (erred inactive side) kills the sane sister (correct 
> active side).
> > > > 
> > > > Primarily to avoid the "kill your sister" syndrome, some 
> > switches were
> > > > designed with 3 or 4 processors on an cross-checking I/O 
> > bus pair.  In that
> > > > case, it would take a 2 out of 3 or 3 out of 4 vote to 
> > shift from the active
> > > > processor.  These systems were far less susceptible to 
> > inactive plane errors
> > > > that were only detected at the I/O bus.
> > > > 
> > > > Note that the only M3UA model that address the above is 
> > the 3+0 or 4+0
> > > > broadcast model.
> > > > 
> > > > When the use of general purpose computing platforms for 
> > telecommunications
> > > > systems started to become popular, there were several 
> > models that emerged.
> > > > 
> > > > First was taking two big iron UNIX hosts and running them 
> > in active/standby.
> > > > In this scenario, one UNIX host would be actively 
> > handling traffic and the
> > > > other would be warm standby.  By "warm" standby, it means 
> > that the device was
> > > > powered and was synchronizing state on a checkpoint 
> > interval or monitoring the
> > > > health of the active hosts with heartbeats.  Shared 
> > memory, shared DASD or
> > > > message passing are the most common mechanisms of 
> > redundancy.  These systems
> > > > are considered "highly available" systems rather than the 
> > "fault tolerant"
> > > > systems used in telecommunications above.  An increased 
> > level of reliability
> > > > is often achieved by using hardware fault tolerant 
> > multiple processor hosts.
> > > > 
> > > > In M3UA, the active/standby model addresses these 
> > systems.  Although M3UA
> > > > provides for a '1+k' model, k is invariably 1.
> > > > 
> > > > The major difficulties with these systems are: a) 
> > switchover latency between
> > > > systems; b) loss of state during switchover; and c) 
> > scalability.  Switchover
> > > > latency is normally improved by decreasing heartbeat 
> > intervals.  State
> > > > synchronization is normally improved by decreasing the 
> > process checkpoint
> > > > interval.  These reductions typically increase cost and 
> > they are normally
> > > > chosen to minimize a cost/performance/reliability metric. 
> >  Scalability was
> > > > typically performed by installing larger more powerful 
> > hosts with increased
> > > > demand.  This was not good, because larger processors 
> > tended to have poorer
> > > > cost/performance ratios that smaller processors.
> > > > 
> > > > Later it was determined that the cost of a complete 
> > redundant host that was
> > > > wasting CPU cycles while inactive could be avoided while 
> > addressing the
> > > > deficiencies of the paired host approach above.  To 
> increase state
> > > > synchronization, state could be spread over a number of 
> > hosts.  If there were
> > > > 'm' hosts, the failure of one host could at worst affect 
> > a proportion, 1/m, of
> > > > the state.  Therefore state synchronization would improve 
> > as m increased.  To
> > > > distribute the state, traffic load would be shared over 
> > the m hosts according
> > > > to state.  This also served the needs of scalability and 
> > provided for closer
> > > > to linear cost scalability, because the number of hosts, 
> > m, could be increased
> > > > while holding the cost/performance ratio of the hosts 
> > constant.  The costs of
> > > > redundancy could also be reduced over the fully redundant 
> > systems described
> > > > above, because redundancy could be achieved using 'm+1' 
> > processors, where 'm'
> > > > processors were required to handle the maximum engineered 
> > load.  For systems
> > > > as small as 'm=2' hosts with equivalent reliability to 
> > that normally
> > > > experienced in the '1+1' active/standby scenario above, 
> > system reliability
> > > > figures could be maintained with less redundancy cost.  
> > It is important to
> > > > note, that in these systems, all 'm+1' hosts are active 
> > and handling only a
> > > > proportion m/(m+1) of their rated capacity.  It is very 
> > seldom that the rated
> > > > capacity was more that 2/3 of the maximum short-term capacity.
> > > > 
> > > > These systems do not solve the heartbeat interval problem of the
> > > > active/standby paired hosts system above.  Typically what 
> > these systems use is
> > > > more sophisticated watchdog approaches rather than 
> > passing heartbeat messages
> > > > between processes on the active and standby hosts.  
> > Memory sharing, DASD
> > > > sharing, message passing, reliable broadcast, voting, and 
> > a plethora of
> > > > reliable distributed systems techniques have been 
> > developed that are useful in
> > > > these systems.
> > > > 
> > > > Adoption of these latter systems changed the requirement 
> > from "duplicated"
> > > > systems to systems with no "single point of failure".
> > > > 
> > > > Now, the 'n+k' load-sharing and broadcast model was 
> > intended to address these
> > > > systems.  Where 'n == m' in the description above.  That 
> > is why we agreed that
> > > > n is the maximum engineered load.  Because that is what 
> > 'm' is.  The k is the
> > > > 1 in the description above, generalized.
> > > > 
> > > > First off, you must understand that it is not the 
> > intention of the model to
> > > > have 'n' ASPs active and 'k' inactive unless 'n=1, k=1'.  
> > This is described in
> > > > section 1.4.4.1:
> > > > 
> > > >   A "1+1" active/standby redundancy is a subset of this model.
> > > > 
> > > > Also no redundancy is possible:
> > > > 
> > > >   A simplex "1+0" model is also supported as a subset, 
> > with no ASP redundancy.
> > > > 
> > > > (The term "simplex" come from the telephone switches 
> > described above where the
> > > > inactive side is taken off-line.  Such a situation is 
> > described as running the
> > > > switch "simplex".)
> > > > 
> > > > The 'n+k' model is to address the 'm+1' distributed 
> > systems described last.
> > > > Rereading section 1.4.4.1:
> > > > 
> > > >   The failover model supports an "n+k" redundancy model, 
> > where "n" APSs is the
> > > >   minimum number of redundant [sic] ASPs required to 
> > handle traffic and "k"
> > > >   ASPs are available to take over for a failed or 
> unavailable ASP.
> > > > 
> > > > Note that the state of the "k" ASPs with respect to an AS 
> > is not stated.  In
> > > > the 'm+1' redundant system described above, the "k" ASPs 
> > are active for the
> > > > AS.  "k" ASPs active in load sharing mode are "available 
> > to take over for a
> > > > failed or unavailable ASP".
> > > > 
> > > > It is typical to run the "k" ASPs active (and I know you 
> > had some questions
> > > > about activating the "k" ASPs, but the point is that they 
> > are normally already
> > > > active), and handling their own share of traffic.  In 
> > this case, each ASPs
> > > > handles n/(n+k) of their engineered capacity.  Were you 
> > actually considering
> > > > deploying such a atypical model where the 'k' ASPs are inactive?
> > > > 
> > > > "k" ASPs can be taken out of service, upgraded, and 
> > brought back into service
> > > > without affecting the maximum engineering traffic 
> > handling capacity of the
> > > > system (in absence of failures during the upgrade).  This 
> > assists in graceful
> > > > software/hardware upgrades on live systems.  (Keeping 
> > them inactive would
> > > > serve no purpose and would, decrease maximum traffic 
> > handling capability,
> > > > increase loss of state on fail-over, and increase 
> > fail-over latency: all good 
> > > > reasons for not deactivating the 'k' ASPs).
> > > > 
> > > > Now, reliability engineering for carrier class systems 
> > takes the view that if
> > > > 50% of the load bearing components in a system fail (stop 
> > operating in any
> > > > way), then the entire system has failed.  If traffic is 
> > stopped above n/2
> > > > processes, the system will have failed prematurely.
> > > > 
> > > > A requirement to withhold traffic < n will always suffer 
> > a prolonged system
> > > > failure, because only n/2 are required to make the system 
> > available.
> > > > Therefore, on startup the reliability of the system is 
> > degraded by the
> > > > negative exponent n/2, regardless of whether n=256 or 
> > n=2; however, increasing
> > > > 'n' increases the damage done to reliability by waiting for 'n'.
> > > > 
> > > > A requirement for the SG to wait for 'n' is unacceptable 
> > from the standpoint
> > > > of the redundancy model that 'n+k' was meant to address, 
> > as well as other
> > > > possible models that it can accommodate.
> > > > 
> > > > > 
> > > > > 
> > > > > > 
> > > > > > Consider the same example with: n=2, k=1, two ASPs failed.
> > > > > > 
> > > > > > Any system that introduces hysteresis to aggravate 
> > and perpetuate an outage in
> > > > > > this fashion is not very useful.
> > > > > > 
> > > > > > In the examples, what would have been a brief 
> > AS-PENDING situation has been
> > > > > > turned into a prolonged and complete system failure.  
> > It is certainly not good
> > > > > > practice to design redundant systems in that fashion 
> > and recommending or
> > > > > > requiring such a design would be a bad thing.
> > > > > > 
> > > > > > To further the example, consider the multiple SG as 
> > STP situation.  One SG
> > > > > > cannot know the state of ASPs for the its mated SG 
> > and therefore cannot
> > > > > > distinguish between the restart scenario and the 
> > start scenario as far as the
> > > > > > AS state is concerned because the SG does not know 
> > the state of the AS itself:
> > > > > > the AS could be active with other SGs.
> > > > > > 
> > > > > > This is the same case for SGP.  Unless AS state is 
> > synchronized across SGPs,
> > > > > > one SGP cannot know the total state of the AS (only 
> > its partition).  There is
> > > > > > no requirement to synchronize AS state across SGP (on 
> > the SGP side, AS-state
> > > > > > is per-SGP), and there is certainly no requirement to 
> > synchronize AS-state
> > > > > > across SGs.
> > > > > > 
> > > > > > So, take the (n=2, k=0) example where there are two 
> > SGs.  Due to a network
> > > > > > partitioning at startup, one ASP can communicate only 
> > with one SG, and the
> > > > > > other ASP can communicate only with the other SG.  
> > Because n=2, neither SG
> > > > > > will start traffic and the system suffers 100% 
> > failure until the network
> > > > > > partitioning ceases.
> > > > > > 
> > > > > > Consider the same example where there are two SGPs 
> > (instead of SGs) and the
> > > > > > two ASPs normally only form one association (e.g., 
> > ASP1-SGP1, ASP2-SGP2).
> > > > > > Note that there is no requirement in the RFC to form 
> > full-mesh associations.
> > > > > > Because the number of active ASPs with each SGP is < 
> > 2, the system will never
> > > > > > start.
> > > > > > 
> > > > > > These are not strange scenarios.  These situations 
> > are all permitted by the
> > > > > > RFC.  Yet, in each situation it is clear that one 
> > cannot recommend that the SG
> > > > > > wait.  The result would be a non-functional system.  
> > There are many more
> > > > > > examples.
> > > > > > 
> > > > > > n+k is a model for redundancy (not for congestion and 
> > flow control).  A system
> > > > > > with an SG that waits for n is, in general, not a 
> > fully redundant system at
> > > > > > all, but is, rather, prone to these failure syndromes.
> > > > > > 
> > > > > > When you look at the situation and apply the 
> > experience of reliability
> > > > > > engineering, the recommendation is that the SG SHOULD 
> > start traffic when the
> > > > > > first ASP goes active.
> > > > > > 
> > > > > > Why wait for 'n' anyway?  Because the ASP cannot 
> > handle some level of traffic?
> > > > > > Any ASP that conforms to MTP-User standards, must 
> > have the standardized
> > > > > > MTP-User (ISUP/SCCP) congestion and flow controls and 
> > must have the SCTP
> > > > > > congestion controls and is therefore able to accept 
> > any amount of traffic
> > > > > > overload.
> > > > > > 
> > > > > > So there is no real reason for the SG to wait for 
> > 'n'.  Nevertheless, the RFC
> > > > > > allows this as an OPTION.  Requiring (MUST) or 
> > recommending it (SHOULD)
> > > > > > regardless of whether the number of active processors 
> > is rising or falling,
> > > > > > whether start, restart or failure, is unacceptable.
> > > > > > 
> > > > > > On the other side, at the ASPs, the ASPs have a 
> > requirement to share state
> > > > > > between them.  They must share circuit state and 
> > transaction state in all
> > > > > > configurations to meet carrier grade.  This is stated 
> > in the RFC.  It is
> > > > > > implicit that the ASPs are capable of knowing with 
> > which SGs, and which SGPs
> > > > > > for which they are active.  They have access to the 
> > total state of the AS that
> > > > > > they collectively represent, that the SG/SGP does 
> > not.  Therefore is it
> > > > > > reasonable for the ASPs to coordinate among 
> > themselves the number of active
> > > > > > ASPs and to ensure that there are sufficient ASPs 
> > active to handle the
> > > > > > expected load before starting traffic.  Because the 
> > ASPs need not respond to
> > > > > > user test traffic (UPT/SST), the ASPs are also in 
> > control of when remote SS7
> > > > > > MTP-User restart traffic toward them.
> > > > > > 
> > > > > > Therefore, the recommendation that the ASPs SHOULD 
> > coordinate among
> > > > > > themselves the number of active ASPs and not start 
> > traffic until a sufficient
> > > > > > number of them are active is sound.  This is not a 
> > MUST, permitting ASPs
> > > > > > that know that they are capable of handling any 
> > overload, to start traffic
> > > > > > immediately, or at some earlier point.
> > > > > > 
> > > > > > Back to the RFC, the RFC sets the normal case where 
> > the SG starts traffic on
> > > > > > the first active ASP but explicitly permits as an 
> > OPTION to wait until it
> > > > > > considers that there are sufficient ASPs active to 
> > handle the offered load.
> > > > > > The RFC also RECOMMENDS that the ASPs coordinate 
> > among themselves the number
> > > > > > of active ASPs so that there are sufficient ASPs to 
> > handle the expected load.
> > > > > > 
> > > > > > The provisions in the RFC are both sound, where the 
> > proposal and section 3.15
> > > > > > is clearly not.
> > > > > > 
> > > > > > Remember when looking at the AS state diagram in the 
> > RFC that this is the
> > > > > > SGP's view of the AS state and not the total state of 
> > the AS, that can only
> > > > > > known to the ASPs for all configurations.  'n' as 
> > defined has no meaning to
> > > > > > the state machine at the SGP because is not required 
> > to know, and often cannot
> > > > > > know, how many ASPs are active for the AS via other 
> > SGPs and SGs, or even via
> > > > > > direct connection to the SS7 network.
> > > > > > 
> > > > > > Some SGs might have special knowledge, or 
> > sophisticated algorithms that allow
> > > > > > it to determine whether a congestive failure will be 
> > the immediate result of
> > >  > > > starting or restarting traffic.  These SGs are 
> > permitted as an
> > > OPTION to wait.
> > > > > > 
> > > > > > The ASPs SHOULD have the knowledge of how many are 
> > required to handle the
> > > > > > expected load, because they must share 
> > circuit/transaction state and they must
> > > > > > have congestion and flow controls as required by the 
> > standards.
> > > > > > 
> > > > > > So again, the RFC is adequate in this regard.  The 
> > current proposal and
> > > > > > section 3.15 are not.
> > > > > RFC is inadequate so we need section 3.15 to explain 
> 'n+k' case.
> > > > 
> > > > Teaching reliability engineering and the design of 
> > redundant systems is
> > > > outside the scope of the RFC and the IG.  People, such as 
> > yourself, are better
> > > > advised to learn the subject matter before attempting to 
> > implement or claim a
> > > > complete understanding of the protocol.
> > > > 
> > > > The current RFC provides the protocol mechanisms and basis for
> > > > interoperability for all of the common approaches to 
> > redundancy in carrier
> > > > class systems, and even provides a general model that can 
> > accommodate
> > > > non-traditional redundancy architectures.
> > > > 
> > > > When confused about how a redundancy architecture works, 
> > don't blame the M3UA
> > > > RFC, blame those responsible for describing redundant 
> > architectures to you.
> > > > 
> > > > In the appendix we stated:
> > > > 
> > > >    M3UA imposes no restrictions as to the exact layout of 
> > the network
> > > >    elements, the message distribution algorithms and the 
> > distribution of
> > > >    signalling processes.  Instead, it provides a 
> > framework and a set of
> > > >    messages that allow for a flexible and scalable 
> > signalling network
> > > >    architecture, aiming to provide reliability and performance.
> > > > 
> > > > Forcing SGs to wait for 'n' ASPs active before activating 
> > is contrary to this.
> > > > The existing "MAY" achieves this flexibility.
> > > > 
> > > > When reading 1.4.4.1, also remember this statement in the 
> > appendix, that used
> > > > to be beside 1.4.4.1.  The description of 'n+k' does not 
> > dictate the state of
> > > > the 'k' or the state of the AS with respect to 'n' to 
> > leave the redundancy
> > > > model flexible, and accommodate '1+1' and '1+0' in the 
> same model.
> > > > 
> > > > Also, please don't presume to tell me to set 'n' to 1, as 
> > it is not a
> > > > arbitrary value, but the number 'm' in the 'm+1' 
> > distributed system in common
> > > > use today.  This number is determined as the minimum 
> > number of ASPs required
> > > > to handle the maximum engineered load, and not, as Tolga 
> > would have, the
> > > > minimum number of ASPs required to handle ANY load.  If 
> > you want your system
> > > > to behave as Tolga proposes, just have your SG exercise 
> > the "MAY choose not to
> > > > direct traffic" option of section 4.3.4.3 of the RFC.
> > > > 
> > > > Section 3.15 and Tolga's proposal should be struck 
> > because they attempt to
> > > > impose a 'n' active 'k' inactive and system cannot 
> > restart (3.15) or start
> > > > (Tolga) without 'n' view on the redundancy model, where 
> > it is clearly the
> > > > intent of the RFC to place no such restrictions on the 
> > model, but to
> > > > accommodate a range of models, that is already achieved 
> > by the RFC text.
> > > > 
> > > > --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
> 

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



From exim@www1.ietf.org  Tue Jul  8 11:57:34 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA28343
	for <sigtran-archive@odin.ietf.org>; Tue, 8 Jul 2003 11:57:34 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Zuq3-0001yK-0l
	for sigtran-archive@odin.ietf.org; Tue, 08 Jul 2003 11:57:07 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h68Fv6jL007559
	for sigtran-archive@odin.ietf.org; Tue, 8 Jul 2003 11:57:06 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Zupy-0001uj-Tu; Tue, 08 Jul 2003 11:57:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ZupU-0001pa-Ss
	for sigtran@optimus.ietf.org; Tue, 08 Jul 2003 11:56:32 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA28281
	for <sigtran@ietf.org>; Tue, 8 Jul 2003 11:56:29 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ZupT-00032S-00
	for sigtran@ietf.org; Tue, 08 Jul 2003 11:56:31 -0400
Received: from rainier.illuminet.com ([12.41.110.52])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ZupS-00031W-00
	for sigtran@ietf.org; Tue, 08 Jul 2003 11:56:31 -0400
Received: from oly1wnexc02.corp.illuminet.com (oly1wnexc02.corp.illuminet.com [10.55.13.13]) by rainier.illuminet.com (8.8.8/8.8.8) with ESMTP id IAA26590; Tue, 8 Jul 2003 08:55:50 -0700 (PDT)
Received: from oly1wnexc01.corp.illuminet.com ([10.55.13.12]) by oly1wnexc02.corp.illuminet.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2653.13)
	id N99YFG68; Tue, 8 Jul 2003 08:55:50 -0700
Received: by oly1wnexc01.corp.illuminet.com with Internet Mail Service (5.5.2653.19)
	id <N9KG6XTD>; Tue, 8 Jul 2003 08:55:50 -0700
Message-ID: <318266A4431BC549B4C4D6FD3EA6E24DBEF0BE@ove1wnexm01.corp.illuminet.com>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "McCandless, Kevin" <KMcCandless@verisign.com>
To: "'Ken Morneault'" <kmorneau@cisco.com>, sigtran@ietf.org
Subject: RE: [Sigtran] Sigtran Plugtest at ETSI 9/03
Date: Tue, 8 Jul 2003 08:55:43 -0700 
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>

Well, wouldn't the lastest version be tested since it is the one that is under last call??

> -----Original Message-----
> From: Ken Morneault [mailto:kmorneau@cisco.com]
> Sent: Monday, July 07, 2003 9:32 AM
> To: sigtran@ietf.org
> Subject: [Sigtran] Sigtran Plugtest at ETSI 9/03
> 
> 
> 
> 
> All:
> 
> Which version of M2PA should be tested at the upcoming
> Sigtran plugtest in 9/03?  Currently, the ETSI Plugtest site
> indicates version 8, but version 9 was just released.  Version
> 6 was tested at the last Plugtest.
> 
> Please email your preference.
> 
> Regards,
> 
> Ken
> 
> 
> _______________________________________________
> 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 exim@www1.ietf.org  Tue Jul  8 12:02:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA28689
	for <sigtran-archive@odin.ietf.org>; Tue, 8 Jul 2003 12:02:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Zuuo-0002eG-5a
	for sigtran-archive@odin.ietf.org; Tue, 08 Jul 2003 12:02:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h68G22Q2010141
	for sigtran-archive@odin.ietf.org; Tue, 8 Jul 2003 12:02:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Zuun-0002dT-AJ; Tue, 08 Jul 2003 12:02:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Zutq-0002LL-9m
	for sigtran@optimus.ietf.org; Tue, 08 Jul 2003 12:01:02 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA28601
	for <sigtran@ietf.org>; Tue, 8 Jul 2003 12:00:58 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19Zutp-00036d-00
	for sigtran@ietf.org; Tue, 08 Jul 2003 12:01:01 -0400
Received: from sj-core-4.cisco.com ([171.68.223.138])
	by ietf-mx with esmtp (Exim 4.12)
	id 19Zutf-000363-00
	for sigtran@ietf.org; Tue, 08 Jul 2003 12:00:59 -0400
Received: from fruitpie.cisco.com (IDENT:mirapoint@fruitpie.cisco.com [64.102.16.27])
	by sj-core-4.cisco.com (8.12.6/8.12.6) with ESMTP id h68G0Gpp011391;
	Tue, 8 Jul 2003 09:00:16 -0700 (PDT)
Received: from kmorneau-w2k01.cisco.com (dhcp-64-100-224-167.cisco.com [64.100.224.167])
	by fruitpie.cisco.com (Mirapoint Messaging Server MOS 3.3.3-GR)
	with ESMTP id AGI24985;
	Tue, 8 Jul 2003 09:00:15 -0700 (PDT)
Message-Id: <4.3.2.7.2.20030708115845.04b62250@cia.cisco.com>
X-Sender: kmorneau@cia.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Tue, 08 Jul 2003 12:00:14 -0400
To: "McCandless, Kevin" <KMcCandless@verisign.com>, sigtran@ietf.org
From: Ken Morneault <kmorneau@cisco.com>
Subject: RE: [Sigtran] Sigtran Plugtest at ETSI 9/03
In-Reply-To: <318266A4431BC549B4C4D6FD3EA6E24DBEF0BE@ove1wnexm01.corp.il
 luminet.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>



Kevin,

That would be ideal but it depends on what can be ready in the
time frame of the Plugtest.  So far, (with only a single response)
it sounds like v9 is ok.

Ken

At 08:55 AM 7/8/2003 -0700, McCandless, Kevin wrote:
>Well, wouldn't the lastest version be tested since it is the one that is under last call??
>
>> -----Original Message-----
>> From: Ken Morneault [mailto:kmorneau@cisco.com]
>> Sent: Monday, July 07, 2003 9:32 AM
>> To: sigtran@ietf.org
>> Subject: [Sigtran] Sigtran Plugtest at ETSI 9/03
>> 
>> 
>> 
>> 
>> All:
>> 
>> Which version of M2PA should be tested at the upcoming
>> Sigtran plugtest in 9/03?  Currently, the ETSI Plugtest site
>> indicates version 8, but version 9 was just released.  Version
>> 6 was tested at the last Plugtest.
>> 
>> Please email your preference.
>> 
>> Regards,
>> 
>> Ken
>> 
>> 
>> _______________________________________________
>> 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 exim@www1.ietf.org  Tue Jul  8 12:57:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA00532
	for <sigtran-archive@odin.ietf.org>; Tue, 8 Jul 2003 12:57:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Zvm2-0005s4-0i
	for sigtran-archive@odin.ietf.org; Tue, 08 Jul 2003 12:57:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h68Gv1dv022548
	for sigtran-archive@odin.ietf.org; Tue, 8 Jul 2003 12:57:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Zvm1-0005rW-7V; Tue, 08 Jul 2003 12:57:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Zvl4-0005qt-CT
	for sigtran@optimus.ietf.org; Tue, 08 Jul 2003 12:56:02 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA00516
	for <sigtran@ietf.org>; Tue, 8 Jul 2003 12:55:58 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19Zvl2-0003ZT-00
	for sigtran@ietf.org; Tue, 08 Jul 2003 12:56:00 -0400
Received: from angelica.ulticom.com ([208.255.120.2] helo=chuckie.dgms.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19Zvl1-0003ZP-00
	for sigtran@ietf.org; Tue, 08 Jul 2003 12:55:59 -0400
Received: from ulticom.com (localhost [127.0.0.1])
	by chuckie.dgms.com (8.9.3/8.9.3) with ESMTP id MAA26797;
	Tue, 8 Jul 2003 12:55:03 -0400 (EDT)
Message-ID: <3F0AF767.6080401@ulticom.com>
Date: Tue, 08 Jul 2003 12:55:03 -0400
From: David Lehmann <lehmann@ulticom.com>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.4) Gecko/20030624
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Ken Morneault <kmorneau@cisco.com>
CC: "McCandless, Kevin" <KMcCandless@verisign.com>, sigtran@ietf.org
Subject: Re: [Sigtran] Sigtran Plugtest at ETSI 9/03
References: <4.3.2.7.2.20030708115845.04b62250@cia.cisco.com>
In-Reply-To: <4.3.2.7.2.20030708115845.04b62250@cia.cisco.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

Ken Morneault wrote:

> 
> Kevin,
> 
> That would be ideal but it depends on what can be ready in the
> time frame of the Plugtest.  So far, (with only a single response)
> it sounds like v9 is ok.

What were the major (functional) differences between v8 and v9?

-- 

David Lehmann                          Ulticom, Inc.
AOL/Yahoo IM: davidULCM                1020 Briggs Road
1-856-787-2729                         Mt. Laurel, NJ 08054   USA


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



From exim@www1.ietf.org  Wed Jul  9 04:16:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA23560
	for <sigtran-archive@odin.ietf.org>; Wed, 9 Jul 2003 04:16:33 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aA7R-0007sk-LC
	for sigtran-archive@odin.ietf.org; Wed, 09 Jul 2003 04:16:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h698G5oM030298
	for sigtran-archive@odin.ietf.org; Wed, 9 Jul 2003 04:16:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aA7O-0007sR-Nq; Wed, 09 Jul 2003 04:16:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aA7J-0007sC-MY
	for sigtran@optimus.ietf.org; Wed, 09 Jul 2003 04:15:57 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA23549
	for <sigtran@ietf.org>; Wed, 9 Jul 2003 04:15:55 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aA7H-0004zy-00
	for sigtran@ietf.org; Wed, 09 Jul 2003 04:15:55 -0400
Received: from zdemail03.zdem.compaq.com ([161.114.112.27])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aA7G-0004zu-00
	for sigtran@ietf.org; Wed, 09 Jul 2003 04:15:54 -0400
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail03.zdem.compaq.com (Postfix) with ESMTP
	id 9C1573CE; Wed,  9 Jul 2003 10:15:24 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Wed, 9 Jul 2003 10:15:24 +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="us-ascii"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] M3UA IG v04
Date: Wed, 9 Jul 2003 10:15:22 +0200
Message-ID: <06CAA4D3753C53408D2CE6340B68444214C461@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA IG v04
Thread-Index: AcNFLmvryjJoQjNzR8a5EmjLQGEI+wAwXurg
From: "Zannin, Francois" <francois.zannin@hp.com>
To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
        <sigtran@ietf.org>
X-OriginalArrivalTime: 09 Jul 2003 08:15:24.0660 (UTC) FILETIME=[3F852B40:01C345F2]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Just some comment about the M3UA IG v04

3.21 NOTIFY messages are missing in Examples section
Figure (section 5.1.2)
	a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the
ASP Up Ack

Figure (section 5.1.3)
	a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the
ASP Up Ack

Figure (section 5.1.4)
	Same remark, the Notify is sent to ASP1, ASP2 and also need to
be sent to=20
      ASP3 after the ASP Up Ack


Another remark about the RC in the ASPTM:
Is the ASP serves several AS are the RC(s) mandatory in the ASP-ACTIVE?
In an old mail without any answer, it was said that :

  Subject: Re: [Sigtran] Fw: absence of NA in ASP Active (march 03)
    Sibi,
    ....
    When Routing Context is not provided, the ASP is activated for the
single=20
    AS for which it is configured or an error returned if the ASP is not

    configured for a single AS.
    --brian

In the RFC 3332, it is said:

   It is not necessary for the ASP to
   include all Routing Contexts of interest in a single ASP Active
   message, thus requesting to become active in all Routing Contexts at
   the same time.  Multiple ASP Active messages MAY be used to activate
   within the Application Servers independently, or in sets.  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.

Given that there is no reference to that item in the IG, is an ASPTM is
received=20
from an ASP that serves several AS, is it considered as an ACTIVE for
ALL AS or=20
must we return an error ?

Francois Z.

-----Original Message-----
From: J.Javier Pastor (EE/EEM) [mailto:j.javier.pastor@ericsson.com]=20
Sent: Tuesday, July 08, 2003 10:50 AM
To: J.Javier Pastor (EE/EEM); sigtran@ietf.org
Subject: RE: [Sigtran] M3UA IG v04



It seems that finally it went through.... (Better late than never) :-D

http://www.ietf.org/internet-drafts/draft-ietf-sigtran-m3ua-implementors
-guide-04.txt

Cheers // Javier.


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



From exim@www1.ietf.org  Wed Jul  9 04:30:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA23757
	for <sigtran-archive@odin.ietf.org>; Wed, 9 Jul 2003 04:30:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aAKx-0008BH-JD
	for sigtran-archive@odin.ietf.org; Wed, 09 Jul 2003 04:30:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h698U3Io031448
	for sigtran-archive@odin.ietf.org; Wed, 9 Jul 2003 04:30:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aAKv-0008As-U8; Wed, 09 Jul 2003 04:30:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aAKg-0008AW-Th
	for sigtran@optimus.ietf.org; Wed, 09 Jul 2003 04:29:47 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA23722
	for <sigtran@ietf.org>; Wed, 9 Jul 2003 04:29:44 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aAKe-00052Z-00
	for sigtran@ietf.org; Wed, 09 Jul 2003 04:29:44 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[imCXU8EDuQE3smXkQ+LT4wR9gkTFFhbs])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aAKd-00052W-00
	for sigtran@ietf.org; Wed, 09 Jul 2003 04:29:43 -0400
Received: from ns.pigworks.openss7.net (IDENT:nyVLmqmWFQKhXRF4uGN6easDraVyza+G@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h698ThX32479;
	Wed, 9 Jul 2003 02:29:43 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h698TgB20512;
	Wed, 9 Jul 2003 02:29:42 -0600
Date: Wed, 9 Jul 2003 02:29:42 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Zannin, Francois" <francois.zannin@hp.com>
Cc: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
        sigtran@ietf.org
Subject: Re: [Sigtran] M3UA IG v04
Message-ID: <20030709022942.B20398@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Zannin, Francois" <francois.zannin@hp.com>,
	"J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
	sigtran@ietf.org
References: <06CAA4D3753C53408D2CE6340B68444214C461@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: <06CAA4D3753C53408D2CE6340B68444214C461@vbeexc02.emea.cpqcorp.net>; from francois.zannin@hp.com on Wed, Jul 09, 2003 at 10:15:22AM +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>

Francois,

Zannin, Francois wrote:                       (Wed, 09 Jul 2003 10:15:22)

-X-snip-X-
> 
> Another remark about the RC in the ASPTM:
> Is the ASP serves several AS are the RC(s) mandatory in the ASP-ACTIVE?
> In an old mail without any answer, it was said that :
> 
>   Subject: Re: [Sigtran] Fw: absence of NA in ASP Active (march 03)
>     Sibi,
>     ....
>     When Routing Context is not provided, the ASP is activated for the
> single 
>     AS for which it is configured or an error returned if the ASP is not
> 
>     configured for a single AS.
>     --brian
> 
> In the RFC 3332, it is said:
> 
>    It is not necessary for the ASP to
>    include all Routing Contexts of interest in a single ASP Active
>    message, thus requesting to become active in all Routing Contexts at
>    the same time.  Multiple ASP Active messages MAY be used to activate
>    within the Application Servers independently, or in sets.  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.
> 
> Given that there is no reference to that item in the IG, is an ASPTM is
> received 
> from an ASP that serves several AS, is it considered as an ACTIVE for
> ALL AS or 
> must we return an error ?

The second passage just says that the RC values do not all need to be included
in the same ASP Active message.  They can be split up into sets of ASP Active
messages.  Also, it reiterates that if there is no RC in the ASP Active
message, that the ASP must be configured at the SGP for the application server.

If there is no RC in the ASPTM message, and no AS is configured for the ASP at
the SGP, then Error("No Configured AS for ASP") as described in section 3.8.1
is sent.

Not placing the RCs in the ASPTM message when there are multiple AS, even if
they are preconfigued at the SGP, is possible according to the RFC, but of
little use because the RC MUST later be included in all DATA and SSNM messages
in this case 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 exim@www1.ietf.org  Wed Jul  9 11:13:57 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA04638
	for <sigtran-archive@odin.ietf.org>; Wed, 9 Jul 2003 11:13:57 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aGdL-0001Ot-Qg
	for sigtran-archive@odin.ietf.org; Wed, 09 Jul 2003 11:13:29 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h69FDRGN005305
	for sigtran-archive@odin.ietf.org; Wed, 9 Jul 2003 11:13:27 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aGdD-0001Bf-CN; Wed, 09 Jul 2003 11:13:19 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aFIh-0002Gd-7p
	for sigtran@optimus.ietf.org; Wed, 09 Jul 2003 09:48:04 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA00784
	for <sigtran@ietf.org>; Wed, 9 Jul 2003 09:47:59 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aFIf-0006mD-00
	for sigtran@ietf.org; Wed, 09 Jul 2003 09:48:01 -0400
Received: from sj-iport-1-in.cisco.com ([171.71.176.70] helo=sj-iport-1.cisco.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19aFIe-0006m5-00
	for sigtran@ietf.org; Wed, 09 Jul 2003 09:48:00 -0400
Received: from fruitpie.cisco.com (IDENT:mirapoint@fruitpie.cisco.com [64.102.16.27])
	by sj-core-5.cisco.com (8.12.9/8.12.6) with ESMTP id h69DlQrm019703;
	Wed, 9 Jul 2003 06:47:27 -0700 (PDT)
Received: from kmorneau-w2k01.cisco.com (dhcp-64-100-224-167.cisco.com [64.100.224.167])
	by fruitpie.cisco.com (Mirapoint Messaging Server MOS 3.3.3-GR)
	with ESMTP id AGL07455;
	Wed, 9 Jul 2003 06:47:25 -0700 (PDT)
Message-Id: <4.3.2.7.2.20030709094527.04d39f00@cia.cisco.com>
X-Sender: kmorneau@cia.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Wed, 09 Jul 2003 09:47:24 -0400
To: David Lehmann <lehmann@ulticom.com>
From: Ken Morneault <kmorneau@cisco.com>
Subject: Re: [Sigtran] Sigtran Plugtest at ETSI 9/03
Cc: "McCandless, Kevin" <KMcCandless@verisign.com>, sigtran@ietf.org
In-Reply-To: <3F0AF767.6080401@ulticom.com>
References: <4.3.2.7.2.20030708115845.04b62250@cia.cisco.com>
 <4.3.2.7.2.20030708115845.04b62250@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>



David,

v9 provides clarifications/enhancements to the text on providing, local processor 
outage, message retrieval, etc. procedures.

Ken

At 12:55 PM 7/8/2003 -0400, David Lehmann wrote:
>Ken Morneault wrote:
>
>>Kevin,
>>That would be ideal but it depends on what can be ready in the
>>time frame of the Plugtest.  So far, (with only a single response)
>>it sounds like v9 is ok.
>
>What were the major (functional) differences between v8 and v9?
>
>-- 
>
>David Lehmann                          Ulticom, Inc.
>AOL/Yahoo IM: davidULCM                1020 Briggs Road
>1-856-787-2729                         Mt. Laurel, NJ 08054   USA


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



From exim@www1.ietf.org  Wed Jul  9 11:13:59 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA04668
	for <sigtran-archive@odin.ietf.org>; Wed, 9 Jul 2003 11:13:58 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aGdL-0001OK-P2
	for sigtran-archive@odin.ietf.org; Wed, 09 Jul 2003 11:13:31 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h69FDROT005304
	for sigtran-archive@odin.ietf.org; Wed, 9 Jul 2003 11:13:27 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aGdE-0001Cb-7T; Wed, 09 Jul 2003 11:13:20 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aFQu-0003Ai-D3
	for sigtran@optimus.ietf.org; Wed, 09 Jul 2003 09:56:32 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA00904
	for <sigtran@ietf.org>; Wed, 9 Jul 2003 09:56:28 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aFQs-0006oG-00
	for sigtran@ietf.org; Wed, 09 Jul 2003 09:56:30 -0400
Received: from sj-iport-3-in.cisco.com ([171.71.176.72] helo=sj-iport-3.cisco.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19aFQr-0006oC-00
	for sigtran@ietf.org; Wed, 09 Jul 2003 09:56:29 -0400
Received: from cisco.com (171.71.177.254)
  by sj-iport-3.cisco.com with ESMTP; 09 Jul 2003 06:58:46 -0700
Received: from fruitpie.cisco.com (IDENT:mirapoint@fruitpie.cisco.com [64.102.16.27])
	by sj-core-2.cisco.com (8.12.9/8.12.6) with ESMTP id h69DttMr006715;
	Wed, 9 Jul 2003 06:55:55 -0700 (PDT)
Received: from kmorneau-w2k01.cisco.com (dhcp-64-100-224-167.cisco.com [64.100.224.167])
	by fruitpie.cisco.com (Mirapoint Messaging Server MOS 3.3.3-GR)
	with ESMTP id AGL09419;
	Wed, 9 Jul 2003 06:55:53 -0700 (PDT)
Message-Id: <4.3.2.7.2.20030709095324.01ba9870@cia.cisco.com>
X-Sender: kmorneau@cia.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Wed, 09 Jul 2003 09:55:53 -0400
To: "Zannin, Francois" <francois.zannin@hp.com>,
        "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
        <sigtran@ietf.org>
From: Ken Morneault <kmorneau@cisco.com>
Subject: RE: [Sigtran] M3UA IG v04
In-Reply-To: <06CAA4D3753C53408D2CE6340B68444214C461@vbeexc02.emea.cpqco
 rp.net>
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>



Francois,

Pls see comments below.

Regards,

Ken

At 10:15 AM 7/9/2003 +0200, Zannin, Francois wrote:
>Just some comment about the M3UA IG v04
>
>3.21 NOTIFY messages are missing in Examples section
>Figure (section 5.1.2)
>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the
>ASP Up Ack
>
>Figure (section 5.1.3)
>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the
>ASP Up Ack
>
>Figure (section 5.1.4)
>        Same remark, the Notify is sent to ASP1, ASP2 and also need to
>be sent to 
>      ASP3 after the ASP Up Ack

NOTIFY is only sent upon "a change in status of a particular Application Server".



>Another remark about the RC in the ASPTM:
>Is the ASP serves several AS are the RC(s) mandatory in the ASP-ACTIVE?
>In an old mail without any answer, it was said that :
>
>  Subject: Re: [Sigtran] Fw: absence of NA in ASP Active (march 03)
>    Sibi,
>    ....
>    When Routing Context is not provided, the ASP is activated for the
>single 
>    AS for which it is configured or an error returned if the ASP is not
>
>    configured for a single AS.
>    --brian
>
>In the RFC 3332, it is said:
>
>   It is not necessary for the ASP to
>   include all Routing Contexts of interest in a single ASP Active
>   message, thus requesting to become active in all Routing Contexts at
>   the same time.  Multiple ASP Active messages MAY be used to activate
>   within the Application Servers independently, or in sets.  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.
>
>Given that there is no reference to that item in the IG, is an ASPTM is
>received 
>from an ASP that serves several AS, is it considered as an ACTIVE for
>ALL AS or 
>must we return an error ?

I agree with Brian's response to this question.

>Francois Z.
>
>-----Original Message-----
>From: J.Javier Pastor (EE/EEM) [mailto:j.javier.pastor@ericsson.com] 
>Sent: Tuesday, July 08, 2003 10:50 AM
>To: J.Javier Pastor (EE/EEM); sigtran@ietf.org
>Subject: RE: [Sigtran] M3UA IG v04
>
>
>
>It seems that finally it went through.... (Better late than never) :-D
>
>http://www.ietf.org/internet-drafts/draft-ietf-sigtran-m3ua-implementors
>-guide-04.txt
>
>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 exim@www1.ietf.org  Wed Jul  9 11:13:59 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA04666
	for <sigtran-archive@odin.ietf.org>; Wed, 9 Jul 2003 11:13:58 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aGdL-0001Oo-QR
	for sigtran-archive@odin.ietf.org; Wed, 09 Jul 2003 11:13:31 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h69FDRDo005306
	for sigtran-archive@odin.ietf.org; Wed, 9 Jul 2003 11:13:27 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aGdH-0001It-Oj; Wed, 09 Jul 2003 11:13:23 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aFvz-0005E0-2E
	for sigtran@optimus.ietf.org; Wed, 09 Jul 2003 10:28:39 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA02773
	for <sigtran@ietf.org>; Wed, 9 Jul 2003 10:28:34 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aFvw-0006wI-00
	for sigtran@ietf.org; Wed, 09 Jul 2003 10:28:36 -0400
Received: from sj-core-4.cisco.com ([171.68.223.138])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aFvq-0006wD-00
	for sigtran@ietf.org; Wed, 09 Jul 2003 10:28:34 -0400
Received: from fruitpie.cisco.com (IDENT:mirapoint@fruitpie.cisco.com [64.102.16.27])
	by sj-core-4.cisco.com (8.12.6/8.12.6) with ESMTP id h69ERqpp005546;
	Wed, 9 Jul 2003 07:27:52 -0700 (PDT)
Received: from kmorneau-w2k01.cisco.com (dhcp-64-100-224-167.cisco.com [64.100.224.167])
	by fruitpie.cisco.com (Mirapoint Messaging Server MOS 3.3.3-GR)
	with ESMTP id AGL17196;
	Wed, 9 Jul 2003 07:27:51 -0700 (PDT)
Message-Id: <4.3.2.7.2.20030709102735.04d24090@cia.cisco.com>
X-Sender: kmorneau@cia.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Wed, 09 Jul 2003 10:27:50 -0400
To: "Zannin, Francois" <francois.zannin@hp.com>,
        "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
        <sigtran@ietf.org>
From: Ken Morneault <kmorneau@cisco.com>
Subject: RE: [Sigtran] M3UA IG v04
In-Reply-To: <06CAA4D3753C53408D2CE6340B68444214C464@vbeexc02.emea.cpqco
 rp.net>
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>



Francois,

Can you send me a ptr to the email you reference.

Thanks,

Ken

At 04:20 PM 7/9/2003 +0200, Zannin, Francois wrote:
>Ken,
>
>My point of view on he question, see hereafter.
>
>Regards,
>
>Francois Z.
>
>-----Original Message-----
>From: Ken Morneault [mailto:kmorneau@cisco.com] 
>Sent: Wednesday, July 09, 2003 3:56 PM
>To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
>Subject: RE: [Sigtran] M3UA IG v04
>
>
>Francois,
>
>Pls see comments below.
>
>Regards,
>
>Ken
>
>At 10:15 AM 7/9/2003 +0200, Zannin, Francois wrote:
>>Just some comment about the M3UA IG v04
>>
>>3.21 NOTIFY messages are missing in Examples section
>>Figure (section 5.1.2)
>>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the 
>>ASP Up Ack
>>
>>Figure (section 5.1.3)
>>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the 
>>ASP Up Ack
>>
>>Figure (section 5.1.4)
>>        Same remark, the Notify is sent to ASP1, ASP2 and also need to 
>>be sent to
>>      ASP3 after the ASP Up Ack
>
>NOTIFY is only sent upon "a change in status of a particular Application
>Server".
>
>OK, but just one more remark, in section 5.1.4, a notify AS-Inactive is
>sent to 
>ASP1 and ASP2 while to transition to ASP-Inactive was effective after
>the first 
>ASP_UP(/ACK) comming from ASP1, so notify to ASP2 should be considerer
>as inopportune.
>
>In last november, it was agreed in a thread that when an ASP was
>transitionned to UP,
>The current AS state was sent to this ASP, whatever could be the state.
>This current 
>State can be used by the management of an ASP to decide if the AS has to
>be activated 
>Immediately or must wait for an AS-PENDING or AS-INACTIVE notification.


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



From exim@www1.ietf.org  Wed Jul  9 11:27:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA05802
	for <sigtran-archive@odin.ietf.org>; Wed, 9 Jul 2003 11:27:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aGqU-0005Bb-JW
	for sigtran-archive@odin.ietf.org; Wed, 09 Jul 2003 11:27:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h69FR2eH019934
	for sigtran-archive@odin.ietf.org; Wed, 9 Jul 2003 11:27:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aGqT-0005BP-Ti; Wed, 09 Jul 2003 11:27:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aGqQ-0005BC-Em
	for sigtran@optimus.ietf.org; Wed, 09 Jul 2003 11:26:58 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA05794
	for <sigtran@ietf.org>; Wed, 9 Jul 2003 11:26:55 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aGqP-0007YY-00
	for sigtran@ietf.org; Wed, 09 Jul 2003 11:26:57 -0400
Received: from zdemail03.zdem.compaq.com ([161.114.112.27])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aGqN-0007Y9-00
	for sigtran@ietf.org; Wed, 09 Jul 2003 11:26:56 -0400
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail03.zdem.compaq.com (Postfix) with ESMTP
	id A9C7F1107; Wed,  9 Jul 2003 17:26:23 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Wed, 9 Jul 2003 17:26:23 +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="us-ascii"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] M3UA IG v04
Date: Wed, 9 Jul 2003 17:26:22 +0200
Message-ID: <06CAA4D3753C53408D2CE6340B68444201488A9C@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA IG v04
Thread-Index: AcNGJkyjivcOst7wRBm+4VeQkgstwQAB/J/w
From: "Zannin, Francois" <francois.zannin@hp.com>
To: "Ken Morneault" <kmorneau@cisco.com>,
        "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
        <sigtran@ietf.org>
X-OriginalArrivalTime: 09 Jul 2003 15:26:23.0653 (UTC) FILETIME=[74AE5D50:01C3462E]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Ken,

Here is a copy of the mail about notify I was refering.

-----Original Message-----
From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]=20
Sent: Wednesday, February 12, 2003 10:40 AM
To: ksrinivas
Cc: sigtran@ietf.org
Subject: [Sigtran] Re: M3UA Notify message


Srini,

ksrinivas wrote:
(Wed, 12 Feb 2003 13:42:16)
> Brian,
>=20
>      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.
>=20
> --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

--=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
-----Original Message-----
From: Ken Morneault [mailto:kmorneau@cisco.com]=20
Sent: Wednesday, July 09, 2003 4:28 PM
To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
Subject: RE: [Sigtran] M3UA IG v04




Francois,

Can you send me a ptr to the email you reference.

Thanks,

Ken

At 04:20 PM 7/9/2003 +0200, Zannin, Francois wrote:
>Ken,
>
>My point of view on he question, see hereafter.
>
>Regards,
>
>Francois Z.
>
>-----Original Message-----
>From: Ken Morneault [mailto:kmorneau@cisco.com]
>Sent: Wednesday, July 09, 2003 3:56 PM
>To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
>Subject: RE: [Sigtran] M3UA IG v04
>
>
>Francois,
>
>Pls see comments below.
>
>Regards,
>
>Ken
>
>At 10:15 AM 7/9/2003 +0200, Zannin, Francois wrote:
>>Just some comment about the M3UA IG v04
>>
>>3.21 NOTIFY messages are missing in Examples section
>>Figure (section 5.1.2)
>>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the
>>ASP Up Ack
>>
>>Figure (section 5.1.3)
>>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the
>>ASP Up Ack
>>
>>Figure (section 5.1.4)
>>        Same remark, the Notify is sent to ASP1, ASP2 and also need to
>>be sent to
>>      ASP3 after the ASP Up Ack
>
>NOTIFY is only sent upon "a change in status of a particular=20
>Application Server".
>
>OK, but just one more remark, in section 5.1.4, a notify AS-Inactive is

>sent to ASP1 and ASP2 while to transition to ASP-Inactive was effective

>after the first
>ASP_UP(/ACK) comming from ASP1, so notify to ASP2 should be considerer
>as inopportune.
>
>In last november, it was agreed in a thread that when an ASP was=20
>transitionned to UP, The current AS state was sent to this ASP,=20
>whatever could be the state. This current
>State can be used by the management of an ASP to decide if the AS has
to
>be activated=20
>Immediately or must wait for an AS-PENDING or AS-INACTIVE notification.


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



From exim@www1.ietf.org  Wed Jul  9 12:00:57 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA04637
	for <sigtran-archive@odin.ietf.org>; Wed, 9 Jul 2003 11:13:57 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aGdL-0001PR-W9
	for sigtran-archive@odin.ietf.org; Wed, 09 Jul 2003 11:13:29 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h69FDRqv005396
	for sigtran-archive@odin.ietf.org; Wed, 9 Jul 2003 11:13:27 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aGdL-0001Ov-Rl; Wed, 09 Jul 2003 11:13:27 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aGYz-0000pc-FX
	for sigtran@optimus.ietf.org; Wed, 09 Jul 2003 11:08:57 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA04037
	for <sigtran@ietf.org>; Wed, 9 Jul 2003 11:08:52 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aGYw-0007EF-00
	for sigtran@ietf.org; Wed, 09 Jul 2003 11:08:54 -0400
Received: from angelica.ulticom.com ([208.255.120.2] helo=chuckie.dgms.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19aGYv-0007EC-00
	for sigtran@ietf.org; Wed, 09 Jul 2003 11:08:53 -0400
Received: from ulticom.com (localhost [127.0.0.1])
	by chuckie.dgms.com (8.9.3/8.9.3) with ESMTP id LAA07154;
	Wed, 9 Jul 2003 11:08:09 -0400 (EDT)
Message-ID: <3F0C2FD9.2020400@ulticom.com>
Date: Wed, 09 Jul 2003 11:08:09 -0400
From: David Lehmann <lehmann@ulticom.com>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.4) Gecko/20030624
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Ken Morneault <kmorneau@cisco.com>
CC: "McCandless, Kevin" <KMcCandless@verisign.com>, sigtran@ietf.org
Subject: Re: [Sigtran] Sigtran Plugtest at ETSI 9/03
References: <4.3.2.7.2.20030708115845.04b62250@cia.cisco.com> <4.3.2.7.2.20030708115845.04b62250@cia.cisco.com> <4.3.2.7.2.20030709094527.04d39f00@cia.cisco.com>
In-Reply-To: <4.3.2.7.2.20030709094527.04d39f00@cia.cisco.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

So the use of v8/v9 at the bakeoff is not a big issue.
The big functional difference from v7/v8 is the FSN
starting point, resulting in a minor coding change.
(Correct me if I am wrong.)

So targeting v9 at the bakeoff should be fine.

-- 

David Lehmann                          Ulticom, Inc.
AOL/Yahoo IM: davidULCM                1020 Briggs Road
1-856-787-2729                         Mt. Laurel, NJ 08054   USA

Ken Morneault wrote:
> 
> David,
> 
> v9 provides clarifications/enhancements to the text on providing, local processor 
> outage, message retrieval, etc. procedures.
> 
> Ken
> 
> At 12:55 PM 7/8/2003 -0400, David Lehmann wrote:
> 
>>Ken Morneault wrote:
>>
>>
>>>Kevin,
>>>That would be ideal but it depends on what can be ready in the
>>>time frame of the Plugtest.  So far, (with only a single response)
>>>it sounds like v9 is ok.
>>
>>What were the major (functional) differences between v8 and v9?
>>
>>-- 
>>
>>David Lehmann                          Ulticom, Inc.
>>AOL/Yahoo IM: davidULCM                1020 Briggs Road
>>1-856-787-2729                         Mt. Laurel, NJ 08054   USA
> 


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



From exim@www1.ietf.org  Wed Jul  9 12:00:58 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA04665
	for <sigtran-archive@odin.ietf.org>; Wed, 9 Jul 2003 11:13:58 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aGdL-0001Op-Qd
	for sigtran-archive@odin.ietf.org; Wed, 09 Jul 2003 11:13:31 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h69FDRNo005303
	for sigtran-archive@odin.ietf.org; Wed, 9 Jul 2003 11:13:27 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aGdH-0001Hp-2k; Wed, 09 Jul 2003 11:13:23 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aFrQ-000580-L9
	for sigtran@optimus.ietf.org; Wed, 09 Jul 2003 10:23:56 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA02701
	for <sigtran@ietf.org>; Wed, 9 Jul 2003 10:23:46 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aFrH-0006v1-00
	for sigtran@ietf.org; Wed, 09 Jul 2003 10:23:47 -0400
Received: from zdemail04.zdem.compaq.com ([161.114.112.28])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aFrF-0006uS-00
	for sigtran@ietf.org; Wed, 09 Jul 2003 10:23:47 -0400
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail04.zdem.compaq.com (Postfix) with ESMTP
	id 5DD68D53; Wed,  9 Jul 2003 16:20:46 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Wed, 9 Jul 2003 16:20:46 +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="us-ascii"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] M3UA IG v04
Date: Wed, 9 Jul 2003 16:20:45 +0200
Message-ID: <06CAA4D3753C53408D2CE6340B68444214C464@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA IG v04
Thread-Index: AcNGIdZrrtJmRDpXQiq8zJIbuSwlJAAAfQ9w
From: "Zannin, Francois" <francois.zannin@hp.com>
To: "Ken Morneault" <kmorneau@cisco.com>,
        "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
        <sigtran@ietf.org>
X-OriginalArrivalTime: 09 Jul 2003 14:20:46.0356 (UTC) FILETIME=[49DE8540:01C34625]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Ken,

My point of view on he question, see hereafter.

Regards,

Francois Z.

-----Original Message-----
From: Ken Morneault [mailto:kmorneau@cisco.com]=20
Sent: Wednesday, July 09, 2003 3:56 PM
To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
Subject: RE: [Sigtran] M3UA IG v04


Francois,

Pls see comments below.

Regards,

Ken

At 10:15 AM 7/9/2003 +0200, Zannin, Francois wrote:
>Just some comment about the M3UA IG v04
>
>3.21 NOTIFY messages are missing in Examples section
>Figure (section 5.1.2)
>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the=20
>ASP Up Ack
>
>Figure (section 5.1.3)
>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the=20
>ASP Up Ack
>
>Figure (section 5.1.4)
>        Same remark, the Notify is sent to ASP1, ASP2 and also need to=20
>be sent to
>      ASP3 after the ASP Up Ack

NOTIFY is only sent upon "a change in status of a particular Application
Server".

OK, but just one more remark, in section 5.1.4, a notify AS-Inactive is
sent to=20
ASP1 and ASP2 while to transition to ASP-Inactive was effective after
the first=20
ASP_UP(/ACK) comming from ASP1, so notify to ASP2 should be considerer
as inopportune.

In last november, it was agreed in a thread that when an ASP was
transitionned to UP,
The current AS state was sent to this ASP, whatever could be the state.
This current=20
State can be used by the management of an ASP to decide if the AS has to
be activated=20
Immediately or must wait for an AS-PENDING or AS-INACTIVE notification.


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



From exim@www1.ietf.org  Wed Jul  9 13:34:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA11434
	for <sigtran-archive@odin.ietf.org>; Wed, 9 Jul 2003 13:34:33 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aIpT-0006Ta-3a
	for sigtran-archive@odin.ietf.org; Wed, 09 Jul 2003 13:34:07 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h69HY7CE024885
	for sigtran-archive@odin.ietf.org; Wed, 9 Jul 2003 13:34:07 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aIpN-0006T4-Bt; Wed, 09 Jul 2003 13:34:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aIpL-0006St-C8
	for sigtran@optimus.ietf.org; Wed, 09 Jul 2003 13:33:59 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA11395
	for <sigtran@ietf.org>; Wed, 9 Jul 2003 13:33:54 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aIpI-00017F-00
	for sigtran@ietf.org; Wed, 09 Jul 2003 13:33:56 -0400
Received: from adsl-209-204-165-125.sonic.net ([209.204.165.125] helo=mail1.adax.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19aIpH-00016r-00
	for sigtran@ietf.org; Wed, 09 Jul 2003 13:33:56 -0400
Received: from fradax (adax [12.0.0.88])
	by mail1.adax.com (Postfix) with ESMTP id 8B9F13D72
	for <sigtran@ietf.org>; Wed,  9 Jul 2003 10:30:23 -0700 (PDT)
Received: from fradax (fradax [11.0.0.88])
	by fradax (8.8.5/8.8.5) with ESMTP id RAA24509
	for <sigtran@ietf.org>; Wed, 9 Jul 2003 17:30:14 GMT
Date: Wed, 9 Jul 2003 10:30:14 -0700 (PDT)
From: Javier Kim <javier@adax.com>
X-X-Sender: javier@fradax
To: sigtran@ietf.org
Subject: Re: [Sigtran] Sigtran Plugtest at ETSI 1-5, Sep 2003
In-Reply-To: <4.3.2.7.2.20030709094527.04d39f00@cia.cisco.com>
Message-ID: <Pine.UW2.4.44.0307090953320.23615-100000@fradax>
X-url: http://www.adax.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,

We believe that only M2PA v9 should be used in the next Plugtest for the
following reasons:

a. M2PA v9 will be used for LC and likely to promoted to RFC status
b. Mixing v8 and v9 will diminish interoperability during the Plugtest
c. Tom G. will publish diff information between V8 and V9 very shortly,
   according to his email to Adax(Richard Knowles). We still have 7 weeks
   to upgrade M2PA to v9.
d. It will be awkward to create an IG based v8 while the v9 is in the
   process of becoming an RFC(hopefully). How useful will it be?

Best regards,

Javier Kim
Adax, Inc.



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



From exim@www1.ietf.org  Wed Jul  9 13:44:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA11752
	for <sigtran-archive@odin.ietf.org>; Wed, 9 Jul 2003 13:44:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aIz4-0007P9-19
	for sigtran-archive@odin.ietf.org; Wed, 09 Jul 2003 13:44:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h69Hi2Jw028457
	for sigtran-archive@odin.ietf.org; Wed, 9 Jul 2003 13:44:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aIz3-0007Oj-7u; Wed, 09 Jul 2003 13:44:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aIy5-0007O7-U8
	for sigtran@optimus.ietf.org; Wed, 09 Jul 2003 13:43:02 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA11703
	for <sigtran@ietf.org>; Wed, 9 Jul 2003 13:42:59 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aIy3-0001Bo-00
	for sigtran@ietf.org; Wed, 09 Jul 2003 13:42:59 -0400
Received: from sj-core-4.cisco.com ([171.68.223.138])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aIy2-0001BP-00
	for sigtran@ietf.org; Wed, 09 Jul 2003 13:42:59 -0400
Received: from fruitpie.cisco.com (IDENT:mirapoint@fruitpie.cisco.com [64.102.16.27])
	by sj-core-4.cisco.com (8.12.6/8.12.6) with ESMTP id h69HgNpp025450;
	Wed, 9 Jul 2003 10:42:24 -0700 (PDT)
Received: from kmorneau-w2k01.cisco.com (dhcp-64-100-224-167.cisco.com [64.100.224.167])
	by fruitpie.cisco.com (Mirapoint Messaging Server MOS 3.3.3-GR)
	with ESMTP id AGL67472;
	Wed, 9 Jul 2003 10:42:22 -0700 (PDT)
Message-Id: <4.3.2.7.2.20030709134138.065f0e80@cia.cisco.com>
X-Sender: kmorneau@cia.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Wed, 09 Jul 2003 13:42:21 -0400
To: "Zannin, Francois" <francois.zannin@hp.com>,
        "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
        <sigtran@ietf.org>
From: Ken Morneault <kmorneau@cisco.com>
Subject: RE: [Sigtran] M3UA IG v04
In-Reply-To: <06CAA4D3753C53408D2CE6340B68444201488A9C@vbeexc02.emea.cpq
 corp.net>
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>



Francois,

I certainly don't have an issue adding this sentence to the IG.  I didn't realize
that we had agreed to do so though.

What do others think?

Regards,

Ken

At 05:26 PM 7/9/2003 +0200, Zannin, Francois wrote:
>Ken,
>
>Here is a copy of the mail about notify I was refering.
>
>-----Original Message-----
>From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
>Sent: Wednesday, February 12, 2003 10:40 AM
>To: ksrinivas
>Cc: sigtran@ietf.org
>Subject: [Sigtran] Re: M3UA Notify message
>
>
>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
>-----Original Message-----
>From: Ken Morneault [mailto:kmorneau@cisco.com] 
>Sent: Wednesday, July 09, 2003 4:28 PM
>To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
>Subject: RE: [Sigtran] M3UA IG v04
>
>
>
>
>Francois,
>
>Can you send me a ptr to the email you reference.
>
>Thanks,
>
>Ken
>
>At 04:20 PM 7/9/2003 +0200, Zannin, Francois wrote:
>>Ken,
>>
>>My point of view on he question, see hereafter.
>>
>>Regards,
>>
>>Francois Z.
>>
>>-----Original Message-----
>>From: Ken Morneault [mailto:kmorneau@cisco.com]
>>Sent: Wednesday, July 09, 2003 3:56 PM
>>To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
>>Subject: RE: [Sigtran] M3UA IG v04
>>
>>
>>Francois,
>>
>>Pls see comments below.
>>
>>Regards,
>>
>>Ken
>>
>>At 10:15 AM 7/9/2003 +0200, Zannin, Francois wrote:
>>>Just some comment about the M3UA IG v04
>>>
>>>3.21 NOTIFY messages are missing in Examples section
>>>Figure (section 5.1.2)
>>>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the
>>>ASP Up Ack
>>>
>>>Figure (section 5.1.3)
>>>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the
>>>ASP Up Ack
>>>
>>>Figure (section 5.1.4)
>>>        Same remark, the Notify is sent to ASP1, ASP2 and also need to
>>>be sent to
>>>      ASP3 after the ASP Up Ack
>>
>>NOTIFY is only sent upon "a change in status of a particular 
>>Application Server".
>>
>>OK, but just one more remark, in section 5.1.4, a notify AS-Inactive is
>
>>sent to ASP1 and ASP2 while to transition to ASP-Inactive was effective
>
>>after the first
>>ASP_UP(/ACK) comming from ASP1, so notify to ASP2 should be considerer
>>as inopportune.
>>
>>In last november, it was agreed in a thread that when an ASP was 
>>transitionned to UP, The current AS state was sent to this ASP, 
>>whatever could be the state. This current
>>State can be used by the management of an ASP to decide if the AS has
>to
>>be activated 
>>Immediately or must wait for an AS-PENDING or AS-INACTIVE notification.


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



From exim@www1.ietf.org  Wed Jul  9 16:51:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA21574
	for <sigtran-archive@odin.ietf.org>; Wed, 9 Jul 2003 16:51:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aLu2-0003ek-Hd
	for sigtran-archive@odin.ietf.org; Wed, 09 Jul 2003 16:51:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h69Kp2oL014042
	for sigtran-archive@odin.ietf.org; Wed, 9 Jul 2003 16:51:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aLu1-0003eE-Gy; Wed, 09 Jul 2003 16:51:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19Zq9z-0002yu-5v
	for sigtran@optimus.ietf.org; Tue, 08 Jul 2003 06:57:23 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA17397
	for <sigtran@ietf.org>; Tue, 8 Jul 2003 06:57:18 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19Zq9v-0000da-00
	for sigtran@ietf.org; Tue, 08 Jul 2003 06:57:19 -0400
Received: from zdemail03.zdem.compaq.com ([161.114.112.27])
	by ietf-mx with esmtp (Exim 4.12)
	id 19Zq9u-0000dQ-00
	for sigtran@ietf.org; Tue, 08 Jul 2003 06:57:18 -0400
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail03.zdem.compaq.com (Postfix) with ESMTP id 1456F108B
	for <sigtran@ietf.org>; Tue,  8 Jul 2003 12:56:39 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Tue, 8 Jul 2003 12:56:38 +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"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] Sigtran Plugtest at ETSI 9/03
Date: Tue, 8 Jul 2003 12:56:37 +0200
Message-ID: <BE462396B4A022449E0EB2DD4753D05402F26906@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] Sigtran Plugtest at ETSI 9/03
Thread-Index: AcNElPqf6pMpZdmVSXuwUoD8wayPtwAqj30g
From: "Anzile, Christophe" <christophe.anzile@hp.com>
To: <sigtran@ietf.org>
X-OriginalArrivalTime: 08 Jul 2003 10:56:38.0620 (UTC) FILETIME=[9B3C91C0:01C3453F]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable


All,
same remark applies for SUA (v14 stated @etsi.org) versus v15 out last =
week.
Christophe.


-----Original Message-----
From: Ken Morneault [mailto:kmorneau@cisco.com]
Sent: Monday, July 07, 2003 4:32 PM
To: sigtran@ietf.org
Subject: [Sigtran] Sigtran Plugtest at ETSI 9/03




All:

Which version of M2PA should be tested at the upcoming
Sigtran plugtest in 9/03?  Currently, the ETSI Plugtest site
indicates version 8, but version 9 was just released.  Version
6 was tested at the last Plugtest.

Please email your preference.

Regards,

Ken


_______________________________________________
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 exim@www1.ietf.org  Thu Jul 10 04:50:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA23067
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 04:50:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aX7t-00040P-46
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 04:50:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6A8o5b5015397
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 04:50:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aX7p-000405-8v; Thu, 10 Jul 2003 04:50:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aX7l-0003zl-HM
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 04:49:57 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA23056
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 04:49:52 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aX7i-0000iO-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 04:49:54 -0400
Received: from penguin-ext.wise.edt.ericsson.se ([193.180.251.47] helo=penguin.al.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19aX7h-0000iK-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 04:49:53 -0400
Received: from esealnt612.al.sw.ericsson.se ([153.88.254.118])
	by penguin.al.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h6A8nIGJ009817;
	Thu, 10 Jul 2003 10:49:18 +0200 (MEST)
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 NW4V8D85; Thu, 10 Jul 2003 10:49:18 +0200
Received: by ESEALNT745.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <N94NNPFJ>; Thu, 10 Jul 2003 10:48:58 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CE9@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
To: "'Zannin, Francois'" <francois.zannin@hp.com>,
        Ken Morneault
	 <kmorneau@cisco.com>, sigtran@ietf.org
Subject: RE: [Sigtran] M3UA IG v04
Date: Thu, 10 Jul 2003 10:48:51 +0200
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Hi Francois,=20

Pls see below....

Cheers // Javier.

> -----Original Message-----
> From: Zannin, Francois [mailto:francois.zannin@hp.com]
> Sent: mi=E9rcoles, 09 de julio de 2003 16:21
> To: Ken Morneault; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> Subject: RE: [Sigtran] M3UA IG v04
>=20
>=20
> Ken,
>=20
> My point of view on he question, see hereafter.
>=20
> Regards,
>=20
> Francois Z.
>=20
> -----Original Message-----
> From: Ken Morneault [mailto:kmorneau@cisco.com]=20
> Sent: Wednesday, July 09, 2003 3:56 PM
> To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> Subject: RE: [Sigtran] M3UA IG v04
>=20
>=20
> Francois,
>=20
> Pls see comments below.
>=20
> Regards,
>=20
> Ken
>=20
> At 10:15 AM 7/9/2003 +0200, Zannin, Francois wrote:
> >Just some comment about the M3UA IG v04
> >
> >3.21 NOTIFY messages are missing in Examples section
> >Figure (section 5.1.2)
> >        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2=20
> after the=20
> >ASP Up Ack
> >
> >Figure (section 5.1.3)
> >        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2=20
> after the=20
> >ASP Up Ack
> >
> >Figure (section 5.1.4)
> >        Same remark, the Notify is sent to ASP1, ASP2 and=20
> also need to=20
> >be sent to
> >      ASP3 after the ASP Up Ack
>=20

It is sent at that point because n=3D2 for that particular example. =
I.e. when 2 ASPs reach the ASP-INACTIVE state, the AS is considered in =
AS-INACTIVE state.

=20

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



From exim@www1.ietf.org  Thu Jul 10 04:51:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA23092
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 04:51:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aX8o-00044A-P0
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 04:51:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6A8p2Mw015624
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 04:51:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aX8o-00043l-3j; Thu, 10 Jul 2003 04:51:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aX8V-00041Z-SZ
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 04:50:43 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA23083
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 04:50:39 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aX8S-0000if-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 04:50:40 -0400
Received: from penguin-ext.wise.edt.ericsson.se ([193.180.251.47] helo=penguin.al.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19aX8R-0000ic-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 04:50:39 -0400
Received: from esealnt613.al.sw.ericsson.se ([153.88.254.125])
	by penguin.al.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h6A8oWGJ010123;
	Thu, 10 Jul 2003 10:50:32 +0200 (MEST)
Received: from ESEALNT744.al.sw.ericsson.se ([153.88.251.4]) by esealnt613.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55)
	id N8XJ2K63; Thu, 10 Jul 2003 10:50:32 +0200
Received: by ESEALNT744.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <N94MG2TR>; Thu, 10 Jul 2003 10:50:36 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CEA@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
To: "'Ken Morneault'" <kmorneau@cisco.com>,
        "Zannin, Francois"
	 <francois.zannin@hp.com>, sigtran@ietf.org
Subject: RE: [Sigtran] M3UA IG v04
Date: Thu, 10 Jul 2003 10:49:55 +0200
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Hi,=20

Neither do I. Let's wait for more opinions...

Cheers // Javier.

> -----Original Message-----
> From: Ken Morneault [mailto:kmorneau@cisco.com]
> Sent: mi=E9rcoles, 09 de julio de 2003 19:42
> To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> Subject: RE: [Sigtran] M3UA IG v04
>=20
>=20
>=20
>=20
> Francois,
>=20
> I certainly don't have an issue adding this sentence to the=20
> IG.  I didn't realize
> that we had agreed to do so though.
>=20
> What do others think?
>=20
> Regards,
>=20
> Ken
>=20
> At 05:26 PM 7/9/2003 +0200, Zannin, Francois wrote:
> >Ken,
> >
> >Here is a copy of the mail about notify I was refering.
> >
> >-----Original Message-----
> >From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]=20
> >Sent: Wednesday, February 12, 2003 10:40 AM
> >To: ksrinivas
> >Cc: sigtran@ietf.org
> >Subject: [Sigtran] Re: M3UA Notify message
> >
> >
> >Srini,
> >
> >ksrinivas wrote:
> >(Wed, 12 Feb 2003 13:42:16)
> >> Brian,
> >>=20
> >>      Can you tell me when an SGP should send Notify(AS-PENDING) =
and
> >> Notify(Insufficient ASP Resources Active in AS)?. I felt=20
> RFC and IG of
> >
> >> M3UA is not clear on these two.
> >>=20
> >> --Srini
> >
> >It think that it is pretty straight-forward.
> >
> >NTFY(Insufficient ASPs) is sent for an AS by an SGP when the=20
> 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=20
> "n" is the
> >minimum number of active ASPs defined for the AS.  So this=20
> notification
> >is only sent if "n" is greater than 1.
> >
> >NTFY(AS-PENDING) is sent for an AS by an SGP when the number=20
> 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=20
> insufficent
> >ASPs available for an AS, there is no notification that there are
> >sufficient ASPs for an AS.  This causes problems for ASP=20
> wishing to use
> >the first notification as in indication to go active.  There=20
> 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=20
> 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=20
> 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=20
> 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=20
> best course of
> >action for the AS.
> >
> >In addition to notifications resulting from AS state=20
> changes, I think we
> >should say in the IG:
> >
> >   The SGP SHOULD send a NTFY message indicating the state=20
> of the AS to
> >   all ASPs newly becoming ASP-INACTIVE from ASP-DOWN for each AS.
> >
> >--brian
> >
> >--=20
> >Brian F. G. Bidulock
> >bidulock@openss7.org
> >http://www.openss7.org/=20
> _______________________________________________
> >Sigtran mailing list
> >Sigtran@ietf.org
> >https://www1.ietf.org/mailman/listinfo/sigtran
> >-----Original Message-----
> >From: Ken Morneault [mailto:kmorneau@cisco.com]=20
> >Sent: Wednesday, July 09, 2003 4:28 PM
> >To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> >Subject: RE: [Sigtran] M3UA IG v04
> >
> >
> >
> >
> >Francois,
> >
> >Can you send me a ptr to the email you reference.
> >
> >Thanks,
> >
> >Ken
> >
> >At 04:20 PM 7/9/2003 +0200, Zannin, Francois wrote:
> >>Ken,
> >>
> >>My point of view on he question, see hereafter.
> >>
> >>Regards,
> >>
> >>Francois Z.
> >>
> >>-----Original Message-----
> >>From: Ken Morneault [mailto:kmorneau@cisco.com]
> >>Sent: Wednesday, July 09, 2003 3:56 PM
> >>To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> >>Subject: RE: [Sigtran] M3UA IG v04
> >>
> >>
> >>Francois,
> >>
> >>Pls see comments below.
> >>
> >>Regards,
> >>
> >>Ken
> >>
> >>At 10:15 AM 7/9/2003 +0200, Zannin, Francois wrote:
> >>>Just some comment about the M3UA IG v04
> >>>
> >>>3.21 NOTIFY messages are missing in Examples section
> >>>Figure (section 5.1.2)
> >>>        a NOTIFY(AS-INACTIVE) must be sent from SGP to=20
> ASP2 after the
> >>>ASP Up Ack
> >>>
> >>>Figure (section 5.1.3)
> >>>        a NOTIFY(AS-INACTIVE) must be sent from SGP to=20
> ASP2 after the
> >>>ASP Up Ack
> >>>
> >>>Figure (section 5.1.4)
> >>>        Same remark, the Notify is sent to ASP1, ASP2 and=20
> also need to
> >>>be sent to
> >>>      ASP3 after the ASP Up Ack
> >>
> >>NOTIFY is only sent upon "a change in status of a particular=20
> >>Application Server".
> >>
> >>OK, but just one more remark, in section 5.1.4, a notify=20
> AS-Inactive is
> >
> >>sent to ASP1 and ASP2 while to transition to ASP-Inactive=20
> was effective
> >
> >>after the first
> >>ASP_UP(/ACK) comming from ASP1, so notify to ASP2 should be=20
> considerer
> >>as inopportune.
> >>
> >>In last november, it was agreed in a thread that when an ASP was=20
> >>transitionned to UP, The current AS state was sent to this ASP,=20
> >>whatever could be the state. This current
> >>State can be used by the management of an ASP to decide if=20
> the AS has
> >to
> >>be activated=20
> >>Immediately or must wait for an AS-PENDING or AS-INACTIVE=20
> notification.
>=20

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



From exim@www1.ietf.org  Thu Jul 10 05:35:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA23932
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 05:35:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aXpP-0006g4-2l
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 05:35:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6A9Z3Dc025669
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 05:35:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aXpN-0006fm-Ac; Thu, 10 Jul 2003 05:35:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aXox-0006ez-Qo
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 05:34:35 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA23889
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 05:34:30 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aXou-0000ws-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 05:34:32 -0400
Received: from zdemail03.zdem.compaq.com ([161.114.112.27])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aXot-0000wK-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 05:34:31 -0400
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail03.zdem.compaq.com (Postfix) with ESMTP
	id BEB761262; Thu, 10 Jul 2003 11:34:02 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Thu, 10 Jul 2003 11:34:02 +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"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] M3UA IG v04
Date: Thu, 10 Jul 2003 11:33:56 +0200
Message-ID: <06CAA4D3753C53408D2CE6340B68444214C465@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA IG v04
Thread-Index: AcNGwDMOACZivnN8Rey9jXkfoIh4YgABNOMA
From: "Zannin, Francois" <francois.zannin@hp.com>
To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
        "Ken Morneault" <kmorneau@cisco.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 10 Jul 2003 09:34:02.0378 (UTC) FILETIME=[65E98EA0:01C346C6]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Javier,

See comment inline,

Regards,

Francois Z.

-----Original Message-----
From: J.Javier Pastor (EE/EEM) [mailto:j.javier.pastor@ericsson.com]=20
Sent: Thursday, July 10, 2003 10:49 AM
To: Zannin, Francois; Ken Morneault; sigtran@ietf.org
Subject: RE: [Sigtran] M3UA IG v04


Hi Francois,=20

Pls see below....

Cheers // Javier.

> -----Original Message-----
> From: Zannin, Francois [mailto:francois.zannin@hp.com]
> Sent: mi=E9rcoles, 09 de julio de 2003 16:21
> To: Ken Morneault; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> Subject: RE: [Sigtran] M3UA IG v04
>=20
>=20
> Ken,
>=20
> My point of view on he question, see hereafter.
>=20
> Regards,
>=20
> Francois Z.
>=20
> -----Original Message-----
> From: Ken Morneault [mailto:kmorneau@cisco.com]
> Sent: Wednesday, July 09, 2003 3:56 PM
> To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> Subject: RE: [Sigtran] M3UA IG v04
>=20
>=20
> Francois,
>=20
> Pls see comments below.
>=20
> Regards,
>=20
> Ken
>=20
> At 10:15 AM 7/9/2003 +0200, Zannin, Francois wrote:
> >Just some comment about the M3UA IG v04
> >
> >3.21 NOTIFY messages are missing in Examples section
> >Figure (section 5.1.2)
> >        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2
> after the
> >ASP Up Ack
> >
> >Figure (section 5.1.3)
> >        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2
> after the
> >ASP Up Ack
> >
> >Figure (section 5.1.4)
> >        Same remark, the Notify is sent to ASP1, ASP2 and
> also need to
> >be sent to
> >      ASP3 after the ASP Up Ack
>=20

# It is sent at that point because n=3D2 for that particular example. =
I.e. when 2 ASPs reach the ASP-INACTIVE state, # the AS is considered in =
AS-INACTIVE state.

I can understand that SG(p) can hold the Notification while there are =
not enough ASP UP to operate the minimum "n" ASP potentially ready to =
"Activate" the AS. But if the AS is still not Active when another ASP =
becomes "UP", a do not see any reason not to send it the current state =
"inactive" for the AS.

If we look in the RFC, it is in charge of the ASP to coordonate the =
activation of the AS. The SG cannot decide which ASP have to send the =
ASP-Active for a particular AS (by holding back the notify to other =
ASP).

RFC 3332
  4.3.4.3 ASP Active Procedures
   ...
   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.
   ...=20


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



From exim@www1.ietf.org  Thu Jul 10 06:39:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA25919
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 06:39:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aYpL-0003AS-FE
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 06:39:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AAd393012170
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 06:39:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aYpI-00039a-K6; Thu, 10 Jul 2003 06:39:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aYoQ-00038b-Gz
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 06:38:06 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA25893
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 06:38:00 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aYoM-0001Vs-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 06:38:02 -0400
Received: from auds953.usa.alcatel.com ([143.209.238.6])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aYoL-0001Vo-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 06:38:01 -0400
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6AAbKLX028230;
	Thu, 10 Jul 2003 05:37:20 -0500 (CDT)
Received: from sun2594.ssd.usa.alcatel.com (sun2594.ssd.usa.alcatel.com [143.209.250.252])
	by ssd.usa.alcatel.com (8.12.8/8.12.8) with ESMTP id h6AAaO1C009240;
	Thu, 10 Jul 2003 05:36:24 -0500 (CDT)
Date: Thu, 10 Jul 2003 05:36:24 -0500 (CDT)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "Zannin, Francois" <francois.zannin@hp.com>
cc: Ken Morneault <kmorneau@cisco.com>,
        "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
        sigtran@ietf.org
Subject: RE: [Sigtran] M3UA IG v04
In-Reply-To: <06CAA4D3753C53408D2CE6340B68444201488A9C@vbeexc02.emea.cpqcorp.net>
Message-ID: <Pine.SOL.4.10.10307100453470.25952-100000@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>


Zannin,

  Just an example, say an operator configures an AS1 as set of 3 ASPes.
  They are namely known as ASP1,ASP2 and ASP3(backup).

  Initially ASP1 and ASP2 send ASP_ACTIVE to SGP  so SGP sends a
  NTFY(AS_ACTIVE) thus results in AS1 to be in ACTIVE state.

  Where is the question of sending NTFY(Sufficient ASPs active)?
  People mean to say that backup ASP also sends an ASP_ACTIVE message
  to SGP? if it so, on what condition it is being sent? Is there
  any valid conditions that backup ASP or some new(!!!??!amazing!!) ASP
  to send ASP_ACTIVE message? 


--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, 9 Jul 2003, Zannin, Francois wrote:

> Ken,
> 
> Here is a copy of the mail about notify I was refering.
> 
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> Sent: Wednesday, February 12, 2003 10:40 AM
> To: ksrinivas
> Cc: sigtran@ietf.org
> Subject: [Sigtran] Re: M3UA Notify message
> 
> 
> 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
> -----Original Message-----
> From: Ken Morneault [mailto:kmorneau@cisco.com] 
> Sent: Wednesday, July 09, 2003 4:28 PM
> To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> Subject: RE: [Sigtran] M3UA IG v04
> 
> 
> 
> 
> Francois,
> 
> Can you send me a ptr to the email you reference.
> 
> Thanks,
> 
> Ken
> 
> At 04:20 PM 7/9/2003 +0200, Zannin, Francois wrote:
> >Ken,
> >
> >My point of view on he question, see hereafter.
> >
> >Regards,
> >
> >Francois Z.
> >
> >-----Original Message-----
> >From: Ken Morneault [mailto:kmorneau@cisco.com]
> >Sent: Wednesday, July 09, 2003 3:56 PM
> >To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> >Subject: RE: [Sigtran] M3UA IG v04
> >
> >
> >Francois,
> >
> >Pls see comments below.
> >
> >Regards,
> >
> >Ken
> >
> >At 10:15 AM 7/9/2003 +0200, Zannin, Francois wrote:
> >>Just some comment about the M3UA IG v04
> >>
> >>3.21 NOTIFY messages are missing in Examples section
> >>Figure (section 5.1.2)
> >>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the
> >>ASP Up Ack
> >>
> >>Figure (section 5.1.3)
> >>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the
> >>ASP Up Ack
> >>
> >>Figure (section 5.1.4)
> >>        Same remark, the Notify is sent to ASP1, ASP2 and also need to
> >>be sent to
> >>      ASP3 after the ASP Up Ack
> >
> >NOTIFY is only sent upon "a change in status of a particular 
> >Application Server".
> >
> >OK, but just one more remark, in section 5.1.4, a notify AS-Inactive is
> 
> >sent to ASP1 and ASP2 while to transition to ASP-Inactive was effective
> 
> >after the first
> >ASP_UP(/ACK) comming from ASP1, so notify to ASP2 should be considerer
> >as inopportune.
> >
> >In last november, it was agreed in a thread that when an ASP was 
> >transitionned to UP, The current AS state was sent to this ASP, 
> >whatever could be the state. This current
> >State can be used by the management of an ASP to decide if the AS has
> to
> >be activated 
> >Immediately or must wait for an AS-PENDING or AS-INACTIVE notification.
> 
> 
> _______________________________________________
> 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 exim@www1.ietf.org  Thu Jul 10 06:55:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA26323
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 06:55:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aZ4o-0004E0-Rx
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 06:55:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AAt2nS016234
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 06:55:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aZ4n-0004DX-KF; Thu, 10 Jul 2003 06:55:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aZ4d-0004Cx-8I
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 06:54:51 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA26302
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 06:54:44 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aZ4Z-0001dB-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 06:54:47 -0400
Received: from albatross-ext.wise.edt.ericsson.se ([193.180.251.49] helo=albatross.tn.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19aZ4Y-0001d8-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 06:54:46 -0400
Received: from esealnt611.al.sw.ericsson.se ([153.88.254.121])
	by albatross.tn.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h6AAsfIi025021;
	Thu, 10 Jul 2003 12:54:41 +0200 (MEST)
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 3S62D1D9; Thu, 10 Jul 2003 12:57:16 +0200
Received: by ESEALNT745.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <N94NNRMF>; Thu, 10 Jul 2003 12:54:21 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CED@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
To: "'Zannin, Francois'" <francois.zannin@hp.com>,
        Ken Morneault
	 <kmorneau@cisco.com>, sigtran@ietf.org
Subject: RE: [Sigtran] M3UA IG v04
Date: Thu, 10 Jul 2003 12:54:14 +0200
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Hi,=20

Pls, see my answers inline..

Cheers // JAvier.

> -----Original Message-----
> From: Zannin, Francois [mailto:francois.zannin@hp.com]
> Sent: jueves, 10 de julio de 2003 11:34
> To: J.Javier Pastor (EE/EEM); Ken Morneault; sigtran@ietf.org
> Subject: RE: [Sigtran] M3UA IG v04
>=20
>=20
> Javier,
>=20
> See comment inline,
>=20
> Regards,
>=20
> Francois Z.
>=20
> -----Original Message-----
> From: J.Javier Pastor (EE/EEM) [mailto:j.javier.pastor@ericsson.com]=20
> Sent: Thursday, July 10, 2003 10:49 AM
> To: Zannin, Francois; Ken Morneault; sigtran@ietf.org
> Subject: RE: [Sigtran] M3UA IG v04
>=20
>=20
> Hi Francois,=20
>=20
> Pls see below....
>=20
> Cheers // Javier.
>=20
> > -----Original Message-----
> > From: Zannin, Francois [mailto:francois.zannin@hp.com]
> > Sent: mi=E9rcoles, 09 de julio de 2003 16:21
> > To: Ken Morneault; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA IG v04
> >=20
> >=20
> > Ken,
> >=20
> > My point of view on he question, see hereafter.
> >=20
> > Regards,
> >=20
> > Francois Z.
> >=20
> > -----Original Message-----
> > From: Ken Morneault [mailto:kmorneau@cisco.com]
> > Sent: Wednesday, July 09, 2003 3:56 PM
> > To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA IG v04
> >=20
> >=20
> > Francois,
> >=20
> > Pls see comments below.
> >=20
> > Regards,
> >=20
> > Ken
> >=20
> > At 10:15 AM 7/9/2003 +0200, Zannin, Francois wrote:
> > >Just some comment about the M3UA IG v04
> > >
> > >3.21 NOTIFY messages are missing in Examples section
> > >Figure (section 5.1.2)
> > >        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2
> > after the
> > >ASP Up Ack
> > >
> > >Figure (section 5.1.3)
> > >        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2
> > after the
> > >ASP Up Ack
> > >
> > >Figure (section 5.1.4)
> > >        Same remark, the Notify is sent to ASP1, ASP2 and
> > also need to
> > >be sent to
> > >      ASP3 after the ASP Up Ack
> >=20
>=20
> # It is sent at that point because n=3D2 for that particular=20
> example. I.e. when 2 ASPs reach the ASP-INACTIVE state, # the=20
> AS is considered in AS-INACTIVE state.
>=20
> I can understand that SG(p) can hold the Notification while=20
> there are not enough ASP UP to operate the minimum "n" ASP=20
> potentially ready to "Activate" the AS. But if the AS is=20
> still not Active when another ASP becomes "UP", a do not see=20
> any reason not to send it the current state "inactive" for the AS.
>=20
The reason was to follow the RFC and it says that NTFY is sent when the =
AS state changes. As n=3D2 for the general case AS becomes AS-INACTIVE =
after two ASP becomes INACTIVE. At that moment the AS changes state. =
According to the RFC the NTFY is sent to all the ASPs that are not ion =
DOWN state. This is what the example is showing.=20

> If we look in the RFC, it is in charge of the ASP to=20
> coordonate the activation of the AS. The SG cannot decide=20
> which ASP have to send the ASP-Active for a particular AS (by=20
> holding back the notify to other ASP).

Agreed. The SGP just sends the NTFY to those ASPs that are in INACTIVE =
or ACTIVE state. But it does not decide which ASP has to become ACTIVE =
or INACTIVE.

>=20
> RFC 3332
>   4.3.4.3 ASP Active Procedures
>    ...
>    For the n+k redundancy case, ASPs which are in that AS should
>    coordinate among themselves the number of active ASPs in=20
> the AS, and
>    should start sending traffic only after n ASPs are active.
>    ...=20
>=20

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



From exim@www1.ietf.org  Thu Jul 10 07:14:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA26731
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 07:14:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aZNC-0005IC-RW
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 07:14:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ABE2mH020335
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 07:14:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aZNA-0005Hj-TY; Thu, 10 Jul 2003 07:14:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aZN3-0005HP-Oc
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 07:13:54 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA26720
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 07:13:50 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aZN2-0001kp-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 07:13:52 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[t2Yoc4JkqBVg2eCDplfba8buf21L39qj])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aZN1-0001kl-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 07:13:51 -0400
Received: from ns.pigworks.openss7.net (IDENT:LolxyiN7UUEDT+dl0fdYAGtOZkFCpyZ/@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6ABDgX24200;
	Thu, 10 Jul 2003 05:13:42 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6ABDfR15155;
	Thu, 10 Jul 2003 05:13:41 -0600
Date: Thu, 10 Jul 2003 05:13:41 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
Cc: "'Zannin, Francois'" <francois.zannin@hp.com>,
        Ken Morneault <kmorneau@cisco.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA IG v04
Message-ID: <20030710051341.A14388@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
	"'Zannin, Francois'" <francois.zannin@hp.com>,
	Ken Morneault <kmorneau@cisco.com>, sigtran@ietf.org
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CED@eestqnt105.es.eu.ericsson.se>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CED@eestqnt105.es.eu.ericsson.se>; from j.javier.pastor@ericsson.com on Thu, Jul 10, 2003 at 12:54:14PM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6ABDgX24200
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

J.Javier,

J.Javier Pastor (EE/EEM) wrote:                                          =
         (Thu, 10 Jul 2003 12:54:14)
> Hi,=20
>=20
> Pls, see my answers inline..
>=20
> Cheers // JAvier.
>=20
> > -----Original Message-----
> > From: Zannin, Francois [mailto:francois.zannin@hp.com]
> > Sent: jueves, 10 de julio de 2003 11:34
> > To: J.Javier Pastor (EE/EEM); Ken Morneault; sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA IG v04
> >=20
> >=20
> > Javier,
> >=20
> > See comment inline,
> >=20
> > Regards,
> >=20
> > Francois Z.
> >=20
> > -----Original Message-----
> > From: J.Javier Pastor (EE/EEM) [mailto:j.javier.pastor@ericsson.com]=20
> > Sent: Thursday, July 10, 2003 10:49 AM
> > To: Zannin, Francois; Ken Morneault; sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA IG v04
> >=20
> >=20
> > Hi Francois,=20
> >=20
> > Pls see below....
> >=20
> > Cheers // Javier.
> >=20
> > > -----Original Message-----
> > > From: Zannin, Francois [mailto:francois.zannin@hp.com]
> > > Sent: mi=E9rcoles, 09 de julio de 2003 16:21
> > > To: Ken Morneault; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> > > Subject: RE: [Sigtran] M3UA IG v04
> > >=20
> > >=20
> > > Ken,
> > >=20
> > > My point of view on he question, see hereafter.
> > >=20
> > > Regards,
> > >=20
> > > Francois Z.
> > >=20
> > > -----Original Message-----
> > > From: Ken Morneault [mailto:kmorneau@cisco.com]
> > > Sent: Wednesday, July 09, 2003 3:56 PM
> > > To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> > > Subject: RE: [Sigtran] M3UA IG v04
> > >=20
> > >=20
> > > Francois,
> > >=20
> > > Pls see comments below.
> > >=20
> > > Regards,
> > >=20
> > > Ken
> > >=20
> > > At 10:15 AM 7/9/2003 +0200, Zannin, Francois wrote:
> > > >Just some comment about the M3UA IG v04
> > > >
> > > >3.21 NOTIFY messages are missing in Examples section
> > > >Figure (section 5.1.2)
> > > >        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2
> > > after the
> > > >ASP Up Ack
> > > >
> > > >Figure (section 5.1.3)
> > > >        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2
> > > after the
> > > >ASP Up Ack
> > > >
> > > >Figure (section 5.1.4)
> > > >        Same remark, the Notify is sent to ASP1, ASP2 and
> > > also need to
> > > >be sent to
> > > >      ASP3 after the ASP Up Ack
> > >=20
> >=20
> > # It is sent at that point because n=3D2 for that particular=20
> > example. I.e. when 2 ASPs reach the ASP-INACTIVE state, # the=20
> > AS is considered in AS-INACTIVE state.
> >=20
> > I can understand that SG(p) can hold the Notification while=20
> > there are not enough ASP UP to operate the minimum "n" ASP=20
> > potentially ready to "Activate" the AS. But if the AS is=20
> > still not Active when another ASP becomes "UP", a do not see=20
> > any reason not to send it the current state "inactive" for the AS.
> >=20
> The reason was to follow the RFC and it says that NTFY is sent when the=
 AS
> state changes. As n=3D2 for the general case AS becomes AS-INACTIVE aft=
er two
> ASP becomes INACTIVE.

Huh?  Section 4.3.4.1 ASP Up Procedures says:

   When an ASP Up message is received at an SGP and internally the
   remote ASP is in the ASP-DOWN state and not considered locked out for
   local management reasons, the SGP marks the remote ASP in the state
   ASP-INACTIVE and informs Layer Management with an M-ASP_Up indication
   primitive.

Also:

   AS-INACTIVE: The Application Server is available but no application
   traffic is active (i.e., one or more related ASPs are in the ASP-
   INACTIVE state, but none in the ASP-ACTIVE state).  The recovery
   timer T(r) is not running or has expired.
 =20
So the ASP-INACTIVE should be sent when the first ASP comes up.

Now I know that you propose to change this in 3.14 (which has not receive=
d
WG approval at all), but there is no reason to change this.  No traffic i=
s
flowing.

Also, if you follow what Francois has pointed to, the example should
look like this (if you are choosing to withhold AS-ACTIVE):

        SGP                 ASP1                ASP2                ASP3
          |                   |                   |                   |
          |<------ASP Up------|                   |                   |
          |-----ASP Up Ack--->|                   |                   |
          |                   |                   |                   |
          |NTFY(AS-INACTIVE)->|                   |                   |
          |                   |                   |                   |
          |<-------------------------ASP Up-------|                   |
          |------------------------ASP Up Ack---->|                   |
          |                   |                   |                   |
          |------------------NOTIFY(AS-INACTIVE)->|                   |
          |                   |                   |                   |
          |<--------------------------------------------ASP Up--------|
          |--------------------------------------------ASP Up Ack---->|
          |                   |                   |                   |
          |--------------------------------------NOTIFY(AS-INACTIVE)->|
          |                   |                   |                   |
          |                   |                   |                   |
          |<--ASP Act (Ldshr)-|                   |                   |
          |----ASP Act Ack--->|                   |                   |
          |                   |                   |                   |
          |NTFY(Insuf. ASPs)->|                   |                   |
          |--------------------NTFY(Insuf. ASPs)->|                   |
          |----------------------------------------NTFY(Insuf. ASPs)->|
          |                   |                   |                   |
          |<-------------------ASP Act. (Ldshr)---|                   |
          |----------------------ASP Act Ack----->|                   |
          |                   |                   |                   |
          |--NTFY(AS-ACTIVE)->|                   |                   |
          |--------------------NOTIFY(AS-ACTIVE)->|                   |
          |----------------------------------------NOTIFY(AS-ACTIVE)->|
          |                   |                   |                   |
          |                   |                   |                   |

If you are not choosing to withhold AS-ACTIVE (the normal case that you
are trying to change with 3.14) it should look like this:

        SGP                 ASP1                ASP2                ASP3
          |                   |                   |                   |
          |<------ASP Up------|                   |                   |
          |-----ASP Up Ack--->|                   |                   |
          |                   |                   |                   |
          |NTFY(AS-INACTIVE)->|                   |                   |
          |                   |                   |                   |
          |<-------------------------ASP Up-------|                   |
          |------------------------ASP Up Ack---->|                   |
          |                   |                   |                   |
          |------------------NOTIFY(AS-INACTIVE)->|                   |
          |                   |                   |                   |
          |<--------------------------------------------ASP Up--------|
          |--------------------------------------------ASP Up Ack---->|
          |                   |                   |                   |
          |--------------------------------------NOTIFY(AS-INACTIVE)->|
          |                   |                   |                   |
          |                   |                   |                   |
          |<--ASP Act (Ldshr)-|                   |                   |
          |----ASP Act Ack--->|                   |                   |
          |                   |                   |                   |
          |--NTFY(AS-ACTIVE)->|                   |                   |
          |NTFY(Insuf. ASPs)->|                   |                   |
          |                   |                   |                   |
          |--------------------NOTIFY(AS-ACTIVE)->|                   |
          |--------------------NTFY(Insuf. ASPs)->|                   |
          |                   |                   |                   |
          |----------------------------------------NOTIFY(AS-ACTIVE)->|
          |----------------------------------------NTFY(Insuf. ASPs)->|
          |                   |                   |                   |
          |<-------------------ASP Act. (Ldshr)---|                   |
          |----------------------ASP Act Ack----->|                   |
          |                   |                   |                   |
          |                   |                   |                   |
          |                   |                   |                   |

This begs the question of a "Sufficient ASPs" notification that Francois
mentions, that does not exist but was mentioned a number of times.

Greg originally drew 5.1.4 as activating at about the same time to avoid
the question of withholding AS-ACTIVE.  It was discussing 5.1.4 that made
us come to the determination that the SGP MAY withhold AS-ACTIVE.

--brian


> At that moment the AS changes state. According to the
> RFC the NTFY is sent to all the ASPs that are not ion DOWN state. This =
is
> what the example is showing.=20
>=20
> > If we look in the RFC, it is in charge of the ASP to=20
> > coordonate the activation of the AS. The SG cannot decide=20
> > which ASP have to send the ASP-Active for a particular AS (by=20
> > holding back the notify to other ASP).
>=20
> Agreed. The SGP just sends the NTFY to those ASPs that are in INACTIVE =
or
> ACTIVE state. But it does not decide which ASP has to become ACTIVE or
> INACTIVE.
>=20
> >=20
> > RFC 3332
> >   4.3.4.3 ASP Active Procedures
> >    ...
> >    For the n+k redundancy case, ASPs which are in that AS should
> >    coordinate among themselves the number of active ASPs in=20
> > the AS, and
> >    should start sending traffic only after n ASPs are active.
> >    ...=20
> >=20
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--=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



From exim@www1.ietf.org  Thu Jul 10 07:17:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA26862
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 07:17:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aZQ5-0005U5-Ln
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 07:17:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ABH1ZT021077
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 07:17:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aZQ5-0005Tr-Im; Thu, 10 Jul 2003 07:17:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aZPQ-0005QG-Sd
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 07:16:20 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA26833
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 07:16:18 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aZPQ-0001mT-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 07:16:20 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[UxgLj7ipwsTrKPNE2qyg/ZQkz4cCNOES])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aZPP-0001mQ-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 07:16:19 -0400
Received: from ns.pigworks.openss7.net (IDENT:3oaSyK3kttZ6YqWM98vG2qGmlYtZMt7A@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6ABGHX24240;
	Thu, 10 Jul 2003 05:16:17 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6ABGHu15202;
	Thu, 10 Jul 2003 05:16:17 -0600
Date: Thu, 10 Jul 2003 05:16:16 -0600
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>,
        Ken Morneault <kmorneau@cisco.com>,
        "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
        sigtran@ietf.org
Subject: Re: [Sigtran] M3UA IG v04
Message-ID: <20030710051616.B14388@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>,
	Ken Morneault <kmorneau@cisco.com>,
	"J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
	sigtran@ietf.org
References: <06CAA4D3753C53408D2CE6340B68444201488A9C@vbeexc02.emea.cpqcorp.net> <Pine.SOL.4.10.10307100453470.25952-100000@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.4.10.10307100453470.25952-100000@sun2594.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Jul 10, 2003 at 05:36:24AM -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>

Samuel,

Are you meaning the "Sufficient ASPs" message that I mentioned some
time ago?

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 10 Jul 2003 05:36:24)
> 
> Zannin,
> 
>   Just an example, say an operator configures an AS1 as set of 3 ASPes.
>   They are namely known as ASP1,ASP2 and ASP3(backup).
> 
>   Initially ASP1 and ASP2 send ASP_ACTIVE to SGP  so SGP sends a
>   NTFY(AS_ACTIVE) thus results in AS1 to be in ACTIVE state.
> 
>   Where is the question of sending NTFY(Sufficient ASPs active)?
>   People mean to say that backup ASP also sends an ASP_ACTIVE message
>   to SGP? if it so, on what condition it is being sent? Is there
>   any valid conditions that backup ASP or some new(!!!??!amazing!!) ASP
>   to send ASP_ACTIVE message? 
> 
> 
> --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, 9 Jul 2003, Zannin, Francois wrote:
> 
> > Ken,
> > 
> > Here is a copy of the mail about notify I was refering.
> > 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> > Sent: Wednesday, February 12, 2003 10:40 AM
> > To: ksrinivas
> > Cc: sigtran@ietf.org
> > Subject: [Sigtran] Re: M3UA Notify message
> > 
> > 
> > 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
> > -----Original Message-----
> > From: Ken Morneault [mailto:kmorneau@cisco.com] 
> > Sent: Wednesday, July 09, 2003 4:28 PM
> > To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA IG v04
> > 
> > 
> > 
> > 
> > Francois,
> > 
> > Can you send me a ptr to the email you reference.
> > 
> > Thanks,
> > 
> > Ken
> > 
> > At 04:20 PM 7/9/2003 +0200, Zannin, Francois wrote:
> > >Ken,
> > >
> > >My point of view on he question, see hereafter.
> > >
> > >Regards,
> > >
> > >Francois Z.
> > >
> > >-----Original Message-----
> > >From: Ken Morneault [mailto:kmorneau@cisco.com]
> > >Sent: Wednesday, July 09, 2003 3:56 PM
> > >To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> > >Subject: RE: [Sigtran] M3UA IG v04
> > >
> > >
> > >Francois,
> > >
> > >Pls see comments below.
> > >
> > >Regards,
> > >
> > >Ken
> > >
> > >At 10:15 AM 7/9/2003 +0200, Zannin, Francois wrote:
> > >>Just some comment about the M3UA IG v04
> > >>
> > >>3.21 NOTIFY messages are missing in Examples section
> > >>Figure (section 5.1.2)
> > >>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the
> > >>ASP Up Ack
> > >>
> > >>Figure (section 5.1.3)
> > >>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the
> > >>ASP Up Ack
> > >>
> > >>Figure (section 5.1.4)
> > >>        Same remark, the Notify is sent to ASP1, ASP2 and also need to
> > >>be sent to
> > >>      ASP3 after the ASP Up Ack
> > >
> > >NOTIFY is only sent upon "a change in status of a particular 
> > >Application Server".
> > >
> > >OK, but just one more remark, in section 5.1.4, a notify AS-Inactive is
> > 
> > >sent to ASP1 and ASP2 while to transition to ASP-Inactive was effective
> > 
> > >after the first
> > >ASP_UP(/ACK) comming from ASP1, so notify to ASP2 should be considerer
> > >as inopportune.
> > >
> > >In last november, it was agreed in a thread that when an ASP was 
> > >transitionned to UP, The current AS state was sent to this ASP, 
> > >whatever could be the state. This current
> > >State can be used by the management of an ASP to decide if the AS has
> > to
> > >be activated 
> > >Immediately or must wait for an AS-PENDING or AS-INACTIVE notification.
> > 
> > 
> > _______________________________________________
> > 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 exim@www1.ietf.org  Thu Jul 10 07:27:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA27046
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 07:27:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aZZm-0006Im-BJ
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 07:27:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ABR2K6024217
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 07:27:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aZZl-0006IU-Fd; Thu, 10 Jul 2003 07:27:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aZYz-0006HQ-9V
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 07:26:13 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA27025
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 07:26:10 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aZYy-0001pO-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 07:26:12 -0400
Received: from auds951.usa.alcatel.com ([143.209.238.80])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aZYx-0001oy-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 07:26:12 -0400
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6ABPLwA021423;
	Thu, 10 Jul 2003 06:25:21 -0500 (CDT)
Received: from sun2594.ssd.usa.alcatel.com (sun2594.ssd.usa.alcatel.com [143.209.250.252])
	by ssd.usa.alcatel.com (8.12.8/8.12.8) with ESMTP id h6ABOZ1C016057;
	Thu, 10 Jul 2003 06:24:35 -0500 (CDT)
Date: Thu, 10 Jul 2003 06:24:35 -0500 (CDT)
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>,
        Ken Morneault <kmorneau@cisco.com>,
        "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
        sigtran@ietf.org
Subject: Re: [Sigtran] M3UA IG v04
In-Reply-To: <20030710051616.B14388@openss7.org>
Message-ID: <Pine.SOL.4.10.10307100622480.25952-100000@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,

 Yes,i meant to say that NTFY(Sufficient ASPs active) is not required.
 (hope this is the main discussion)

--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, 10 Jul 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Are you meaning the "Sufficient ASPs" message that I mentioned some
> time ago?
> 
> --brian
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 10 Jul 2003 05:36:24)
> > 
> > Zannin,
> > 
> >   Just an example, say an operator configures an AS1 as set of 3 ASPes.
> >   They are namely known as ASP1,ASP2 and ASP3(backup).
> > 
> >   Initially ASP1 and ASP2 send ASP_ACTIVE to SGP  so SGP sends a
> >   NTFY(AS_ACTIVE) thus results in AS1 to be in ACTIVE state.
> > 
> >   Where is the question of sending NTFY(Sufficient ASPs active)?
> >   People mean to say that backup ASP also sends an ASP_ACTIVE message
> >   to SGP? if it so, on what condition it is being sent? Is there
> >   any valid conditions that backup ASP or some new(!!!??!amazing!!) ASP
> >   to send ASP_ACTIVE message? 
> > 
> > 
> > --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, 9 Jul 2003, Zannin, Francois wrote:
> > 
> > > Ken,
> > > 
> > > Here is a copy of the mail about notify I was refering.
> > > 
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> > > Sent: Wednesday, February 12, 2003 10:40 AM
> > > To: ksrinivas
> > > Cc: sigtran@ietf.org
> > > Subject: [Sigtran] Re: M3UA Notify message
> > > 
> > > 
> > > 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
> > > -----Original Message-----
> > > From: Ken Morneault [mailto:kmorneau@cisco.com] 
> > > Sent: Wednesday, July 09, 2003 4:28 PM
> > > To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> > > Subject: RE: [Sigtran] M3UA IG v04
> > > 
> > > 
> > > 
> > > 
> > > Francois,
> > > 
> > > Can you send me a ptr to the email you reference.
> > > 
> > > Thanks,
> > > 
> > > Ken
> > > 
> > > At 04:20 PM 7/9/2003 +0200, Zannin, Francois wrote:
> > > >Ken,
> > > >
> > > >My point of view on he question, see hereafter.
> > > >
> > > >Regards,
> > > >
> > > >Francois Z.
> > > >
> > > >-----Original Message-----
> > > >From: Ken Morneault [mailto:kmorneau@cisco.com]
> > > >Sent: Wednesday, July 09, 2003 3:56 PM
> > > >To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> > > >Subject: RE: [Sigtran] M3UA IG v04
> > > >
> > > >
> > > >Francois,
> > > >
> > > >Pls see comments below.
> > > >
> > > >Regards,
> > > >
> > > >Ken
> > > >
> > > >At 10:15 AM 7/9/2003 +0200, Zannin, Francois wrote:
> > > >>Just some comment about the M3UA IG v04
> > > >>
> > > >>3.21 NOTIFY messages are missing in Examples section
> > > >>Figure (section 5.1.2)
> > > >>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the
> > > >>ASP Up Ack
> > > >>
> > > >>Figure (section 5.1.3)
> > > >>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the
> > > >>ASP Up Ack
> > > >>
> > > >>Figure (section 5.1.4)
> > > >>        Same remark, the Notify is sent to ASP1, ASP2 and also need to
> > > >>be sent to
> > > >>      ASP3 after the ASP Up Ack
> > > >
> > > >NOTIFY is only sent upon "a change in status of a particular 
> > > >Application Server".
> > > >
> > > >OK, but just one more remark, in section 5.1.4, a notify AS-Inactive is
> > > 
> > > >sent to ASP1 and ASP2 while to transition to ASP-Inactive was effective
> > > 
> > > >after the first
> > > >ASP_UP(/ACK) comming from ASP1, so notify to ASP2 should be considerer
> > > >as inopportune.
> > > >
> > > >In last november, it was agreed in a thread that when an ASP was 
> > > >transitionned to UP, The current AS state was sent to this ASP, 
> > > >whatever could be the state. This current
> > > >State can be used by the management of an ASP to decide if the AS has
> > > to
> > > >be activated 
> > > >Immediately or must wait for an AS-PENDING or AS-INACTIVE notification.
> > > 
> > > 
> > > _______________________________________________
> > > 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 exim@www1.ietf.org  Thu Jul 10 07:37:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA27301
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 07:37:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aZjT-000700-G8
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 07:37:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ABb3lo026893
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 07:37:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aZjS-0006yx-7p; Thu, 10 Jul 2003 07:37:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aZj5-0006uz-Hc
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 07:36:39 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA27261
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 07:36:36 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aZj4-0001sJ-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 07:36:38 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[IDyra1nGHamegImneUaT0AmOVY4HEWDe])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aZj2-0001sG-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 07:36:36 -0400
Received: from ns.pigworks.openss7.net (IDENT:YgW0nywA2Ig3rfkqXVxo1WFbUOcGXgcX@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6ABaYX24671;
	Thu, 10 Jul 2003 05:36:34 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6ABaXh15581;
	Thu, 10 Jul 2003 05:36:33 -0600
Date: Thu, 10 Jul 2003 05:36:33 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04
Message-ID: <20030710053633.C14388@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CCA@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: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CCA@eestqnt105.es.eu.ericsson.se>; from j.javier.pastor@ericsson.com on Thu, Jul 03, 2003 at 11:45:57AM +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>

Javier,

Your "new possibility" is the norm in the RFC.  You seem to want reverse
the sense of the RFC.  The RFC says that we activate on the 1st ASP and
MAY withhold until the 'n'th ASP.  3.14 says that we SHOULD withhold to
the 'n'th ASP and MAY activate on the first.

The reason for this change escapes me.  There is no justification to
reverse the option.

Please, answer me this:

  Who has an ASP that is so broken that it needs the SGP to wait
  for 'n' of them to become active?

  Please identify yourselves!


There are several other places in the RFC that assume activation on the
first ASP.  It is sewn through and through the RFC that loadsharing and
broadcast is performed across the active APSs (regardless of n).

Section 3.7.1:
                                                  ...  In Loadshare
      mode, the ASP will share in the traffic distribution with any
      other currently active ASPs.  In Broadcast mode, the ASP will
      receive the same messages as any other currently active ASP.

(Notice the word "any", twice, above.)

Section 4.1.1

   From the list of ASPs within the AS table, an ASP in the ASP-ACTIVE
   state is selected and a DATA message is constructed and issued on the
   corresponding SCTP association.  If more than one ASP is in the ASP-
   ACTIVE state (i.e., traffic is to be loadshared across more than one
   ASP), one of the ASPs in the ASP-ACTIVE state is selected from the
   list.  If the ASPs are in Broadcast Mode, all active ASPs will be
   selected and the message sent to each of the active ASPs.

(Note above that there is no 'n' restriction on selecting ASPs to receive
traffic.)

Section 4.3.4.3:

   In the case of a Loadshare mode AS, reception of an ASP Active
   message at an SGP or IPSP causes the direction of traffic to the ASP
   sending the ASP Active message, in addition to all the other ASPs
   that are currently active in the AS.

Section 4.3.4.4:

   In the case of a Loadshare mode AS, the SGP moves the ASP to the
   ASP-INACTIVE state and the AS traffic is reallocated across the
   remaining ASPs in the state ASP-ACTIVE, as per the loadsharing


Also, note that there is no limit to T(r).  So, if it was set infinite,
or just rather large, the start case is transformed to the restart case.


Changing the trantion from the AS-DOWN to AS-INACTIVE state at the SGP
to wait for 'n' ASP-INACTIVE ASPs is even more pointless than waiting
for 'n' ASP-ACTIVE ASPs.  No traffic is flowing.  The RFC always has us
move to the AS-INACTIVE state on the first ASP-INACTIVE ASP.

What in the world did you think would necessitate changing this?

--brian


Javier Pastor-Balbas (ECE) wrote:                                                   (Thu, 03 Jul 2003 11:45:57)
> 
> Hi guys,
> 
> We tried to submit the next IG version before the deadline, but I got
> confused since I thought it ended at 12:00 instead of 9:00. So the server
> bounced it. I'll submit the doc again after the IETF meeting. In the
> meantime you can find it at:
> 
> http://standards.ericsson.net/javier/draft-ietf-sigtran-m3ua-implementors-guide-04.txt
> 
> We've tried to include all the different opinions in here. Although it seems
> that the case of waiting till n ASPs is the preferred one, we have included
> the new possibility of a smooth start for those of you who want to activate
> the AS when the first ASP gets to ACTIVE. At the end it will be the
> operator's decission how to proceed during the activation phase. Hopefully
> everyone agrees on this wording...
> 
> Cheers // Javier.
> 
> 
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: martes, 01 de julio de 2003 20:18
> > To: Samuel Dur D. Jeyaseelan
> > Cc: sigtran@ietf.org
> > Subject: Re: [Sigtran] Re:n+k issue
> > 
> > 
> > Samuel,
> > 
> > Perhaps you could answer my questions below first.  Please be
> > specific.
> > 
> > --brian
> > 
> > Samuel Dur D. Jeyaseelan wrote:                               
> >                       (Tue, 01 Jul 2003 06:09:34)
> > > 
> > > Brian,
> > > 
> > >   Can you bit elaborate your n=256 example (ANSI-SLS) with 
> > real-time 
> > >   scenario?(please write to the point)
> > > 
> > > --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, 1 Jul 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Samuel,
> > > > 
> > > > Samuel Dur D. Jeyaseelan wrote:                      
> > (Mon, 30 Jun 2003 09:44:01)
> > > > > 
> > > > > Brian,
> > > > > 
> > > > > On Sat, 28 Jun 2003, Brian F. G. Bidulock wrote:
> > > > > 
> > > > > > Samuel,
> > > > > > 
> > > > > > I should point out that the proposal to require that 
> > the SG wait for 'n' to
> > > > > > start traffic does not have much merit.
> > > > > > 
> > > > > > Example: n=256, k=1
> > > > > > 
> > > > > > 2 ASPs fail, permanently and need to be manually replaced.
> > > > > > 
> > > > > > A temporary network failure causes SCTP associations 
> > to be lost to the
> > > > > > remaining 255 ASPs.  They move to the ASP down state 
> > and immediately
> > > > > > re-establish SCTP associations and move to the 
> > ASP-Active state.  Because the
> > > > > > number of ASPs is < n, the SG will not restart 
> > traffic for some time.
> > > > > 
> > > > > Your example leads to the following figure.
> > > > > 
> > > > >          SG1                            AS1
> > > > >      
> > > > >  SS7 MSG +------+----------------------- ASP (1)
> > > > > ---------| SGP1 |----------------------- ASP (2)
> > > > >          |      |----------------------- ....
> > > > >          +------+----------------------- ASP (256)
> > > > >             |
> > > > >             | 
> > > > >             +---------------------------ASP(k)
> > > > > 
> > > > > 1. First of all craft will not configure 1 ASP as 
> > backup for 256 Active
> > > > >    ASPs, if he does, it means to be a poor network 
> > configuration.
> > > > 
> > > > Well, first of all, craft does not normally choose the 
> > number of ASPs used for
> > > > backup, engineering does.  Second, k=1 is more than 
> > adequate for n=256 where
> > > > the ASPs are reasonably reliable and have a short 
> > replacement time.
> > > > 
> > > > Regardless, I followed with the same example with n=2 and k=1.
> > > > 
> > > > > 
> > > > > 2. Consider ASP as an MGC, your example leads to 256 
> > MGCs and they are 
> > > > >    controlled by a single SGP.A good network will have 
> > multiple SGPs to
> > > > >    share the POTENTIAL traffic across ASPs.
> > > > 
> > > > You seem to misunderstand the concept of redundancy in 
> > M3UA: there are not
> > > > multiple SGPs or ASPs because of traffic load.  It is 
> > easy enough to handle
> > > > the traffic of a small to medium sized country through a 
> > single processor.
> > > > 
> > > > Also, you seem to misunderstand that an SGP and an ASP 
> > are not processors.
> > > > They are processes.  One host processor could have many 
> > (even 256) ASPs or
> > > > SGPs (or both for that matter).
> > > > 
> > > > > 
> > > > > 3. BASIC INTENTION OF 'n+k' is,just to make sure that 
> > incoming messages
> > > > >    ARE EFFICIENTLY TRANSMITTED TO DESTINATION ASPs WITH 
> > THE HELP OF
> > > > >    MINIMUM NUMBER OF ASSOCIATIONS.
> > > > 
> > > > No.  The intention of 'n+k' is to provide one of the more 
> > recently accepted
> > > > models for reliability using redundancy (n = m+1, k=0 
> > being the typical system
> > > > deployed in the field today, m usually being greater than 
> > 2).  Up until
> > > > recently, only 2m (or m+m) systems were considered 
> > carrier class.  Some
> > > > systems were 4m (with 3 out of 4 vote).
> > > > 
> > > > You should take some time to look at redundancy and 
> > reliability requirements
> > > > for carrier grade systems for an understanding of how to 
> > calculate and provide
> > > > the reliability requirements expected by carriers.  There 
> > are good standards
> > > > written on the topic.  See the Telcordia NEBS and 
> > reliability specs, for
> > > > example.  If you have an RBOC RFP handy, take a look at 
> > the hardware and
> > > > software reliability requirements sections or talk to the 
> > engineer in your
> > > > company responsible for answering those sections.
> > > > 
> > > > > 
> > > > >    Say for example
> > > > > 
> > > > >    Inefficient configuration -just an example
> > > > >    ------------------------------------------
> > > > >   
> > > > >             SLS         DEST
> > > > >             =================
> > > > >              0           ASP1
> > > > >              1           ASP1
> > > > >              2           ASP1
> > > > >              3           ASP1
> > > > >              4           ASP1
> > > > >              5           ASP1
> > > > >              6           ASP1
> > > > >              .            .
> > > > >              .            .
> > > > >              15          ASP1
> > > > >              
> > > > >              
> > > > >                SG                            AS1
> > > > > sls(0,1..15 )+----+  sls(0,1..15)         +------+
> > > > > -------------|SGP1|-----------------------| ASP1 |
> > > > >              |    |              ASSOC-1  +------+
> > > > >              +----+                     
> > > > > 
> > > > >                       Fig - 1
> > > > > 
> > > > > In this figure, all incoming messages are transmitted 
> > over ASSOC-1 towards
> > > > > ASP1.Here there are two potential problems 
> > > > > 
> > > > >        1. Total incoming traffic is dependent on a single ASP
> > > > >        2. ASP1 is always overloaded, messages are 
> > processed one by one
> > > > >           from the transmission buffer which ultimately 
> > introduces time
> > > > >           delay(example last arrived msg will be 
> > processed at the end).
> > > > 
> > > > The above was not in my examples.  My examples had n > 1.
> > > > 
> > > > > 
> > > > > 
> > > > > Efficient configuration
> > > > > -----------------------  
> > > > > 
> > > > >              SLS         DEST
> > > > >             =================
> > > > >              0           ASP1
> > > > >              1           ASP1
> > > > >              2           ASP1
> > > > >              3           ASP1
> > > > >              4           ASP1
> > > > >              5           ASP1
> > > > >              6           ASP1
> > > > >              7           ASP1
> > > > >              8           ASP2
> > > > >              9           ASP2
> > > > >              .            .
> > > > >              .            .
> > > > >              15          ASP2
> > > > > 
> > > > >                
> > > > >                        
> > > > >                        
> > > > >               SG                             AS1
> > > > > sls(0,1..15 )+----+   sls(0,1,..7)ASSOC-1 +------+
> > > > > -------------|SGP1|-----------------------| ASP1 |
> > > > >              |    |--------+              +------+
> > > > >              +----+        | sls(8,9..15) +------+
> > > > >                            +--------------| ASP2 |
> > > > >                                  ASSOC-2  +------+
> > > > >                         
> > > > >                    fig - 2
> > > > > 
> > > > > In this example incoming traffic is distributed between 
> > ASP1 and ASP2.
> > > > > 
> > > > > IN ORDER TO HANDLE THE HEAVY INCOMING TRAFFIC, SGP1 
> > NEEDS MINIMUM OF
> > > > > 2 ACTIVE ASPs BY SETTING n=2.
> > > > 
> > > > Do they not comply to SS7 ISUP/SCCP flow and congestion 
> > controls?  Are they
> > > > running at full capacity in this arrangement?  Are these 
> > your ASPs?  Do you
> > > > design your systems this way?
> > > > 
> > > > Take a look at the reliability engineering.  If either of 
> > these ASPs is
> > > > handling its maximum rated capacity, it will have to be 
> > reliable to .99999
> > > > This is because, if one ASP fails, 50% of the system has 
> > failed.  A 50% system
> > > > failure is deemed a "system" failure rather than a 
> > "component" failure in
> > > > reliability engineering for carrier grade systems.  
> > Considering proscribed
> > > > replacement times on the ASP, this would be difficult to 
> > achieve unless both
> > > > ASPs were hardware fault tolerant platforms.
> > > > 
> > > > Nevertheless, the RFC defines ASPs as a process, not a 
> > processor.  There is
> > > > nothing restricting the two ASPs in your diagram from 
> > being processes on the
> > > > same processor.  In that case, when ASP2 is down, ASP1 
> > has double the
> > > > processing capacity as it had before.
> > > > 
> > > > You certainly cannot make statements as you have made here without
> > > > presupposing an implementation architecture outside the 
> > scope of the standard.
> > > > 
> > > > Besides, the RFC permits you to configure your SGP (and 
> > why do you only have
> > > > one of them?) so that it waits for 'n' ASPs to become 
> > active.  You MAY do
> > > > this.  Just don't require everybody else to follow your 
> > approach.  (Actually,
> > > > if the above is your approach, don't require anyone else 
> > to follow it.)
> > > > 
> > > > Several other points: ANSI T1.111.4/2000 SLS values are 
> > 8-bits.  That means
> > > > there are 256 SLS values (gee, I had 256 ASPs in my example).
> > > > 
> > > > For another thing, sequenced traffic flows are not only 
> > based on SLS.  In an
> > > > ITU-T network, each remote signalling point has 16 
> > independent traffic flows
> > > > for each AS.  The traffic flows from each remote 
> > signalling point is
> > > > independent from the traffic flows of other remote 
> > signalling points.
> > > > Therefore, if there are 400 remote signalling points, 
> > there will be 6400
> > > > independent traffic flows that the SG can use for 
> > distribution of AS traffic
> > > > for one AS over as many as 6400 ASPs without risking 
> > mis-sequencing of
> > > > messages.  For an ANSI T1.111.4/2000 network using 8-bit 
> > SLS values with 400
> > > > remote signalling points, there would be 102400 
> > independent traffic flows to
> > > > the AS that the SG can use to distribute traffic to one 
> > AS over as many as
> > > > 102400 ASPs without risking message mis-sequencing.
> > > > 
> > > > > 
> > > > > 
> > > > > Your example with n=256 is completely a wrong 
> > interpretaion of 'n+k'
> > > > > redundancy case.
> > > > 
> > > > Quite the opposite, you clearly misunderstand redundancy 
> > and reliability in
> > > > carrier grade systems and obviously have little 
> > experience with them.
> > > > 
> > > > Here's a little primer for you:
> > > > 
> > > > Carrier grade systems were originally designed as 
> > hot-standby, fully redundant
> > > > systems.  For every processor actively handling traffic, 
> > there was a hot
> > > > standby processor also actively handling the same 
> > traffic.  This typically
> > > > required special bus arrangements whereby the I/O bus of 
> > one processor was
> > > > actively controlling peripheral devices, while the hot 
> > standby processor was
> > > > issuing I/O bus output that was not terminated to the I/O 
> > bus.  The bus also
> > > > normally performed a comparison of the outputs from the 
> > active side and the
> > > > outputs from the inactive side.  Peripheral components, 
> > such as switch fabric
> > > > were also fully redundant (duplicated).  Switching 
> > fabrics are separated into
> > > > an A-plane and a B-plane in much the same fashion as the 
> > SPCS front-end I/O
> > > > bus.
> > > > 
> > > > Note that active/standby in M3UA does not address the 
> > above model.  2+0
> > > > broadcast does.
> > > > 
> > > > The problem with such a systems was what we used to call 
> > the "kill your
> > > > sister" syndrome.  A failure on the inactive side only 
> > detectable to the I/O
> > > > bus could cause a switch from the active to inactive 
> > side, resulting in the
> > > > switch running on the erroneous inactive side.  We called 
> > it the "kill your
> > > > sister" syndrome because it is not so surprising that the 
> > psychotic sister
> > > > (erred inactive side) kills the sane sister (correct active side).
> > > > 
> > > > Primarily to avoid the "kill your sister" syndrome, some 
> > switches were
> > > > designed with 3 or 4 processors on an cross-checking I/O 
> > bus pair.  In that
> > > > case, it would take a 2 out of 3 or 3 out of 4 vote to 
> > shift from the active
> > > > processor.  These systems were far less susceptible to 
> > inactive plane errors
> > > > that were only detected at the I/O bus.
> > > > 
> > > > Note that the only M3UA model that address the above is 
> > the 3+0 or 4+0
> > > > broadcast model.
> > > > 
> > > > When the use of general purpose computing platforms for 
> > telecommunications
> > > > systems started to become popular, there were several 
> > models that emerged.
> > > > 
> > > > First was taking two big iron UNIX hosts and running them 
> > in active/standby.
> > > > In this scenario, one UNIX host would be actively 
> > handling traffic and the
> > > > other would be warm standby.  By "warm" standby, it means 
> > that the device was
> > > > powered and was synchronizing state on a checkpoint 
> > interval or monitoring the
> > > > health of the active hosts with heartbeats.  Shared 
> > memory, shared DASD or
> > > > message passing are the most common mechanisms of 
> > redundancy.  These systems
> > > > are considered "highly available" systems rather than the 
> > "fault tolerant"
> > > > systems used in telecommunications above.  An increased 
> > level of reliability
> > > > is often achieved by using hardware fault tolerant 
> > multiple processor hosts.
> > > > 
> > > > In M3UA, the active/standby model addresses these 
> > systems.  Although M3UA
> > > > provides for a '1+k' model, k is invariably 1.
> > > > 
> > > > The major difficulties with these systems are: a) 
> > switchover latency between
> > > > systems; b) loss of state during switchover; and c) 
> > scalability.  Switchover
> > > > latency is normally improved by decreasing heartbeat 
> > intervals.  State
> > > > synchronization is normally improved by decreasing the 
> > process checkpoint
> > > > interval.  These reductions typically increase cost and 
> > they are normally
> > > > chosen to minimize a cost/performance/reliability metric. 
> >  Scalability was
> > > > typically performed by installing larger more powerful 
> > hosts with increased
> > > > demand.  This was not good, because larger processors 
> > tended to have poorer
> > > > cost/performance ratios that smaller processors.
> > > > 
> > > > Later it was determined that the cost of a complete 
> > redundant host that was
> > > > wasting CPU cycles while inactive could be avoided while 
> > addressing the
> > > > deficiencies of the paired host approach above.  To increase state
> > > > synchronization, state could be spread over a number of 
> > hosts.  If there were
> > > > 'm' hosts, the failure of one host could at worst affect 
> > a proportion, 1/m, of
> > > > the state.  Therefore state synchronization would improve 
> > as m increased.  To
> > > > distribute the state, traffic load would be shared over 
> > the m hosts according
> > > > to state.  This also served the needs of scalability and 
> > provided for closer
> > > > to linear cost scalability, because the number of hosts, 
> > m, could be increased
> > > > while holding the cost/performance ratio of the hosts 
> > constant.  The costs of
> > > > redundancy could also be reduced over the fully redundant 
> > systems described
> > > > above, because redundancy could be achieved using 'm+1' 
> > processors, where 'm'
> > > > processors were required to handle the maximum engineered 
> > load.  For systems
> > > > as small as 'm=2' hosts with equivalent reliability to 
> > that normally
> > > > experienced in the '1+1' active/standby scenario above, 
> > system reliability
> > > > figures could be maintained with less redundancy cost.  
> > It is important to
> > > > note, that in these systems, all 'm+1' hosts are active 
> > and handling only a
> > > > proportion m/(m+1) of their rated capacity.  It is very 
> > seldom that the rated
> > > > capacity was more that 2/3 of the maximum short-term capacity.
> > > > 
> > > > These systems do not solve the heartbeat interval problem of the
> > > > active/standby paired hosts system above.  Typically what 
> > these systems use is
> > > > more sophisticated watchdog approaches rather than 
> > passing heartbeat messages
> > > > between processes on the active and standby hosts.  
> > Memory sharing, DASD
> > > > sharing, message passing, reliable broadcast, voting, and 
> > a plethora of
> > > > reliable distributed systems techniques have been 
> > developed that are useful in
> > > > these systems.
> > > > 
> > > > Adoption of these latter systems changed the requirement 
> > from "duplicated"
> > > > systems to systems with no "single point of failure".
> > > > 
> > > > Now, the 'n+k' load-sharing and broadcast model was 
> > intended to address these
> > > > systems.  Where 'n == m' in the description above.  That 
> > is why we agreed that
> > > > n is the maximum engineered load.  Because that is what 
> > 'm' is.  The k is the
> > > > 1 in the description above, generalized.
> > > > 
> > > > First off, you must understand that it is not the 
> > intention of the model to
> > > > have 'n' ASPs active and 'k' inactive unless 'n=1, k=1'.  
> > This is described in
> > > > section 1.4.4.1:
> > > > 
> > > >   A "1+1" active/standby redundancy is a subset of this model.
> > > > 
> > > > Also no redundancy is possible:
> > > > 
> > > >   A simplex "1+0" model is also supported as a subset, 
> > with no ASP redundancy.
> > > > 
> > > > (The term "simplex" come from the telephone switches 
> > described above where the
> > > > inactive side is taken off-line.  Such a situation is 
> > described as running the
> > > > switch "simplex".)
> > > > 
> > > > The 'n+k' model is to address the 'm+1' distributed 
> > systems described last.
> > > > Rereading section 1.4.4.1:
> > > > 
> > > >   The failover model supports an "n+k" redundancy model, 
> > where "n" APSs is the
> > > >   minimum number of redundant [sic] ASPs required to 
> > handle traffic and "k"
> > > >   ASPs are available to take over for a failed or unavailable ASP.
> > > > 
> > > > Note that the state of the "k" ASPs with respect to an AS 
> > is not stated.  In
> > > > the 'm+1' redundant system described above, the "k" ASPs 
> > are active for the
> > > > AS.  "k" ASPs active in load sharing mode are "available 
> > to take over for a
> > > > failed or unavailable ASP".
> > > > 
> > > > It is typical to run the "k" ASPs active (and I know you 
> > had some questions
> > > > about activating the "k" ASPs, but the point is that they 
> > are normally already
> > > > active), and handling their own share of traffic.  In 
> > this case, each ASPs
> > > > handles n/(n+k) of their engineered capacity.  Were you 
> > actually considering
> > > > deploying such a atypical model where the 'k' ASPs are inactive?
> > > > 
> > > > "k" ASPs can be taken out of service, upgraded, and 
> > brought back into service
> > > > without affecting the maximum engineering traffic 
> > handling capacity of the
> > > > system (in absence of failures during the upgrade).  This 
> > assists in graceful
> > > > software/hardware upgrades on live systems.  (Keeping 
> > them inactive would
> > > > serve no purpose and would, decrease maximum traffic 
> > handling capability,
> > > > increase loss of state on fail-over, and increase 
> > fail-over latency: all good 
> > > > reasons for not deactivating the 'k' ASPs).
> > > > 
> > > > Now, reliability engineering for carrier class systems 
> > takes the view that if
> > > > 50% of the load bearing components in a system fail (stop 
> > operating in any
> > > > way), then the entire system has failed.  If traffic is 
> > stopped above n/2
> > > > processes, the system will have failed prematurely.
> > > > 
> > > > A requirement to withhold traffic < n will always suffer 
> > a prolonged system
> > > > failure, because only n/2 are required to make the system 
> > available.
> > > > Therefore, on startup the reliability of the system is 
> > degraded by the
> > > > negative exponent n/2, regardless of whether n=256 or 
> > n=2; however, increasing
> > > > 'n' increases the damage done to reliability by waiting for 'n'.
> > > > 
> > > > A requirement for the SG to wait for 'n' is unacceptable 
> > from the standpoint
> > > > of the redundancy model that 'n+k' was meant to address, 
> > as well as other
> > > > possible models that it can accommodate.
> > > > 
> > > > > 
> > > > > 
> > > > > > 
> > > > > > Consider the same example with: n=2, k=1, two ASPs failed.
> > > > > > 
> > > > > > Any system that introduces hysteresis to aggravate 
> > and perpetuate an outage in
> > > > > > this fashion is not very useful.
> > > > > > 
> > > > > > In the examples, what would have been a brief 
> > AS-PENDING situation has been
> > > > > > turned into a prolonged and complete system failure.  
> > It is certainly not good
> > > > > > practice to design redundant systems in that fashion 
> > and recommending or
> > > > > > requiring such a design would be a bad thing.
> > > > > > 
> > > > > > To further the example, consider the multiple SG as 
> > STP situation.  One SG
> > > > > > cannot know the state of ASPs for the its mated SG 
> > and therefore cannot
> > > > > > distinguish between the restart scenario and the 
> > start scenario as far as the
> > > > > > AS state is concerned because the SG does not know 
> > the state of the AS itself:
> > > > > > the AS could be active with other SGs.
> > > > > > 
> > > > > > This is the same case for SGP.  Unless AS state is 
> > synchronized across SGPs,
> > > > > > one SGP cannot know the total state of the AS (only 
> > its partition).  There is
> > > > > > no requirement to synchronize AS state across SGP (on 
> > the SGP side, AS-state
> > > > > > is per-SGP), and there is certainly no requirement to 
> > synchronize AS-state
> > > > > > across SGs.
> > > > > > 
> > > > > > So, take the (n=2, k=0) example where there are two 
> > SGs.  Due to a network
> > > > > > partitioning at startup, one ASP can communicate only 
> > with one SG, and the
> > > > > > other ASP can communicate only with the other SG.  
> > Because n=2, neither SG
> > > > > > will start traffic and the system suffers 100% 
> > failure until the network
> > > > > > partitioning ceases.
> > > > > > 
> > > > > > Consider the same example where there are two SGPs 
> > (instead of SGs) and the
> > > > > > two ASPs normally only form one association (e.g., 
> > ASP1-SGP1, ASP2-SGP2).
> > > > > > Note that there is no requirement in the RFC to form 
> > full-mesh associations.
> > > > > > Because the number of active ASPs with each SGP is < 
> > 2, the system will never
> > > > > > start.
> > > > > > 
> > > > > > These are not strange scenarios.  These situations 
> > are all permitted by the
> > > > > > RFC.  Yet, in each situation it is clear that one 
> > cannot recommend that the SG
> > > > > > wait.  The result would be a non-functional system.  
> > There are many more
> > > > > > examples.
> > > > > > 
> > > > > > n+k is a model for redundancy (not for congestion and 
> > flow control).  A system
> > > > > > with an SG that waits for n is, in general, not a 
> > fully redundant system at
> > > > > > all, but is, rather, prone to these failure syndromes.
> > > > > > 
> > > > > > When you look at the situation and apply the 
> > experience of reliability
> > > > > > engineering, the recommendation is that the SG SHOULD 
> > start traffic when the
> > > > > > first ASP goes active.
> > > > > > 
> > > > > > Why wait for 'n' anyway?  Because the ASP cannot 
> > handle some level of traffic?
> > > > > > Any ASP that conforms to MTP-User standards, must 
> > have the standardized
> > > > > > MTP-User (ISUP/SCCP) congestion and flow controls and 
> > must have the SCTP
> > > > > > congestion controls and is therefore able to accept 
> > any amount of traffic
> > > > > > overload.
> > > > > > 
> > > > > > So there is no real reason for the SG to wait for 
> > 'n'.  Nevertheless, the RFC
> > > > > > allows this as an OPTION.  Requiring (MUST) or 
> > recommending it (SHOULD)
> > > > > > regardless of whether the number of active processors 
> > is rising or falling,
> > > > > > whether start, restart or failure, is unacceptable.
> > > > > > 
> > > > > > On the other side, at the ASPs, the ASPs have a 
> > requirement to share state
> > > > > > between them.  They must share circuit state and 
> > transaction state in all
> > > > > > configurations to meet carrier grade.  This is stated 
> > in the RFC.  It is
> > > > > > implicit that the ASPs are capable of knowing with 
> > which SGs, and which SGPs
> > > > > > for which they are active.  They have access to the 
> > total state of the AS that
> > > > > > they collectively represent, that the SG/SGP does 
> > not.  Therefore is it
> > > > > > reasonable for the ASPs to coordinate among 
> > themselves the number of active
> > > > > > ASPs and to ensure that there are sufficient ASPs 
> > active to handle the
> > > > > > expected load before starting traffic.  Because the 
> > ASPs need not respond to
> > > > > > user test traffic (UPT/SST), the ASPs are also in 
> > control of when remote SS7
> > > > > > MTP-User restart traffic toward them.
> > > > > > 
> > > > > > Therefore, the recommendation that the ASPs SHOULD 
> > coordinate among
> > > > > > themselves the number of active ASPs and not start 
> > traffic until a sufficient
> > > > > > number of them are active is sound.  This is not a 
> > MUST, permitting ASPs
> > > > > > that know that they are capable of handling any 
> > overload, to start traffic
> > > > > > immediately, or at some earlier point.
> > > > > > 
> > > > > > Back to the RFC, the RFC sets the normal case where 
> > the SG starts traffic on
> > > > > > the first active ASP but explicitly permits as an 
> > OPTION to wait until it
> > > > > > considers that there are sufficient ASPs active to 
> > handle the offered load.
> > > > > > The RFC also RECOMMENDS that the ASPs coordinate 
> > among themselves the number
> > > > > > of active ASPs so that there are sufficient ASPs to 
> > handle the expected load.
> > > > > > 
> > > > > > The provisions in the RFC are both sound, where the 
> > proposal and section 3.15
> > > > > > is clearly not.
> > > > > > 
> > > > > > Remember when looking at the AS state diagram in the 
> > RFC that this is the
> > > > > > SGP's view of the AS state and not the total state of 
> > the AS, that can only
> > > > > > known to the ASPs for all configurations.  'n' as 
> > defined has no meaning to
> > > > > > the state machine at the SGP because is not required 
> > to know, and often cannot
> > > > > > know, how many ASPs are active for the AS via other 
> > SGPs and SGs, or even via
> > > > > > direct connection to the SS7 network.
> > > > > > 
> > > > > > Some SGs might have special knowledge, or 
> > sophisticated algorithms that allow
> > > > > > it to determine whether a congestive failure will be 
> > the immediate result of
> > >  > > > starting or restarting traffic.  These SGs are 
> > permitted as an
> > > OPTION to wait.
> > > > > > 
> > > > > > The ASPs SHOULD have the knowledge of how many are 
> > required to handle the
> > > > > > expected load, because they must share 
> > circuit/transaction state and they must
> > > > > > have congestion and flow controls as required by the 
> > standards.
> > > > > > 
> > > > > > So again, the RFC is adequate in this regard.  The 
> > current proposal and
> > > > > > section 3.15 are not.
> > > > > RFC is inadequate so we need section 3.15 to explain 'n+k' case.
> > > > 
> > > > Teaching reliability engineering and the design of 
> > redundant systems is
> > > > outside the scope of the RFC and the IG.  People, such as 
> > yourself, are better
> > > > advised to learn the subject matter before attempting to 
> > implement or claim a
> > > > complete understanding of the protocol.
> > > > 
> > > > The current RFC provides the protocol mechanisms and basis for
> > > > interoperability for all of the common approaches to 
> > redundancy in carrier
> > > > class systems, and even provides a general model that can 
> > accommodate
> > > > non-traditional redundancy architectures.
> > > > 
> > > > When confused about how a redundancy architecture works, 
> > don't blame the M3UA
> > > > RFC, blame those responsible for describing redundant 
> > architectures to you.
> > > > 
> > > > In the appendix we stated:
> > > > 
> > > >    M3UA imposes no restrictions as to the exact layout of 
> > the network
> > > >    elements, the message distribution algorithms and the 
> > distribution of
> > > >    signalling processes.  Instead, it provides a 
> > framework and a set of
> > > >    messages that allow for a flexible and scalable 
> > signalling network
> > > >    architecture, aiming to provide reliability and performance.
> > > > 
> > > > Forcing SGs to wait for 'n' ASPs active before activating 
> > is contrary to this.
> > > > The existing "MAY" achieves this flexibility.
> > > > 
> > > > When reading 1.4.4.1, also remember this statement in the 
> > appendix, that used
> > > > to be beside 1.4.4.1.  The description of 'n+k' does not 
> > dictate the state of
> > > > the 'k' or the state of the AS with respect to 'n' to 
> > leave the redundancy
> > > > model flexible, and accommodate '1+1' and '1+0' in the same model.
> > > > 
> > > > Also, please don't presume to tell me to set 'n' to 1, as 
> > it is not a
> > > > arbitrary value, but the number 'm' in the 'm+1' 
> > distributed system in common
> > > > use today.  This number is determined as the minimum 
> > number of ASPs required
> > > > to handle the maximum engineered load, and not, as Tolga 
> > would have, the
> > > > minimum number of ASPs required to handle ANY load.  If 
> > you want your system
> > > > to behave as Tolga proposes, just have your SG exercise 
> > the "MAY choose not to
> > > > direct traffic" option of section 4.3.4.3 of the RFC.
> > > > 
> > > > Section 3.15 and Tolga's proposal should be struck 
> > because they attempt to
> > > > impose a 'n' active 'k' inactive and system cannot 
> > restart (3.15) or start
> > > > (Tolga) without 'n' view on the redundancy model, where 
> > it is clearly the
> > > > intent of the RFC to place no such restrictions on the 
> > model, but to
> > > > accommodate a range of models, that is already achieved 
> > by the RFC text.
> > > > 
> > > > --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

-- 
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 exim@www1.ietf.org  Thu Jul 10 07:39:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA27428
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 07:39:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aZlN-0007Xo-00
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 07:39:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ABd0ht028990
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 07:39:00 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aZlM-0007XU-Sb; Thu, 10 Jul 2003 07:39:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aZl2-0007WH-Co
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 07:38:40 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA27399
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 07:38:37 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aZl1-0001uN-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 07:38:39 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[bBtwvMdgWlc8IK0bzQmK4JNXmVlb1xiE])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aZl0-0001uK-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 07:38:38 -0400
Received: from ns.pigworks.openss7.net (IDENT:D0rVlm4DZ0lF5ezw0kjMiALsxXUuntFE@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6ABcaX24696;
	Thu, 10 Jul 2003 05:38:36 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6ABcZk15608;
	Thu, 10 Jul 2003 05:38:35 -0600
Date: Thu, 10 Jul 2003 05:38:35 -0600
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>,
        Ken Morneault <kmorneau@cisco.com>,
        "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
        sigtran@ietf.org
Subject: Re: [Sigtran] M3UA IG v04
Message-ID: <20030710053835.D14388@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>,
	Ken Morneault <kmorneau@cisco.com>,
	"J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
	sigtran@ietf.org
References: <20030710051616.B14388@openss7.org> <Pine.SOL.4.10.10307100622480.25952-100000@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.4.10.10307100622480.25952-100000@sun2594.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Jul 10, 2003 at 06:24:35AM -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>

Samuel,

Are you sure?

Don't you mean NTFY("Insufficient ASP Resources Active in AS") ???

There is no notification message in the RFC called: "Sufficient ASPs"...

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 10 Jul 2003 06:24:35)
> 
> Brian,
> 
>  Yes,i meant to say that NTFY(Sufficient ASPs active) is not required.
>  (hope this is the main discussion)
> 
> --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, 10 Jul 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Are you meaning the "Sufficient ASPs" message that I mentioned some
> > time ago?
> > 
> > --brian
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 10 Jul 2003 05:36:24)
> > > 
> > > Zannin,
> > > 
> > >   Just an example, say an operator configures an AS1 as set of 3 ASPes.
> > >   They are namely known as ASP1,ASP2 and ASP3(backup).
> > > 
> > >   Initially ASP1 and ASP2 send ASP_ACTIVE to SGP  so SGP sends a
> > >   NTFY(AS_ACTIVE) thus results in AS1 to be in ACTIVE state.
> > > 
> > >   Where is the question of sending NTFY(Sufficient ASPs active)?
> > >   People mean to say that backup ASP also sends an ASP_ACTIVE message
> > >   to SGP? if it so, on what condition it is being sent? Is there
> > >   any valid conditions that backup ASP or some new(!!!??!amazing!!) ASP
> > >   to send ASP_ACTIVE message? 
> > > 
> > > 
> > > --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, 9 Jul 2003, Zannin, Francois wrote:
> > > 
> > > > Ken,
> > > > 
> > > > Here is a copy of the mail about notify I was refering.
> > > > 
> > > > -----Original Message-----
> > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> > > > Sent: Wednesday, February 12, 2003 10:40 AM
> > > > To: ksrinivas
> > > > Cc: sigtran@ietf.org
> > > > Subject: [Sigtran] Re: M3UA Notify message
> > > > 
> > > > 
> > > > 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
> > > > -----Original Message-----
> > > > From: Ken Morneault [mailto:kmorneau@cisco.com] 
> > > > Sent: Wednesday, July 09, 2003 4:28 PM
> > > > To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> > > > Subject: RE: [Sigtran] M3UA IG v04
> > > > 
> > > > 
> > > > 
> > > > 
> > > > Francois,
> > > > 
> > > > Can you send me a ptr to the email you reference.
> > > > 
> > > > Thanks,
> > > > 
> > > > Ken
> > > > 
> > > > At 04:20 PM 7/9/2003 +0200, Zannin, Francois wrote:
> > > > >Ken,
> > > > >
> > > > >My point of view on he question, see hereafter.
> > > > >
> > > > >Regards,
> > > > >
> > > > >Francois Z.
> > > > >
> > > > >-----Original Message-----
> > > > >From: Ken Morneault [mailto:kmorneau@cisco.com]
> > > > >Sent: Wednesday, July 09, 2003 3:56 PM
> > > > >To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> > > > >Subject: RE: [Sigtran] M3UA IG v04
> > > > >
> > > > >
> > > > >Francois,
> > > > >
> > > > >Pls see comments below.
> > > > >
> > > > >Regards,
> > > > >
> > > > >Ken
> > > > >
> > > > >At 10:15 AM 7/9/2003 +0200, Zannin, Francois wrote:
> > > > >>Just some comment about the M3UA IG v04
> > > > >>
> > > > >>3.21 NOTIFY messages are missing in Examples section
> > > > >>Figure (section 5.1.2)
> > > > >>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the
> > > > >>ASP Up Ack
> > > > >>
> > > > >>Figure (section 5.1.3)
> > > > >>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the
> > > > >>ASP Up Ack
> > > > >>
> > > > >>Figure (section 5.1.4)
> > > > >>        Same remark, the Notify is sent to ASP1, ASP2 and also need to
> > > > >>be sent to
> > > > >>      ASP3 after the ASP Up Ack
> > > > >
> > > > >NOTIFY is only sent upon "a change in status of a particular 
> > > > >Application Server".
> > > > >
> > > > >OK, but just one more remark, in section 5.1.4, a notify AS-Inactive is
> > > > 
> > > > >sent to ASP1 and ASP2 while to transition to ASP-Inactive was effective
> > > > 
> > > > >after the first
> > > > >ASP_UP(/ACK) comming from ASP1, so notify to ASP2 should be considerer
> > > > >as inopportune.
> > > > >
> > > > >In last november, it was agreed in a thread that when an ASP was 
> > > > >transitionned to UP, The current AS state was sent to this ASP, 
> > > > >whatever could be the state. This current
> > > > >State can be used by the management of an ASP to decide if the AS has
> > > > to
> > > > >be activated 
> > > > >Immediately or must wait for an AS-PENDING or AS-INACTIVE notification.
> > > > 
> > > > 
> > > > _______________________________________________
> > > > 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 exim@www1.ietf.org  Thu Jul 10 07:51:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA27861
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 07:51:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aZx2-0000FF-8x
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 07:51:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ABp4ek000933
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 07:51:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aZx0-0000Ea-2H; Thu, 10 Jul 2003 07:51:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aZwr-0000Do-TX
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 07:50:53 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA27801
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 07:50:50 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aZwr-00020A-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 07:50:53 -0400
Received: from auds952.usa.alcatel.com ([143.209.238.7])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aZwq-000200-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 07:50:52 -0400
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6ABoKB3010798
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 06:50:20 -0500 (CDT)
Received: from sun2594.ssd.usa.alcatel.com (sun2594.ssd.usa.alcatel.com [143.209.250.252])
	by ssd.usa.alcatel.com (8.12.8/8.12.8) with ESMTP id h6ABnv1C019329
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 06:49:57 -0500 (CDT)
Date: Thu, 10 Jul 2003 06:49:57 -0500 (CDT)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA IG v04
In-Reply-To: <20030710053835.D14388@openss7.org>
Message-ID: <Pine.SOL.4.10.10307100647330.25952-100000@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,

See the underlined sentence.

--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, 10 Jul 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> Are you sure?
> 
> Don't you mean NTFY("Insufficient ASP Resources Active in AS") ???
> 
> There is no notification message in the RFC called: "Sufficient ASPs"...
> 
> --brian
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 10 Jul 2003 06:24:35)
> > 
> > Brian,
> > 
> >  Yes,i meant to say that NTFY(Sufficient ASPs active) is not required.
> >  (hope this is the main discussion)
> > 
> > --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, 10 Jul 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > Are you meaning the "Sufficient ASPs" message that I mentioned some
> > > time ago?
> > > 
> > > --brian
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 10 Jul 2003 05:36:24)
> > > > 
> > > > Zannin,
> > > > 
> > > >   Just an example, say an operator configures an AS1 as set of 3 ASPes.
> > > >   They are namely known as ASP1,ASP2 and ASP3(backup).
> > > > 
> > > >   Initially ASP1 and ASP2 send ASP_ACTIVE to SGP  so SGP sends a
> > > >   NTFY(AS_ACTIVE) thus results in AS1 to be in ACTIVE state.
> > > > 
> > > >   Where is the question of sending NTFY(Sufficient ASPs active)?
> > > >   People mean to say that backup ASP also sends an ASP_ACTIVE message
> > > >   to SGP? if it so, on what condition it is being sent? Is there
> > > >   any valid conditions that backup ASP or some new(!!!??!amazing!!) ASP
> > > >   to send ASP_ACTIVE message? 
> > > > 
> > > > 
> > > > --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, 9 Jul 2003, Zannin, Francois wrote:
> > > > 
> > > > > Ken,
> > > > > 
> > > > > Here is a copy of the mail about notify I was refering.
> > > > > 
> > > > > -----Original Message-----
> > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> > > > > Sent: Wednesday, February 12, 2003 10:40 AM
> > > > > To: ksrinivas
> > > > > Cc: sigtran@ietf.org
> > > > > Subject: [Sigtran] Re: M3UA Notify message
> > > > > 
> > > > > 
> > > > > 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
> > > > > -----Original Message-----
> > > > > From: Ken Morneault [mailto:kmorneau@cisco.com] 
> > > > > Sent: Wednesday, July 09, 2003 4:28 PM
> > > > > To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> > > > > Subject: RE: [Sigtran] M3UA IG v04
> > > > > 
> > > > > 
> > > > > 
> > > > > 
> > > > > Francois,
> > > > > 
> > > > > Can you send me a ptr to the email you reference.
> > > > > 
> > > > > Thanks,
> > > > > 
> > > > > Ken
> > > > > 
> > > > > At 04:20 PM 7/9/2003 +0200, Zannin, Francois wrote:
> > > > > >Ken,
> > > > > >
> > > > > >My point of view on he question, see hereafter.
> > > > > >
> > > > > >Regards,
> > > > > >
> > > > > >Francois Z.
> > > > > >
> > > > > >-----Original Message-----
> > > > > >From: Ken Morneault [mailto:kmorneau@cisco.com]
> > > > > >Sent: Wednesday, July 09, 2003 3:56 PM
> > > > > >To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> > > > > >Subject: RE: [Sigtran] M3UA IG v04
> > > > > >
> > > > > >
> > > > > >Francois,
> > > > > >
> > > > > >Pls see comments below.
> > > > > >
> > > > > >Regards,
> > > > > >
> > > > > >Ken
> > > > > >
> > > > > >At 10:15 AM 7/9/2003 +0200, Zannin, Francois wrote:
> > > > > >>Just some comment about the M3UA IG v04
> > > > > >>
> > > > > >>3.21 NOTIFY messages are missing in Examples section
> > > > > >>Figure (section 5.1.2)
> > > > > >>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the
> > > > > >>ASP Up Ack
> > > > > >>
> > > > > >>Figure (section 5.1.3)
> > > > > >>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the
> > > > > >>ASP Up Ack
> > > > > >>
> > > > > >>Figure (section 5.1.4)
> > > > > >>        Same remark, the Notify is sent to ASP1, ASP2 and also need to
> > > > > >>be sent to
> > > > > >>      ASP3 after the ASP Up Ack
> > > > > >
> > > > > >NOTIFY is only sent upon "a change in status of a particular 
> > > > > >Application Server".
> > > > > >
> > > > > >OK, but just one more remark, in section 5.1.4, a notify AS-Inactive is
> > > > > 
> > > > > >sent to ASP1 and ASP2 while to transition to ASP-Inactive was effective
> > > > > 
> > > > > >after the first
> > > > > >ASP_UP(/ACK) comming from ASP1, so notify to ASP2 should be considerer
> > > > > >as inopportune.
> > > > > >
> > > > > >In last november, it was agreed in a thread that when an ASP was 
> > > > > >transitionned to UP, The current AS state was sent to this ASP, 
> > > > > >whatever could be the state. This current
> > > > > >State can be used by the management of an ASP to decide if the AS has
> > > > > to
> > > > > >be activated 
> > > > > >Immediately or must wait for an AS-PENDING or AS-INACTIVE notification.
> > > > > 
> > > > > 
> > > > > _______________________________________________
> > > > > 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 exim@www1.ietf.org  Thu Jul 10 08:01:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA28149
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 08:01:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aa6f-0000xl-VL
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 08:01:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AC11OM003701
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 08:01:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aa6e-0000xP-T3; Thu, 10 Jul 2003 08:01:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aa6D-0000wk-NY
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 08:00:33 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA28128
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 08:00:30 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aa6C-00024b-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 08:00:32 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[SgK4nVKbVFXtal9st4A0Z0rc9F3Yq4By])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aa6B-00024X-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 08:00:31 -0400
Received: from ns.pigworks.openss7.net (IDENT:KXRIZHixE+9OkJSa8SbQ5nj2znT1Njbm@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6AC0UX25049;
	Thu, 10 Jul 2003 06:00:30 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6AC0Tg16441;
	Thu, 10 Jul 2003 06:00:29 -0600
Date: Thu, 10 Jul 2003 06:00:29 -0600
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 IG v04
Message-ID: <20030710060029.A16241@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	sigtran@ietf.org
References: <20030710053835.D14388@openss7.org> <Pine.SOL.4.10.10307100647330.25952-100000@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.4.10.10307100647330.25952-100000@sun2594.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Thu, Jul 10, 2003 at 06:49:57AM -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>

Samuel,

It was never added.

I believe Francois was talking about this proposal:

> > > > > >    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.

(Of course this would probably mean that one would have to rewrite the
remainder of the RFC into the absurd section 3.14.)

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 10 Jul 2003 06:49:57)
> 
> Brian,
> 
> See the underlined sentence.
> 
> --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, 10 Jul 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > Are you sure?
> > 
> > Don't you mean NTFY("Insufficient ASP Resources Active in AS") ???
> > 
> > There is no notification message in the RFC called: "Sufficient ASPs"...
> > 
> > --brian
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 10 Jul 2003 06:24:35)
> > > 
> > > Brian,
> > > 
> > >  Yes,i meant to say that NTFY(Sufficient ASPs active) is not required.
> > >  (hope this is the main discussion)
> > > 
> > > --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, 10 Jul 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Samuel,
> > > > 
> > > > Are you meaning the "Sufficient ASPs" message that I mentioned some
> > > > time ago?
> > > > 
> > > > --brian
> > > > 
> > > > Samuel Dur D. Jeyaseelan wrote:                                                     (Thu, 10 Jul 2003 05:36:24)
> > > > > 
> > > > > Zannin,
> > > > > 
> > > > >   Just an example, say an operator configures an AS1 as set of 3 ASPes.
> > > > >   They are namely known as ASP1,ASP2 and ASP3(backup).
> > > > > 
> > > > >   Initially ASP1 and ASP2 send ASP_ACTIVE to SGP  so SGP sends a
> > > > >   NTFY(AS_ACTIVE) thus results in AS1 to be in ACTIVE state.
> > > > > 
> > > > >   Where is the question of sending NTFY(Sufficient ASPs active)?
> > > > >   People mean to say that backup ASP also sends an ASP_ACTIVE message
> > > > >   to SGP? if it so, on what condition it is being sent? Is there
> > > > >   any valid conditions that backup ASP or some new(!!!??!amazing!!) ASP
> > > > >   to send ASP_ACTIVE message? 
> > > > > 
> > > > > 
> > > > > --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, 9 Jul 2003, Zannin, Francois wrote:
> > > > > 
> > > > > > Ken,
> > > > > > 
> > > > > > Here is a copy of the mail about notify I was refering.
> > > > > > 
> > > > > > -----Original Message-----
> > > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> > > > > > Sent: Wednesday, February 12, 2003 10:40 AM
> > > > > > To: ksrinivas
> > > > > > Cc: sigtran@ietf.org
> > > > > > Subject: [Sigtran] Re: M3UA Notify message
> > > > > > 
> > > > > > 
> > > > > > 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
> > > > > > -----Original Message-----
> > > > > > From: Ken Morneault [mailto:kmorneau@cisco.com] 
> > > > > > Sent: Wednesday, July 09, 2003 4:28 PM
> > > > > > To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> > > > > > Subject: RE: [Sigtran] M3UA IG v04
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > > Francois,
> > > > > > 
> > > > > > Can you send me a ptr to the email you reference.
> > > > > > 
> > > > > > Thanks,
> > > > > > 
> > > > > > Ken
> > > > > > 
> > > > > > At 04:20 PM 7/9/2003 +0200, Zannin, Francois wrote:
> > > > > > >Ken,
> > > > > > >
> > > > > > >My point of view on he question, see hereafter.
> > > > > > >
> > > > > > >Regards,
> > > > > > >
> > > > > > >Francois Z.
> > > > > > >
> > > > > > >-----Original Message-----
> > > > > > >From: Ken Morneault [mailto:kmorneau@cisco.com]
> > > > > > >Sent: Wednesday, July 09, 2003 3:56 PM
> > > > > > >To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> > > > > > >Subject: RE: [Sigtran] M3UA IG v04
> > > > > > >
> > > > > > >
> > > > > > >Francois,
> > > > > > >
> > > > > > >Pls see comments below.
> > > > > > >
> > > > > > >Regards,
> > > > > > >
> > > > > > >Ken
> > > > > > >
> > > > > > >At 10:15 AM 7/9/2003 +0200, Zannin, Francois wrote:
> > > > > > >>Just some comment about the M3UA IG v04
> > > > > > >>
> > > > > > >>3.21 NOTIFY messages are missing in Examples section
> > > > > > >>Figure (section 5.1.2)
> > > > > > >>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the
> > > > > > >>ASP Up Ack
> > > > > > >>
> > > > > > >>Figure (section 5.1.3)
> > > > > > >>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after the
> > > > > > >>ASP Up Ack
> > > > > > >>
> > > > > > >>Figure (section 5.1.4)
> > > > > > >>        Same remark, the Notify is sent to ASP1, ASP2 and also need to
> > > > > > >>be sent to
> > > > > > >>      ASP3 after the ASP Up Ack
> > > > > > >
> > > > > > >NOTIFY is only sent upon "a change in status of a particular 
> > > > > > >Application Server".
> > > > > > >
> > > > > > >OK, but just one more remark, in section 5.1.4, a notify AS-Inactive is
> > > > > > 
> > > > > > >sent to ASP1 and ASP2 while to transition to ASP-Inactive was effective
> > > > > > 
> > > > > > >after the first
> > > > > > >ASP_UP(/ACK) comming from ASP1, so notify to ASP2 should be considerer
> > > > > > >as inopportune.
> > > > > > >
> > > > > > >In last november, it was agreed in a thread that when an ASP was 
> > > > > > >transitionned to UP, The current AS state was sent to this ASP, 
> > > > > > >whatever could be the state. This current
> > > > > > >State can be used by the management of an ASP to decide if the AS has
> > > > > > to
> > > > > > >be activated 
> > > > > > >Immediately or must wait for an AS-PENDING or AS-INACTIVE notification.
> > > > > > 
> > > > > > 
> > > > > > _______________________________________________
> > > > > > 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

-- 
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 exim@www1.ietf.org  Thu Jul 10 08:04:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA28260
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 08:04:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aa9Z-00016j-QL
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 08:04:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AC41lc004253
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 08:04:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aa9Z-00016M-Du; Thu, 10 Jul 2003 08:04:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aa8y-00015l-9u
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 08:03:24 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA28210
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 08:03:20 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aa8x-00025N-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 08:03:23 -0400
Received: from penguin-ext.wise.edt.ericsson.se ([193.180.251.47] helo=penguin.al.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19aa8w-00025K-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 08:03:22 -0400
Received: from esealnt610.al.sw.ericsson.se ([153.88.254.120])
	by penguin.al.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h6AC3KGJ028392;
	Thu, 10 Jul 2003 14:03:20 +0200 (MEST)
Received: from ESEALNT745.al.sw.ericsson.se ([153.88.251.5]) by esealnt610.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55)
	id 3S6BRXPN; Thu, 10 Jul 2003 14:04:14 +0200
Received: by ESEALNT745.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <N94NNSLJ>; Thu, 10 Jul 2003 14:02:59 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF0@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA IG v04
Date: Thu, 10 Jul 2003 14:02:53 +0200
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, 

Pls see my comments inline...

Cheers // Javier.

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: jueves, 10 de julio de 2003 13:37
> To: J.Javier Pastor (EE/EEM)
> Cc: 'sigtran@ietf.org'
> Subject: Re: [Sigtran] M3UA IG v04
> 
> 
> Javier,
> 
> Your "new possibility" is the norm in the RFC.  You seem to 
> want reverse
> the sense of the RFC.  The RFC says that we activate on the 
> 1st ASP and
> MAY withhold until the 'n'th ASP.  3.14 says that we SHOULD 
> withhold to
> the 'n'th ASP and MAY activate on the first.
> 
It is a "new possibility" considering the base of the rough consensus it was reached in the mailing list regarding the AS States. The introduction of smooth start was in order to have all the opinions included.

It's nothing about reversing the RFC. It is simply that there are several understandings of what the RFC says and we should clarify all the conflicts. It seems that most people have the understanding of "n" ASPs are needed as a general rule...

> The reason for this change escapes me.  There is no justification to
> reverse the option.

Just the consensus in the mailing list. 

> 
> Please, answer me this:
> 
>   Who has an ASP that is so broken that it needs the SGP to wait
>   for 'n' of them to become active?
> 
>   Please identify yourselves!
> 
> 
> There are several other places in the RFC that assume 
> activation on the
> first ASP.  It is sewn through and through the RFC that 
> loadsharing and
> broadcast is performed across the active APSs (regardless of n).

The RFC has different views on this, and this is the point that we needed to clarify.


> 
> Section 3.7.1:
>                                                   ...  In Loadshare
>       mode, the ASP will share in the traffic distribution with any
>       other currently active ASPs.  In Broadcast mode, the ASP will
>       receive the same messages as any other currently active ASP.
> 
> (Notice the word "any", twice, above.)
> 
> Section 4.1.1
> 
>    From the list of ASPs within the AS table, an ASP in the ASP-ACTIVE
>    state is selected and a DATA message is constructed and 
> issued on the
>    corresponding SCTP association.  If more than one ASP is 
> in the ASP-
>    ACTIVE state (i.e., traffic is to be loadshared across 
> more than one
>    ASP), one of the ASPs in the ASP-ACTIVE state is selected from the
>    list.  If the ASPs are in Broadcast Mode, all active ASPs will be
>    selected and the message sent to each of the active ASPs.
> 
> (Note above that there is no 'n' restriction on selecting 
> ASPs to receive
> traffic.)
> 
> Section 4.3.4.3:
> 
>    In the case of a Loadshare mode AS, reception of an ASP Active
>    message at an SGP or IPSP causes the direction of traffic 
> to the ASP
>    sending the ASP Active message, in addition to all the other ASPs
>    that are currently active in the AS.
> 
> Section 4.3.4.4:
> 
>    In the case of a Loadshare mode AS, the SGP moves the ASP to the
>    ASP-INACTIVE state and the AS traffic is reallocated across the
>    remaining ASPs in the state ASP-ACTIVE, as per the loadsharing
> 
> 
> Also, note that there is no limit to T(r).  So, if it was set 
> infinite,
> or just rather large, the start case is transformed to the 
> restart case.
> 
> 
> Changing the trantion from the AS-DOWN to AS-INACTIVE state at the SGP
> to wait for 'n' ASP-INACTIVE ASPs is even more pointless than waiting
> for 'n' ASP-ACTIVE ASPs.  No traffic is flowing.  The RFC 
> always has us
> move to the AS-INACTIVE state on the first ASP-INACTIVE ASP.
> 
> What in the world did you think would necessitate changing this?
> 

If you have any other proposal that reach consensus and clarify the RFC regarding this concern, we will place it in the IG.

Cheers // Javier.


> --brian
> 
> 
> Javier Pastor-Balbas (ECE) wrote:                             
>                       (Thu, 03 Jul 2003 11:45:57)
> > 
> > Hi guys,
> > 
> > We tried to submit the next IG version before the deadline, 
> but I got
> > confused since I thought it ended at 12:00 instead of 9:00. 
> So the server
> > bounced it. I'll submit the doc again after the IETF meeting. In the
> > meantime you can find it at:
> > 
> > 
> http://standards.ericsson.net/javier/draft-ietf-sigtran-m3ua-i
> mplementors-guide-04.txt
> > 
> > We've tried to include all the different opinions in here. 
> Although it seems
> > that the case of waiting till n ASPs is the preferred one, 
> we have included
> > the new possibility of a smooth start for those of you who 
> want to activate
> > the AS when the first ASP gets to ACTIVE. At the end it will be the
> > operator's decission how to proceed during the activation 
> phase. Hopefully
> > everyone agrees on this wording...
> > 
> > Cheers // Javier.
> > 
> > 
> > 
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: martes, 01 de julio de 2003 20:18
> > > To: Samuel Dur D. Jeyaseelan
> > > Cc: sigtran@ietf.org
> > > Subject: Re: [Sigtran] Re:n+k issue
> > > 
> > > 
> > > Samuel,
> > > 
> > > Perhaps you could answer my questions below first.  Please be
> > > specific.
> > > 
> > > --brian
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                               
> > >                       (Tue, 01 Jul 2003 06:09:34)
> > > > 
> > > > Brian,
> > > > 
> > > >   Can you bit elaborate your n=256 example (ANSI-SLS) with 
> > > real-time 
> > > >   scenario?(please write to the point)
> > > > 
> > > > --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, 1 Jul 2003, Brian F. G. Bidulock wrote:
> > > > 
> > > > > Samuel,
> > > > > 
> > > > > Samuel Dur D. Jeyaseelan wrote:                      
> > > (Mon, 30 Jun 2003 09:44:01)
> > > > > > 
> > > > > > Brian,
> > > > > > 
> > > > > > On Sat, 28 Jun 2003, Brian F. G. Bidulock wrote:
> > > > > > 
> > > > > > > Samuel,
> > > > > > > 
> > > > > > > I should point out that the proposal to require that 
> > > the SG wait for 'n' to
> > > > > > > start traffic does not have much merit.
> > > > > > > 
> > > > > > > Example: n=256, k=1
> > > > > > > 
> > > > > > > 2 ASPs fail, permanently and need to be manually replaced.
> > > > > > > 
> > > > > > > A temporary network failure causes SCTP associations 
> > > to be lost to the
> > > > > > > remaining 255 ASPs.  They move to the ASP down state 
> > > and immediately
> > > > > > > re-establish SCTP associations and move to the 
> > > ASP-Active state.  Because the
> > > > > > > number of ASPs is < n, the SG will not restart 
> > > traffic for some time.
> > > > > > 
> > > > > > Your example leads to the following figure.
> > > > > > 
> > > > > >          SG1                            AS1
> > > > > >      
> > > > > >  SS7 MSG +------+----------------------- ASP (1)
> > > > > > ---------| SGP1 |----------------------- ASP (2)
> > > > > >          |      |----------------------- ....
> > > > > >          +------+----------------------- ASP (256)
> > > > > >             |
> > > > > >             | 
> > > > > >             +---------------------------ASP(k)
> > > > > > 
> > > > > > 1. First of all craft will not configure 1 ASP as 
> > > backup for 256 Active
> > > > > >    ASPs, if he does, it means to be a poor network 
> > > configuration.
> > > > > 
> > > > > Well, first of all, craft does not normally choose the 
> > > number of ASPs used for
> > > > > backup, engineering does.  Second, k=1 is more than 
> > > adequate for n=256 where
> > > > > the ASPs are reasonably reliable and have a short 
> > > replacement time.
> > > > > 
> > > > > Regardless, I followed with the same example with n=2 and k=1.
> > > > > 
> > > > > > 
> > > > > > 2. Consider ASP as an MGC, your example leads to 256 
> > > MGCs and they are 
> > > > > >    controlled by a single SGP.A good network will have 
> > > multiple SGPs to
> > > > > >    share the POTENTIAL traffic across ASPs.
> > > > > 
> > > > > You seem to misunderstand the concept of redundancy in 
> > > M3UA: there are not
> > > > > multiple SGPs or ASPs because of traffic load.  It is 
> > > easy enough to handle
> > > > > the traffic of a small to medium sized country through a 
> > > single processor.
> > > > > 
> > > > > Also, you seem to misunderstand that an SGP and an ASP 
> > > are not processors.
> > > > > They are processes.  One host processor could have many 
> > > (even 256) ASPs or
> > > > > SGPs (or both for that matter).
> > > > > 
> > > > > > 
> > > > > > 3. BASIC INTENTION OF 'n+k' is,just to make sure that 
> > > incoming messages
> > > > > >    ARE EFFICIENTLY TRANSMITTED TO DESTINATION ASPs WITH 
> > > THE HELP OF
> > > > > >    MINIMUM NUMBER OF ASSOCIATIONS.
> > > > > 
> > > > > No.  The intention of 'n+k' is to provide one of the more 
> > > recently accepted
> > > > > models for reliability using redundancy (n = m+1, k=0 
> > > being the typical system
> > > > > deployed in the field today, m usually being greater than 
> > > 2).  Up until
> > > > > recently, only 2m (or m+m) systems were considered 
> > > carrier class.  Some
> > > > > systems were 4m (with 3 out of 4 vote).
> > > > > 
> > > > > You should take some time to look at redundancy and 
> > > reliability requirements
> > > > > for carrier grade systems for an understanding of how to 
> > > calculate and provide
> > > > > the reliability requirements expected by carriers.  There 
> > > are good standards
> > > > > written on the topic.  See the Telcordia NEBS and 
> > > reliability specs, for
> > > > > example.  If you have an RBOC RFP handy, take a look at 
> > > the hardware and
> > > > > software reliability requirements sections or talk to the 
> > > engineer in your
> > > > > company responsible for answering those sections.
> > > > > 
> > > > > > 
> > > > > >    Say for example
> > > > > > 
> > > > > >    Inefficient configuration -just an example
> > > > > >    ------------------------------------------
> > > > > >   
> > > > > >             SLS         DEST
> > > > > >             =================
> > > > > >              0           ASP1
> > > > > >              1           ASP1
> > > > > >              2           ASP1
> > > > > >              3           ASP1
> > > > > >              4           ASP1
> > > > > >              5           ASP1
> > > > > >              6           ASP1
> > > > > >              .            .
> > > > > >              .            .
> > > > > >              15          ASP1
> > > > > >              
> > > > > >              
> > > > > >                SG                            AS1
> > > > > > sls(0,1..15 )+----+  sls(0,1..15)         +------+
> > > > > > -------------|SGP1|-----------------------| ASP1 |
> > > > > >              |    |              ASSOC-1  +------+
> > > > > >              +----+                     
> > > > > > 
> > > > > >                       Fig - 1
> > > > > > 
> > > > > > In this figure, all incoming messages are transmitted 
> > > over ASSOC-1 towards
> > > > > > ASP1.Here there are two potential problems 
> > > > > > 
> > > > > >        1. Total incoming traffic is dependent on a 
> single ASP
> > > > > >        2. ASP1 is always overloaded, messages are 
> > > processed one by one
> > > > > >           from the transmission buffer which ultimately 
> > > introduces time
> > > > > >           delay(example last arrived msg will be 
> > > processed at the end).
> > > > > 
> > > > > The above was not in my examples.  My examples had n > 1.
> > > > > 
> > > > > > 
> > > > > > 
> > > > > > Efficient configuration
> > > > > > -----------------------  
> > > > > > 
> > > > > >              SLS         DEST
> > > > > >             =================
> > > > > >              0           ASP1
> > > > > >              1           ASP1
> > > > > >              2           ASP1
> > > > > >              3           ASP1
> > > > > >              4           ASP1
> > > > > >              5           ASP1
> > > > > >              6           ASP1
> > > > > >              7           ASP1
> > > > > >              8           ASP2
> > > > > >              9           ASP2
> > > > > >              .            .
> > > > > >              .            .
> > > > > >              15          ASP2
> > > > > > 
> > > > > >                
> > > > > >                        
> > > > > >                        
> > > > > >               SG                             AS1
> > > > > > sls(0,1..15 )+----+   sls(0,1,..7)ASSOC-1 +------+
> > > > > > -------------|SGP1|-----------------------| ASP1 |
> > > > > >              |    |--------+              +------+
> > > > > >              +----+        | sls(8,9..15) +------+
> > > > > >                            +--------------| ASP2 |
> > > > > >                                  ASSOC-2  +------+
> > > > > >                         
> > > > > >                    fig - 2
> > > > > > 
> > > > > > In this example incoming traffic is distributed between 
> > > ASP1 and ASP2.
> > > > > > 
> > > > > > IN ORDER TO HANDLE THE HEAVY INCOMING TRAFFIC, SGP1 
> > > NEEDS MINIMUM OF
> > > > > > 2 ACTIVE ASPs BY SETTING n=2.
> > > > > 
> > > > > Do they not comply to SS7 ISUP/SCCP flow and congestion 
> > > controls?  Are they
> > > > > running at full capacity in this arrangement?  Are these 
> > > your ASPs?  Do you
> > > > > design your systems this way?
> > > > > 
> > > > > Take a look at the reliability engineering.  If either of 
> > > these ASPs is
> > > > > handling its maximum rated capacity, it will have to be 
> > > reliable to .99999
> > > > > This is because, if one ASP fails, 50% of the system has 
> > > failed.  A 50% system
> > > > > failure is deemed a "system" failure rather than a 
> > > "component" failure in
> > > > > reliability engineering for carrier grade systems.  
> > > Considering proscribed
> > > > > replacement times on the ASP, this would be difficult to 
> > > achieve unless both
> > > > > ASPs were hardware fault tolerant platforms.
> > > > > 
> > > > > Nevertheless, the RFC defines ASPs as a process, not a 
> > > processor.  There is
> > > > > nothing restricting the two ASPs in your diagram from 
> > > being processes on the
> > > > > same processor.  In that case, when ASP2 is down, ASP1 
> > > has double the
> > > > > processing capacity as it had before.
> > > > > 
> > > > > You certainly cannot make statements as you have made 
> here without
> > > > > presupposing an implementation architecture outside the 
> > > scope of the standard.
> > > > > 
> > > > > Besides, the RFC permits you to configure your SGP (and 
> > > why do you only have
> > > > > one of them?) so that it waits for 'n' ASPs to become 
> > > active.  You MAY do
> > > > > this.  Just don't require everybody else to follow your 
> > > approach.  (Actually,
> > > > > if the above is your approach, don't require anyone else 
> > > to follow it.)
> > > > > 
> > > > > Several other points: ANSI T1.111.4/2000 SLS values are 
> > > 8-bits.  That means
> > > > > there are 256 SLS values (gee, I had 256 ASPs in my example).
> > > > > 
> > > > > For another thing, sequenced traffic flows are not only 
> > > based on SLS.  In an
> > > > > ITU-T network, each remote signalling point has 16 
> > > independent traffic flows
> > > > > for each AS.  The traffic flows from each remote 
> > > signalling point is
> > > > > independent from the traffic flows of other remote 
> > > signalling points.
> > > > > Therefore, if there are 400 remote signalling points, 
> > > there will be 6400
> > > > > independent traffic flows that the SG can use for 
> > > distribution of AS traffic
> > > > > for one AS over as many as 6400 ASPs without risking 
> > > mis-sequencing of
> > > > > messages.  For an ANSI T1.111.4/2000 network using 8-bit 
> > > SLS values with 400
> > > > > remote signalling points, there would be 102400 
> > > independent traffic flows to
> > > > > the AS that the SG can use to distribute traffic to one 
> > > AS over as many as
> > > > > 102400 ASPs without risking message mis-sequencing.
> > > > > 
> > > > > > 
> > > > > > 
> > > > > > Your example with n=256 is completely a wrong 
> > > interpretaion of 'n+k'
> > > > > > redundancy case.
> > > > > 
> > > > > Quite the opposite, you clearly misunderstand redundancy 
> > > and reliability in
> > > > > carrier grade systems and obviously have little 
> > > experience with them.
> > > > > 
> > > > > Here's a little primer for you:
> > > > > 
> > > > > Carrier grade systems were originally designed as 
> > > hot-standby, fully redundant
> > > > > systems.  For every processor actively handling traffic, 
> > > there was a hot
> > > > > standby processor also actively handling the same 
> > > traffic.  This typically
> > > > > required special bus arrangements whereby the I/O bus of 
> > > one processor was
> > > > > actively controlling peripheral devices, while the hot 
> > > standby processor was
> > > > > issuing I/O bus output that was not terminated to the I/O 
> > > bus.  The bus also
> > > > > normally performed a comparison of the outputs from the 
> > > active side and the
> > > > > outputs from the inactive side.  Peripheral components, 
> > > such as switch fabric
> > > > > were also fully redundant (duplicated).  Switching 
> > > fabrics are separated into
> > > > > an A-plane and a B-plane in much the same fashion as the 
> > > SPCS front-end I/O
> > > > > bus.
> > > > > 
> > > > > Note that active/standby in M3UA does not address the 
> > > above model.  2+0
> > > > > broadcast does.
> > > > > 
> > > > > The problem with such a systems was what we used to call 
> > > the "kill your
> > > > > sister" syndrome.  A failure on the inactive side only 
> > > detectable to the I/O
> > > > > bus could cause a switch from the active to inactive 
> > > side, resulting in the
> > > > > switch running on the erroneous inactive side.  We called 
> > > it the "kill your
> > > > > sister" syndrome because it is not so surprising that the 
> > > psychotic sister
> > > > > (erred inactive side) kills the sane sister (correct 
> active side).
> > > > > 
> > > > > Primarily to avoid the "kill your sister" syndrome, some 
> > > switches were
> > > > > designed with 3 or 4 processors on an cross-checking I/O 
> > > bus pair.  In that
> > > > > case, it would take a 2 out of 3 or 3 out of 4 vote to 
> > > shift from the active
> > > > > processor.  These systems were far less susceptible to 
> > > inactive plane errors
> > > > > that were only detected at the I/O bus.
> > > > > 
> > > > > Note that the only M3UA model that address the above is 
> > > the 3+0 or 4+0
> > > > > broadcast model.
> > > > > 
> > > > > When the use of general purpose computing platforms for 
> > > telecommunications
> > > > > systems started to become popular, there were several 
> > > models that emerged.
> > > > > 
> > > > > First was taking two big iron UNIX hosts and running them 
> > > in active/standby.
> > > > > In this scenario, one UNIX host would be actively 
> > > handling traffic and the
> > > > > other would be warm standby.  By "warm" standby, it means 
> > > that the device was
> > > > > powered and was synchronizing state on a checkpoint 
> > > interval or monitoring the
> > > > > health of the active hosts with heartbeats.  Shared 
> > > memory, shared DASD or
> > > > > message passing are the most common mechanisms of 
> > > redundancy.  These systems
> > > > > are considered "highly available" systems rather than the 
> > > "fault tolerant"
> > > > > systems used in telecommunications above.  An increased 
> > > level of reliability
> > > > > is often achieved by using hardware fault tolerant 
> > > multiple processor hosts.
> > > > > 
> > > > > In M3UA, the active/standby model addresses these 
> > > systems.  Although M3UA
> > > > > provides for a '1+k' model, k is invariably 1.
> > > > > 
> > > > > The major difficulties with these systems are: a) 
> > > switchover latency between
> > > > > systems; b) loss of state during switchover; and c) 
> > > scalability.  Switchover
> > > > > latency is normally improved by decreasing heartbeat 
> > > intervals.  State
> > > > > synchronization is normally improved by decreasing the 
> > > process checkpoint
> > > > > interval.  These reductions typically increase cost and 
> > > they are normally
> > > > > chosen to minimize a cost/performance/reliability metric. 
> > >  Scalability was
> > > > > typically performed by installing larger more powerful 
> > > hosts with increased
> > > > > demand.  This was not good, because larger processors 
> > > tended to have poorer
> > > > > cost/performance ratios that smaller processors.
> > > > > 
> > > > > Later it was determined that the cost of a complete 
> > > redundant host that was
> > > > > wasting CPU cycles while inactive could be avoided while 
> > > addressing the
> > > > > deficiencies of the paired host approach above.  To 
> increase state
> > > > > synchronization, state could be spread over a number of 
> > > hosts.  If there were
> > > > > 'm' hosts, the failure of one host could at worst affect 
> > > a proportion, 1/m, of
> > > > > the state.  Therefore state synchronization would improve 
> > > as m increased.  To
> > > > > distribute the state, traffic load would be shared over 
> > > the m hosts according
> > > > > to state.  This also served the needs of scalability and 
> > > provided for closer
> > > > > to linear cost scalability, because the number of hosts, 
> > > m, could be increased
> > > > > while holding the cost/performance ratio of the hosts 
> > > constant.  The costs of
> > > > > redundancy could also be reduced over the fully redundant 
> > > systems described
> > > > > above, because redundancy could be achieved using 'm+1' 
> > > processors, where 'm'
> > > > > processors were required to handle the maximum engineered 
> > > load.  For systems
> > > > > as small as 'm=2' hosts with equivalent reliability to 
> > > that normally
> > > > > experienced in the '1+1' active/standby scenario above, 
> > > system reliability
> > > > > figures could be maintained with less redundancy cost.  
> > > It is important to
> > > > > note, that in these systems, all 'm+1' hosts are active 
> > > and handling only a
> > > > > proportion m/(m+1) of their rated capacity.  It is very 
> > > seldom that the rated
> > > > > capacity was more that 2/3 of the maximum short-term capacity.
> > > > > 
> > > > > These systems do not solve the heartbeat interval 
> problem of the
> > > > > active/standby paired hosts system above.  Typically what 
> > > these systems use is
> > > > > more sophisticated watchdog approaches rather than 
> > > passing heartbeat messages
> > > > > between processes on the active and standby hosts.  
> > > Memory sharing, DASD
> > > > > sharing, message passing, reliable broadcast, voting, and 
> > > a plethora of
> > > > > reliable distributed systems techniques have been 
> > > developed that are useful in
> > > > > these systems.
> > > > > 
> > > > > Adoption of these latter systems changed the requirement 
> > > from "duplicated"
> > > > > systems to systems with no "single point of failure".
> > > > > 
> > > > > Now, the 'n+k' load-sharing and broadcast model was 
> > > intended to address these
> > > > > systems.  Where 'n == m' in the description above.  That 
> > > is why we agreed that
> > > > > n is the maximum engineered load.  Because that is what 
> > > 'm' is.  The k is the
> > > > > 1 in the description above, generalized.
> > > > > 
> > > > > First off, you must understand that it is not the 
> > > intention of the model to
> > > > > have 'n' ASPs active and 'k' inactive unless 'n=1, k=1'.  
> > > This is described in
> > > > > section 1.4.4.1:
> > > > > 
> > > > >   A "1+1" active/standby redundancy is a subset of this model.
> > > > > 
> > > > > Also no redundancy is possible:
> > > > > 
> > > > >   A simplex "1+0" model is also supported as a subset, 
> > > with no ASP redundancy.
> > > > > 
> > > > > (The term "simplex" come from the telephone switches 
> > > described above where the
> > > > > inactive side is taken off-line.  Such a situation is 
> > > described as running the
> > > > > switch "simplex".)
> > > > > 
> > > > > The 'n+k' model is to address the 'm+1' distributed 
> > > systems described last.
> > > > > Rereading section 1.4.4.1:
> > > > > 
> > > > >   The failover model supports an "n+k" redundancy model, 
> > > where "n" APSs is the
> > > > >   minimum number of redundant [sic] ASPs required to 
> > > handle traffic and "k"
> > > > >   ASPs are available to take over for a failed or 
> unavailable ASP.
> > > > > 
> > > > > Note that the state of the "k" ASPs with respect to an AS 
> > > is not stated.  In
> > > > > the 'm+1' redundant system described above, the "k" ASPs 
> > > are active for the
> > > > > AS.  "k" ASPs active in load sharing mode are "available 
> > > to take over for a
> > > > > failed or unavailable ASP".
> > > > > 
> > > > > It is typical to run the "k" ASPs active (and I know you 
> > > had some questions
> > > > > about activating the "k" ASPs, but the point is that they 
> > > are normally already
> > > > > active), and handling their own share of traffic.  In 
> > > this case, each ASPs
> > > > > handles n/(n+k) of their engineered capacity.  Were you 
> > > actually considering
> > > > > deploying such a atypical model where the 'k' ASPs 
> are inactive?
> > > > > 
> > > > > "k" ASPs can be taken out of service, upgraded, and 
> > > brought back into service
> > > > > without affecting the maximum engineering traffic 
> > > handling capacity of the
> > > > > system (in absence of failures during the upgrade).  This 
> > > assists in graceful
> > > > > software/hardware upgrades on live systems.  (Keeping 
> > > them inactive would
> > > > > serve no purpose and would, decrease maximum traffic 
> > > handling capability,
> > > > > increase loss of state on fail-over, and increase 
> > > fail-over latency: all good 
> > > > > reasons for not deactivating the 'k' ASPs).
> > > > > 
> > > > > Now, reliability engineering for carrier class systems 
> > > takes the view that if
> > > > > 50% of the load bearing components in a system fail (stop 
> > > operating in any
> > > > > way), then the entire system has failed.  If traffic is 
> > > stopped above n/2
> > > > > processes, the system will have failed prematurely.
> > > > > 
> > > > > A requirement to withhold traffic < n will always suffer 
> > > a prolonged system
> > > > > failure, because only n/2 are required to make the system 
> > > available.
> > > > > Therefore, on startup the reliability of the system is 
> > > degraded by the
> > > > > negative exponent n/2, regardless of whether n=256 or 
> > > n=2; however, increasing
> > > > > 'n' increases the damage done to reliability by 
> waiting for 'n'.
> > > > > 
> > > > > A requirement for the SG to wait for 'n' is unacceptable 
> > > from the standpoint
> > > > > of the redundancy model that 'n+k' was meant to address, 
> > > as well as other
> > > > > possible models that it can accommodate.
> > > > > 
> > > > > > 
> > > > > > 
> > > > > > > 
> > > > > > > Consider the same example with: n=2, k=1, two ASPs failed.
> > > > > > > 
> > > > > > > Any system that introduces hysteresis to aggravate 
> > > and perpetuate an outage in
> > > > > > > this fashion is not very useful.
> > > > > > > 
> > > > > > > In the examples, what would have been a brief 
> > > AS-PENDING situation has been
> > > > > > > turned into a prolonged and complete system failure.  
> > > It is certainly not good
> > > > > > > practice to design redundant systems in that fashion 
> > > and recommending or
> > > > > > > requiring such a design would be a bad thing.
> > > > > > > 
> > > > > > > To further the example, consider the multiple SG as 
> > > STP situation.  One SG
> > > > > > > cannot know the state of ASPs for the its mated SG 
> > > and therefore cannot
> > > > > > > distinguish between the restart scenario and the 
> > > start scenario as far as the
> > > > > > > AS state is concerned because the SG does not know 
> > > the state of the AS itself:
> > > > > > > the AS could be active with other SGs.
> > > > > > > 
> > > > > > > This is the same case for SGP.  Unless AS state is 
> > > synchronized across SGPs,
> > > > > > > one SGP cannot know the total state of the AS (only 
> > > its partition).  There is
> > > > > > > no requirement to synchronize AS state across SGP (on 
> > > the SGP side, AS-state
> > > > > > > is per-SGP), and there is certainly no requirement to 
> > > synchronize AS-state
> > > > > > > across SGs.
> > > > > > > 
> > > > > > > So, take the (n=2, k=0) example where there are two 
> > > SGs.  Due to a network
> > > > > > > partitioning at startup, one ASP can communicate only 
> > > with one SG, and the
> > > > > > > other ASP can communicate only with the other SG.  
> > > Because n=2, neither SG
> > > > > > > will start traffic and the system suffers 100% 
> > > failure until the network
> > > > > > > partitioning ceases.
> > > > > > > 
> > > > > > > Consider the same example where there are two SGPs 
> > > (instead of SGs) and the
> > > > > > > two ASPs normally only form one association (e.g., 
> > > ASP1-SGP1, ASP2-SGP2).
> > > > > > > Note that there is no requirement in the RFC to form 
> > > full-mesh associations.
> > > > > > > Because the number of active ASPs with each SGP is < 
> > > 2, the system will never
> > > > > > > start.
> > > > > > > 
> > > > > > > These are not strange scenarios.  These situations 
> > > are all permitted by the
> > > > > > > RFC.  Yet, in each situation it is clear that one 
> > > cannot recommend that the SG
> > > > > > > wait.  The result would be a non-functional system.  
> > > There are many more
> > > > > > > examples.
> > > > > > > 
> > > > > > > n+k is a model for redundancy (not for congestion and 
> > > flow control).  A system
> > > > > > > with an SG that waits for n is, in general, not a 
> > > fully redundant system at
> > > > > > > all, but is, rather, prone to these failure syndromes.
> > > > > > > 
> > > > > > > When you look at the situation and apply the 
> > > experience of reliability
> > > > > > > engineering, the recommendation is that the SG SHOULD 
> > > start traffic when the
> > > > > > > first ASP goes active.
> > > > > > > 
> > > > > > > Why wait for 'n' anyway?  Because the ASP cannot 
> > > handle some level of traffic?
> > > > > > > Any ASP that conforms to MTP-User standards, must 
> > > have the standardized
> > > > > > > MTP-User (ISUP/SCCP) congestion and flow controls and 
> > > must have the SCTP
> > > > > > > congestion controls and is therefore able to accept 
> > > any amount of traffic
> > > > > > > overload.
> > > > > > > 
> > > > > > > So there is no real reason for the SG to wait for 
> > > 'n'.  Nevertheless, the RFC
> > > > > > > allows this as an OPTION.  Requiring (MUST) or 
> > > recommending it (SHOULD)
> > > > > > > regardless of whether the number of active processors 
> > > is rising or falling,
> > > > > > > whether start, restart or failure, is unacceptable.
> > > > > > > 
> > > > > > > On the other side, at the ASPs, the ASPs have a 
> > > requirement to share state
> > > > > > > between them.  They must share circuit state and 
> > > transaction state in all
> > > > > > > configurations to meet carrier grade.  This is stated 
> > > in the RFC.  It is
> > > > > > > implicit that the ASPs are capable of knowing with 
> > > which SGs, and which SGPs
> > > > > > > for which they are active.  They have access to the 
> > > total state of the AS that
> > > > > > > they collectively represent, that the SG/SGP does 
> > > not.  Therefore is it
> > > > > > > reasonable for the ASPs to coordinate among 
> > > themselves the number of active
> > > > > > > ASPs and to ensure that there are sufficient ASPs 
> > > active to handle the
> > > > > > > expected load before starting traffic.  Because the 
> > > ASPs need not respond to
> > > > > > > user test traffic (UPT/SST), the ASPs are also in 
> > > control of when remote SS7
> > > > > > > MTP-User restart traffic toward them.
> > > > > > > 
> > > > > > > Therefore, the recommendation that the ASPs SHOULD 
> > > coordinate among
> > > > > > > themselves the number of active ASPs and not start 
> > > traffic until a sufficient
> > > > > > > number of them are active is sound.  This is not a 
> > > MUST, permitting ASPs
> > > > > > > that know that they are capable of handling any 
> > > overload, to start traffic
> > > > > > > immediately, or at some earlier point.
> > > > > > > 
> > > > > > > Back to the RFC, the RFC sets the normal case where 
> > > the SG starts traffic on
> > > > > > > the first active ASP but explicitly permits as an 
> > > OPTION to wait until it
> > > > > > > considers that there are sufficient ASPs active to 
> > > handle the offered load.
> > > > > > > The RFC also RECOMMENDS that the ASPs coordinate 
> > > among themselves the number
> > > > > > > of active ASPs so that there are sufficient ASPs to 
> > > handle the expected load.
> > > > > > > 
> > > > > > > The provisions in the RFC are both sound, where the 
> > > proposal and section 3.15
> > > > > > > is clearly not.
> > > > > > > 
> > > > > > > Remember when looking at the AS state diagram in the 
> > > RFC that this is the
> > > > > > > SGP's view of the AS state and not the total state of 
> > > the AS, that can only
> > > > > > > known to the ASPs for all configurations.  'n' as 
> > > defined has no meaning to
> > > > > > > the state machine at the SGP because is not required 
> > > to know, and often cannot
> > > > > > > know, how many ASPs are active for the AS via other 
> > > SGPs and SGs, or even via
> > > > > > > direct connection to the SS7 network.
> > > > > > > 
> > > > > > > Some SGs might have special knowledge, or 
> > > sophisticated algorithms that allow
> > > > > > > it to determine whether a congestive failure will be 
> > > the immediate result of
> > > >  > > > starting or restarting traffic.  These SGs are 
> > > permitted as an
> > > > OPTION to wait.
> > > > > > > 
> > > > > > > The ASPs SHOULD have the knowledge of how many are 
> > > required to handle the
> > > > > > > expected load, because they must share 
> > > circuit/transaction state and they must
> > > > > > > have congestion and flow controls as required by the 
> > > standards.
> > > > > > > 
> > > > > > > So again, the RFC is adequate in this regard.  The 
> > > current proposal and
> > > > > > > section 3.15 are not.
> > > > > > RFC is inadequate so we need section 3.15 to 
> explain 'n+k' case.
> > > > > 
> > > > > Teaching reliability engineering and the design of 
> > > redundant systems is
> > > > > outside the scope of the RFC and the IG.  People, such as 
> > > yourself, are better
> > > > > advised to learn the subject matter before attempting to 
> > > implement or claim a
> > > > > complete understanding of the protocol.
> > > > > 
> > > > > The current RFC provides the protocol mechanisms and basis for
> > > > > interoperability for all of the common approaches to 
> > > redundancy in carrier
> > > > > class systems, and even provides a general model that can 
> > > accommodate
> > > > > non-traditional redundancy architectures.
> > > > > 
> > > > > When confused about how a redundancy architecture works, 
> > > don't blame the M3UA
> > > > > RFC, blame those responsible for describing redundant 
> > > architectures to you.
> > > > > 
> > > > > In the appendix we stated:
> > > > > 
> > > > >    M3UA imposes no restrictions as to the exact layout of 
> > > the network
> > > > >    elements, the message distribution algorithms and the 
> > > distribution of
> > > > >    signalling processes.  Instead, it provides a 
> > > framework and a set of
> > > > >    messages that allow for a flexible and scalable 
> > > signalling network
> > > > >    architecture, aiming to provide reliability and 
> performance.
> > > > > 
> > > > > Forcing SGs to wait for 'n' ASPs active before activating 
> > > is contrary to this.
> > > > > The existing "MAY" achieves this flexibility.
> > > > > 
> > > > > When reading 1.4.4.1, also remember this statement in the 
> > > appendix, that used
> > > > > to be beside 1.4.4.1.  The description of 'n+k' does not 
> > > dictate the state of
> > > > > the 'k' or the state of the AS with respect to 'n' to 
> > > leave the redundancy
> > > > > model flexible, and accommodate '1+1' and '1+0' in 
> the same model.
> > > > > 
> > > > > Also, please don't presume to tell me to set 'n' to 1, as 
> > > it is not a
> > > > > arbitrary value, but the number 'm' in the 'm+1' 
> > > distributed system in common
> > > > > use today.  This number is determined as the minimum 
> > > number of ASPs required
> > > > > to handle the maximum engineered load, and not, as Tolga 
> > > would have, the
> > > > > minimum number of ASPs required to handle ANY load.  If 
> > > you want your system
> > > > > to behave as Tolga proposes, just have your SG exercise 
> > > the "MAY choose not to
> > > > > direct traffic" option of section 4.3.4.3 of the RFC.
> > > > > 
> > > > > Section 3.15 and Tolga's proposal should be struck 
> > > because they attempt to
> > > > > impose a 'n' active 'k' inactive and system cannot 
> > > restart (3.15) or start
> > > > > (Tolga) without 'n' view on the redundancy model, where 
> > > it is clearly the
> > > > > intent of the RFC to place no such restrictions on the 
> > > model, but to
> > > > > accommodate a range of models, that is already achieved 
> > > by the RFC text.
> > > > > 
> > > > > --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
> 
> -- 
> 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 exim@www1.ietf.org  Thu Jul 10 08:20:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA28640
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 08:20:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aaP4-0002YO-Os
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 08:20:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ACK2ck009809
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 08:20:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aaP3-0002Y1-Nl; Thu, 10 Jul 2003 08:20:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aaP1-0002XQ-LW
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 08:19:59 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA28634
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 08:19:56 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aaP0-0002Bz-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 08:19:58 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[jq++wF957p8jQxfZ1WmUvkTFqJ029iO7])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aaOz-0002Bq-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 08:19:57 -0400
Received: from ns.pigworks.openss7.net (IDENT:LCIHHCZMvbEgysW61nYORG8IsRcGOTUI@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6ACJuX25316;
	Thu, 10 Jul 2003 06:19:56 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6ACJtU16819;
	Thu, 10 Jul 2003 06:19:55 -0600
Date: Thu, 10 Jul 2003 06:19:55 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04
Message-ID: <20030710061955.A16716@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF0@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: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF0@eestqnt105.es.eu.ericsson.se>; from j.javier.pastor@ericsson.com on Thu, Jul 10, 2003 at 02:02:53PM +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>

J.Javier,

J.Javier Pastor (EE/EEM) wrote:                                                   (Thu, 10 Jul 2003 14:02:53)
> Brian, 
> 
> Pls see my comments inline...
> 
> Cheers // Javier.
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: jueves, 10 de julio de 2003 13:37
> > To: J.Javier Pastor (EE/EEM)
> > Cc: 'sigtran@ietf.org'
> > Subject: Re: [Sigtran] M3UA IG v04
> > 
> > 
> > Javier,
> > 
> > Your "new possibility" is the norm in the RFC.  You seem to 
> > want reverse
> > the sense of the RFC.  The RFC says that we activate on the 
> > 1st ASP and
> > MAY withhold until the 'n'th ASP.  3.14 says that we SHOULD 
> > withhold to
> > the 'n'th ASP and MAY activate on the first.
> > 
> It is a "new possibility" considering the base of the rough consensus it was
> reached in the mailing list regarding the AS States. The introduction of
> smooth start was in order to have all the opinions included.
> 
> It's nothing about reversing the RFC. It is simply that there are several
> understandings of what the RFC says and we should clarify all the conflicts.
> It seems that most people have the understanding of "n" ASPs are needed as a
> general rule...
> 
> > The reason for this change escapes me.  There is no justification to
> > reverse the option.
> 
> Just the consensus in the mailing list. 

That is not correct.  The history of consensus on the mailing list
as I have already detaied is that the RFC should remain as it is.

Please point to the mails where you feel this consensus was acheived.

> 
> > 
> > Please, answer me this:
> > 
> >   Who has an ASP that is so broken that it needs the SGP to wait
> >   for 'n' of them to become active?
> > 
> >   Please identify yourselves!

Do you have an ASP that needs this reversal?

Who does?  Please, can anyone on the mailing list identify themselves
as having an ASP that requires the SGP to wait?  Who are you?

Please identify yourselves!

> > 
> > 
> > There are several other places in the RFC that assume 
> > activation on the
> > first ASP.  It is sewn through and through the RFC that 
> > loadsharing and
> > broadcast is performed across the active APSs (regardless of n).
> 
> The RFC has different views on this, and this is the point that we needed to clarify.

I thought you just said that you were not reversing the RFC?

Which is it?

The RFC clearly states in these places that if there is one ASP
active in loadshare or broadcast, it receives traffic, but that
the SGP MAY wait.

You are changing that.

--brian

> 
> 
> > 
> > Section 3.7.1:
> >                                                   ...  In Loadshare
> >       mode, the ASP will share in the traffic distribution with any
> >       other currently active ASPs.  In Broadcast mode, the ASP will
> >       receive the same messages as any other currently active ASP.
> > 
> > (Notice the word "any", twice, above.)
> > 
> > Section 4.1.1
> > 
> >    From the list of ASPs within the AS table, an ASP in the ASP-ACTIVE
> >    state is selected and a DATA message is constructed and 
> > issued on the
> >    corresponding SCTP association.  If more than one ASP is 
> > in the ASP-
> >    ACTIVE state (i.e., traffic is to be loadshared across 
> > more than one
> >    ASP), one of the ASPs in the ASP-ACTIVE state is selected from the
> >    list.  If the ASPs are in Broadcast Mode, all active ASPs will be
> >    selected and the message sent to each of the active ASPs.
> > 
> > (Note above that there is no 'n' restriction on selecting 
> > ASPs to receive
> > traffic.)
> > 
> > Section 4.3.4.3:
> > 
> >    In the case of a Loadshare mode AS, reception of an ASP Active
> >    message at an SGP or IPSP causes the direction of traffic 
> > to the ASP
> >    sending the ASP Active message, in addition to all the other ASPs
> >    that are currently active in the AS.
> > 
> > Section 4.3.4.4:
> > 
> >    In the case of a Loadshare mode AS, the SGP moves the ASP to the
> >    ASP-INACTIVE state and the AS traffic is reallocated across the
> >    remaining ASPs in the state ASP-ACTIVE, as per the loadsharing
> > 
> > 
> > Also, note that there is no limit to T(r).  So, if it was set 
> > infinite,
> > or just rather large, the start case is transformed to the 
> > restart case.
> > 
> > 
> > Changing the trantion from the AS-DOWN to AS-INACTIVE state at the SGP
> > to wait for 'n' ASP-INACTIVE ASPs is even more pointless than waiting
> > for 'n' ASP-ACTIVE ASPs.  No traffic is flowing.  The RFC 
> > always has us
> > move to the AS-INACTIVE state on the first ASP-INACTIVE ASP.
> > 
> > What in the world did you think would necessitate changing this?
> > 
> 
> If you have any other proposal that reach consensus and clarify the RFC
> regarding this concern, we will place it in the IG.

I believe that you placed it in the IG without consensus in the first place.
Where did we discuss this at all?  Can you point to an email or thread
discussing this topic?  Who proposed it?  When?  When was it agreed?  By who?

--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 exim@www1.ietf.org  Thu Jul 10 08:23:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA28728
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 08:23:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aaRx-0002ue-PA
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 08:23:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ACN1RT011186
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 08:23:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aaRx-0002uB-Gp; Thu, 10 Jul 2003 08:23:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aaRZ-0002ti-3i
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 08:22:37 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA28674
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 08:22:33 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aaRX-0002CV-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 08:22:35 -0400
Received: from zdemail03.zdem.compaq.com ([161.114.112.27])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aaRW-0002CS-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 08:22:34 -0400
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail03.zdem.compaq.com (Postfix) with ESMTP
	id DF6B2132C; Thu, 10 Jul 2003 14:21:57 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Thu, 10 Jul 2003 14:21:57 +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="us-ascii"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] M3UA IG v04
Date: Thu, 10 Jul 2003 14:21:56 +0200
Message-ID: <06CAA4D3753C53408D2CE6340B68444214C466@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA IG v04
Thread-Index: AcNGz2a6xfdhesJeTAaydjbWmS0SEwADAmzQ
From: "Zannin, Francois" <francois.zannin@hp.com>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: "Ken Morneault" <kmorneau@cisco.com>,
        "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
        <sigtran@ietf.org>
X-OriginalArrivalTime: 10 Jul 2003 12:21:57.0737 (UTC) FILETIME=[DB4B5590:01C346DD]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Let's continue on your example,

When an ASP starts, it sends an ASP-UP, and then receives an ASP-UP-ACK,
Then two possibilities,=20
    1) Sends an ASP-Active while receiving the NOTIFY xxx
    2) Receive the NOTIFY xxx and then sends the ASP-Active

If there is not enough ASP active for the AS, the SG(p) answer with an
ASP-Active_Ack and then everything is fine, the SG(p) start traffic for
the AS.
If there are enough ASP Already Active for the AS, the SG(p) do not send
the ASP-Active_Ack. Then this "cleaver" ASP will continue to send
ASP-Active and stop after some time (x retry of the Active).

If an "Active ASP" fails, then the SG sends an "Insufficient ASP Active"
that can be used by all "Inactive ASP" to restart the Activate procedure
until the number is ASP is sufficient.

With this kind of solution, No "new m3ua messages", no waste of time in
sending eternally ASP_Active, This is not the role of the SG to chose
which ASP have to be used base on notification, I guess this can be an
"amazing" solution!

Regards,
Francois.

-----Original Message-----
From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]=20
Sent: Thursday, July 10, 2003 12:36 PM
To: Zannin, Francois
Cc: Ken Morneault; J.Javier Pastor (EE/EEM); sigtran@ietf.org
Subject: RE: [Sigtran] M3UA IG v04



Zannin,

  Just an example, say an operator configures an AS1 as set of 3 ASPes.
  They are namely known as ASP1,ASP2 and ASP3(backup).

  Initially ASP1 and ASP2 send ASP_ACTIVE to SGP  so SGP sends a
  NTFY(AS_ACTIVE) thus results in AS1 to be in ACTIVE state.

  Where is the question of sending NTFY(Sufficient ASPs active)?
  People mean to say that backup ASP also sends an ASP_ACTIVE message
  to SGP? if it so, on what condition it is being sent? Is there
  any valid conditions that backup ASP or some new(!!!??!amazing!!) ASP
  to send ASP_ACTIVE message?=20


--Samuel=20
=20

  Alcatel USA, 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, 9 Jul 2003, Zannin, Francois wrote:

> Ken,
>=20
> Here is a copy of the mail about notify I was refering.
>=20
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Wednesday, February 12, 2003 10:40 AM
> To: ksrinivas
> Cc: sigtran@ietf.org
> Subject: [Sigtran] Re: M3UA Notify message
>=20
>=20
> Srini,
>=20
> ksrinivas wrote:
> (Wed, 12 Feb 2003 13:42:16)
> > Brian,
> >=20
> >      Can you tell me when an SGP should send Notify(AS-PENDING) and=20
> > Notify(Insufficient ASP Resources Active in AS)?. I felt RFC and IG=20
> > of
>=20
> > M3UA is not clear on these two.
> >=20
> > --Srini
>=20
> It think that it is pretty straight-forward.
>=20
> 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=20
> zero, (as a result of ASP failures or deactivations); where "n" is the

> minimum number of active ASPs defined for the AS.  So this=20
> notification is only sent if "n" is greater than 1.
>=20
> NTFY(AS-PENDING) is sent for an AS by an SGP when the number of ASPs=20
> in the ASP-ACTIVE state for the AS falls to zero (as a result of ASP=20
> failures or deactivations).
>=20
> Although there is the notification to indicate when there is=20
> insufficent ASPs available for an AS, there is no notification that=20
> there are sufficient ASPs for an AS.  This causes problems for ASP=20
> wishing to use the first notification as in indication to go active. =20
> There should be a NTFY("Sufficient ASPs active") to notify that the=20
> opposite of insufficient ASPs active is now true.
>=20
> I also send these two notifications to any ASP that transitions from=20
> 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=20
> expired. I also send other state notifications for the AS in this case

> as well (i.e, NTFY(AS-ACTIVE), NTFY(AS-INACTIVE)).
>=20
> As there is no procedure for a newly inactive ASP to audit the state=20
> of the AS, this helps ASPs which are just coming up for an AS to=20
> immediately determine the state of the AS and their next best course=20
> of action for the AS.
>=20
> In addition to notifications resulting from AS state changes, I think=20
> we should say in the IG:
>=20
>    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.
>=20
> --brian
>=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
> -----Original Message-----
> From: Ken Morneault [mailto:kmorneau@cisco.com]=20
> Sent: Wednesday, July 09, 2003 4:28 PM
> To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> Subject: RE: [Sigtran] M3UA IG v04
>=20
>=20
>=20
>=20
> Francois,
>=20
> Can you send me a ptr to the email you reference.
>=20
> Thanks,
>=20
> Ken
>=20
> At 04:20 PM 7/9/2003 +0200, Zannin, Francois wrote:
> >Ken,
> >
> >My point of view on he question, see hereafter.
> >
> >Regards,
> >
> >Francois Z.
> >
> >-----Original Message-----
> >From: Ken Morneault [mailto:kmorneau@cisco.com]
> >Sent: Wednesday, July 09, 2003 3:56 PM
> >To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> >Subject: RE: [Sigtran] M3UA IG v04
> >
> >
> >Francois,
> >
> >Pls see comments below.
> >
> >Regards,
> >
> >Ken
> >
> >At 10:15 AM 7/9/2003 +0200, Zannin, Francois wrote:
> >>Just some comment about the M3UA IG v04
> >>
> >>3.21 NOTIFY messages are missing in Examples section
> >>Figure (section 5.1.2)
> >>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after=20
> >>the ASP Up Ack
> >>
> >>Figure (section 5.1.3)
> >>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after=20
> >>the ASP Up Ack
> >>
> >>Figure (section 5.1.4)
> >>        Same remark, the Notify is sent to ASP1, ASP2 and also need=20
> >>to be sent to
> >>      ASP3 after the ASP Up Ack
> >
> >NOTIFY is only sent upon "a change in status of a particular
> >Application Server".
> >
> >OK, but just one more remark, in section 5.1.4, a notify AS-Inactive=20
> >is
>=20
> >sent to ASP1 and ASP2 while to transition to ASP-Inactive was=20
> >effective
>=20
> >after the first
> >ASP_UP(/ACK) comming from ASP1, so notify to ASP2 should be=20
> >considerer as inopportune.
> >
> >In last november, it was agreed in a thread that when an ASP was
> >transitionned to UP, The current AS state was sent to this ASP,=20
> >whatever could be the state. This current
> >State can be used by the management of an ASP to decide if the AS has
> to
> >be activated
> >Immediately or must wait for an AS-PENDING or AS-INACTIVE
notification.
>=20
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
>=20
>=20


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



From exim@www1.ietf.org  Thu Jul 10 08:37:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA29059
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 08:37:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aafW-0003XY-7O
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 08:37:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ACb214013588
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 08:37:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aafV-0003Wn-Cd; Thu, 10 Jul 2003 08:37:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aaej-0003Pu-SW
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 08:36:13 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA29034
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 08:36:10 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aaei-0002IH-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 08:36:12 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[7q01ajn58ljvZWtMGfh2z2D4SO/2alUq])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aaeh-0002IE-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 08:36:12 -0400
Received: from ns.pigworks.openss7.net (IDENT:35NcklBICWWBI/8Zfj6wmk8NnGgQtwvA@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6ACaAX25596;
	Thu, 10 Jul 2003 06:36:10 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6ACaAJ17261;
	Thu, 10 Jul 2003 06:36:10 -0600
Date: Thu, 10 Jul 2003 06:36:10 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04
Message-ID: <20030710063610.B16716@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CCA@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: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CCA@eestqnt105.es.eu.ericsson.se>; from j.javier.pastor@ericsson.com on Thu, Jul 03, 2003 at 11:45:57AM +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>

Javier,

'n' is defined as the number of ASPs that are needed to handle
the maximum engineered load of the AS.

This is the total load from the network correct?  As in the
traffic delivered to the AS, from all SGP, all SG.  Correct?

What has that to do with the number of ASPs active for particular
SGP?

There can be more than on SGP.  Correct?

There can be more than one SG.  Correct?

An ASP can be active for any number of SGPs in any number of SGs.  Correct?

SGPs in an SG should be able to transfer traffic between them.  Correct?

In general, what has the number of ASPs active for one SGP have
to do with the traffic handling capacity of an AS?


Consider the following (n+k == 2+0): (Figure 5).
Consider first that SGP1 and SGP2 are in the same SG.


     ________                  ________   
____|        |________________|        |
    |  SGP1  |        ________|  ASP1  |
    |________|       /        |________|
     ________       /          ________
    |        |_____/          |        |
____|  SGP2  |________________|  ASP2  |
    |________|                |________|


3.14 tells me that I SHOULD NOT direct traffic from the
newtork from SGP1 to ASP1 on startup of ASP1 with SGP1.

Does that include passing traffic from SGP1 via SGP2?

Is this because ASP1 cannot handle the load?

What should SGP1 do?

Now, consider that SGP1 and SGP2 are in different SGs.

Should SGP1 tell the network that the AS is inaccessible?

Should SGP1 direct traffic over C-links to SGP2 that will
ultimately be the same a passing traffic to ASP1?

Why should not SGP1 direct traffic to ASP1?

I don't think that 3.14 is very clear.  'n' has nothing to do
with the handling capacity from a given SGP to an AS.  It has
to do with the overall engineered maximum traffic that 'n' ASPs
can handle for the AS.  In even the more trivial cases the SGP
cannot follow the recommendation.

Please explain.

--brian

Javier Pastor-Balbas (ECE) wrote:                                                   (Thu, 03 Jul 2003 11:45:57)
> 
> Hi guys,
> 
> We tried to submit the next IG version before the deadline, but I got
> confused since I thought it ended at 12:00 instead of 9:00. So the server
> bounced it. I'll submit the doc again after the IETF meeting. In the
> meantime you can find it at:
> 
> http://standards.ericsson.net/javier/draft-ietf-sigtran-m3ua-implementors-guide-04.txt
> 
> We've tried to include all the different opinions in here. Although it seems
> that the case of waiting till n ASPs is the preferred one, we have included
> the new possibility of a smooth start for those of you who want to activate
> the AS when the first ASP gets to ACTIVE. At the end it will be the
> operator's decission how to proceed during the activation phase. Hopefully
> everyone agrees on this wording...
> 
> 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 exim@www1.ietf.org  Thu Jul 10 08:53:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA29957
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 08:53:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aav2-000537-1l
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 08:53:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ACr4L2019408
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 08:53:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aauz-00052e-J0; Thu, 10 Jul 2003 08:53:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aauX-000524-Ec
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 08:52:33 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA29932
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 08:52:29 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aauV-0002WE-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 08:52:31 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[rcHS1VKReUk3aSh6zNODuaqt7/JtVPoG])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aauU-0002W7-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 08:52:30 -0400
Received: from ns.pigworks.openss7.net (IDENT:mK++lCnk5yHwSDGTQIbv3SaGp3Z8zqet@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6ACqSX25798;
	Thu, 10 Jul 2003 06:52:28 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6ACqSw17456;
	Thu, 10 Jul 2003 06:52:28 -0600
Date: Thu, 10 Jul 2003 06:52:27 -0600
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>,
        Ken Morneault <kmorneau@cisco.com>,
        "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
        sigtran@ietf.org
Subject: Re: [Sigtran] M3UA IG v04
Message-ID: <20030710065227.C16716@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>,
	Ken Morneault <kmorneau@cisco.com>,
	"J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
	sigtran@ietf.org
References: <06CAA4D3753C53408D2CE6340B68444214C466@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: <06CAA4D3753C53408D2CE6340B68444214C466@vbeexc02.emea.cpqcorp.net>; from francois.zannin@hp.com on Thu, Jul 10, 2003 at 02:21:56PM +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>

Zannin,,

Zannin, Francois wrote:                                                            (Thu, 10 Jul 2003 14:21:56)
> Let's continue on your example,
> 
> When an ASP starts, it sends an ASP-UP, and then receives an ASP-UP-ACK,
> Then two possibilities, 
>     1) Sends an ASP-Active while receiving the NOTIFY xxx
>     2) Receive the NOTIFY xxx and then sends the ASP-Active
> 
> If there is not enough ASP active for the AS, the SG(p) answer with an
> ASP-Active_Ack and then everything is fine, the SG(p) start traffic for
> the AS.
> If there are enough ASP Already Active for the AS, the SG(p) do not send
> the ASP-Active_Ack. Then this "cleaver" ASP will continue to send
> ASP-Active and stop after some time (x retry of the Active).

I don't think it would be a good idea to ignore the ASP-Active.  The SGP
has three choices on receiving an ASP-Active: send ASP Active Ack,
send Error("Management Blocking") or send ASP Inactive Ack.  Discarding
the ASP Active and causing the ASP to fall back on OPTIONAL timers is
not a good idea.

Also, it is not a good idea for the SGP to attempt to decide which ASP
is to be activated in this way and which one is to be denied.  There
are dependencies with ASPs going active due to the state that they must
share.  For example: this might not be a possible situation from the
ASP's perspective:

     SGP1 - ASP1 Active
     SGP1 - ASP2 Active
     SGP1 - ASP3 Inactive

     SGP2 - ASP1 Inactive
     SGP2 - ASP2 Active
     SGP2 - ASP3 Active

Because this might force the ASPs to inefficiently share call state.

The RFC warns (Appendix A) that the loadsharing scheme might need to send all
messages pertaining to the same call or transaction to the same ASP under
normal operation.  This might not be possible in the above arrangement.

> If an "Active ASP" fails, then the SG sends an "Insufficient ASP Active"
> that can be used by all "Inactive ASP" to restart the Activate procedure
> until the number is ASP is sufficient.

The RFC also warns in section 4.3.4.5: (Notify Procedures)

   In the case where a Notify message ("AS-PENDING") message is sent by
   an SGP that now has no ASPs active to service the traffic, or where a
   Notify ("Insufficient ASP resources active in AS") message is sent in
   the Loadshare or Broadcast mode, the Notify message does not
   explicitly compel the ASP(s) receiving the message to become active.
   The ASPs remain in control of what (and when) traffic action is
   taken.

It sounds like you are expecting the opposite: that sending the
NTFY("Insufficient ASP resources active in AS") is expected to compell
the ASPs to initiate the ASP Active Procedure.

> 
> With this kind of solution, No "new m3ua messages", no waste of time in
> sending eternally ASP_Active, This is not the role of the SG to chose
> which ASP have to be used base on notification, I guess this can be an
> "amazing" solution!

Yes, amazingly bad.  This is precisely one of the reasons we said:

   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.

The ASPs know which one of them is active and which one is backup for
any given call/trasaction state.  They should decide which are active
and which are not.

The SGP should only refuse ASP Active under management blocking or failure.

--brian

> 
> Regards,
> Francois.
> 
> -----Original Message-----
> From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com] 
> Sent: Thursday, July 10, 2003 12:36 PM
> To: Zannin, Francois
> Cc: Ken Morneault; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> Subject: RE: [Sigtran] M3UA IG v04
> 
> 
> 
> Zannin,
> 
>   Just an example, say an operator configures an AS1 as set of 3 ASPes.
>   They are namely known as ASP1,ASP2 and ASP3(backup).
> 
>   Initially ASP1 and ASP2 send ASP_ACTIVE to SGP  so SGP sends a
>   NTFY(AS_ACTIVE) thus results in AS1 to be in ACTIVE state.
> 
>   Where is the question of sending NTFY(Sufficient ASPs active)?
>   People mean to say that backup ASP also sends an ASP_ACTIVE message
>   to SGP? if it so, on what condition it is being sent? Is there
>   any valid conditions that backup ASP or some new(!!!??!amazing!!) ASP
>   to send ASP_ACTIVE message? 
> 
> 
> --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, 9 Jul 2003, Zannin, Francois wrote:
> 
> > Ken,
> > 
> > Here is a copy of the mail about notify I was refering.
> > 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: Wednesday, February 12, 2003 10:40 AM
> > To: ksrinivas
> > Cc: sigtran@ietf.org
> > Subject: [Sigtran] Re: M3UA Notify message
> > 
> > 
> > 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
> > -----Original Message-----
> > From: Ken Morneault [mailto:kmorneau@cisco.com] 
> > Sent: Wednesday, July 09, 2003 4:28 PM
> > To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA IG v04
> > 
> > 
> > 
> > 
> > Francois,
> > 
> > Can you send me a ptr to the email you reference.
> > 
> > Thanks,
> > 
> > Ken
> > 
> > At 04:20 PM 7/9/2003 +0200, Zannin, Francois wrote:
> > >Ken,
> > >
> > >My point of view on he question, see hereafter.
> > >
> > >Regards,
> > >
> > >Francois Z.
> > >
> > >-----Original Message-----
> > >From: Ken Morneault [mailto:kmorneau@cisco.com]
> > >Sent: Wednesday, July 09, 2003 3:56 PM
> > >To: Zannin, Francois; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> > >Subject: RE: [Sigtran] M3UA IG v04
> > >
> > >
> > >Francois,
> > >
> > >Pls see comments below.
> > >
> > >Regards,
> > >
> > >Ken
> > >
> > >At 10:15 AM 7/9/2003 +0200, Zannin, Francois wrote:
> > >>Just some comment about the M3UA IG v04
> > >>
> > >>3.21 NOTIFY messages are missing in Examples section
> > >>Figure (section 5.1.2)
> > >>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after 
> > >>the ASP Up Ack
> > >>
> > >>Figure (section 5.1.3)
> > >>        a NOTIFY(AS-INACTIVE) must be sent from SGP to ASP2 after 
> > >>the ASP Up Ack
> > >>
> > >>Figure (section 5.1.4)
> > >>        Same remark, the Notify is sent to ASP1, ASP2 and also need 
> > >>to be sent to
> > >>      ASP3 after the ASP Up Ack
> > >
> > >NOTIFY is only sent upon "a change in status of a particular
> > >Application Server".
> > >
> > >OK, but just one more remark, in section 5.1.4, a notify AS-Inactive 
> > >is
> > 
> > >sent to ASP1 and ASP2 while to transition to ASP-Inactive was 
> > >effective
> > 
> > >after the first
> > >ASP_UP(/ACK) comming from ASP1, so notify to ASP2 should be 
> > >considerer as inopportune.
> > >
> > >In last november, it was agreed in a thread that when an ASP was
> > >transitionned to UP, The current AS state was sent to this ASP, 
> > >whatever could be the state. This current
> > >State can be used by the management of an ASP to decide if the AS has
> > to
> > >be activated
> > >Immediately or must wait for an AS-PENDING or AS-INACTIVE
> notification.
> > 
> > 
> > _______________________________________________
> > 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 exim@www1.ietf.org  Thu Jul 10 09:17:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA00670
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 09:17:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19abIF-0006mU-Ui
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 09:17:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ADH3qk026055
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 09:17:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19abIE-0006m0-21; Thu, 10 Jul 2003 09:17:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19abHM-0006k3-Pf
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 09:16:08 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA00630
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 09:16:04 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19abHK-0002ex-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 09:16:06 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[s73Dn5rwp2YQ+hijdPCfFhC8V+Ytak3i])
	by ietf-mx with esmtp (Exim 4.12)
	id 19abHK-0002et-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 09:16:06 -0400
Received: from ns.pigworks.openss7.net (IDENT:sEDaZpunZVIul/VbZ7OHCi69rm0NcdYe@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6ADG4X26366
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 07:16:04 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6ADG3L17864
	for sigtran@ietf.org; Thu, 10 Jul 2003 07:16:03 -0600
Date: Thu, 10 Jul 2003 07:16:03 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04 AS State Machine
Message-ID: <20030710071603.A17772@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "'sigtran@ietf.org'" <sigtran@ietf.org>
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CCA@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: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CCA@eestqnt105.es.eu.ericsson.se>; from j.javier.pastor@ericsson.com on Thu, Jul 03, 2003 at 11:45:57AM +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>

Javier,

The AS State Machine illustrated in 3.14 is per-SGP.  Correct?

SGP can communicate with each other in an SG.  Correct?

Why should an SGP pend when the last ASP goes ASP-INACTIVE for
an AS with a given SGP?

Cannot the SGP redirect traffic via another SGP that has ASPs
active for the AS?

Example:

        _______                    _______
       |       |______x    x______|       |
       | SGP1  |         _________| ASP1  |
       |_______|        /         |_______|
        _______        /           _______
       |       |______/           |       |
       | SGP2  |__________________| ASP2  |
       |_______|                  |_______|


When ASP1 goes inactive (or down) w.r.t SGP1, cannot
SGP1 direct traffic via SGP2?

What is the state of the AS at SGP1?

In section 3.14 AC2PN says that:

  AC2PN: the last ASP in the ASP-ACTIVE state moves to
  ASP-INACTIVE or ASP-DOWN states, causing the number of ASPs
  in the ASP-ACTIVE [sic] drop below 1.

Is that the number of ASPs for SGP1?  for SGP2? for the SG?
for all (both) SGs?

In the RFC the AS state machine was per-SGP.  Is the 3.14 state
machine per-SGP?

   IA2AC: one ASP moves to [the] ASP-ACTIVE [state], causing
   [the] number of ASPs in the ASP-ACTIVE state to be n.

When SGP1 receives an ASP Active and sends an ASP Active Ack
to ASP1, how many ASPs are active for the AS?   1?  2?

Please clarify.

--brian


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



From exim@www1.ietf.org  Thu Jul 10 09:22:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA00897
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 09:22:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19abN4-0006zO-Vy
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 09:22:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ADM2jX026845
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 09:22:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19abN3-0006yQ-Hk; Thu, 10 Jul 2003 09:22:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19abMX-0006xt-82
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 09:21:29 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA00841
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 09:21:24 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19abMV-0002h0-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 09:21:27 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[HMX7sbpabsuCfRPPiZIXcNhllbOSw9R3])
	by ietf-mx with esmtp (Exim 4.12)
	id 19abMU-0002gx-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 09:21:26 -0400
Received: from ns.pigworks.openss7.net (IDENT:qK5f06mKe7c5wU5bEhiElO/8+tvy0bwr@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6ADLPX26424;
	Thu, 10 Jul 2003 07:21:25 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6ADLPx17955;
	Thu, 10 Jul 2003 07:21:25 -0600
Date: Thu, 10 Jul 2003 07:21:24 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04 Section 3.14
Message-ID: <20030710072124.B17772@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CCA@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: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CCA@eestqnt105.es.eu.ericsson.se>; from j.javier.pastor@ericsson.com on Thu, Jul 03, 2003 at 11:45:57AM +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>

Javier,

   ---------
   New text: (Section 4.3.4.3, for both loadsharing and broadcast)
   ---------

   At start-up or re-start phases, an SGP or IPSP, upon reception of an
   ASP Active message for the first ASP in a Loadshare AS, SHOULD NOT
   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.

How can the SGP know when there are sufficient resources?

'n'?  How do I count ASPs active for comparison against 'n' at one SGP
in multiple SGs?

Please provide a definition of, and criteria for, sufficient resources to
go with the recommendation.

--brian



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



From exim@www1.ietf.org  Thu Jul 10 09:31:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA01310
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 09:31:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19abVl-0007kF-F2
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 09:31:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ADV1t0029765
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 09:31:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19abVl-0007jq-1w; Thu, 10 Jul 2003 09:31:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19abUo-0007Sw-Vw
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 09:30:03 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA01259
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 09:29:58 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19abUn-0002lN-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 09:30:01 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[MkYoEKp3qrk9bgbudcmJMyLH3opYI9r+])
	by ietf-mx with esmtp (Exim 4.12)
	id 19abUm-0002lD-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 09:30:00 -0400
Received: from ns.pigworks.openss7.net (IDENT:MMSwU1kBnXEHsWL+rtoj5qJZ9s95Lt3M@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6ADTwX26503;
	Thu, 10 Jul 2003 07:29:58 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6ADTwm18015;
	Thu, 10 Jul 2003 07:29:58 -0600
Date: Thu, 10 Jul 2003 07:29:58 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04 Section 3.8
Message-ID: <20030710072958.C17772@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CCA@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: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CCA@eestqnt105.es.eu.ericsson.se>; from j.javier.pastor@ericsson.com on Thu, Jul 03, 2003 at 11:45:57AM +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>

Javier,

Section 3.8 says:

   5.4 Auditing examples

   5.4.1 SG State: Uncongested / Unavailable

          ASP                          SGP
          ---                          ---
           |  -------- DAUD --------->  |
           |  <------ SCON(0) --------  |
           |  <------- DUNA ----------  |

What state of a destination is Uncongested/Unavailable?  There is no
such state in Q.704.  Uncongested routesets are by definition available.
Inaccessible routesets have no congestion state by definition.

Did you think that a routeset could be in an unavailble and congested state?

What is the use of the SCON(0)?  Why is it being sent?

   5.4.4 SG state: Uncongested / Unavailable

          ASP                          SGP
          ---                          ---
           |  -------- DAUD --------->  |
           |  <------- DUNA ----------  |

What is the difference between 5.4.1 and 5.4.4?  They both say
Uncongested/Available.



   [...]. Where the SGP maintains the congestion status of the SS7
   destination, the SGP MUST additionally respond with an SCON message
   before the DAVA or DRST message.  If the SS7 destination is

When is the SGP required to maintain congestion status?  What is the
criteria for an ASP to determine that an SGP should maintain congestion
status and is violating this MUST.  Why must the ASP require the SGP
to send SCON in this circumstance?  What interoperability concern is the
MUST addressing?

Please explain.

--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 exim@www1.ietf.org  Thu Jul 10 09:34:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA01410
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 09:34:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19abYf-00081z-OQ
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 09:34:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ADY1lk030850
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 09:34:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19abYf-000811-Bf; Thu, 10 Jul 2003 09:34:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19abYH-00080H-HO
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 09:33:37 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA01376
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 09:33:32 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19abYF-0002mv-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 09:33:35 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[VTrweewvKe++/o3ijc601L2Nc3TKUjpP])
	by ietf-mx with esmtp (Exim 4.12)
	id 19abYE-0002ms-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 09:33:35 -0400
Received: from ns.pigworks.openss7.net (IDENT:Fd5zdryFsh5JbUgXTO+SM222sKENWJ3s@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6ADXXX26636;
	Thu, 10 Jul 2003 07:33:33 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6ADXX218110;
	Thu, 10 Jul 2003 07:33:33 -0600
Date: Thu, 10 Jul 2003 07:33:33 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
Message-ID: <20030710073333.D17772@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CCA@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: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CCA@eestqnt105.es.eu.ericsson.se>; from j.javier.pastor@ericsson.com on Thu, Jul 03, 2003 at 11:45:57AM +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>

Javier,

3.18.3 Solution Description

   Clarify that the ASP needs to indicate when the congestion level has
   changed (including abatement).  Further, the ASP peer can maintain
   a timer, if desired, in case the ASP fails to indicate congestion
   abatement.

Where are the procedures for this timer described?  This is the only place
that the timer is mentioned.  Please describe the timer or remove the sentence
saying that it has been described...

--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 exim@www1.ietf.org  Thu Jul 10 09:34:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA01415
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 09:34:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19abYg-00082e-3C
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 09:34:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ADY268030890
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 09:34:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19abYf-000826-VM; Thu, 10 Jul 2003 09:34:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19abYc-00080e-Dm
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 09:33:58 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA01383
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 09:33:53 -0400 (EDT)
From: john.loughney@nokia.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19abYa-0002n6-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 09:33:56 -0400
Received: from mgw-x4.nokia.com ([131.228.20.27])
	by ietf-mx with esmtp (Exim 4.12)
	id 19abYZ-0002n3-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 09:33:55 -0400
Received: from esvir04nok.ntc.nokia.com (esvir04nokt.ntc.nokia.com [172.21.143.36])
	by mgw-x4.nokia.com (Switch-2.2.6/Switch-2.2.6) with ESMTP id h6ADXsk11610
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 16:33:54 +0300 (EET DST)
Received: from esebh002.NOE.Nokia.com (unverified) by esvir04nok.ntc.nokia.com
 (Content Technologies SMTPRS 4.2.5) with ESMTP id <T63597889feac158f24076@esvir04nok.ntc.nokia.com>;
 Thu, 10 Jul 2003 16:33:54 +0300
Received: from esebe002.NOE.Nokia.com ([172.21.138.17]) by esebh002.NOE.Nokia.com with Microsoft SMTPSVC(5.0.2195.6139);
	 Thu, 10 Jul 2003 16:33:52 +0300
Received: from esebe023.NOE.Nokia.com ([172.21.138.115]) by esebe002.NOE.Nokia.com with Microsoft SMTPSVC(5.0.2195.6139);
	 Thu, 10 Jul 2003 16:33:52 +0300
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"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Date: Thu, 10 Jul 2003 16:33:51 +0300
Message-ID: <DADF50F5EC506B41A0F375ABEB32063658F116@esebe023.ntc.nokia.com>
Thread-Topic: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Thread-Index: AcNBeNCN+OPV66NiRz2QUXHtMtAv4wAB2wyQAVENuwAACNlU0A==
To: <christophe.anzile@hp.com>
Cc: <sigtran@ietf.org>
X-OriginalArrivalTime: 10 Jul 2003 13:33:52.0482 (UTC) FILETIME=[E7152C20:01C346E7]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Hi,

That could be reasonable - anyone have problems with this?

John

> -----Original Message-----
> From: ext Anzile, Christophe [mailto:christophe.anzile@hp.com]
> Sent: 10 July, 2003 12:26
> To: Loughney John (NRC/Helsinki)
> Cc: sigtran@ietf.org
> Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
>=20
>=20
>=20
> Hi John,
>=20
> section 3.5.2 :=20
> it would be judicious to add an optional ASP-ID parameter in=20
> the UP-ACK message,
> especially for IPSP communication (think of single exchange).
>=20
> Regards,
> Christophe.
>=20

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



From exim@www1.ietf.org  Thu Jul 10 09:42:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA01764
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 09:42:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19abgR-0000Y6-5O
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 09:42:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ADg301002089
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 09:42:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19abgQ-0000XS-2k; Thu, 10 Jul 2003 09:42:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19abg3-0000X9-QM
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 09:41:39 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA01689
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 09:41:35 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19abg1-0002se-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 09:41:38 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[WRWk2iDLslytwkW0fg0/S3s5kfsjerqd])
	by ietf-mx with esmtp (Exim 4.12)
	id 19abg1-0002sb-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 09:41:37 -0400
Received: from ns.pigworks.openss7.net (IDENT:NJNkULcTpjc8Jcx1ugNdvGIjNvtLgavL@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6ADfaX26735;
	Thu, 10 Jul 2003 07:41:36 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6ADfaj18249;
	Thu, 10 Jul 2003 07:41:36 -0600
Date: Thu, 10 Jul 2003 07:41:35 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04 Section 3.9
Message-ID: <20030710074135.E17772@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CCA@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: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CCA@eestqnt105.es.eu.ericsson.se>; from j.javier.pastor@ericsson.com on Thu, Jul 03, 2003 at 11:45:57AM +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>

Javier,

Section 3.9 says:

          ASP                          SGP
          ---                          ---
           |  ------ ASPIA (RC1)----->  |
           |  <----  ASPIA Ack -------  |
           |  -----DEREG REQ (RC1)--->  |
           |  <----DEREG RSP (RC1)----  |
           |  -------ASPIA (RC1)----->  |

   What should SG do?

The new text does not explicity answer this question.

What does the SGP send in this case?

As RC values are per-SGP, it appears to me that it must be
Error("Invalid Routing Context").

Correct?

--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 exim@www1.ietf.org  Thu Jul 10 09:45:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA01879
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 09:45:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19abjJ-0000nd-NF
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 09:45:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ADj1AI003051
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 09:45:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19abjJ-0000n6-GX; Thu, 10 Jul 2003 09:45:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19abj6-0000mp-Ee
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 09:44:48 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA01867
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 09:44:43 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19abj4-0002vb-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 09:44:46 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[UlM6azQZ8lxu7zUbR1OdqDy3xT6ePTba])
	by ietf-mx with esmtp (Exim 4.12)
	id 19abj3-0002vY-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 09:44:45 -0400
Received: from ns.pigworks.openss7.net (IDENT:1nMM+qU5gWwbGFr8Aoq5MxfPGWHwSyAY@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6ADijX26765;
	Thu, 10 Jul 2003 07:44:45 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6ADii018278;
	Thu, 10 Jul 2003 07:44:44 -0600
Date: Thu, 10 Jul 2003 07:44:44 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: john.loughney@nokia.com
Cc: christophe.anzile@hp.com, sigtran@ietf.org
Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Message-ID: <20030710074444.F17772@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: john.loughney@nokia.com, christophe.anzile@hp.com,
	sigtran@ietf.org
References: <DADF50F5EC506B41A0F375ABEB32063658F116@esebe023.ntc.nokia.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: <DADF50F5EC506B41A0F375ABEB32063658F116@esebe023.ntc.nokia.com>; from john.loughney@nokia.com on Thu, Jul 10, 2003 at 04:33:51PM +0300
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,

We discussed this before an nobody had a real problem with it.

However, more than just adding the optional parameter to the
ASP Up Ack, it also needs to be described how one should use it.
Althoug a lot of uses were discussed, nobody put together a
coherent proposal for its use.

Does someone have a simple proposal for it use?

--brian

John Loughney wrote:                                                                  (Thu, 10 Jul 2003 16:33:51)
> Hi,
> 
> That could be reasonable - anyone have problems with this?
> 
> John
> 
> > -----Original Message-----
> > From: ext Anzile, Christophe [mailto:christophe.anzile@hp.com]
> > Sent: 10 July, 2003 12:26
> > To: Loughney John (NRC/Helsinki)
> > Cc: sigtran@ietf.org
> > Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
> > 
> > 
> > 
> > Hi John,
> > 
> > section 3.5.2 : 
> > it would be judicious to add an optional ASP-ID parameter in 
> > the UP-ACK message,
> > especially for IPSP communication (think of single exchange).
> > 
> > Regards,
> > Christophe.
> > 
> 
> _______________________________________________
> 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 exim@www1.ietf.org  Thu Jul 10 11:04:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA06127
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 11:04:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19acxp-0007be-8G
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 11:04:06 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AF45ZE029234
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 11:04:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19acxl-0007aC-HT; Thu, 10 Jul 2003 11:04:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19acx4-0007Oc-GH
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 11:03:18 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA06061
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 11:03:12 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19acx1-0003go-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 11:03:15 -0400
Received: from zdemail04.zdem.compaq.com ([161.114.112.28])
	by ietf-mx with esmtp (Exim 4.12)
	id 19acx0-0003gH-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 11:03:14 -0400
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail04.zdem.compaq.com (Postfix) with ESMTP
	id 284A2C35; Thu, 10 Jul 2003 17:02:36 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Thu, 10 Jul 2003 17:02:35 +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="us-ascii"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] M3UA IG v04
Date: Thu, 10 Jul 2003 17:02:34 +0200
Message-ID: <06CAA4D3753C53408D2CE6340B68444214C467@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA IG v04
Thread-Index: AcNG4mi2+BEf/W9xTXCWsphDXVEdSQAAMebQ
From: "Zannin, Francois" <francois.zannin@hp.com>
To: <bidulock@openss7.org>
Cc: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
        "Ken Morneault" <kmorneau@cisco.com>,
        "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
        <sigtran@ietf.org>
X-OriginalArrivalTime: 10 Jul 2003 15:02:35.0940 (UTC) FILETIME=[4C1C6640:01C346F4]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Hi,

See after...

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]=20
> Sent: Thursday, July 10, 2003 2:52 PM
> To: Zannin, Francois
> Cc: Samuel Dur D. Jeyaseelan; Ken Morneault; J.Javier Pastor=20
> (EE/EEM); sigtran@ietf.org
> Subject: Re: [Sigtran] M3UA IG v04
>=20
>=20
> Zannin,,
>=20
> Zannin, Francois wrote:                                      =20
>                      (Thu, 10 Jul 2003 14:21:56)
> > Let's continue on your example,
> >=20
> > When an ASP starts, it sends an ASP-UP, and then receives an=20
> > ASP-UP-ACK, Then two possibilities,
> >     1) Sends an ASP-Active while receiving the NOTIFY xxx
> >     2) Receive the NOTIFY xxx and then sends the ASP-Active
> >=20
> > If there is not enough ASP active for the AS, the SG(p)  answer with
an=20
> > ASP-Active_Ack and then everything is fine, the SG(p) start traffic=20
> > for the AS. If there are enough ASP Already Active for the AS, the=20
> > SG(p) do not send the ASP-Active_Ack. Then this "cleaver" ASP will=20
> > continue to send ASP-Active and stop after some time (x  retry of
the=20
> > Active).
>=20
> I don't think it would be a good idea to ignore the=20
> ASP-Active.  The SGP has three choices on receiving an=20
> ASP-Active: send ASP Active Ack, send Error("Management=20
> Blocking") or send ASP Inactive Ack.  Discarding the ASP=20
> Active and causing the ASP to fall back on OPTIONAL timers is=20
> not a good idea.

This idea to block the ASP-Active completion in order to indicate to the
remote
ASP that there are enough ASP active if we understand that "n" is the
maximum
Of ASP that the SG want to be ative.=20
The SG can choose the acknowledge all ASP... That's not the my issue...

>=20
> Also, it is not a good idea for the SGP to attempt to decide=20
> which ASP is to be activated in this way and which one is to=20
> be denied.  There are dependencies with ASPs going active due=20
> to the state that they must share.  For example: this might=20
> not be a possible situation from the ASP's perspective:
>=20
>      SGP1 - ASP1 Active
>      SGP1 - ASP2 Active
>      SGP1 - ASP3 Inactive
>=20
>      SGP2 - ASP1 Inactive
>      SGP2 - ASP2 Active
>      SGP2 - ASP3 Active
>=20
> Because this might force the ASPs to inefficiently share call state.

This example demonstrate that the NOTIFY sent by the SG cannot be used
to=20
decide if an ASP has to send an ASP-Active or not. The example of the IG

Shows that the SG send a notify to ASP1 and ASP2 and NOT ASP3...=20
Why ASP3? Has the SG to take this decision?
Now, it is in charge all the ASP to coordonate themselves to decide
which
ASP must send the ASP-Active.

BUT if the SG send the notification to all ASP, it is possible to
imagine=20
That someone create a service delivered by several independent small ASP

and that Only uses the indication coming from the SG for automatic
service
Activation, recovery...

>=20
> The RFC warns (Appendix A) that the loadsharing scheme might=20
> need to send all messages pertaining to the same call or=20
> transaction to the same ASP under normal operation.  This=20
> might not be possible in the above arrangement.
>=20
> > If an "Active ASP" fails, then the SG sends an "Insufficient ASP=20
> > Active" that can be used by all "Inactive ASP" to restart  the
Activate=20
> > procedure until the number is ASP is sufficient.
>=20
> The RFC also warns in section 4.3.4.5: (Notify Procedures)
>=20
>    In the case where a Notify message ("AS-PENDING") message  is sent
by
>    an SGP that now has no ASPs active to service the traffic,  or
where a
>    Notify ("Insufficient ASP resources active in AS") message  is sent
in
>    the Loadshare or Broadcast mode, the Notify message does not
>    explicitly compel the ASP(s) receiving the message to  become
active.
>    The ASPs remain in control of what (and when) traffic action is
>    taken.

You are right ! But this ASP can use this notification... For example, a

notify PENDING can be used for an automatic activation of a backup
ASP...

>=20
> It sounds like you are expecting the opposite: that sending=20
> the NTFY("Insufficient ASP resources active in AS") is=20
> expected to compell the ASPs to initiate the ASP Active Procedure.

This notification can be use by an ASP to trigg an automatic activation.

>=20
> >=20
> > With this kind of solution, No "new m3ua messages", no  waste of
time=20
> > in sending eternally ASP_Active, This is not the role of the SG to=20
> > chose which ASP have to be used base on notification, I  guess this
can=20
> > be an "amazing" solution!
>=20
> Yes, amazingly bad.  This is precisely one of the reasons we said:
>=20

My "amazing" was refering to the previous mail and the impossibility to=20
elaborate an automatic activation of an ASP...=20
      NOT to this wonderfull solution ;-)

>    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.
>=20

For n+k model, OK for a simple service served by an Active standby
independent
model, we don't need a big communication between ASP.

> The ASPs know which one of them is active and which one is=20
> backup for any given call/trasaction state.  They should=20
> decide which are active and which are not.
>=20
> The SGP should only refuse ASP Active under management=20
> blocking or failure.
>=20

That's not the goal of my issue, I just want to be notified when I am an

ASP and I becomes UP to an SG where I don't know if the service I a
offer
Is already active or not.

> --brian
>=20
> >=20
> > Regards,
> > Francois.
> >=20
> > -----Original Message-----
> > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > Sent: Thursday, July 10, 2003 12:36 PM
> > To: Zannin, Francois
> > Cc: Ken Morneault; J.Javier Pastor (EE/EEM); sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA IG v04
> >=20
> >=20
> >=20
> > Zannin,
> >=20
> >   Just an example, say an operator configures an AS1 as set  of 3
ASPes.
> >   They are namely known as ASP1,ASP2 and ASP3(backup).
> >=20
> >   Initially ASP1 and ASP2 send ASP_ACTIVE to SGP  so SGP sends a
> >   NTFY(AS_ACTIVE) thus results in AS1 to be in ACTIVE state.
> >=20
> >   Where is the question of sending NTFY(Sufficient ASPs active)?
> >   People mean to say that backup ASP also sends an  ASP_ACTIVE
message
> >   to SGP? if it so, on what condition it is being sent? Is there
> >   any valid conditions that backup ASP or some  new(!!!??!amazing!!)
ASP
> >   to send ASP_ACTIVE message?
> >=20
> >=20
> > --Samuel
> > =20
> >=20
> >   Alcatel USA, Inc.                  Internet:
> > <userid>@ssd.usa.alcatel.com
> >   1000 Coit Road, Plano, Texas 75075>=20

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



From exim@www1.ietf.org  Thu Jul 10 11:09:26 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA06348
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 11:09:26 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ad2c-0008Ri-3e
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 11:09:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AF92Ll032446
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 11:09:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ad2a-0008Qu-Le; Thu, 10 Jul 2003 11:09:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ad1j-0008Mh-F5
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 11:08:07 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA06312
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 11:08:01 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ad1g-0003l7-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 11:08:04 -0400
Received: from web41809.mail.yahoo.com ([66.218.93.143])
	by ietf-mx with smtp (Exim 4.12)
	id 19ad1f-0003kx-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 11:08:03 -0400
Message-ID: <20030710150722.8707.qmail@web41809.mail.yahoo.com>
Received: from [65.198.40.51] by web41809.mail.yahoo.com via HTTP; Thu, 10 Jul 2003 10:07:22 CDT
Date: Thu, 10 Jul 2003 10:07:22 -0500 (CDT)
From: =?iso-8859-1?q?Pablo=20Salinas?= <pablopesb@yahoo.com>
To: sigtran@ietf.org
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id LAA06313
Subject: [Sigtran] SCTP association in a SGP scenario
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Hi there,
          Could anybody tell me if this situation is
possible?

|--------------SGP-----------------------||----MG---|

IID 1 --AS1----ASP1-(65.198.40.50:3005)-SCTP1---ASP1
IID 2 --/  \---ASP2-(65.198.40.50:3006)-SCTP2---ASP2

IID 3 --AS2----ASP3-(65.198.40.50:3007)-SCTP2---ASP3

I mean, the sctp addresses in the SGP are IP:Port
So that, for one Ethernet card, I could have many SCTP
associations. Could this be possible?

Hope anyone can answer my question.

                thanks,
                        Pablo Salinas

_________________________________________________________
Do You Yahoo!?
Informaci=F3n de Estados Unidos y Am=E9rica Latina, en Yahoo! Noticias.
Vis=EDtanos en http://noticias.espanol.yahoo.com

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



From exim@www1.ietf.org  Thu Jul 10 11:16:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA06912
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 11:16:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ad9N-0000sc-SF
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 11:16:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AFG1kE003375
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 11:16:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ad9N-0000rW-Ep; Thu, 10 Jul 2003 11:16:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ad8X-0000ez-56
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 11:15:09 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA06766
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 11:15:04 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ad8W-0003rC-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 11:15:08 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[W+mtBRjEVZtFEZKgYShsTyEeUuz2gCMz])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ad8V-0003qw-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 11:15:07 -0400
Received: from ns.pigworks.openss7.net (IDENT:mBL4Hs/9t00APw1lN77jGyaEbbkB7alu@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6AFF0X28059;
	Thu, 10 Jul 2003 09:15:00 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6AFF0k19575;
	Thu, 10 Jul 2003 09:15:00 -0600
Date: Thu, 10 Jul 2003 09:15:00 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04  Section 3.14
Message-ID: <20030710091500.A19145@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CCA@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: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CCA@eestqnt105.es.eu.ericsson.se>; from j.javier.pastor@ericsson.com on Thu, Jul 03, 2003 at 11:45:57AM +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>

Javier,

Section 3.14 says:

   IA2DN: One ASP moves to ASP-DOWN causing that the number of ASPs in
   either ASP-ACTIVE or ASP-INACTIVE drops below n.

Am I correct to say that if n=5 and there are 4 ASPs in the ASP-INACTIVE
and 1 ASP in the ASP-ACTIVE state, and then the 1 active ASP moves to
the ASP-DOWN state and T(r) expires that the AS is to move to the AS-DOWN
state?  As follows:

  If T(r) expires before an ASP becomes ASP-ACTIVE, the SGP MAY stop
  queuing messages and discards all
  previously queued messages. The AS will move to the AS-INACTIVE state
  if at least the number of ASPs in ASP-INACTIVE sum n, otherwise it
  will move to AS-DOWN state.

Now, that is an AS state change and we must send a NTFY to all the INACTIVE
ASPs (the 4 left inactive) of the state change.  But from Section 3.8.2 of
the RFC, there is no message to notify of an AS-DOWN state:

         1    reserved
         2    Application Server Inactive (AS-INACTIVE)
         3    Application Server Active (AS-ACTIVE)
         4    Application Server Pending (AS-PENDING)

You see, we removed it because there was never an instance in which it
would be sent.  You have inserted an instance again.

Aren't you going to have to define a Notify message (the Reserved value)
for "Application Server Down (AS-DOWN)" so that the INACTIVE ASPs can
be properly notified by the SGP?

Particularly when the RFC says in Section 4.3.4.5: (Notify Procedures)

   A Notify message reflecting a change in the AS state MUST be sent to
   all ASPs in the AS, except those in the ASP-DOWN state, with
   appropriate Status Information and any ASP Identifier of the failed
   ASP.  At the ASP, Layer Management is informed with an M-NOTIFY
   indication primitive.  The Notify message must be sent whether the AS
   state change was a result of an ASP failure or reception of an ASP
   State management (ASPSM) / ASP Traffic Management (ASPTM) message.
   In the second case, the Notify message MUST be sent after any related
   acknowledgement messages  (e.g., ASP Up Ack, ASP Down Ack, ASP Active
   Ack, or ASP Inactive Ack).

The above tells me that I MUST send a Notify message becase of the AS
state change, but there is no message to inform about AS-DOWN state
transition.

What is the purpose of doing this?  What is the protocol difference between
the AS-DOWN state and the AS-INACTIVE state?  What is it that the SGP
does or doesn't do in the one state that it doesn't or does do in the other?

I don't recall this being discussed at all.

Please clarify.

-- 
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 exim@www1.ietf.org  Thu Jul 10 11:27:32 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA07384
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 11:27:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19adK5-0001RR-CJ
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 11:27:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AFR5u3005537
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 11:27:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19adK1-0001Qx-Sa; Thu, 10 Jul 2003 11:27:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19adJN-0001QT-MW
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 11:26:21 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA07318
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 11:26:17 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19adJM-0003yr-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 11:26:20 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[GyOTD4BJkV3EJxO2WlAhOJbZQg5Ypm37])
	by ietf-mx with esmtp (Exim 4.12)
	id 19adJL-0003yo-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 11:26:20 -0400
Received: from ns.pigworks.openss7.net (IDENT:gvmmmvyoQKNO39EXyV5GQBfVyp07mh/P@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6AFQIX28198;
	Thu, 10 Jul 2003 09:26:18 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6AFQHX19749;
	Thu, 10 Jul 2003 09:26:17 -0600
Date: Thu, 10 Jul 2003 09:26:17 -0600
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>,
        Ken Morneault <kmorneau@cisco.com>,
        "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
        sigtran@ietf.org
Subject: Re: [Sigtran] M3UA IG v04
Message-ID: <20030710092617.B19145@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>,
	Ken Morneault <kmorneau@cisco.com>,
	"J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
	sigtran@ietf.org
References: <06CAA4D3753C53408D2CE6340B68444214C467@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: <06CAA4D3753C53408D2CE6340B68444214C467@vbeexc02.emea.cpqcorp.net>; from francois.zannin@hp.com on Thu, Jul 10, 2003 at 05:02:34PM +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>

Francois,

Zannin, Francois wrote:                                                            (Thu, 10 Jul 2003 17:02:34)

-X-snip-X-
> 
> > 
> > It sounds like you are expecting the opposite: that sending 
> > the NTFY("Insufficient ASP resources active in AS") is 
> > expected to compell the ASPs to initiate the ASP Active Procedure.
> 
> This notification can be use by an ASP to trigg an automatic activation.

It could, but the SGP should not assume that the ASP will perform this
automatic activation.

> 
> > 
> > > 
> > > With this kind of solution, No "new m3ua messages", no  waste of
> time 
> > > in sending eternally ASP_Active, This is not the role of the SG to 
> > > chose which ASP have to be used base on notification, I  guess this
> can 
> > > be an "amazing" solution!
> > 
> > Yes, amazingly bad.  This is precisely one of the reasons we said:
> > 
> 
> My "amazing" was refering to the previous mail and the impossibility to 
> elaborate an automatic activation of an ASP... 
>       NOT to this wonderfull solution ;-)
> 
> >    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.
> > 
> 
> For n+k model, OK for a simple service served by an Active standby
> independent
> model, we don't need a big communication between ASP.

active/standy does not send "Insufficient ASPs ...".

> 
> > The ASPs know which one of them is active and which one is 
> > backup for any given call/trasaction state.  They should 
> > decide which are active and which are not.
> > 
> > The SGP should only refuse ASP Active under management 
> > blocking or failure.
> > 
> 
> That's not the goal of my issue, I just want to be notified when I am an
> ASP and I becomes UP to an SG where I don't know if the service I a
> offer
> Is already active or not.

I agree that the newly ASP-INACTIVE ASP should be notified of the state of
the AS, but not in an attempt to compel or rely on the ASP invoking any
procedure.  Simply to provide the information to the ASP that it might not
otherwise have, particularly if it is temporarily partitioned from the
other ASPs.

The RFC permits you to do that now.  It does not say that the SGP cannot
send Notify to newly active ASPs.  This is permitted.  What the proposal
was for was to way SHOULD.  That is, to get all SGPs to send Notify to
newly INACTIVE ASPs.

So we agree for different reasons.

Put the following lines in the IG added to the "Notify Procedures" and then
your ASP can at least rely on the SGP sending the Notifies when it move from
DOWN to INACTIVE for an AS (including registrations):

  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 exim@www1.ietf.org  Thu Jul 10 11:51:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA08350
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 11:51:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19adhI-0003Cs-Lc
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 11:51:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AFp436012305
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 11:51:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19adhF-0003BZ-Gv; Thu, 10 Jul 2003 11:51:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19adgh-0003Aj-Dd
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 11:50:27 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA08267
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 11:50:22 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19adez-0004Bv-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 11:48:41 -0400
Received: from angelica.ulticom.com ([208.255.120.2] helo=chuckie.dgms.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19adey-0004Bc-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 11:48:40 -0400
Received: from navy.fr.ulticom.com (localhost [127.0.0.1])
	by chuckie.dgms.com (8.9.3/8.9.3) with ESMTP id LAA27234
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 11:48:08 -0400 (EDT)
Received: from PCGASTAUDL (pc-gastaud-l.fr.ulticom.com [172.26.32.88])
	by navy.fr.ulticom.com (8.11.6+Sun/8.11.6) with SMTP id h6AFidE03021
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 17:44:39 +0200 (MEST)
From: "Valerie Gastaud" <valerie.gastaud@ulticom.com>
To: "Sigtran" <sigtran@ietf.org>
Date: Thu, 10 Jul 2003 17:48:35 +0200
Message-ID: <NDBBIHAOCKGAMIKNLALOCECODOAA.valerie.gastaud@ulticom.com>
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 IMO, Build 9.0.6604 (9.0.2911.0)
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106
Importance: Normal
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] M3UA - RC in SSNM messages
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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


Is there any reason why M3UA requires to insert all the Routing Contexts in
the SSNM messages when the message applies to all the AS; while in the same
situation, M3UA allows not to insert any RC in the ASPAC and ASPIA messages.

Valerie



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



From exim@www1.ietf.org  Thu Jul 10 11:57:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA08552
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 11:57:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19adn4-0003xK-5G
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 11:57:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AFv2BN015194
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 11:57:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19adn2-0003wu-Vl; Thu, 10 Jul 2003 11:57:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19adn2-0003wj-2J
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 11:57:00 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA08546
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 11:56:55 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19adn0-0004HW-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 11:56:58 -0400
Received: from zdemail03.zdem.compaq.com ([161.114.112.27])
	by ietf-mx with esmtp (Exim 4.12)
	id 19admz-0004HL-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 11:56:58 -0400
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail03.zdem.compaq.com (Postfix) with ESMTP
	id 56DBA1170; Thu, 10 Jul 2003 17:56:27 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Thu, 10 Jul 2003 17:56:27 +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="us-ascii"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] M3UA IG v04
Date: Thu, 10 Jul 2003 17:56:26 +0200
Message-ID: <06CAA4D3753C53408D2CE6340B68444214C468@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA IG v04
Thread-Index: AcNEzM4Fp+YUPNYwQ6uk5Y6EjHquegCLWctw
From: "Zannin, Francois" <francois.zannin@hp.com>
To: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>,
        <sigtran@ietf.org>
Cc: "Ken Morneault" <kmorneau@cisco.com>
X-OriginalArrivalTime: 10 Jul 2003 15:56:27.0159 (UTC) FILETIME=[D2114A70:01C346FB]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

All,

One more question about IPSP.=20
If an AS is registered only with a PC and the IPSP is able to handle
traffic=20
for several SI, what could be the answer of the IPSP if it receive a
message=20
for a SI that is not available.=20
Is it possible for an IPSP to send a DUPU (a SSNM) for this specific SI
to=20
the originating IPSP? Or is this configuration prohibited (PC Only) if
the=20
IPSP serve more than one SI?

Regards,

Francois.

> -----Original Message-----
> From: Javier Pastor-Balbas (ECE)=20
> [mailto:j.javier.pastor@ericsson.com]=20
> Sent: Thursday, July 03, 2003 10:59 AM
> To: sigtran@ietf.org
> Cc: 'bidulock@openss7.org'; Samuel Dur D. Jeyaseelan
> Subject: [Sigtran] M3UA IG v04
>=20
>=20
>=20
> Hi guys,
>=20
> We tried to submit the next IG version before the deadline,=20
> but I got confused since I thought it ended at 12:00 instead=20
> of 9:00. So the server bounced it. I'll submit the doc again=20
> after the IETF meeting. In the meantime you can find it at:
>=20
> http://standards.ericsson.net/javier/draft-ietf-sigtran-m3ua-i
> mplementors-guide-04.txt
>=20
> We've tried to include all the different opinions in here.=20
> Although it seems that the case of waiting till n ASPs is the=20
> preferred one, we have included the new possibility of a=20
> smooth start for those of you who want to activate the AS=20
> when the first ASP gets to ACTIVE. At the end it will be the=20
> operator's decission how to proceed during the activation=20
> phase. Hopefully everyone agrees on this wording...
>=20
> Cheers // Javier.
>=20
>=20
>=20

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



From exim@www1.ietf.org  Thu Jul 10 12:16:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA10106
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 12:16:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ae5V-00078i-1s
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 12:16:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AGG5Ye027444
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 12:16:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ae5S-000780-Fg; Thu, 10 Jul 2003 12:16:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ae5I-00077G-6R
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 12:15:52 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA09993
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 12:15:47 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ae5G-0004cA-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 12:15:50 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[oJ3eQxLh5hVjoL7GB1/w/XMgClDNEo/D])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ae5F-0004c7-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 12:15:49 -0400
Received: from ns.pigworks.openss7.net (IDENT:ZkK7+FmOFChFIMWgwzqLqPIXvPf6d0wO@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6AGFlX28940;
	Thu, 10 Jul 2003 10:15:47 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6AGFlN20751;
	Thu, 10 Jul 2003 10:15:47 -0600
Date: Thu, 10 Jul 2003 10:15:47 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04 Section 3.20
Message-ID: <20030710101547.C19145@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CCA@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: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CCA@eestqnt105.es.eu.ericsson.se>; from j.javier.pastor@ericsson.com on Thu, Jul 03, 2003 at 11:45:57AM +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>

Javier,

Section 3.20 has the following new text:

   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.

This is provided for by the DAUD procedure in the case of Isolation
in Section 4.5.3 of the RFC.

We discussed this and said that the SGP could send DUNA(*) if it wanted
to try and thwart traffic, but it is not necessary.  The SGP can always
rely on responsive procedures instead (responding to DATA with DUNA for
the inaccessible destination).  Also, the audit procedure allows the
ASP to audit after it becomes newly active whether the MTP is restarting
or not.

Also, there is the situation where the SG serves more than one point code.
In that case, the SG may transfer messages between point codes in the
IP domain while being in MTP Restart w.r.t. the SS7 domain.
That is why 4.6 MTP3 Restart 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.

Rather than DUNA(*).  MTP Restart does not necessarily affect all SS7
point codes, particularly those in the IP domain.

Change SHOULD to MAY and change DUNA(*) to "DUNA messages for the concerned
destinations".

--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 exim@www1.ietf.org  Thu Jul 10 12:23:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA10568
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 12:23:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aeCE-0007c7-1G
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 12:23:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AGN2DG029249
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 12:23:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aeCC-0007bW-V4; Thu, 10 Jul 2003 12:23:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aeBY-0007b7-WA
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 12:22:21 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA10498
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 12:22:15 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aeBX-0004kD-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 12:22:19 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[dGR3xeiT1FG7X/N5QJzWgbGm+RhHSfhg])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aeBW-0004kA-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 12:22:18 -0400
Received: from ns.pigworks.openss7.net (IDENT:5bpywITihopNsQMiV3qqXvJNgJgs8LM1@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6AGMHX29011;
	Thu, 10 Jul 2003 10:22:17 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6AGMGm20804;
	Thu, 10 Jul 2003 10:22:16 -0600
Date: Thu, 10 Jul 2003 10:22:16 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Valerie Gastaud <valerie.gastaud@ulticom.com>
Cc: Sigtran <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA - RC in SSNM messages
Message-ID: <20030710102216.D19145@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Valerie Gastaud <valerie.gastaud@ulticom.com>,
	Sigtran <sigtran@ietf.org>
References: <NDBBIHAOCKGAMIKNLALOCECODOAA.valerie.gastaud@ulticom.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: <NDBBIHAOCKGAMIKNLALOCECODOAA.valerie.gastaud@ulticom.com>; from valerie.gastaud@ulticom.com on Thu, Jul 10, 2003 at 05:48:35PM +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>

Valerie,

M3UA does not require one to insert RCs in all SSNM message:

Section 3.4.1 of the RFC says:

   Routing Context: n x 32-bits (unsigned integer)

      The optional Routing Context parameter contains the Routing
      Context values associated with the DUNA message.  Where a Routing
      Key has not been coordinated between the SGP and ASP, sending of
      Routing Context is not required.  Where multiple Routing Keys and
      Routing Contexts are used across a common association, the Routing
      Context(s) MUST be sent to identify the concerned traffic flows
      for which the DUNA message applies, assisting in outgoing traffic
      management and internal distribution of MTP-PAUSE indications to
      MTP3-Users at the receiver.

This is really no different than the ASPTM messages.

--brian

Valerie Gastaud wrote:                                                             (Thu, 10 Jul 2003 17:48:35)
> 
> Is there any reason why M3UA requires to insert all the Routing Contexts in
> the SSNM messages when the message applies to all the AS; while in the same
> situation, M3UA allows not to insert any RC in the ASPAC and ASPIA messages.
> 
> Valerie
> 
> 
> 
> _______________________________________________
> 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 exim@www1.ietf.org  Thu Jul 10 12:35:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA11012
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 12:35:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aeNr-00086p-8X
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 12:35:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AGZ339031165
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 12:35:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aeNp-00086Q-5G; Thu, 10 Jul 2003 12:35:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aeNi-000868-Nj
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 12:34:55 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA11005
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 12:34:49 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aeNh-0004rJ-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 12:34:53 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[UTHBxaBQR3Efsu27nhMAoK5UUeK3QEEh])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aeNg-0004rG-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 12:34:52 -0400
Received: from ns.pigworks.openss7.net (IDENT:4UOxy7HXK0ax3XKZZQh/jj+PBHIA0mB5@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6AGYiX29216;
	Thu, 10 Jul 2003 10:34:44 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6AGYiQ21105;
	Thu, 10 Jul 2003 10:34:44 -0600
Date: Thu, 10 Jul 2003 10:34:44 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Zannin, Francois" <francois.zannin@hp.com>
Cc: "Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>,
        sigtran@ietf.org, Ken Morneault <kmorneau@cisco.com>
Subject: Re: [Sigtran] M3UA IG v04
Message-ID: <20030710103444.E19145@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Zannin, Francois" <francois.zannin@hp.com>,
	"Javier Pastor-Balbas (ECE)" <j.javier.pastor@ericsson.com>,
	sigtran@ietf.org, Ken Morneault <kmorneau@cisco.com>
References: <06CAA4D3753C53408D2CE6340B68444214C468@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: <06CAA4D3753C53408D2CE6340B68444214C468@vbeexc02.emea.cpqcorp.net>; from francois.zannin@hp.com on Thu, Jul 10, 2003 at 05:56:26PM +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>

Francois,

The SS7 standards say that if an MTP-User accepts transfer traffic for
an entire point code (without associating itself with an MTP-SAP) then
it is responsible for user-part flow control.

There is, therefore, not MTP-STATUS Request primitive, just an MTP-STATUS
Indication primitive.

The MTP-User that is responsible for an entire point code must provide the
MTP-SAP (SI value) on all MTP-TRANSFER Request primitives anyway, so sending a
UPU is possible.

That is, to perform user-part flow control, rather than sending DUPU, it
should send a DATA message containing a UPU.

This permits an MTP implementation to transfer the message without
implementing some special MTP-STATUS Request primitive.  Inside the SGP
implementation, the only response to a DUPU would be sending the identical
UPU anyway, on the M3UA side of the NIF (because MTP does not have a
MTP-STATUS Indication primitive).

That is also the case for IPSP SE ASP side with Point-Code AS.

OTOH IPSP DE as defined in the IG needs to send DUPU, because both IPSPs
act as SGPs.

--brian

Zannin, Francois wrote:                                                            (Thu, 10 Jul 2003 17:56:26)
> All,
> 
> One more question about IPSP. 
> If an AS is registered only with a PC and the IPSP is able to handle
> traffic 
> for several SI, what could be the answer of the IPSP if it receive a
> message 
> for a SI that is not available. 
> Is it possible for an IPSP to send a DUPU (a SSNM) for this specific SI
> to 
> the originating IPSP? Or is this configuration prohibited (PC Only) if
> the 
> IPSP serve more than one SI?
> 
> Regards,
> 
> Francois.
> 
> > -----Original Message-----
> > From: Javier Pastor-Balbas (ECE) 
> > [mailto:j.javier.pastor@ericsson.com] 
> > Sent: Thursday, July 03, 2003 10:59 AM
> > To: sigtran@ietf.org
> > Cc: 'bidulock@openss7.org'; Samuel Dur D. Jeyaseelan
> > Subject: [Sigtran] M3UA IG v04
> > 
> > 
> > 
> > Hi guys,
> > 
> > We tried to submit the next IG version before the deadline, 
> > but I got confused since I thought it ended at 12:00 instead 
> > of 9:00. So the server bounced it. I'll submit the doc again 
> > after the IETF meeting. In the meantime you can find it at:
> > 
> > http://standards.ericsson.net/javier/draft-ietf-sigtran-m3ua-i
> > mplementors-guide-04.txt
> > 
> > We've tried to include all the different opinions in here. 
> > Although it seems that the case of waiting till n ASPs is the 
> > preferred one, we have included the new possibility of a 
> > smooth start for those of you who want to activate the AS 
> > when the first ASP gets to ACTIVE. At the end it will be the 
> > operator's decission how to proceed during the activation 
> > phase. Hopefully everyone agrees on this wording...
> > 
> > Cheers // Javier.
> > 
> > 
> > 
> 
> _______________________________________________
> 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 exim@www1.ietf.org  Thu Jul 10 12:48:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA11683
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 12:48:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aeaR-0000xk-9l
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 12:48:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AGm38b003679
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 12:48:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aeaP-0000wk-UH; Thu, 10 Jul 2003 12:48:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aeZz-0000v1-MX
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 12:47:35 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA11553
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 12:47:30 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aeZx-00051T-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 12:47:33 -0400
Received: from web12805.mail.yahoo.com ([216.136.174.40])
	by ietf-mx with smtp (Exim 4.12)
	id 19aeZx-00051K-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 12:47:33 -0400
Message-ID: <20030710164731.14220.qmail@web12805.mail.yahoo.com>
Received: from [208.255.120.2] by web12805.mail.yahoo.com via HTTP; Thu, 10 Jul 2003 09:47:31 PDT
Date: Thu, 10 Jul 2003 09:47:31 -0700 (PDT)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] M3UA - RC in SSNM messages
To: sigtran@ietf.org
In-Reply-To: <20030710102216.D19145@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:
> Valerie,
> 
> M3UA does not require one to insert RCs in all SSNM
> message:
> 
> Section 3.4.1 of the RFC says:
> 
>    Routing Context: n x 32-bits (unsigned integer)
> 
>       The optional Routing Context parameter
> contains the Routing
>       Context values associated with the DUNA
> message.  Where a Routing
>       Key has not been coordinated between the SGP
> and ASP, sending of
>       Routing Context is not required.  Where
> multiple Routing Keys and
>       Routing Contexts are used across a common
> association, the Routing
>       Context(s) MUST be sent to identify the
> concerned traffic flows
>       for which the DUNA message applies, assisting
> in outgoing traffic
>       management and internal distribution of
> MTP-PAUSE indications to
>       MTP3-Users at the receiver.
> 
> This is really no different than the ASPTM messages.
[TOLGA] IMO it is not the same. An ASPAC with no RC
inside is applicable for configured AS -so even
multiple AS is served via the same association, one
can send ASPAC with no RC-, but RC is mandatory for
SSNM, traffic for more than one AS is carried over the
same association.
> 
> --brian
> 
> Valerie Gastaud wrote:                              
>                               (Thu, 10 Jul 2003
> 17:48:35)
> > 
> > Is there any reason why M3UA requires to insert
> all the Routing Contexts in
> > the SSNM messages when the message applies to all
> the AS; while in the same
> > situation, M3UA allows not to insert any RC in the
> ASPAC and ASPIA messages.
> > 
> > Valerie
> > 
> > 
> > 
> > _______________________________________________
> > 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


__________________________________
Do you Yahoo!?
SBC Yahoo! DSL - Now only $29.95 per month!
http://sbc.yahoo.com

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



From exim@www1.ietf.org  Thu Jul 10 12:58:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA12063
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 12:58:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aek6-000294-DG
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 12:58:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AGw2np008225
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 12:58:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aek5-00028Q-PT; Thu, 10 Jul 2003 12:58:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aejk-00027j-1D
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 12:57:40 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA12037
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 12:57:34 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aeji-0005AH-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 12:57:38 -0400
Received: from web12802.mail.yahoo.com ([216.136.174.37])
	by ietf-mx with smtp (Exim 4.12)
	id 19aejh-0005AE-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 12:57:37 -0400
Message-ID: <20030710165736.79598.qmail@web12802.mail.yahoo.com>
Received: from [208.255.120.2] by web12802.mail.yahoo.com via HTTP; Thu, 10 Jul 2003 09:57:36 PDT
Date: Thu, 10 Jul 2003 09:57:36 -0700 (PDT)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] M3UA IG v04
To: sigtran@ietf.org
In-Reply-To: <20030710103444.E19145@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/Francois,

Just my 2 cents below....

   Tolga
--- "Brian F. G. Bidulock" <bidulock@openss7.org>
wrote:
> This permits an MTP implementation to transfer the
> message without
> implementing some special MTP-STATUS Request
> primitive.  Inside the SGP
> implementation, the only response to a DUPU would be
> sending the identical
> UPU anyway, on the M3UA side of the NIF (because MTP
> does not have a
> MTP-STATUS Indication primitive).
> 
> That is also the case for IPSP SE ASP side with
> Point-Code AS.
[TOLGA]For SE IPSP, IMO there is no need for UPU/DUPU
etc... SE IPSP is two AS communicating each other. AS
is available or not, never partly available. If one
chooses to use PC granularity, then when on user part
goes down, IMO the whole AS is INACTIVE. If one wants
finer granularity, then AS should be defined
accordingly.
> 
> OTOH IPSP DE as defined in the IG needs to send
> DUPU, because both IPSPs
> act as SGPs.
[TOLGA] I think the same for DE IPSP case as defined
in IG.
> 
> --brian
> 
> Zannin, Francois wrote:                             
>                               (Thu, 10 Jul 2003
> 17:56:26)
> > All,
> > 
> > One more question about IPSP. 
> > If an AS is registered only with a PC and the IPSP
> is able to handle
> > traffic 
> > for several SI, what could be the answer of the
> IPSP if it receive a
> > message 
> > for a SI that is not available. 
> > Is it possible for an IPSP to send a DUPU (a SSNM)
> for this specific SI
> > to 
> > the originating IPSP? Or is this configuration
> prohibited (PC Only) if
> > the 
> > IPSP serve more than one SI?


__________________________________
Do you Yahoo!?
SBC Yahoo! DSL - Now only $29.95 per month!
http://sbc.yahoo.com

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



From exim@www1.ietf.org  Thu Jul 10 13:37:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA13889
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 13:37:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19afLr-0005Va-Tm
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 13:37:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AHb37x021163
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 13:37:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19afLo-0005US-Pg; Thu, 10 Jul 2003 13:37:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19afLE-0005SR-Fg
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 13:36:24 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA13591
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 13:36:20 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19afLC-0005UE-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 13:36:22 -0400
Received: from web12807.mail.yahoo.com ([216.136.174.42])
	by ietf-mx with smtp (Exim 4.12)
	id 19afLB-0005U4-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 13:36:21 -0400
Message-ID: <20030710173614.8249.qmail@web12807.mail.yahoo.com>
Received: from [208.255.120.2] by web12807.mail.yahoo.com via HTTP; Thu, 10 Jul 2003 10:36:14 PDT
Date: Thu, 10 Jul 2003 10:36:14 -0700 (PDT)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] M3UA IG v04
To: sigtran@ietf.org
In-Reply-To: <20030710165736.79598.qmail@web12802.mail.yahoo.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>

After thinking a bit more, I believe also for DE IPSP
case as defined in IG, there is no need to send DUPU.
Afterall, it behaves like 2 SGP/ASPs. Like there is no
need for DUPU from ASP to SGP, I think the same should
be true for DE IPSP. So, DE IPSP should also send
ASPIA in such a case, IMO.

DUPU makes sense for SG-SG communication, which mimics
MTP3/MTP3 communication, instead of MTP3/MTP3 user
part communication.
  
   Tolga
--- Tolga Asveren <tolga_asveren@yahoo.com> wrote:
> Brian/Francois,
> 
> Just my 2 cents below....
> 
>    Tolga
> --- "Brian F. G. Bidulock" <bidulock@openss7.org>
> wrote:
> > This permits an MTP implementation to transfer the
> > message without
> > implementing some special MTP-STATUS Request
> > primitive.  Inside the SGP
> > implementation, the only response to a DUPU would
> be
> > sending the identical
> > UPU anyway, on the M3UA side of the NIF (because
> MTP
> > does not have a
> > MTP-STATUS Indication primitive).
> > 
> > That is also the case for IPSP SE ASP side with
> > Point-Code AS.
> [TOLGA]For SE IPSP, IMO there is no need for
> UPU/DUPU
> etc... SE IPSP is two AS communicating each other.
> AS
> is available or not, never partly available. If one
> chooses to use PC granularity, then when on user
> part
> goes down, IMO the whole AS is INACTIVE. If one
> wants
> finer granularity, then AS should be defined
> accordingly.
> > 
> > OTOH IPSP DE as defined in the IG needs to send
> > DUPU, because both IPSPs
> > act as SGPs.
> [TOLGA] I think the same for DE IPSP case as defined
> in IG.
> > 
> > --brian
> > 
> > Zannin, Francois wrote:                           
>  
> >                               (Thu, 10 Jul 2003
> > 17:56:26)
> > > All,
> > > 
> > > One more question about IPSP. 
> > > If an AS is registered only with a PC and the
> IPSP
> > is able to handle
> > > traffic 
> > > for several SI, what could be the answer of the
> > IPSP if it receive a
> > > message 
> > > for a SI that is not available. 
> > > Is it possible for an IPSP to send a DUPU (a
> SSNM)
> > for this specific SI
> > > to 
> > > the originating IPSP? Or is this configuration
> > prohibited (PC Only) if
> > > the 
> > > IPSP serve more than one SI?
> 
> 
> __________________________________
> Do you Yahoo!?
> SBC Yahoo! DSL - Now only $29.95 per month!
> http://sbc.yahoo.com
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran


__________________________________
Do you Yahoo!?
SBC Yahoo! DSL - Now only $29.95 per month!
http://sbc.yahoo.com

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



From exim@www1.ietf.org  Thu Jul 10 13:40:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA13958
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 13:40:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19afOl-0006HH-6y
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 13:40:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AHe3ZY024120
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 13:40:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19afOj-0006Gu-Bd; Thu, 10 Jul 2003 13:40:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19afNo-0006Dd-SR
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 13:39:04 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA13949
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 13:39:00 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19afNm-0005WR-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 13:39:02 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[w93y3jbZO7TYOtZ/RaH/DzpHLnEBJmCj])
	by ietf-mx with esmtp (Exim 4.12)
	id 19afNl-0005WN-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 13:39:01 -0400
Received: from ns.pigworks.openss7.net (IDENT:Oo/Kjb37Ne0P6AFaHgoZ5alNVvGUdhL9@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6AHcwX30069;
	Thu, 10 Jul 2003 11:38:58 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6AHcw722451;
	Thu, 10 Jul 2003 11:38:58 -0600
Date: Thu, 10 Jul 2003 11:38:58 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tolga Asveren <tolga_asveren@yahoo.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA IG v04
Message-ID: <20030710113858.H19145@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030710165736.79598.qmail@web12802.mail.yahoo.com> <20030710173614.8249.qmail@web12807.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: <20030710173614.8249.qmail@web12807.mail.yahoo.com>; from tolga_asveren@yahoo.com on Thu, Jul 10, 2003 at 10:36:14AM -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>

Tolga,

IPSP DE in the IG is defined as acting in the role of
two SGPs.  SGPs already send DUPU and it must therefore
be permitted for IPSPs in DE.

--brian

Tolga Asveren wrote:                                                                 (Thu, 10 Jul 2003 10:36:14)
> After thinking a bit more, I believe also for DE IPSP
> case as defined in IG, there is no need to send DUPU.
> Afterall, it behaves like 2 SGP/ASPs. Like there is no
> need for DUPU from ASP to SGP, I think the same should
> be true for DE IPSP. So, DE IPSP should also send
> ASPIA in such a case, IMO.
> 
> DUPU makes sense for SG-SG communication, which mimics
> MTP3/MTP3 communication, instead of MTP3/MTP3 user
> part communication.
>   
>    Tolga
> --- Tolga Asveren <tolga_asveren@yahoo.com> wrote:
> > Brian/Francois,
> > 
> > Just my 2 cents below....
> > 
> >    Tolga
> > --- "Brian F. G. Bidulock" <bidulock@openss7.org>
> > wrote:
> > > This permits an MTP implementation to transfer the
> > > message without
> > > implementing some special MTP-STATUS Request
> > > primitive.  Inside the SGP
> > > implementation, the only response to a DUPU would
> > be
> > > sending the identical
> > > UPU anyway, on the M3UA side of the NIF (because
> > MTP
> > > does not have a
> > > MTP-STATUS Indication primitive).
> > > 
> > > That is also the case for IPSP SE ASP side with
> > > Point-Code AS.
> > [TOLGA]For SE IPSP, IMO there is no need for
> > UPU/DUPU
> > etc... SE IPSP is two AS communicating each other.
> > AS
> > is available or not, never partly available. If one
> > chooses to use PC granularity, then when on user
> > part
> > goes down, IMO the whole AS is INACTIVE. If one
> > wants
> > finer granularity, then AS should be defined
> > accordingly.
> > > 
> > > OTOH IPSP DE as defined in the IG needs to send
> > > DUPU, because both IPSPs
> > > act as SGPs.
> > [TOLGA] I think the same for DE IPSP case as defined
> > in IG.
> > > 
> > > --brian
> > > 
> > > Zannin, Francois wrote:                           
> >  
> > >                               (Thu, 10 Jul 2003
> > > 17:56:26)
> > > > All,
> > > > 
> > > > One more question about IPSP. 
> > > > If an AS is registered only with a PC and the
> > IPSP
> > > is able to handle
> > > > traffic 
> > > > for several SI, what could be the answer of the
> > > IPSP if it receive a
> > > > message 
> > > > for a SI that is not available. 
> > > > Is it possible for an IPSP to send a DUPU (a
> > SSNM)
> > > for this specific SI
> > > > to 
> > > > the originating IPSP? Or is this configuration
> > > prohibited (PC Only) if
> > > > the 
> > > > IPSP serve more than one SI?
> > 
> > 
> > __________________________________
> > Do you Yahoo!?
> > SBC Yahoo! DSL - Now only $29.95 per month!
> > http://sbc.yahoo.com
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> 
> __________________________________
> Do you Yahoo!?
> SBC Yahoo! DSL - Now only $29.95 per month!
> http://sbc.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 exim@www1.ietf.org  Thu Jul 10 13:47:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA14227
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 13:47:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19afVX-0006mc-1U
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 13:47:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AHl3RR026066
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 13:47:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19afVW-0006m5-D6; Thu, 10 Jul 2003 13:47:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19afVC-0006lc-Tl
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 13:46:42 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA14214
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 13:46:38 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19afVA-0005ca-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 13:46:40 -0400
Received: from web12808.mail.yahoo.com ([216.136.174.43])
	by ietf-mx with smtp (Exim 4.12)
	id 19afV9-0005cU-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 13:46:39 -0400
Message-ID: <20030710174637.63767.qmail@web12808.mail.yahoo.com>
Received: from [208.255.120.2] by web12808.mail.yahoo.com via HTTP; Thu, 10 Jul 2003 10:46:37 PDT
Date: Thu, 10 Jul 2003 10:46:37 -0700 (PDT)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] M3UA IG v04
To: sigtran@ietf.org
In-Reply-To: <20030710113858.H19145@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,
> 
> IPSP DE in the IG is defined as acting in the role
> of
> two SGPs.  SGPs already send DUPU and it must
> therefore
> be permitted for IPSPs in DE.
[TOLGA] Yes, SGPs send DUPU to ASPs but to my
understanding, Francois's case would correspond
sending DUPU from ASP to SGP.
As I said, my interpretation of DE IPSP in the IG is a
combined SGP/ASP. On ASP part of IPSP, the user part
goes down, then I would expect inactivating the
corresponding AS.
> 
> --brian
> 
> Tolga Asveren wrote:                                
>                                 (Thu, 10 Jul 2003
> 10:36:14)
> > After thinking a bit more, I believe also for DE
> IPSP
> > case as defined in IG, there is no need to send
> DUPU.
> > Afterall, it behaves like 2 SGP/ASPs. Like there
> is no
> > need for DUPU from ASP to SGP, I think the same
> should
> > be true for DE IPSP. So, DE IPSP should also send
> > ASPIA in such a case, IMO.
> > 
> > DUPU makes sense for SG-SG communication, which
> mimics
> > MTP3/MTP3 communication, instead of MTP3/MTP3 user
> > part communication.
> >   
> >    Tolga
> > --- Tolga Asveren <tolga_asveren@yahoo.com> wrote:
> > > Brian/Francois,
> > > 
> > > Just my 2 cents below....
> > > 
> > >    Tolga
> > > --- "Brian F. G. Bidulock"
> <bidulock@openss7.org>
> > > wrote:
> > > > This permits an MTP implementation to transfer
> the
> > > > message without
> > > > implementing some special MTP-STATUS Request
> > > > primitive.  Inside the SGP
> > > > implementation, the only response to a DUPU
> would
> > > be
> > > > sending the identical
> > > > UPU anyway, on the M3UA side of the NIF
> (because
> > > MTP
> > > > does not have a
> > > > MTP-STATUS Indication primitive).
> > > > 
> > > > That is also the case for IPSP SE ASP side
> with
> > > > Point-Code AS.
> > > [TOLGA]For SE IPSP, IMO there is no need for
> > > UPU/DUPU
> > > etc... SE IPSP is two AS communicating each
> other.
> > > AS
> > > is available or not, never partly available. If
> one
> > > chooses to use PC granularity, then when on user
> > > part
> > > goes down, IMO the whole AS is INACTIVE. If one
> > > wants
> > > finer granularity, then AS should be defined
> > > accordingly.
> > > > 
> > > > OTOH IPSP DE as defined in the IG needs to
> send
> > > > DUPU, because both IPSPs
> > > > act as SGPs.
> > > [TOLGA] I think the same for DE IPSP case as
> defined
> > > in IG.
> > > > 
> > > > --brian
> > > > 
> > > > Zannin, Francois wrote:                       
>    
> > >  
> > > >                               (Thu, 10 Jul
> 2003
> > > > 17:56:26)
> > > > > All,
> > > > > 
> > > > > One more question about IPSP. 
> > > > > If an AS is registered only with a PC and
> the
> > > IPSP
> > > > is able to handle
> > > > > traffic 
> > > > > for several SI, what could be the answer of
> the
> > > > IPSP if it receive a
> > > > > message 
> > > > > for a SI that is not available. 
> > > > > Is it possible for an IPSP to send a DUPU (a
> > > SSNM)
> > > > for this specific SI
> > > > > to 
> > > > > the originating IPSP? Or is this
> configuration
> > > > prohibited (PC Only) if
> > > > > the 
> > > > > IPSP serve more than one SI?
> > > 
> > > 
> > > __________________________________
> > > Do you Yahoo!?
> > > SBC Yahoo! DSL - Now only $29.95 per month!
> > > http://sbc.yahoo.com
> > > 
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > 
> > 
> > __________________________________
> > Do you Yahoo!?
> > SBC Yahoo! DSL - Now only $29.95 per month!
> > http://sbc.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


__________________________________
Do you Yahoo!?
SBC Yahoo! DSL - Now only $29.95 per month!
http://sbc.yahoo.com

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



From exim@www1.ietf.org  Thu Jul 10 13:57:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA14494
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 13:57:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19affC-0007cS-EF
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 13:57:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AHv2sA029284
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 13:57:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19affB-0007cD-6P; Thu, 10 Jul 2003 13:57:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19afeD-0007IE-SL
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 13:56:01 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA14419
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 13:55:57 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19afeB-0005gq-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 13:55:59 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[1yABHeZ2elM45sd6NFQOb3pXPpPDvblr])
	by ietf-mx with esmtp (Exim 4.12)
	id 19afeA-0005gm-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 13:55:58 -0400
Received: from ns.pigworks.openss7.net (IDENT:xT5YmFONuYQ3+kA8jsAx6GQj79b7nx+6@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6AHtvX30281;
	Thu, 10 Jul 2003 11:55:57 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6AHtvW22786;
	Thu, 10 Jul 2003 11:55:57 -0600
Date: Thu, 10 Jul 2003 11:55:56 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tolga Asveren <tolga_asveren@yahoo.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - RC in SSNM messages
Message-ID: <20030710115556.I19145@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030710102216.D19145@openss7.org> <20030710164731.14220.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: <20030710164731.14220.qmail@web12805.mail.yahoo.com>; from tolga_asveren@yahoo.com on Thu, Jul 10, 2003 at 09:47:31AM -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>

Tolga,

Tolga Asveren wrote:                                                                 (Thu, 10 Jul 2003 09:47:31)
> Brian,
> --- "Brian F. G. Bidulock" <bidulock@openss7.org>
> wrote:
> > Valerie,
> > 
> > M3UA does not require one to insert RCs in all SSNM
> > message:
> > 
> > Section 3.4.1 of the RFC says:
> > 
> >    Routing Context: n x 32-bits (unsigned integer)
> > 
> >       The optional Routing Context parameter contains the Routing Context
> >       values associated with the DUNA message.  Where a Routing Key has
> >       not been coordinated between the SGP and ASP, sending of Routing
> >       Context is not required.  Where multiple Routing Keys and Routing
> >       Contexts are used across a common association, the Routing
> >       Context(s) MUST be sent to identify the concerned traffic flows for
> >       which the DUNA message applies, assisting in outgoing traffic
> >       management and internal distribution of MTP-PAUSE indications to
> >       MTP3-Users at the receiver.
> > 
> > This is really no different than the ASPTM messages.
> [TOLGA] IMO it is not the same. An ASPAC with no RC
> inside is applicable for configured AS -so even
> multiple AS is served via the same association, one
> can send ASPAC with no RC-, but RC is mandatory for
> SSNM, traffic for more than one AS is carried over the
> same association.

Section 4.3.4.3:

                    ...  In the case where an ASP wishes to process the
   traffic for more than one Application Server across a common SCTP
   association, the ASP Active message(s) SHOULD contain a list of one
   or more Routing Contexts to indicate for which Application Servers
   the ASP Active message applies. ...

How is this so different?

3.4.1 says MUST for SSNM (as it does for DATA) whereas 4.3.5.3 says SHOULD
for ASP Active (ASPTM).

Valerie asked why M3UA allowed you to not insert any RC in ASPAC and ASPIA
when it was required in SSNM.

Well, the answer is that it SHOULD be included in ASPTM as well according
to the passage above.

Also, if you are configuring multiple AS over the same SCTP association and
then activating without an RC, you are going against the SHOULD above, and
you are on your own.

--brian


> > 
> > --brian
> > 
> > Valerie Gastaud wrote:                              
> >                               (Thu, 10 Jul 2003
> > 17:48:35)
> > > 
> > > Is there any reason why M3UA requires to insert all the Routing Contexts in
> > > the SSNM messages when the message applies to all the AS; while in the same
> > > situation, M3UA allows not to insert any RC in the ASPAC and ASPIA messages.
> > > 
> > > Valerie
> > > 
> > > 
> > > 
> > > _______________________________________________
> > > 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
> 
> 
> __________________________________
> Do you Yahoo!?
> SBC Yahoo! DSL - Now only $29.95 per month!
> http://sbc.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 exim@www1.ietf.org  Thu Jul 10 14:01:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA14635
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 14:01:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19afj3-00089f-O7
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 14:01:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AI11c8031326
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 14:01:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19afj3-000891-4j; Thu, 10 Jul 2003 14:01:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19afih-00088F-00
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 14:00:39 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA14611
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 14:00:34 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19afie-0005k8-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 14:00:36 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[nf+ZAIRgC2vpskGGjte/k5lxtvNkDnvi])
	by ietf-mx with esmtp (Exim 4.12)
	id 19afid-0005k5-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 14:00:36 -0400
Received: from ns.pigworks.openss7.net (IDENT:OMy2d3kGL6R3WUUbCzAJa5iO1atCXimF@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6AI0YX30432;
	Thu, 10 Jul 2003 12:00:34 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6AI0YQ22918;
	Thu, 10 Jul 2003 12:00:34 -0600
Date: Thu, 10 Jul 2003 12:00:33 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tolga Asveren <tolga_asveren@yahoo.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA IG v04
Message-ID: <20030710120033.J19145@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030710113858.H19145@openss7.org> <20030710174637.63767.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: <20030710174637.63767.qmail@web12808.mail.yahoo.com>; from tolga_asveren@yahoo.com on Thu, Jul 10, 2003 at 10:46:37AM -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>

Tolga,

Tolga Asveren wrote:                                                                 (Thu, 10 Jul 2003 10:46:37)
> Brian,
> --- "Brian F. G. Bidulock" <bidulock@openss7.org>
> wrote:
> > Tolga,
> > 
> > IPSP DE in the IG is defined as acting in the role
> > of
> > two SGPs.  SGPs already send DUPU and it must
> > therefore
> > be permitted for IPSPs in DE.
> [TOLGA] Yes, SGPs send DUPU to ASPs but to my
> understanding, Francois's case would correspond
> sending DUPU from ASP to SGP.
> As I said, my interpretation of DE IPSP in the IG is a
> combined SGP/ASP. On ASP part of IPSP, the user part
> goes down, then I would expect inactivating the
> corresponding AS.

Yes, that's what we understood and agreed, however, that is
not what is presented in the IG: Read the IG again:
New text in section 3.11.2:

   In section 4.3.1: ASP/IPSP States, only the SGP-ASP and the IPSP SE
   scenarios are described. For the IPSP DE model, both IPSPs MUST
   follow the SGP side of the SGP-ASP procedures.

That looks pretty plain to me.  According to the IG, DE IPSPs can
send DUPU (they are actually required above to do so when the SGP
procedure dictates).

--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 exim@www1.ietf.org  Thu Jul 10 14:11:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA14954
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 14:11:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19afsl-0000ba-SX
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 14:11:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AIB3BE002325
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 14:11:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19afsk-0000bG-1T; Thu, 10 Jul 2003 14:11:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19afsE-0000aX-70
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 14:10:30 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA14903
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 14:10:24 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19afs7-0005oC-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 14:10:23 -0400
Received: from web12802.mail.yahoo.com ([216.136.174.37])
	by ietf-mx with smtp (Exim 4.12)
	id 19afs6-0005o6-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 14:10:22 -0400
Message-ID: <20030710181020.93050.qmail@web12802.mail.yahoo.com>
Received: from [208.255.120.2] by web12802.mail.yahoo.com via HTTP; Thu, 10 Jul 2003 11:10:20 PDT
Date: Thu, 10 Jul 2003 11:10:20 -0700 (PDT)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] M3UA - RC in SSNM messages
To: sigtran@ietf.org
In-Reply-To: <20030710115556.I19145@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,

I was referring to 4.3.4.3 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."
This enables the option to send ASPAC with no RC, even
if multiple AS traffic is carried over one
association. I don't think, this is permitted for
SSNM.

    Tolga

--- "Brian F. G. Bidulock" <bidulock@openss7.org>
wrote:
> Tolga,
> 
> Tolga Asveren wrote:                                
>                                 (Thu, 10 Jul 2003
> 09:47:31)
> > Brian,
> > --- "Brian F. G. Bidulock" <bidulock@openss7.org>
> > wrote:
> > > Valerie,
> > > 
> > > M3UA does not require one to insert RCs in all
> SSNM
> > > message:
> > > 
> > > Section 3.4.1 of the RFC says:
> > > 
> > >    Routing Context: n x 32-bits (unsigned
> integer)
> > > 
> > >       The optional Routing Context parameter
> contains the Routing Context
> > >       values associated with the DUNA message. 
> Where a Routing Key has
> > >       not been coordinated between the SGP and
> ASP, sending of Routing
> > >       Context is not required.  Where multiple
> Routing Keys and Routing
> > >       Contexts are used across a common
> association, the Routing
> > >       Context(s) MUST be sent to identify the
> concerned traffic flows for
> > >       which the DUNA message applies, assisting
> in outgoing traffic
> > >       management and internal distribution of
> MTP-PAUSE indications to
> > >       MTP3-Users at the receiver.
> > > 
> > > This is really no different than the ASPTM
> messages.
> > [TOLGA] IMO it is not the same. An ASPAC with no
> RC
> > inside is applicable for configured AS -so even
> > multiple AS is served via the same association,
> one
> > can send ASPAC with no RC-, but RC is mandatory
> for
> > SSNM, traffic for more than one AS is carried over
> the
> > same association.
> 
> Section 4.3.4.3:
> 
>                     ...  In the case where an ASP
> wishes to process the
>    traffic for more than one Application Server
> across a common SCTP
>    association, the ASP Active message(s) SHOULD
> contain a list of one
>    or more Routing Contexts to indicate for which
> Application Servers
>    the ASP Active message applies. ...
> 
> How is this so different?
> 
> 3.4.1 says MUST for SSNM (as it does for DATA)
> whereas 4.3.5.3 says SHOULD
> for ASP Active (ASPTM).
> 
> Valerie asked why M3UA allowed you to not insert any
> RC in ASPAC and ASPIA
> when it was required in SSNM.
> 
> Well, the answer is that it SHOULD be included in
> ASPTM as well according
> to the passage above.
> 
> Also, if you are configuring multiple AS over the
> same SCTP association and
> then activating without an RC, you are going against
> the SHOULD above, and
> you are on your own.
> 
> --brian
> 
> 
> > > 
> > > --brian
> > > 
> > > Valerie Gastaud wrote:                          
>    
> > >                               (Thu, 10 Jul 2003
> > > 17:48:35)
> > > > 
> > > > Is there any reason why M3UA requires to
> insert all the Routing Contexts in
> > > > the SSNM messages when the message applies to
> all the AS; while in the same
> > > > situation, M3UA allows not to insert any RC in
> the ASPAC and ASPIA messages.
> > > > 
> > > > Valerie
> > > > 
> > > > 
> > > > 
> > > >
> _______________________________________________
> > > > 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
> > 
> > 
> > __________________________________
> > Do you Yahoo!?
> > SBC Yahoo! DSL - Now only $29.95 per month!
> > http://sbc.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


__________________________________
Do you Yahoo!?
SBC Yahoo! DSL - Now only $29.95 per month!
http://sbc.yahoo.com

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



From exim@www1.ietf.org  Thu Jul 10 14:59:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA16414
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 14:59:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19agdD-0003tH-N7
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 14:59:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AIx3rw014950
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 14:59:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19agdB-0003sr-8z; Thu, 10 Jul 2003 14:59:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19agcr-0003rP-IJ
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 14:58:41 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA16402
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 14:58:36 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19agco-0006EC-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 14:58:38 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[Zizcx4UTu4FWElogDj2W7vTE50wDfb/u])
	by ietf-mx with esmtp (Exim 4.12)
	id 19agcn-0006E9-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 14:58:37 -0400
Received: from ns.pigworks.openss7.net (IDENT:o5VuLNI5VJkAH/I/mJIm3UpBglT86WRg@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6AIwaX31180;
	Thu, 10 Jul 2003 12:58:36 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6AIwZZ23760;
	Thu, 10 Jul 2003 12:58:35 -0600
Date: Thu, 10 Jul 2003 12:58:35 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tolga Asveren <tolga_asveren@yahoo.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - RC in SSNM messages
Message-ID: <20030710125835.K19145@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030710115556.I19145@openss7.org> <20030710181020.93050.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: <20030710181020.93050.qmail@web12802.mail.yahoo.com>; from tolga_asveren@yahoo.com on Thu, Jul 10, 2003 at 11:10:20AM -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>

Tolga,

Tolga Asveren wrote:                                                                 (Thu, 10 Jul 2003 11:10:20)
> Brian,
> 
> I was referring to 4.3.4.3 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."
> This enables the option to send ASPAC with no RC, even
> if multiple AS traffic is carried over one
> association. I don't think, this is permitted for
> SSNM.

It permits a single AS in fitting with the SHOULD I quoted.  Multiple
configured AS for the same association would be against the SHOULD.
So you are still on your own.

--brian

> 
>     Tolga
> 
> --- "Brian F. G. Bidulock" <bidulock@openss7.org>
> wrote:
> > Tolga,
> > 
> > Tolga Asveren wrote:                                
> >                                 (Thu, 10 Jul 2003
> > 09:47:31)
> > > Brian,
> > > --- "Brian F. G. Bidulock" <bidulock@openss7.org>
> > > wrote:
> > > > Valerie,
> > > > 
> > > > M3UA does not require one to insert RCs in all
> > SSNM
> > > > message:
> > > > 
> > > > Section 3.4.1 of the RFC says:
> > > > 
> > > >    Routing Context: n x 32-bits (unsigned
> > integer)
> > > > 
> > > >       The optional Routing Context parameter
> > contains the Routing Context
> > > >       values associated with the DUNA message. 
> > Where a Routing Key has
> > > >       not been coordinated between the SGP and
> > ASP, sending of Routing
> > > >       Context is not required.  Where multiple
> > Routing Keys and Routing
> > > >       Contexts are used across a common
> > association, the Routing
> > > >       Context(s) MUST be sent to identify the
> > concerned traffic flows for
> > > >       which the DUNA message applies, assisting
> > in outgoing traffic
> > > >       management and internal distribution of
> > MTP-PAUSE indications to
> > > >       MTP3-Users at the receiver.
> > > > 
> > > > This is really no different than the ASPTM
> > messages.
> > > [TOLGA] IMO it is not the same. An ASPAC with no
> > RC
> > > inside is applicable for configured AS -so even
> > > multiple AS is served via the same association,
> > one
> > > can send ASPAC with no RC-, but RC is mandatory
> > for
> > > SSNM, traffic for more than one AS is carried over
> > the
> > > same association.
> > 
> > Section 4.3.4.3:
> > 
> >                     ...  In the case where an ASP
> > wishes to process the
> >    traffic for more than one Application Server
> > across a common SCTP
> >    association, the ASP Active message(s) SHOULD
> > contain a list of one
> >    or more Routing Contexts to indicate for which
> > Application Servers
> >    the ASP Active message applies. ...
> > 
> > How is this so different?
> > 
> > 3.4.1 says MUST for SSNM (as it does for DATA)
> > whereas 4.3.5.3 says SHOULD
> > for ASP Active (ASPTM).
> > 
> > Valerie asked why M3UA allowed you to not insert any
> > RC in ASPAC and ASPIA
> > when it was required in SSNM.
> > 
> > Well, the answer is that it SHOULD be included in
> > ASPTM as well according
> > to the passage above.
> > 
> > Also, if you are configuring multiple AS over the
> > same SCTP association and
> > then activating without an RC, you are going against
> > the SHOULD above, and
> > you are on your own.
> > 
> > --brian
> > 
> > 
> > > > 
> > > > --brian
> > > > 
> > > > Valerie Gastaud wrote:                          
> >    
> > > >                               (Thu, 10 Jul 2003
> > > > 17:48:35)
> > > > > 
> > > > > Is there any reason why M3UA requires to
> > insert all the Routing Contexts in
> > > > > the SSNM messages when the message applies to
> > all the AS; while in the same
> > > > > situation, M3UA allows not to insert any RC in
> > the ASPAC and ASPIA messages.
> > > > > 
> > > > > Valerie
> > > > > 
> > > > > 
> > > > > 
> > > > >
> > _______________________________________________
> > > > > 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
> > > 
> > > 
> > > __________________________________
> > > Do you Yahoo!?
> > > SBC Yahoo! DSL - Now only $29.95 per month!
> > > http://sbc.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
> 
> 
> __________________________________
> Do you Yahoo!?
> SBC Yahoo! DSL - Now only $29.95 per month!
> http://sbc.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 exim@www1.ietf.org  Thu Jul 10 15:09:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA17214
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 15:09:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19agmt-0005LD-KJ
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 15:09:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AJ93bJ020511
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 15:09:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19agms-0005Kd-Oz; Thu, 10 Jul 2003 15:09:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19agml-0005K8-2E
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 15:08:55 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA17145
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 15:08:49 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19agmi-0006IP-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 15:08:52 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[H7gd1UOKD4mjVvoS+HS5xpirjGYJHfhH])
	by ietf-mx with esmtp (Exim 4.12)
	id 19agmh-0006IM-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 15:08:51 -0400
Received: from ns.pigworks.openss7.net (IDENT:sLTwEY5KG3tq1ajbZ66PnxuvGgeq6J+i@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6AJ8oX31406;
	Thu, 10 Jul 2003 13:08:50 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6AJ8nf24131;
	Thu, 10 Jul 2003 13:08:49 -0600
Date: Thu, 10 Jul 2003 13:08:49 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Anzile, Christophe" <christophe.anzile@hp.com>
Cc: john.loughney@nokia.com, sigtran@ietf.org
Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Message-ID: <20030710130849.L19145@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Anzile, Christophe" <christophe.anzile@hp.com>,
	john.loughney@nokia.com, sigtran@ietf.org
References: <BE462396B4A022449E0EB2DD4753D05402F2695C@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: <BE462396B4A022449E0EB2DD4753D05402F2695C@vbeexc02.emea.cpqcorp.net>; from christophe.anzile@hp.com on Thu, Jul 10, 2003 at 05:35:08PM +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>

Christophe,

IPSP SE is an ASP/SGP relationship and only the ASP side need send
ASP Id.

IPSP DE both send ASPAC and the ASP Id is not needed in ASPUP Ack.

So how would the IPSP use this?

You say to include it in some messages below, but not what the
receiver of the ASPUP Ack is expected (or able) to do with the value.

What would the SE IPSP receiving the ASPUp Ack (remember, acting in ASP
mode) do with it?  There are no procedures defined for an ASP to use
an "SGP Id" if you will...

So, I'm an SE IPSP acting in ASP mode.  I send you an ASPUP and you
send me an ASPUP Ack with an ASP Id in it.  What am I to do with it?
I never send Notify (you do), so what do I do with it?  What use is it?

--brian

Anzile, Christophe wrote:                                                          (Thu, 10 Jul 2003 17:35:08)
> My proposal would be :
> 
> The optional ASP Identifier parameter is specifically useful for IPSP 
> communication. In that case the IPSP answering the ASP Up message MAY 
> include its own ASP Identifier value, so that it can give back its ASP
> Identifier to the demanding IPSP in case of a single UP/UP-Ack exchange.
> For standard AS-SG communication, this parameter MUST NOT be used. 
> 
> 
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Thursday, July 10, 2003 3:45 PM
> To: john.loughney@nokia.com
> Cc: Anzile, Christophe; sigtran@ietf.org
> Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
> 
> 
> John,
> 
> We discussed this before an nobody had a real problem with it.
> 
> However, more than just adding the optional parameter to the
> ASP Up Ack, it also needs to be described how one should use it.
> Althoug a lot of uses were discussed, nobody put together a
> coherent proposal for its use.
> 
> Does someone have a simple proposal for it use?
> 
> --brian
> 
> John Loughney wrote:                                                                  (Thu, 10 Jul 2003 16:33:51)
> > Hi,
> > 
> > That could be reasonable - anyone have problems with this?
> > 
> > John
> > 
> > > -----Original Message-----
> > > From: ext Anzile, Christophe [mailto:christophe.anzile@hp.com]
> > > Sent: 10 July, 2003 12:26
> > > To: Loughney John (NRC/Helsinki)
> > > Cc: sigtran@ietf.org
> > > Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
> > > 
> > > 
> > > 
> > > Hi John,
> > > 
> > > section 3.5.2 : 
> > > it would be judicious to add an optional ASP-ID parameter in 
> > > the UP-ACK message,
> > > especially for IPSP communication (think of single exchange).
> > > 
> > > Regards,
> > > Christophe.
> > > 
> > 
> > _______________________________________________
> > 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 exim@www1.ietf.org  Thu Jul 10 15:14:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA17871
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 15:14:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19agri-0005wM-CO
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 15:14:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AJE2ZZ022816
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 15:14:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19agrh-0005vl-Ln; Thu, 10 Jul 2003 15:14:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19agqq-0005td-2m
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 15:13:08 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA17703
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 15:13:04 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19agqo-0006M3-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 15:13:06 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[tPRPa+QICXtLKvfoOHbaKhkcBuW+Ff0q])
	by ietf-mx with esmtp (Exim 4.12)
	id 19agqn-0006M0-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 15:13:06 -0400
Received: from ns.pigworks.openss7.net (IDENT:AdTlhDigDMgPjNh+3y5KyRsME3skPS6S@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6AJD4X31461;
	Thu, 10 Jul 2003 13:13:04 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6AJD4F24213;
	Thu, 10 Jul 2003 13:13:04 -0600
Date: Thu, 10 Jul 2003 13:13:04 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Desiderio, Didier" <didier.desiderio@hp.com>
Cc: john.loughney@nokia.com, sigtran@ietf.org
Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Message-ID: <20030710131304.M19145@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Desiderio, Didier" <didier.desiderio@hp.com>,
	john.loughney@nokia.com, sigtran@ietf.org
References: <BE462396B4A022449E0EB2DD4753D054022578E8@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: <BE462396B4A022449E0EB2DD4753D054022578E8@vbeexc02.emea.cpqcorp.net>; from didier.desiderio@hp.com on Fri, Jul 04, 2003 at 03:10:45PM +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>

Didier,

Yes, it should be Mandatory.

--brian

Desiderio, Didier wrote:                                                        (Fri, 04 Jul 2003 15:10:45)
> 
> 
> Some comments...
> 
> 
> 
> section 3.2.2 Connectionless Data Response (CLDR) 
> 
> 
> 
> If we refer to
> 
> ANSI T1.112 chap 4 & 4.11 
> 
>      T1.112.3-26 Table 11A/T1.112.3
> 
> ITU-T Q713 Chap 3.16, 3.20 and 4.11
> 
> 
> 
> Data parameter seems to be mandatory in UDTS, XUDTS and LUDTS.
> 
> Isn't it possible to have the DATA parameter of the CLDR message 
> 
> mandatory or semi optional (at least mandatory in case of ASP-SGP 
> 
> communication) and not only optional.
> 
> 
> 
> Regards
> 
> 
> 
> Didier
> 
> _______________________________________________
> 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 exim@www1.ietf.org  Thu Jul 10 15:18:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA18190
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 15:18:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19agva-000692-AN
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 15:18:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AJI2Ae023599
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 15:18:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19agvZ-00068T-GC; Thu, 10 Jul 2003 15:18:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19agvK-000680-F2
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 15:17:46 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA18170
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 15:17:42 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19agvJ-0006OM-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 15:17:45 -0400
Received: from rainier.illuminet.com ([12.41.110.52])
	by ietf-mx with esmtp (Exim 4.12)
	id 19agvI-0006OE-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 15:17:44 -0400
Received: from oly1wnexc02.corp.illuminet.com (oly1wnexc02.corp.illuminet.com [10.55.13.13]) by rainier.illuminet.com (8.8.8/8.8.8) with ESMTP id MAA18925; Thu, 10 Jul 2003 12:16:15 -0700 (PDT)
Received: by oly1wnexc02.corp.illuminet.com with Internet Mail Service (5.5.2653.19)
	id <N99YGC7B>; Thu, 10 Jul 2003 12:16:14 -0700
Message-ID: <6B830E7F37A7CF4A9806AE9D98AF42B5332432@atwinex01.corp.illuminet.com>
From: "Doss, James" <JDoss@verisign.com>
To: "'john.loughney@nokia.com'" <john.loughney@nokia.com>,
        christophe.anzile@hp.com
Cc: sigtran@ietf.org
Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Date: Thu, 10 Jul 2003 12:16:05 -0700
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 John,
The ASP-ID as an optional parameter would help to distinguish the 
ACK.
James Doss
VeriSign Inc



-----Original Message-----
From: john.loughney@nokia.com [mailto:john.loughney@nokia.com]
Sent: Thursday, July 10, 2003 9:34 AM
To: christophe.anzile@hp.com
Cc: sigtran@ietf.org
Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt


Hi,

That could be reasonable - anyone have problems with this?

John

> -----Original Message-----
> From: ext Anzile, Christophe [mailto:christophe.anzile@hp.com]
> Sent: 10 July, 2003 12:26
> To: Loughney John (NRC/Helsinki)
> Cc: sigtran@ietf.org
> Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
> 
> 
> 
> Hi John,
> 
> section 3.5.2 : 
> it would be judicious to add an optional ASP-ID parameter in 
> the UP-ACK message,
> especially for IPSP communication (think of single exchange).
> 
> Regards,
> Christophe.
> 

_______________________________________________
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 exim@www1.ietf.org  Thu Jul 10 15:28:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA18547
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 15:28:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ah5H-00077D-Nk
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 15:28:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AJS3kL027350
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 15:28:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ah5F-00076n-Rn; Thu, 10 Jul 2003 15:28:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ah5A-00076Z-QN
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 15:27:56 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA18527
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 15:27:53 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ah59-0006Tv-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 15:27:55 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[irGa+nR4ksOLlZjq+2AenBEoRSuZsV8J])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ah58-0006Ts-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 15:27:54 -0400
Received: from ns.pigworks.openss7.net (IDENT:AtNwP/vuZgcmB7nH5Q+4Lm3D//pHSfTy@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6AJRrX31623;
	Thu, 10 Jul 2003 13:27:53 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6AJRrj24511;
	Thu, 10 Jul 2003 13:27:53 -0600
Date: Thu, 10 Jul 2003 13:27:53 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Doss, James" <JDoss@verisign.com>
Cc: "'john.loughney@nokia.com'" <john.loughney@nokia.com>,
        christophe.anzile@hp.com, sigtran@ietf.org
Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Message-ID: <20030710132753.A24498@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Doss, James" <JDoss@verisign.com>,
	"'john.loughney@nokia.com'" <john.loughney@nokia.com>,
	christophe.anzile@hp.com, sigtran@ietf.org
References: <6B830E7F37A7CF4A9806AE9D98AF42B5332432@atwinex01.corp.illuminet.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: <6B830E7F37A7CF4A9806AE9D98AF42B5332432@atwinex01.corp.illuminet.com>; from JDoss@verisign.com on Thu, Jul 10, 2003 at 12:16:05PM -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>

James,

Distinguish it from what?  Other ASP Up Ack's???  There should
only be one ASP Up Ack on an SCTP association...

Please explain.

--brian

Doss, James wrote:                                                                   (Thu, 10 Jul 2003 12:16:05)
> Hi John,
> The ASP-ID as an optional parameter would help to distinguish the 
> ACK.
> James Doss
> VeriSign Inc
> 
> 
> 
> -----Original Message-----
> From: john.loughney@nokia.com [mailto:john.loughney@nokia.com]
> Sent: Thursday, July 10, 2003 9:34 AM
> To: christophe.anzile@hp.com
> Cc: sigtran@ietf.org
> Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
> 
> 
> Hi,
> 
> That could be reasonable - anyone have problems with this?
> 
> John
> 
> > -----Original Message-----
> > From: ext Anzile, Christophe [mailto:christophe.anzile@hp.com]
> > Sent: 10 July, 2003 12:26
> > To: Loughney John (NRC/Helsinki)
> > Cc: sigtran@ietf.org
> > Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
> > 
> > 
> > 
> > Hi John,
> > 
> > section 3.5.2 : 
> > it would be judicious to add an optional ASP-ID parameter in 
> > the UP-ACK message,
> > especially for IPSP communication (think of single exchange).
> > 
> > Regards,
> > Christophe.
> > 
> 
> _______________________________________________
> 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 exim@www1.ietf.org  Thu Jul 10 15:29:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA18603
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 15:29:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ah6C-0007Ln-PC
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 15:29:00 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AJT0ck028249
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 15:29:00 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ah6C-0007LX-Jj; Thu, 10 Jul 2003 15:29:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ah5X-0007IH-5n
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 15:28:19 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA18537
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 15:28:15 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ah5V-0006UD-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 15:28:17 -0400
Received: from rainier.illuminet.com ([12.41.110.52])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ah5U-0006Tr-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 15:28:17 -0400
Received: from oly1wnexc02.corp.illuminet.com (oly1wnexc02.corp.illuminet.com [10.55.13.13]) by rainier.illuminet.com (8.8.8/8.8.8) with ESMTP id MAA19408; Thu, 10 Jul 2003 12:27:20 -0700 (PDT)
Received: by oly1wnexc02.corp.illuminet.com with Internet Mail Service (5.5.2653.19)
	id <N99YGDAD>; Thu, 10 Jul 2003 12:27:19 -0700
Message-ID: <6B830E7F37A7CF4A9806AE9D98AF42B5332433@atwinex01.corp.illuminet.com>
From: "Doss, James" <JDoss@verisign.com>
To: sigtran@ietf.org
Cc: "'john.loughney@nokia.com'" <john.loughney@nokia.com>
Date: Thu, 10 Jul 2003 12:27:16 -0700
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C34719.45B17C80"
Subject: [Sigtran] SUA Question - Trasaction ID based Routing
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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.

------_=_NextPart_001_01C34719.45B17C80
Content-Type: text/plain;
	charset="iso-8859-1"

Hi John,
 
In SUA spec draft version 15.
 
With reference to Page 112 . SCTP Associations Fig 2 Signaling Gateway Architecture.
 
Will the transaction ID based routing the responses to the ASP originated the transaction.
 
Ex scenario:
 
ASP1 from AS3 originated a transaction using SG1 towards SS7 network.
 
The Response for this  QUERY hits SG2 (Assuming SG1 and SG2 are mated).
 
Question 1:
 
At this condition will SG2 route this message back to ASP1.
 
Question 2: 
 
Will SG1 and SG2 share the information about Open Transactions originated from the ASPs(1..n).
Or Else how the Transaction ID based routing (key ) will work.
 
Please correct me if I understood the routing in a wrong way.
 
Thanks and regards
James Doss
Sr Software Engineer
VeriSign Inc,
Norcross, GA
 
 
 

------_=_NextPart_001_01C34719.45B17C80
Content-Type: text/html;
	charset="iso-8859-1"

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">


<META content="MSHTML 5.50.4522.1800" name=GENERATOR></HEAD>
<BODY style="COLOR: #000000; FONT-FAMILY: Arial">
<DIV><SPAN class=907371719-10072003><FONT size=2>Hi John,</FONT></SPAN></DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2>In SUA spec draft version 
15.</FONT></SPAN></DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2>With reference to Page 112 . 
SCTP Associations Fig 2 Signaling Gateway Architecture.</FONT></SPAN></DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2>Will the transaction ID based 
routing the responses to the ASP originated the transaction.</FONT></SPAN></DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2>Ex 
scenario:</FONT></SPAN></DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2>ASP1 from AS3 originated a 
transaction using SG1 towards SS7 network.</FONT></SPAN></DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2>The Response for 
this&nbsp;&nbsp;QUERY hits SG2 (Assuming SG1 and SG2 are 
mated).</FONT></SPAN></DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2>Question 1:</FONT></SPAN></DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2>At this condition will SG2 
route this message back to ASP1.</FONT></SPAN></DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2>Question 2: 
</FONT></SPAN></DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2>Will SG1 and SG2 share the 
information about Open Transactions originated from the 
ASPs(1..n).</FONT></SPAN></DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2>Or Else how the Transaction ID 
based routing (key ) will work.</FONT></SPAN></DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2>Please correct me if I 
understood the routing in a wrong way.</FONT></SPAN></DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2>Thanks and 
regards</FONT></SPAN></DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2>James Doss</FONT></SPAN></DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2>Sr Software 
Engineer</FONT></SPAN></DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2>VeriSign 
Inc,</FONT></SPAN></DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2>Norcross, 
GA</FONT></SPAN></DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=907371719-10072003><FONT size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=907371719-10072003><FONT 
size=2></FONT></SPAN>&nbsp;</DIV></BODY></HTML>

------_=_NextPart_001_01C34719.45B17C80--

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



From exim@www1.ietf.org  Thu Jul 10 15:33:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA18741
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 15:33:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahA6-0007df-KB
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 15:33:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AJX2GA029356
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 15:33:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahA5-0007dB-IJ; Thu, 10 Jul 2003 15:33:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ah9r-0007c8-Ak
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 15:32:47 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA18695
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 15:32:43 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ah9q-0006Wm-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 15:32:46 -0400
Received: from zdemail04.zdem.compaq.com ([161.114.112.28])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ah9o-0006Wc-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 15:32:45 -0400
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail04.zdem.compaq.com (Postfix) with ESMTP
	id 25F6EFDC; Thu, 10 Jul 2003 21:32:08 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Thu, 10 Jul 2003 21:32:07 +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"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Date: Thu, 10 Jul 2003 21:32:06 +0200
Message-ID: <06CAA4D3753C53408D2CE6340B68444214C46A@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Thread-Index: AcNHFuqZuBYIPLH4SnC/ENZMG2aPHAAAOYsg
From: "Zannin, Francois" <francois.zannin@hp.com>
To: <bidulock@openss7.org>, "Anzile, Christophe" <christophe.anzile@hp.com>
Cc: <john.loughney@nokia.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 10 Jul 2003 19:32:08.0003 (UTC) FILETIME=[F3696130:01C34719]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Brian,

My understanding of the SE for M3UA (and I guess SUA) was that the =
single exchange of ASPTM (and others) request and acknowledgement was =
enough to Activate (or deactivate) Two AS identified by the same RC.
This do not remove the need to send the Notify to the service provider =
for an AS (the remote IPSP)!

Ex: If one IPSP can see one remote AS as Active, it sends the notify to =
all IPSP that offers the same AS. And this is true for all IPSP.=20
Now, if we consider that in single exchange mode, one IPSP is playing =
the role of ASP and the other one the role of SG, the behavior is not =
the same!=20

By reading the RFC, I can understand that the SE is just a shortcut to =
the DE where two AS uses the same RC on the same association, and one =
send the Active only if it is ready to handle  traffic and the second =
answer with the Active_Ack only if it is also ready to handle traffic. =
This is not the same thing as a simple SG/ASP emulation between two =
IPSP!!!

Can this point of the SE be more precisely detailed!!!

Regards,

Francois

-----Original Message-----
From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
Sent: Thursday, July 10, 2003 9:09 PM
To: Anzile, Christophe
Cc: john.loughney@nokia.com; sigtran@ietf.org
Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt


Christophe,

IPSP SE is an ASP/SGP relationship and only the ASP side need send
ASP Id.

IPSP DE both send ASPAC and the ASP Id is not needed in ASPUP Ack.

So how would the IPSP use this?

You say to include it in some messages below, but not what the
receiver of the ASPUP Ack is expected (or able) to do with the value.

What would the SE IPSP receiving the ASPUp Ack (remember, acting in ASP
mode) do with it?  There are no procedures defined for an ASP to use
an "SGP Id" if you will...

So, I'm an SE IPSP acting in ASP mode.  I send you an ASPUP and you
send me an ASPUP Ack with an ASP Id in it.  What am I to do with it?
I never send Notify (you do), so what do I do with it?  What use is it?

--brian

Anzile, Christophe wrote:                                                =
          (Thu, 10 Jul 2003 17:35:08)
> My proposal would be :
>=20
> The optional ASP Identifier parameter is specifically useful for IPSP=20
> communication. In that case the IPSP answering the ASP Up message MAY=20
> include its own ASP Identifier value, so that it can give back its ASP
> Identifier to the demanding IPSP in case of a single UP/UP-Ack =
exchange.
> For standard AS-SG communication, this parameter MUST NOT be used.=20
>=20
>=20
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Thursday, July 10, 2003 3:45 PM
> To: john.loughney@nokia.com
> Cc: Anzile, Christophe; sigtran@ietf.org
> Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
>=20
>=20
> John,
>=20
> We discussed this before an nobody had a real problem with it.
>=20
> However, more than just adding the optional parameter to the
> ASP Up Ack, it also needs to be described how one should use it.
> Althoug a lot of uses were discussed, nobody put together a
> coherent proposal for its use.
>=20
> Does someone have a simple proposal for it use?
>=20
> --brian
>=20
> John Loughney wrote:                                                   =
               (Thu, 10 Jul 2003 16:33:51)
> > Hi,
> >=20
> > That could be reasonable - anyone have problems with this?
> >=20
> > John
> >=20
> > > -----Original Message-----
> > > From: ext Anzile, Christophe [mailto:christophe.anzile@hp.com]
> > > Sent: 10 July, 2003 12:26
> > > To: Loughney John (NRC/Helsinki)
> > > Cc: sigtran@ietf.org
> > > Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
> > >=20
> > >=20
> > >=20
> > > Hi John,
> > >=20
> > > section 3.5.2 :=20
> > > it would be judicious to add an optional ASP-ID parameter in=20
> > > the UP-ACK message,
> > > especially for IPSP communication (think of single exchange).
> > >=20
> > > Regards,
> > > Christophe.
> > >=20
> >=20
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
>=20
> --=20
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/

--=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

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



From exim@www1.ietf.org  Thu Jul 10 15:39:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA18877
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 15:39:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahFt-0008Rf-W9
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 15:39:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AJd1G3032454
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 15:39:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahFt-0008RL-9u; Thu, 10 Jul 2003 15:39:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahEy-0008MA-QD
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 15:38:05 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA18853
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 15:38:00 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ahEx-0006ZI-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 15:38:03 -0400
Received: from rainier.illuminet.com ([12.41.110.52])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ahEw-0006Z1-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 15:38:02 -0400
Received: from oly1wnexc02.corp.illuminet.com (oly1wnexc02.corp.illuminet.com [10.55.13.13]) by rainier.illuminet.com (8.8.8/8.8.8) with ESMTP id MAA19838; Thu, 10 Jul 2003 12:36:16 -0700 (PDT)
Received: from oly1wnexc01.corp.illuminet.com ([10.55.13.12]) by oly1wnexc02.corp.illuminet.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2653.13)
	id N99YGDD9; Thu, 10 Jul 2003 12:36:15 -0700
Received: by oly1wnexc01.corp.illuminet.com with Internet Mail Service (5.5.2653.19)
	id <N9KG74SA>; Thu, 10 Jul 2003 12:36:14 -0700
Message-ID: <6B830E7F37A7CF4A9806AE9D98AF42B5332434@atwinex01.corp.illuminet.com>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "Doss, James" <JDoss@verisign.com>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>,
        "Doss, James"
	 <JDoss@verisign.com>
Cc: "'john.loughney@nokia.com'" <john.loughney@nokia.com>,
        christophe.anzile@hp.com, sigtran@ietf.org
Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Date: Thu, 10 Jul 2003 12:36:05 -0700
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 Brian,
In our present system we use 4 SCTP associations 2 port per (AS) server (1+1) to meet the traffic.
Presently I have one (application) ASP per server.
We are planning to increase to another ASP and we would like to use 1 or few ASPs to share the
same SCTP associations. In this scenario if the ASP ACK with out an ID will be confusing.
I am assuming more than one ASP applications can share the SCTP ports.
I do not find any reason to tie an SCTP port for an ASP.

Correct me if I am wrong.
James


-----Original Message-----
From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
Sent: Thursday, July 10, 2003 3:28 PM
To: Doss, James
Cc: 'john.loughney@nokia.com'; christophe.anzile@hp.com;
sigtran@ietf.org
Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt


James,

Distinguish it from what?  Other ASP Up Ack's???  There should
only be one ASP Up Ack on an SCTP association...

Please explain.

--brian

Doss, James wrote:                                                                   (Thu, 10 Jul 2003 12:16:05)
> Hi John,
> The ASP-ID as an optional parameter would help to distinguish the 
> ACK.
> James Doss
> VeriSign Inc
> 
> 
> 
> -----Original Message-----
> From: john.loughney@nokia.com [mailto:john.loughney@nokia.com]
> Sent: Thursday, July 10, 2003 9:34 AM
> To: christophe.anzile@hp.com
> Cc: sigtran@ietf.org
> Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
> 
> 
> Hi,
> 
> That could be reasonable - anyone have problems with this?
> 
> John
> 
> > -----Original Message-----
> > From: ext Anzile, Christophe [mailto:christophe.anzile@hp.com]
> > Sent: 10 July, 2003 12:26
> > To: Loughney John (NRC/Helsinki)
> > Cc: sigtran@ietf.org
> > Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
> > 
> > 
> > 
> > Hi John,
> > 
> > section 3.5.2 : 
> > it would be judicious to add an optional ASP-ID parameter in 
> > the UP-ACK message,
> > especially for IPSP communication (think of single exchange).
> > 
> > Regards,
> > Christophe.
> > 
> 
> _______________________________________________
> 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 exim@www1.ietf.org  Thu Jul 10 15:47:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA19119
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 15:47:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahNe-0000xU-6J
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 15:47:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AJl2KB003669
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 15:47:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahNd-0000wt-Tk; Thu, 10 Jul 2003 15:47:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahND-0000vo-6A
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 15:46:35 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA19090
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 15:46:31 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ahNB-0006dd-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 15:46:33 -0400
Received: from web12809.mail.yahoo.com ([216.136.174.136])
	by ietf-mx with smtp (Exim 4.12)
	id 19ahNA-0006dP-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 15:46:32 -0400
Message-ID: <20030710194632.63165.qmail@web12809.mail.yahoo.com>
Received: from [208.255.120.2] by web12809.mail.yahoo.com via HTTP; Thu, 10 Jul 2003 12:46:32 PDT
Date: Thu, 10 Jul 2003 12:46:32 -0700 (PDT)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] M3UA IG v04
To: sigtran@ietf.org
In-Reply-To: <20030710120033.J19145@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..]
> > [TOLGA] Yes, SGPs send DUPU to ASPs but to my
> > understanding, Francois's case would correspond
> > sending DUPU from ASP to SGP.
> > As I said, my interpretation of DE IPSP in the IG
> is a
> > combined SGP/ASP. On ASP part of IPSP, the user
> part
> > goes down, then I would expect inactivating the
> > corresponding AS.
> 
> Yes, that's what we understood and agreed, however,
> that is
> not what is presented in the IG: Read the IG again:
> New text in section 3.11.2:
> 
>    In section 4.3.1: ASP/IPSP States, only the
> SGP-ASP and the IPSP SE
>    scenarios are described. For the IPSP DE model,
> both IPSPs MUST
>    follow the SGP side of the SGP-ASP procedures.
> 
> That looks pretty plain to me.  According to the IG,
> DE IPSPs can
> send DUPU (they are actually required above to do so
> when the SGP
> procedure dictates).
[TOLGA]IMO, the sentence you refer might really be
mislaeading, but I don't think it should make the SGP
send DUPU for the case we are describing. Let's say we
have IPSP1 and IPSP2. If one user part is down on
IPSP2, then it should be ASP part of IPSP2 taking some
action and sending ASPIA, not SGP part of IPSP2.
> 
> --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!?
The New Yahoo! Search - Faster. Easier. Bingo.
http://search.yahoo.com

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



From exim@www1.ietf.org  Thu Jul 10 15:49:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA19201
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 15:49:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahPb-0001Jt-Sj
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 15:49:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AJn3Rg005066
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 15:49:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahPY-0001JN-OD; Thu, 10 Jul 2003 15:49:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahOr-00017d-6v
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 15:48:17 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA19158
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 15:48:13 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ahOp-0006fA-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 15:48:15 -0400
Received: from web12804.mail.yahoo.com ([216.136.174.39])
	by ietf-mx with smtp (Exim 4.12)
	id 19ahOp-0006f7-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 15:48:15 -0400
Message-ID: <20030710194814.20444.qmail@web12804.mail.yahoo.com>
Received: from [208.255.120.2] by web12804.mail.yahoo.com via HTTP; Thu, 10 Jul 2003 12:48:14 PDT
Date: Thu, 10 Jul 2003 12:48:14 -0700 (PDT)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: Re: [Sigtran] M3UA - RC in SSNM messages
To: sigtran@ietf.org
In-Reply-To: <20030710125835.K19145@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:                                
>                                 (Thu, 10 Jul 2003
> 11:10:20)
> > Brian,
> > 
> > I was referring to 4.3.4.3 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."
> > This enables the option to send ASPAC with no RC,
> even
> > if multiple AS traffic is carried over one
> > association. I don't think, this is permitted for
> > SSNM.
> 
> It permits a single AS in fitting with the SHOULD I
> quoted.  Multiple
> configured AS for the same association would be
> against the SHOULD.
> So you are still on your own.
[TOLGA]Then why does it say "Application Server(s)"
                                                ^ 
I am pretty sure, what is meant here (In fact even
just two days ago, there was a qustion related with
this)
> 
> --brian
> 
> > 
> >     Tolga
> > 
> > --- "Brian F. G. Bidulock" <bidulock@openss7.org>
> > wrote:
> > > Tolga,
> > > 
> > > Tolga Asveren wrote:                            
>    
> > >                                 (Thu, 10 Jul
> 2003
> > > 09:47:31)
> > > > Brian,
> > > > --- "Brian F. G. Bidulock"
> <bidulock@openss7.org>
> > > > wrote:
> > > > > Valerie,
> > > > > 
> > > > > M3UA does not require one to insert RCs in
> all
> > > SSNM
> > > > > message:
> > > > > 
> > > > > Section 3.4.1 of the RFC says:
> > > > > 
> > > > >    Routing Context: n x 32-bits (unsigned
> > > integer)
> > > > > 
> > > > >       The optional Routing Context parameter
> > > contains the Routing Context
> > > > >       values associated with the DUNA
> message. 
> > > Where a Routing Key has
> > > > >       not been coordinated between the SGP
> and
> > > ASP, sending of Routing
> > > > >       Context is not required.  Where
> multiple
> > > Routing Keys and Routing
> > > > >       Contexts are used across a common
> > > association, the Routing
> > > > >       Context(s) MUST be sent to identify
> the
> > > concerned traffic flows for
> > > > >       which the DUNA message applies,
> assisting
> > > in outgoing traffic
> > > > >       management and internal distribution
> of
> > > MTP-PAUSE indications to
> > > > >       MTP3-Users at the receiver.
> > > > > 
> > > > > This is really no different than the ASPTM
> > > messages.
> > > > [TOLGA] IMO it is not the same. An ASPAC with
> no
> > > RC
> > > > inside is applicable for configured AS -so
> even
> > > > multiple AS is served via the same
> association,
> > > one
> > > > can send ASPAC with no RC-, but RC is
> mandatory
> > > for
> > > > SSNM, traffic for more than one AS is carried
> over
> > > the
> > > > same association.
> > > 
> > > Section 4.3.4.3:
> > > 
> > >                     ...  In the case where an
> ASP
> > > wishes to process the
> > >    traffic for more than one Application Server
> > > across a common SCTP
> > >    association, the ASP Active message(s) SHOULD
> > > contain a list of one
> > >    or more Routing Contexts to indicate for
> which
> > > Application Servers
> > >    the ASP Active message applies. ...
> > > 
> > > How is this so different?
> > > 
> > > 3.4.1 says MUST for SSNM (as it does for DATA)
> > > whereas 4.3.5.3 says SHOULD
> > > for ASP Active (ASPTM).
> > > 
> > > Valerie asked why M3UA allowed you to not insert
> any
> > > RC in ASPAC and ASPIA
> > > when it was required in SSNM.
> > > 
> > > Well, the answer is that it SHOULD be included
> in
> > > ASPTM as well according
> > > to the passage above.
> > > 
> > > Also, if you are configuring multiple AS over
> the
> > > same SCTP association and
> > > then activating without an RC, you are going
> against
> > > the SHOULD above, and
> > > you are on your own.
> > > 
> > > --brian
> > > 
> > > 
> > > > > 
> > > > > --brian
> > > > > 
> > > > > Valerie Gastaud wrote:                      
>    
> > >    
> > > > >                               (Thu, 10 Jul
> 2003
> > > > > 17:48:35)
> > > > > > 
> > > > > > Is there any reason why M3UA requires to
> > > insert all the Routing Contexts in
> > > > > > the SSNM messages when the message applies
> to
> > > all the AS; while in the same
> > > > > > situation, M3UA allows not to insert any
> RC in
> > > the ASPAC and ASPIA messages.
> > > > > > 
> > > > > > Valerie
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > >
> > > _______________________________________________
> > > > > > 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
> > > > 
> > > > 
> > > > __________________________________
> > > > Do you Yahoo!?
> > > > SBC Yahoo! DSL - Now only $29.95 per month!
> > > > http://sbc.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
> 
=== message truncated ===


__________________________________
Do you Yahoo!?
SBC Yahoo! DSL - Now only $29.95 per month!
http://sbc.yahoo.com

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



From exim@www1.ietf.org  Thu Jul 10 16:05:32 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA19724
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 16:05:32 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahf6-0002MQ-Ut
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 16:05:06 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AK54A2009075
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 16:05:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahf3-0002Lw-20; Thu, 10 Jul 2003 16:05:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahed-0002Fb-W3
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 16:04:36 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA19665
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 16:04:31 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ahec-0006nH-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 16:04:34 -0400
Received: from zdemail03.zdem.compaq.com ([161.114.112.27])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aheb-0006mK-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 16:04:33 -0400
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail03.zdem.compaq.com (Postfix) with ESMTP
	id ADFA7117B; Thu, 10 Jul 2003 22:02:36 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Thu, 10 Jul 2003 22:02:36 +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"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] M3UA IG v04
Date: Thu, 10 Jul 2003 22:02:35 +0200
Message-ID: <06CAA4D3753C53408D2CE6340B68444201488AA4@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA IG v04
Thread-Index: AcNHHC9rHesBFnlmQs+xb8/F7GLslAAAGHxA
From: "Zannin, Francois" <francois.zannin@hp.com>
To: "Tolga Asveren" <tolga_asveren@yahoo.com>, <sigtran@ietf.org>
Cc: "Brian F. G. Bidulock" <bidulock@openss7.org>
X-OriginalArrivalTime: 10 Jul 2003 20:02:36.0566 (UTC) FILETIME=[3551F360:01C3471E]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

tolga,

see inline,

Regards,

Francois.

> -----Original Message-----
> From: Tolga Asveren [mailto:tolga_asveren@yahoo.com]
> Sent: Thursday, July 10, 2003 9:47 PM
> To: sigtran@ietf.org
> Subject: Re: [Sigtran] M3UA IG v04
>=20
>=20
> Brian,
> [..snip..]
> > > [TOLGA] Yes, SGPs send DUPU to ASPs but to my
> > > understanding, Francois's case would correspond
> > > sending DUPU from ASP to SGP.
> > > As I said, my interpretation of DE IPSP in the IG
> > is a
> > > combined SGP/ASP. On ASP part of IPSP, the user
> > part
> > > goes down, then I would expect inactivating the
> > > corresponding AS.
> >=20
> > Yes, that's what we understood and agreed, however,
> > that is
> > not what is presented in the IG: Read the IG again:
> > New text in section 3.11.2:
> >=20
> >    In section 4.3.1: ASP/IPSP States, only the
> > SGP-ASP and the IPSP SE
> >    scenarios are described. For the IPSP DE model,
> > both IPSPs MUST
> >    follow the SGP side of the SGP-ASP procedures.
> >=20
> > That looks pretty plain to me.  According to the IG,
> > DE IPSPs can
> > send DUPU (they are actually required above to do so
> > when the SGP
> > procedure dictates).
> [TOLGA]IMO, the sentence you refer might really be
> mislaeading, but I don't think it should make the SGP
> send DUPU for the case we are describing. Let's say we
> have IPSP1 and IPSP2. If one user part is down on
> IPSP2, then it should be ASP part of IPSP2 taking some
> action and sending ASPIA, not SGP part of IPSP2.

The case I was refering first, was a platform hosting only one AS but
internally made of several services (SI), not always available at the =
same
time.
If one SI is not available but the others are still "usable", if the AS
is "Inactivated", all SI will be prohibited! In that case the "IPSP" MAY
(that's my question) send a DUPU for the concerned SI but the traffic=20
for the others SI is still possible.
If IPSP cannot send a DUPU (or an UPU encoded), this mean that an AS
identified by a PC only CANNOT corretly play several SI.

So my question, is the DUPU allowed between two IPSP ?=20
   (or for a same reason between an AS and a SG(p) ? )

> > --brian
> >=20
> > --=20
> > Brian F. G. Bidulock
> > bidulock@openss7.org
> > http://www.openss7.org/
> >=20
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
>=20
>=20
> __________________________________
> Do you Yahoo!?
> The New Yahoo! Search - Faster. Easier. Bingo.
> http://search.yahoo.com
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>=20

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



From exim@www1.ietf.org  Thu Jul 10 16:13:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA19982
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 16:13:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahmo-0003ON-S5
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 16:13:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AKD2Yo013031
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 16:13:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahmn-0003Ne-Qi; Thu, 10 Jul 2003 16:13:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahme-0003N9-A3
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 16:12:52 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA19961
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 16:12:47 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ahmc-0006ri-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 16:12:50 -0400
Received: from web12803.mail.yahoo.com ([216.136.174.38])
	by ietf-mx with smtp (Exim 4.12)
	id 19ahmb-0006rf-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 16:12:49 -0400
Message-ID: <20030710201248.86479.qmail@web12803.mail.yahoo.com>
Received: from [208.255.120.2] by web12803.mail.yahoo.com via HTTP; Thu, 10 Jul 2003 13:12:48 PDT
Date: Thu, 10 Jul 2003 13:12:48 -0700 (PDT)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: RE: [Sigtran] M3UA IG v04
To: sigtran@ietf.org
In-Reply-To: <06CAA4D3753C53408D2CE6340B68444201488AA4@vbeexc02.emea.cpqcorp.net>
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>

Francois,

IMO, for the scenario you describe, sending ASPIA is
the correct behaviour. Yes, you won't be able to use
healthy user parts in such a case, but if you have
that level of granularity, I would think, RKs should
be defined accordingly(for example for your example
PC/SI RKs)
The same applies to ASP/SGP case. There is no DUPU
from ASP to SGP direction.
So, IMO, DUPU for IPSP case is not necessary.

     Tolga
--- "Zannin, Francois" <francois.zannin@hp.com> wrote:
> tolga,
> 
> see inline,
> 
> Regards,
> 
> Francois.
> 
> > -----Original Message-----
> > From: Tolga Asveren
> [mailto:tolga_asveren@yahoo.com]
> > Sent: Thursday, July 10, 2003 9:47 PM
> > To: sigtran@ietf.org
> > Subject: Re: [Sigtran] M3UA IG v04
> > 
> > 
> > Brian,
> > [..snip..]
> > > > [TOLGA] Yes, SGPs send DUPU to ASPs but to my
> > > > understanding, Francois's case would
> correspond
> > > > sending DUPU from ASP to SGP.
> > > > As I said, my interpretation of DE IPSP in the
> IG
> > > is a
> > > > combined SGP/ASP. On ASP part of IPSP, the
> user
> > > part
> > > > goes down, then I would expect inactivating
> the
> > > > corresponding AS.
> > > 
> > > Yes, that's what we understood and agreed,
> however,
> > > that is
> > > not what is presented in the IG: Read the IG
> again:
> > > New text in section 3.11.2:
> > > 
> > >    In section 4.3.1: ASP/IPSP States, only the
> > > SGP-ASP and the IPSP SE
> > >    scenarios are described. For the IPSP DE
> model,
> > > both IPSPs MUST
> > >    follow the SGP side of the SGP-ASP
> procedures.
> > > 
> > > That looks pretty plain to me.  According to the
> IG,
> > > DE IPSPs can
> > > send DUPU (they are actually required above to
> do so
> > > when the SGP
> > > procedure dictates).
> > [TOLGA]IMO, the sentence you refer might really be
> > mislaeading, but I don't think it should make the
> SGP
> > send DUPU for the case we are describing. Let's
> say we
> > have IPSP1 and IPSP2. If one user part is down on
> > IPSP2, then it should be ASP part of IPSP2 taking
> some
> > action and sending ASPIA, not SGP part of IPSP2.
> 
> The case I was refering first, was a platform
> hosting only one AS but
> internally made of several services (SI), not always
> available at the same
> time.
> If one SI is not available but the others are still
> "usable", if the AS
> is "Inactivated", all SI will be prohibited! In that
> case the "IPSP" MAY
> (that's my question) send a DUPU for the concerned
> SI but the traffic 
> for the others SI is still possible.
> If IPSP cannot send a DUPU (or an UPU encoded), this
> mean that an AS
> identified by a PC only CANNOT corretly play several
> SI.
> 
> So my question, is the DUPU allowed between two IPSP
> ? 
>    (or for a same reason between an AS and a SG(p) ?
> )
> 
> > > --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!?
> > The New Yahoo! Search - Faster. Easier. Bingo.
> > http://search.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


__________________________________
Do you Yahoo!?
SBC Yahoo! DSL - Now only $29.95 per month!
http://sbc.yahoo.com

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



From exim@www1.ietf.org  Thu Jul 10 16:14:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20042
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 16:14:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahnm-0003Zz-46
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 16:14:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AKE2Br013752
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 16:14:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahnl-0003Zi-Vi; Thu, 10 Jul 2003 16:14:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahnN-0003Wl-2y
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 16:13:41 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20000
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 16:13:32 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ahnL-0006sL-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 16:13:35 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[I5W8iPEiC4/3Fo/MDVmhdN76dimBz+nr])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ahnK-0006sH-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 16:13:34 -0400
Received: from ns.pigworks.openss7.net (IDENT:nvTZ9SSIcTLittHdsCpd375wbfDmd74E@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6AKDSX32386;
	Thu, 10 Jul 2003 14:13:28 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6AKDSt25408;
	Thu, 10 Jul 2003 14:13:28 -0600
Date: Thu, 10 Jul 2003 14:13:28 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Doss, James" <JDoss@verisign.com>
Cc: sigtran@ietf.org, "'john.loughney@nokia.com'" <john.loughney@nokia.com>
Subject: Re: [Sigtran] SUA Question - Trasaction ID based Routing
Message-ID: <20030710141328.B24498@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Doss, James" <JDoss@verisign.com>, sigtran@ietf.org,
	"'john.loughney@nokia.com'" <john.loughney@nokia.com>
References: <6B830E7F37A7CF4A9806AE9D98AF42B5332433@atwinex01.corp.illuminet.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: <6B830E7F37A7CF4A9806AE9D98AF42B5332433@atwinex01.corp.illuminet.com>; from JDoss@verisign.com on Thu, Jul 10, 2003 at 12:27:16PM -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>

James,

Doss, James wrote:                                                                   (Thu, 10 Jul 2003 12:27:16)
> 
>    Hi John,
> 
> 
> 
>    In SUA spec draft version 15.
> 
> 
> 
>    With reference to Page 112 . SCTP Associations Fig 2 Signaling Gateway
>    Architecture.

Page 112 would be the Author's Addresses...

> 
> 
> 
>    Will  the  transaction  ID  based  routing  the  responses  to the ASP
>    originated the transaction.
> 
> 
> 
>    Ex scenario:
> 
> 
> 
>    ASP1 from AS3 originated a transaction using SG1 towards SS7 network.
> 
> 
> 
>    The  Response  for  this  QUERY  hits  SG2  (Assuming  SG1 and SG2 are
>    mated).
> 
> 
> 
>    Question 1:
> 
> 
> 
>    At this condition will SG2 route this message back to ASP1.
> 
> 
> 
>    Question 2:
> 
> 
> 
>    Will  SG1  and  SG2  share  the  information  about  Open Transactions
>    originated from the ASPs(1..n).
> 
>    Or Else how the Transaction ID based routing (key ) will work.
> 
> 
> 
>    Please correct me if I understood the routing in a wrong way.

The TID Label approach described in 4.7.2 and 4.7.3 and A.3.1 will work for
simple (and correct) Query/Response pairs in protocol class 0 with operations
class 3 (no return on error and no rejects or aborts) if you have sufficient
control of the TCAP application that you can control the range of TIDs that
it assigns per ASP.  Protocol class 0 or 1 and operations class 1 or 2 (return
on error, rejects and aborts) will fail because the returned error does not
have sufficient information to determine the ASP from TID (because the ASP's
TID is not present in the message), or if you cannot strictly control the way
that the TCAP layer assigns TIDs at an ASP.

So, it works for only the most trivial of applications.  For more
sophisticated applications, see draft-bidulock-sigtran-tua-01.txt

A superior registration procedure was proposed whereby the ASP could inform
both SGs of the ASP ownership of a TID on the BEGIN/QWP/QWOP, but the broken
TID Label approach slithered its way into SUA.

The DRN Label approach is broken worse.  For serious TCAP applications, use
TUA.

The last thread was "SUA-09: SCCP Connections and TCAP Transactions" from
Jan 2002.  (At that time it was too late to make changes--hah!)

Before that there is a discussion in "SUA Decision needed - handlin multiple
SGs" where it was determined that broken procedures are ok if those broken
procedures have a couple of people to nod their heads at each other.

See especially "[SUA Issue 14] Changes and addtions required for decoupled
operation."  These problems go back to August 2001.

Perhaps Verwimp is still around to defend the Label approach, or perhaps
not.

--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 exim@www1.ietf.org  Thu Jul 10 16:24:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20340
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 16:24:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahxS-0004I6-D1
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 16:24:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AKO2No016487
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 16:24:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahxR-0004Hf-FL; Thu, 10 Jul 2003 16:24:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ahwx-0004H0-Nx
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 16:23:31 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20306
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 16:23:27 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ahwv-0006y4-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 16:23:29 -0400
Received: from zdemail03.zdem.compaq.com ([161.114.112.27])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ahwu-0006xd-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 16:23:29 -0400
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail03.zdem.compaq.com (Postfix) with ESMTP
	id CA541124C; Thu, 10 Jul 2003 22:22:58 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Thu, 10 Jul 2003 22:22:58 +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"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] M3UA IG v04
Date: Thu, 10 Jul 2003 22:22:57 +0200
Message-ID: <06CAA4D3753C53408D2CE6340B68444201488AA5@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA IG v04
Thread-Index: AcNHH9TFZUI9SDJzSVGVNrY7rW62lQAAKSag
From: "Zannin, Francois" <francois.zannin@hp.com>
To: "Tolga Asveren" <tolga_asveren@yahoo.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 10 Jul 2003 20:22:58.0628 (UTC) FILETIME=[0DB9D040:01C34721]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

OK, now, is the DUPU received on a SG or an IPSP considered as a =
protocol error?
This has to be clearly stated in the RFC (or IG)
Same remark for the SCON received from an IPSP, is it possible to =
receive a SCON
or must the IPSP use the SCTP congestion to handle the congestion on a =
remote IPSP.

Francois.

> -----Original Message-----
> From: Tolga Asveren [mailto:tolga_asveren@yahoo.com]
> Sent: Thursday, July 10, 2003 10:13 PM
> To: sigtran@ietf.org
> Subject: RE: [Sigtran] M3UA IG v04
>=20
>=20
> Francois,
>=20
> IMO, for the scenario you describe, sending ASPIA is
> the correct behaviour. Yes, you won't be able to use
> healthy user parts in such a case, but if you have
> that level of granularity, I would think, RKs should
> be defined accordingly(for example for your example
> PC/SI RKs)
> The same applies to ASP/SGP case. There is no DUPU
> from ASP to SGP direction.
> So, IMO, DUPU for IPSP case is not necessary.
>=20
>      Tolga
> --- "Zannin, Francois" <francois.zannin@hp.com> wrote:
> > tolga,
> >=20
> > see inline,
> >=20
> > Regards,
> >=20
> > Francois.
> >=20
> > > -----Original Message-----
> > > From: Tolga Asveren
> > [mailto:tolga_asveren@yahoo.com]
> > > Sent: Thursday, July 10, 2003 9:47 PM
> > > To: sigtran@ietf.org
> > > Subject: Re: [Sigtran] M3UA IG v04
> > >=20
> > >=20
> > > Brian,
> > > [..snip..]
> > > > > [TOLGA] Yes, SGPs send DUPU to ASPs but to my
> > > > > understanding, Francois's case would
> > correspond
> > > > > sending DUPU from ASP to SGP.
> > > > > As I said, my interpretation of DE IPSP in the
> > IG
> > > > is a
> > > > > combined SGP/ASP. On ASP part of IPSP, the
> > user
> > > > part
> > > > > goes down, then I would expect inactivating
> > the
> > > > > corresponding AS.
> > > >=20
> > > > Yes, that's what we understood and agreed,
> > however,
> > > > that is
> > > > not what is presented in the IG: Read the IG
> > again:
> > > > New text in section 3.11.2:
> > > >=20
> > > >    In section 4.3.1: ASP/IPSP States, only the
> > > > SGP-ASP and the IPSP SE
> > > >    scenarios are described. For the IPSP DE
> > model,
> > > > both IPSPs MUST
> > > >    follow the SGP side of the SGP-ASP
> > procedures.
> > > >=20
> > > > That looks pretty plain to me.  According to the
> > IG,
> > > > DE IPSPs can
> > > > send DUPU (they are actually required above to
> > do so
> > > > when the SGP
> > > > procedure dictates).
> > > [TOLGA]IMO, the sentence you refer might really be
> > > mislaeading, but I don't think it should make the
> > SGP
> > > send DUPU for the case we are describing. Let's
> > say we
> > > have IPSP1 and IPSP2. If one user part is down on
> > > IPSP2, then it should be ASP part of IPSP2 taking
> > some
> > > action and sending ASPIA, not SGP part of IPSP2.
> >=20
> > The case I was refering first, was a platform
> > hosting only one AS but
> > internally made of several services (SI), not always
> > available at the same
> > time.
> > If one SI is not available but the others are still
> > "usable", if the AS
> > is "Inactivated", all SI will be prohibited! In that
> > case the "IPSP" MAY
> > (that's my question) send a DUPU for the concerned
> > SI but the traffic=20
> > for the others SI is still possible.
> > If IPSP cannot send a DUPU (or an UPU encoded), this
> > mean that an AS
> > identified by a PC only CANNOT corretly play several
> > SI.
> >=20
> > So my question, is the DUPU allowed between two IPSP
> > ?=20
> >    (or for a same reason between an AS and a SG(p) ?
> > )
> >=20
> > > > --brian
> > > >=20
> > > > --=20
> > > > Brian F. G. Bidulock
> > > > bidulock@openss7.org
> > > > http://www.openss7.org/
> > > >=20
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >=20
> > >=20
> > > __________________________________
> > > Do you Yahoo!?
> > > The New Yahoo! Search - Faster. Easier. Bingo.
> > > http://search.yahoo.com
> > >=20
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >=20
> >=20
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
>=20
>=20
> __________________________________
> Do you Yahoo!?
> SBC Yahoo! DSL - Now only $29.95 per month!
> http://sbc.yahoo.com
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>=20

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



From exim@www1.ietf.org  Thu Jul 10 16:27:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20421
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 16:27:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ai0L-0004QV-Ku
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 16:27:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AKR1um017010
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 16:27:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ai0L-0004QF-HF; Thu, 10 Jul 2003 16:27:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ai0I-0004Pv-Gq
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 16:26:58 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20404
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 16:26:53 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ai0G-00070F-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 16:26:56 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[aVFVyWaxsEE1tmkHW+pKBpgV5TxiLk6h])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ai0F-00070C-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 16:26:56 -0400
Received: from ns.pigworks.openss7.net (IDENT:CoInk70H8cRTIOg+tioxhVPp4cQZY8hQ@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6AKQsX32529;
	Thu, 10 Jul 2003 14:26:54 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6AKQs525528;
	Thu, 10 Jul 2003 14:26:54 -0600
Date: Thu, 10 Jul 2003 14:26:54 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tolga Asveren <tolga_asveren@yahoo.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - RC in SSNM messages
Message-ID: <20030710142654.C24498@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030710125835.K19145@openss7.org> <20030710194814.20444.qmail@web12804.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: <20030710194814.20444.qmail@web12804.mail.yahoo.com>; from tolga_asveren@yahoo.com on Thu, Jul 10, 2003 at 12:48:14PM -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>

Tolga,


Tolga Asveren wrote:                                                                 (Thu, 10 Jul 2003 12:48:14)
> Brian,
> --- "Brian F. G. Bidulock" <bidulock@openss7.org>
> wrote:
> > Tolga,
> > 
> > Tolga Asveren wrote:                                
> >                                 (Thu, 10 Jul 2003
> > 11:10:20)
> > > Brian,
> > > 
> > > I was referring to 4.3.4.3 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."
> > > This enables the option to send ASPAC with no RC,
> > even
> > > if multiple AS traffic is carried over one
> > > association. I don't think, this is permitted for
> > > SSNM.
> > 
> > It permits a single AS in fitting with the SHOULD I
> > quoted.  Multiple
> > configured AS for the same association would be
> > against the SHOULD.
> > So you are still on your own.
> [TOLGA]Then why does it say "Application Server(s)"
>                                                 ^ 
> I am pretty sure, what is meant here (In fact even
> just two days ago, there was a qustion related with
> this)

If you recall, the (s) predates the RC in SSNM and DATA
discussion.  The SHOULD was put in after, at the same
time as the RC passages in SSNM and DATA.  If you have
multiple AS on the same association you need to put RC
in all messages.  This was one reason we changed "Optional"
to "Conditional' for RC throughout the document.  The
same condition applies in all cases.

--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 exim@www1.ietf.org  Thu Jul 10 16:30:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20573
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 16:30:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ai3G-0004j5-Km
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 16:30:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AKU2UN018158
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 16:30:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ai3F-0004il-Sd; Thu, 10 Jul 2003 16:30:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ai39-0004iR-3f
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 16:29:55 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20539
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 16:29:50 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ai37-00071o-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 16:29:53 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[XemgZ6nWKzRZng/X+657ubycy2q4uuyv])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ai36-00071l-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 16:29:52 -0400
Received: from ns.pigworks.openss7.net (IDENT:XCMtUlzqx+rS5d26psJy+TWUXwZ3hzuF@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6AKTpX32562;
	Thu, 10 Jul 2003 14:29:51 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6AKToB25571;
	Thu, 10 Jul 2003 14:29:50 -0600
Date: Thu, 10 Jul 2003 14:29:50 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Doss, James" <JDoss@verisign.com>
Cc: "'john.loughney@nokia.com'" <john.loughney@nokia.com>,
        christophe.anzile@hp.com, sigtran@ietf.org
Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Message-ID: <20030710142950.D24498@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Doss, James" <JDoss@verisign.com>,
	"'john.loughney@nokia.com'" <john.loughney@nokia.com>,
	christophe.anzile@hp.com, sigtran@ietf.org
References: <6B830E7F37A7CF4A9806AE9D98AF42B5332434@atwinex01.corp.illuminet.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: <6B830E7F37A7CF4A9806AE9D98AF42B5332434@atwinex01.corp.illuminet.com>; from JDoss@verisign.com on Thu, Jul 10, 2003 at 12:36:05PM -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>

James,

1 SCTP Association <=> 1 ASP.  The protocol is not intended to
support multiple ASPs per SCTP association.

--brian

Doss, James wrote:                                                                   (Thu, 10 Jul 2003 12:36:05)
> Hi Brian,
> In our present system we use 4 SCTP associations 2 port per (AS) server (1+1) to meet the traffic.
> Presently I have one (application) ASP per server.
> We are planning to increase to another ASP and we would like to use 1 or few ASPs to share the
> same SCTP associations. In this scenario if the ASP ACK with out an ID will be confusing.
> I am assuming more than one ASP applications can share the SCTP ports.
> I do not find any reason to tie an SCTP port for an ASP.
> 
> Correct me if I am wrong.
> James
> 
> 
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Thursday, July 10, 2003 3:28 PM
> To: Doss, James
> Cc: 'john.loughney@nokia.com'; christophe.anzile@hp.com;
> sigtran@ietf.org
> Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
> 
> 
> James,
> 
> Distinguish it from what?  Other ASP Up Ack's???  There should
> only be one ASP Up Ack on an SCTP association...
> 
> Please explain.
> 
> --brian
> 
> Doss, James wrote:                                                                   (Thu, 10 Jul 2003 12:16:05)
> > Hi John,
> > The ASP-ID as an optional parameter would help to distinguish the 
> > ACK.
> > James Doss
> > VeriSign Inc
> > 
> > 
> > 
> > -----Original Message-----
> > From: john.loughney@nokia.com [mailto:john.loughney@nokia.com]
> > Sent: Thursday, July 10, 2003 9:34 AM
> > To: christophe.anzile@hp.com
> > Cc: sigtran@ietf.org
> > Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
> > 
> > 
> > Hi,
> > 
> > That could be reasonable - anyone have problems with this?
> > 
> > John
> > 
> > > -----Original Message-----
> > > From: ext Anzile, Christophe [mailto:christophe.anzile@hp.com]
> > > Sent: 10 July, 2003 12:26
> > > To: Loughney John (NRC/Helsinki)
> > > Cc: sigtran@ietf.org
> > > Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
> > > 
> > > 
> > > 
> > > Hi John,
> > > 
> > > section 3.5.2 : 
> > > it would be judicious to add an optional ASP-ID parameter in 
> > > the UP-ACK message,
> > > especially for IPSP communication (think of single exchange).
> > > 
> > > Regards,
> > > Christophe.
> > > 
> > 
> > _______________________________________________
> > 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 exim@www1.ietf.org  Thu Jul 10 16:35:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20739
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 16:35:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ai87-0005RE-Dd
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 16:35:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AKZ3RQ020898
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 16:35:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ai85-0005Qv-Pb; Thu, 10 Jul 2003 16:35:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ai7a-0005PF-V5
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 16:34:31 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20712
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 16:34:26 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ai7Z-00074a-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 16:34:29 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[Zv0An8fq5AdzeWG9KKwWMdGE1QWdLZmd])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ai7Y-00074X-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 16:34:28 -0400
Received: from ns.pigworks.openss7.net (IDENT:qlc4xanbgvXlHAO8yRjGqVdXWCnQZyBY@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6AKYRX32703;
	Thu, 10 Jul 2003 14:34:27 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6AKYQN25730;
	Thu, 10 Jul 2003 14:34:26 -0600
Date: Thu, 10 Jul 2003 14:34:26 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Zannin, Francois" <francois.zannin@hp.com>
Cc: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA IG v04
Message-ID: <20030710143426.E24498@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Zannin, Francois" <francois.zannin@hp.com>,
	Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <06CAA4D3753C53408D2CE6340B68444201488AA5@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: <06CAA4D3753C53408D2CE6340B68444201488AA5@vbeexc02.emea.cpqcorp.net>; from francois.zannin@hp.com on Thu, Jul 10, 2003 at 10:22:57PM +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>

Francois,

I don't think IPSP is that well defined yet.  We were supposed to address
it in an extension draft.  Some corrections have been made in the IG, but
a lot of work remains to be done.  We spent some time defining SG-SG
operation which makes more sense that the SE or DE models, but there wasn't
enough interest to advance the work.

--brian

Zannin, Francois wrote:                                                            (Thu, 10 Jul 2003 22:22:57)
> OK, now, is the DUPU received on a SG or an IPSP considered as a protocol error?
> This has to be clearly stated in the RFC (or IG)
> Same remark for the SCON received from an IPSP, is it possible to receive a SCON
> or must the IPSP use the SCTP congestion to handle the congestion on a remote IPSP.
> 
> Francois.
> 
> > -----Original Message-----
> > From: Tolga Asveren [mailto:tolga_asveren@yahoo.com]
> > Sent: Thursday, July 10, 2003 10:13 PM
> > To: sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA IG v04
> > 
> > 
> > Francois,
> > 
> > IMO, for the scenario you describe, sending ASPIA is
> > the correct behaviour. Yes, you won't be able to use
> > healthy user parts in such a case, but if you have
> > that level of granularity, I would think, RKs should
> > be defined accordingly(for example for your example
> > PC/SI RKs)
> > The same applies to ASP/SGP case. There is no DUPU
> > from ASP to SGP direction.
> > So, IMO, DUPU for IPSP case is not necessary.
> > 
> >      Tolga
> > --- "Zannin, Francois" <francois.zannin@hp.com> wrote:
> > > tolga,
> > > 
> > > see inline,
> > > 
> > > Regards,
> > > 
> > > Francois.
> > > 
> > > > -----Original Message-----
> > > > From: Tolga Asveren
> > > [mailto:tolga_asveren@yahoo.com]
> > > > Sent: Thursday, July 10, 2003 9:47 PM
> > > > To: sigtran@ietf.org
> > > > Subject: Re: [Sigtran] M3UA IG v04
> > > > 
> > > > 
> > > > Brian,
> > > > [..snip..]
> > > > > > [TOLGA] Yes, SGPs send DUPU to ASPs but to my
> > > > > > understanding, Francois's case would
> > > correspond
> > > > > > sending DUPU from ASP to SGP.
> > > > > > As I said, my interpretation of DE IPSP in the
> > > IG
> > > > > is a
> > > > > > combined SGP/ASP. On ASP part of IPSP, the
> > > user
> > > > > part
> > > > > > goes down, then I would expect inactivating
> > > the
> > > > > > corresponding AS.
> > > > > 
> > > > > Yes, that's what we understood and agreed,
> > > however,
> > > > > that is
> > > > > not what is presented in the IG: Read the IG
> > > again:
> > > > > New text in section 3.11.2:
> > > > > 
> > > > >    In section 4.3.1: ASP/IPSP States, only the
> > > > > SGP-ASP and the IPSP SE
> > > > >    scenarios are described. For the IPSP DE
> > > model,
> > > > > both IPSPs MUST
> > > > >    follow the SGP side of the SGP-ASP
> > > procedures.
> > > > > 
> > > > > That looks pretty plain to me.  According to the
> > > IG,
> > > > > DE IPSPs can
> > > > > send DUPU (they are actually required above to
> > > do so
> > > > > when the SGP
> > > > > procedure dictates).
> > > > [TOLGA]IMO, the sentence you refer might really be
> > > > mislaeading, but I don't think it should make the
> > > SGP
> > > > send DUPU for the case we are describing. Let's
> > > say we
> > > > have IPSP1 and IPSP2. If one user part is down on
> > > > IPSP2, then it should be ASP part of IPSP2 taking
> > > some
> > > > action and sending ASPIA, not SGP part of IPSP2.
> > > 
> > > The case I was refering first, was a platform
> > > hosting only one AS but
> > > internally made of several services (SI), not always
> > > available at the same
> > > time.
> > > If one SI is not available but the others are still
> > > "usable", if the AS
> > > is "Inactivated", all SI will be prohibited! In that
> > > case the "IPSP" MAY
> > > (that's my question) send a DUPU for the concerned
> > > SI but the traffic 
> > > for the others SI is still possible.
> > > If IPSP cannot send a DUPU (or an UPU encoded), this
> > > mean that an AS
> > > identified by a PC only CANNOT corretly play several
> > > SI.
> > > 
> > > So my question, is the DUPU allowed between two IPSP
> > > ? 
> > >    (or for a same reason between an AS and a SG(p) ?
> > > )
> > > 
> > > > > --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!?
> > > > The New Yahoo! Search - Faster. Easier. Bingo.
> > > > http://search.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
> > 
> > 
> > __________________________________
> > Do you Yahoo!?
> > SBC Yahoo! DSL - Now only $29.95 per month!
> > http://sbc.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

-- 
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 exim@www1.ietf.org  Thu Jul 10 16:37:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20776
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 16:37:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aiA1-0005h8-P6
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 16:37:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AKb1w1021877
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 16:37:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aiA1-0005gk-HW; Thu, 10 Jul 2003 16:37:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aiA0-0005gF-71
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 16:37:00 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20769
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 16:36:55 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ai9y-00075H-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 16:36:58 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[pXW1PgmRTBeqygcxD5c4El9qWmyqWe7L])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ai9x-00075E-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 16:36:57 -0400
Received: from ns.pigworks.openss7.net (IDENT:PAjJu8VbTfHyzSm+2J+BVBUwY2FljMeF@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6AKauX32743;
	Thu, 10 Jul 2003 14:36:56 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6AKaue25831;
	Thu, 10 Jul 2003 14:36:56 -0600
Date: Thu, 10 Jul 2003 14:36:55 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Pablo Salinas <pablopesb@yahoo.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] SCTP association in a SGP scenario
Message-ID: <20030710143655.F24498@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Pablo Salinas <pablopesb@yahoo.com>, sigtran@ietf.org
References: <20030710150722.8707.qmail@web41809.mail.yahoo.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20030710150722.8707.qmail@web41809.mail.yahoo.com>; from pablopesb@yahoo.com on Thu, Jul 10, 2003 at 10:07:22AM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6AKauX32743
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Pablo,

Yes, SCTP can not only form many associations from a single
interface, it can form many associations from a single IP:Port.
It is no different that TCP in that regard.

Multihoming adds some restrictions "across" cards.

--brian


Pablo Salinas wrote:                                                     =
            (Thu, 10 Jul 2003 10:07:22)
> Hi there,
>           Could anybody tell me if this situation is
> possible?
>=20
> |--------------SGP-----------------------||----MG---|
>=20
> IID 1 --AS1----ASP1-(65.198.40.50:3005)-SCTP1---ASP1
> IID 2 --/  \---ASP2-(65.198.40.50:3006)-SCTP2---ASP2
>=20
> IID 3 --AS2----ASP3-(65.198.40.50:3007)-SCTP2---ASP3
>=20
> I mean, the sctp addresses in the SGP are IP:Port
> So that, for one Ethernet card, I could have many SCTP
> associations. Could this be possible?
>=20
> Hope anyone can answer my question.
>=20
>                 thanks,
>                         Pablo Salinas
>=20
> _________________________________________________________
> Do You Yahoo!?
> Informaci=F3n de Estados Unidos y Am=E9rica Latina, en Yahoo! Noticias.
> Vis=EDtanos en http://noticias.espanol.yahoo.com
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--=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



From exim@www1.ietf.org  Thu Jul 10 16:41:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20942
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 16:41:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aiDt-0006LS-HE
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 16:41:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6AKf1ZA024385
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 16:41:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aiDt-0006L1-6w; Thu, 10 Jul 2003 16:41:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aiDK-0006HC-Mm
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 16:40:26 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20870
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 16:40:21 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aiDI-00077G-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 16:40:24 -0400
Received: from web12801.mail.yahoo.com ([216.136.174.36])
	by ietf-mx with smtp (Exim 4.12)
	id 19aiDH-000777-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 16:40:23 -0400
Message-ID: <20030710204022.33375.qmail@web12801.mail.yahoo.com>
Received: from [208.255.120.2] by web12801.mail.yahoo.com via HTTP; Thu, 10 Jul 2003 13:40:22 PDT
Date: Thu, 10 Jul 2003 13:40:22 -0700 (PDT)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: RE: [Sigtran] M3UA IG v04
To: sigtran@ietf.org
In-Reply-To: <06CAA4D3753C53408D2CE6340B68444201488AA5@vbeexc02.emea.cpqcorp.net>
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>

Francois,
--- "Zannin, Francois" <francois.zannin@hp.com> wrote:
> OK, now, is the DUPU received on a SG or an IPSP
> considered as a protocol error?
> This has to be clearly stated in the RFC (or IG)
> Same remark for the SCON received from an IPSP, is
> it possible to receive a SCON
> or must the IPSP use the SCTP congestion to handle
> the congestion on a remote IPSP.
[TOLGA] Again I would think, SCON is not necessary for
IPSP communication. 
In fact I think, it could be nice to have an ASPTM to
inform about congestion between peers, because IMO it
is more elegant than waiting the congestion to
propogate to the remote SCTP -maybe using ECN might
help- but I am representing the minority on this issue
(a minority of 1 person :) ) 
> 
> Francois.
> 
> > -----Original Message-----
> > From: Tolga Asveren
> [mailto:tolga_asveren@yahoo.com]
> > Sent: Thursday, July 10, 2003 10:13 PM
> > To: sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA IG v04
> > 
> > 
> > Francois,
> > 
> > IMO, for the scenario you describe, sending ASPIA
> is
> > the correct behaviour. Yes, you won't be able to
> use
> > healthy user parts in such a case, but if you have
> > that level of granularity, I would think, RKs
> should
> > be defined accordingly(for example for your
> example
> > PC/SI RKs)
> > The same applies to ASP/SGP case. There is no DUPU
> > from ASP to SGP direction.
> > So, IMO, DUPU for IPSP case is not necessary.
> > 
> >      Tolga
> > --- "Zannin, Francois" <francois.zannin@hp.com>
> wrote:
> > > tolga,
> > > 
> > > see inline,
> > > 
> > > Regards,
> > > 
> > > Francois.
> > > 
> > > > -----Original Message-----
> > > > From: Tolga Asveren
> > > [mailto:tolga_asveren@yahoo.com]
> > > > Sent: Thursday, July 10, 2003 9:47 PM
> > > > To: sigtran@ietf.org
> > > > Subject: Re: [Sigtran] M3UA IG v04
> > > > 
> > > > 
> > > > Brian,
> > > > [..snip..]
> > > > > > [TOLGA] Yes, SGPs send DUPU to ASPs but to
> my
> > > > > > understanding, Francois's case would
> > > correspond
> > > > > > sending DUPU from ASP to SGP.
> > > > > > As I said, my interpretation of DE IPSP in
> the
> > > IG
> > > > > is a
> > > > > > combined SGP/ASP. On ASP part of IPSP, the
> > > user
> > > > > part
> > > > > > goes down, then I would expect
> inactivating
> > > the
> > > > > > corresponding AS.
> > > > > 
> > > > > Yes, that's what we understood and agreed,
> > > however,
> > > > > that is
> > > > > not what is presented in the IG: Read the IG
> > > again:
> > > > > New text in section 3.11.2:
> > > > > 
> > > > >    In section 4.3.1: ASP/IPSP States, only
> the
> > > > > SGP-ASP and the IPSP SE
> > > > >    scenarios are described. For the IPSP DE
> > > model,
> > > > > both IPSPs MUST
> > > > >    follow the SGP side of the SGP-ASP
> > > procedures.
> > > > > 
> > > > > That looks pretty plain to me.  According to
> the
> > > IG,
> > > > > DE IPSPs can
> > > > > send DUPU (they are actually required above
> to
> > > do so
> > > > > when the SGP
> > > > > procedure dictates).
> > > > [TOLGA]IMO, the sentence you refer might
> really be
> > > > mislaeading, but I don't think it should make
> the
> > > SGP
> > > > send DUPU for the case we are describing.
> Let's
> > > say we
> > > > have IPSP1 and IPSP2. If one user part is down
> on
> > > > IPSP2, then it should be ASP part of IPSP2
> taking
> > > some
> > > > action and sending ASPIA, not SGP part of
> IPSP2.
> > > 
> > > The case I was refering first, was a platform
> > > hosting only one AS but
> > > internally made of several services (SI), not
> always
> > > available at the same
> > > time.
> > > If one SI is not available but the others are
> still
> > > "usable", if the AS
> > > is "Inactivated", all SI will be prohibited! In
> that
> > > case the "IPSP" MAY
> > > (that's my question) send a DUPU for the
> concerned
> > > SI but the traffic 
> > > for the others SI is still possible.
> > > If IPSP cannot send a DUPU (or an UPU encoded),
> this
> > > mean that an AS
> > > identified by a PC only CANNOT corretly play
> several
> > > SI.
> > > 
> > > So my question, is the DUPU allowed between two
> IPSP
> > > ? 
> > >    (or for a same reason between an AS and a
> SG(p) ?
> > > )
> > > 
> > > > > --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!?
> > > > The New Yahoo! Search - Faster. Easier. Bingo.
> > > > http://search.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
> > 
> > 
> > __________________________________
> > Do you Yahoo!?
> > SBC Yahoo! DSL - Now only $29.95 per month!
> > http://sbc.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


__________________________________
Do you Yahoo!?
SBC Yahoo! DSL - Now only $29.95 per month!
http://sbc.yahoo.com

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



From exim@www1.ietf.org  Thu Jul 10 17:15:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA22096
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 17:15:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aikr-0000wh-6T
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 17:15:06 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ALF5dW003601
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 17:15:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aikn-0000vf-M0; Thu, 10 Jul 2003 17:15:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19adSQ-0001uz-VG
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 11:35:43 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA07736
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 11:35:38 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19adSP-00044J-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 11:35:42 -0400
Received: from zdemail03.zdem.compaq.com ([161.114.112.27])
	by ietf-mx with esmtp (Exim 4.12)
	id 19adSP-000440-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 11:35:41 -0400
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail03.zdem.compaq.com (Postfix) with ESMTP
	id 7AD111363; Thu, 10 Jul 2003 17:35:09 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Thu, 10 Jul 2003 17:35:09 +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"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Date: Thu, 10 Jul 2003 17:35:08 +0200
Message-ID: <BE462396B4A022449E0EB2DD4753D05402F2695C@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Thread-Index: AcNG6Xg0WbYN6nqqRIqaAXIBnObB6AADaDAQ
From: "Anzile, Christophe" <christophe.anzile@hp.com>
To: <bidulock@openss7.org>, <john.loughney@nokia.com>
Cc: <sigtran@ietf.org>
X-OriginalArrivalTime: 10 Jul 2003 15:35:09.0300 (UTC) FILETIME=[D8677B40:01C346F8]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

My proposal would be :

The optional ASP Identifier parameter is specifically useful for IPSP=20
communication. In that case the IPSP answering the ASP Up message MAY=20
include its own ASP Identifier value, so that it can give back its ASP
Identifier to the demanding IPSP in case of a single UP/UP-Ack exchange.
For standard AS-SG communication, this parameter MUST NOT be used.=20


-----Original Message-----
From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
Sent: Thursday, July 10, 2003 3:45 PM
To: john.loughney@nokia.com
Cc: Anzile, Christophe; sigtran@ietf.org
Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt


John,

We discussed this before an nobody had a real problem with it.

However, more than just adding the optional parameter to the
ASP Up Ack, it also needs to be described how one should use it.
Althoug a lot of uses were discussed, nobody put together a
coherent proposal for its use.

Does someone have a simple proposal for it use?

--brian

John Loughney wrote:                                                     =
             (Thu, 10 Jul 2003 16:33:51)
> Hi,
>=20
> That could be reasonable - anyone have problems with this?
>=20
> John
>=20
> > -----Original Message-----
> > From: ext Anzile, Christophe [mailto:christophe.anzile@hp.com]
> > Sent: 10 July, 2003 12:26
> > To: Loughney John (NRC/Helsinki)
> > Cc: sigtran@ietf.org
> > Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
> >=20
> >=20
> >=20
> > Hi John,
> >=20
> > section 3.5.2 :=20
> > it would be judicious to add an optional ASP-ID parameter in=20
> > the UP-ACK message,
> > especially for IPSP communication (think of single exchange).
> >=20
> > Regards,
> > Christophe.
> >=20
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--=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



From exim@www1.ietf.org  Thu Jul 10 18:01:08 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA22095
	for <sigtran-archive@odin.ietf.org>; Thu, 10 Jul 2003 17:15:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aikr-0000we-66
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 17:15:06 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ALF5Ka003602
	for sigtran-archive@odin.ietf.org; Thu, 10 Jul 2003 17:15:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aikn-0000vV-4C; Thu, 10 Jul 2003 17:15:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aXgl-0006Qv-8z
	for sigtran@optimus.ietf.org; Thu, 10 Jul 2003 05:26:07 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA23710
	for <sigtran@ietf.org>; Thu, 10 Jul 2003 05:26:01 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aXgh-0000tQ-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 05:26:03 -0400
Received: from zdemail04.zdem.compaq.com ([161.114.112.28])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aXgg-0000tI-00
	for sigtran@ietf.org; Thu, 10 Jul 2003 05:26:03 -0400
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail04.zdem.compaq.com (Postfix) with ESMTP
	id 83CA3EA8; Thu, 10 Jul 2003 11:25:33 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Thu, 10 Jul 2003 11:25:33 +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"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Date: Thu, 10 Jul 2003 11:25:31 +0200
Message-ID: <BE462396B4A022449E0EB2DD4753D05402F26948@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Thread-Index: AcNBeNCN+OPV66NiRz2QUXHtMtAv4wAB2wyQAVENuwA=
From: "Anzile, Christophe" <christophe.anzile@hp.com>
To: <john.loughney@nokia.com>
Cc: <sigtran@ietf.org>
X-OriginalArrivalTime: 10 Jul 2003 09:25:33.0300 (UTC) FILETIME=[367A6B40:01C346C5]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable


Hi John,

section 3.5.2 :=20
it would be judicious to add an optional ASP-ID parameter in the UP-ACK =
message,
especially for IPSP communication (think of single exchange).

Regards,
Christophe.

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



From exim@www1.ietf.org  Fri Jul 11 02:56:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA18603
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 02:56:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19arp5-0003x4-S3
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 02:56:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B6u3hA015189
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 02:56:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19arp3-0003wY-Ro; Fri, 11 Jul 2003 02:56:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aroY-0003vX-9e
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 02:55:30 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA18583
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 02:55:26 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aroU-0002wr-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 02:55:26 -0400
Received: from zdemail04.zdem.compaq.com ([161.114.112.28])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aroT-0002wU-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 02:55:25 -0400
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail04.zdem.compaq.com (Postfix) with ESMTP
	id A8FE1EE5; Fri, 11 Jul 2003 08:54:46 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Fri, 11 Jul 2003 08:54:46 +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="us-ascii"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] M3UA IG v04
Date: Fri, 11 Jul 2003 08:54:45 +0200
Message-ID: <06CAA4D3753C53408D2CE6340B68444201488AA6@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA IG v04
Thread-Index: AcNHI7nEHxP6YZYxRSmJ3aRzlS+qGwAVO0uA
From: "Zannin, Francois" <francois.zannin@hp.com>
To: "Tolga Asveren" <tolga_asveren@yahoo.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 11 Jul 2003 06:54:46.0539 (UTC) FILETIME=[5099FDB0:01C34779]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Tolga,

IN the case of IPSP (but also SGP), if we send an ASPTM (INACTIVE I
guess) would result in the impossibility to exchange messages between
IPSP and then the congestion that can be resorbed by exchanges messages
for already open transaction is eliminated. I don't think ASPTM can be
the best solution to controle the congestion.

Francois.

> -----Original Message-----
> From: Tolga Asveren [mailto:tolga_asveren@yahoo.com]=20
> Sent: Thursday, July 10, 2003 10:40 PM
> To: sigtran@ietf.org
> Subject: RE: [Sigtran] M3UA IG v04
>=20
>=20
> Francois,
> --- "Zannin, Francois" <francois.zannin@hp.com> wrote:
> > OK, now, is the DUPU received on a SG or an IPSP
> > considered as a protocol error?
> > This has to be clearly stated in the RFC (or IG)
> > Same remark for the SCON received from an IPSP, is
> > it possible to receive a SCON
> > or must the IPSP use the SCTP congestion to handle
> > the congestion on a remote IPSP.
> [TOLGA] Again I would think, SCON is not necessary for
> IPSP communication.=20
> In fact I think, it could be nice to have an ASPTM to
> inform about congestion between peers, because IMO it
> is more elegant than waiting the congestion to
> propogate to the remote SCTP -maybe using ECN might
> help- but I am representing the minority on this issue
> (a minority of 1 person :) )=20
> >=20
> > Francois.
> >=20
> > > -----Original Message-----
> > > From: Tolga Asveren
> > [mailto:tolga_asveren@yahoo.com]
> > > Sent: Thursday, July 10, 2003 10:13 PM
> > > To: sigtran@ietf.org
> > > Subject: RE: [Sigtran] M3UA IG v04
> > >=20
> > >=20
> > > Francois,
> > >=20
> > > IMO, for the scenario you describe, sending ASPIA
> > is
> > > the correct behaviour. Yes, you won't be able to
> > use
> > > healthy user parts in such a case, but if you have
> > > that level of granularity, I would think, RKs
> > should
> > > be defined accordingly(for example for your
> > example
> > > PC/SI RKs)
> > > The same applies to ASP/SGP case. There is no DUPU
> > > from ASP to SGP direction.
> > > So, IMO, DUPU for IPSP case is not necessary.
> > >=20
> > >      Tolga
> > > --- "Zannin, Francois" <francois.zannin@hp.com>
> > wrote:
> > > > tolga,
> > > >=20
> > > > see inline,
> > > >=20
> > > > Regards,
> > > >=20
> > > > Francois.
> > > >=20
> > > > > -----Original Message-----
> > > > > From: Tolga Asveren
> > > > [mailto:tolga_asveren@yahoo.com]
> > > > > Sent: Thursday, July 10, 2003 9:47 PM
> > > > > To: sigtran@ietf.org
> > > > > Subject: Re: [Sigtran] M3UA IG v04
> > > > >=20
> > > > >=20
> > > > > Brian,
> > > > > [..snip..]
> > > > > > > [TOLGA] Yes, SGPs send DUPU to ASPs but to
> > my
> > > > > > > understanding, Francois's case would
> > > > correspond
> > > > > > > sending DUPU from ASP to SGP.
> > > > > > > As I said, my interpretation of DE IPSP in
> > the
> > > > IG
> > > > > > is a
> > > > > > > combined SGP/ASP. On ASP part of IPSP, the
> > > > user
> > > > > > part
> > > > > > > goes down, then I would expect
> > inactivating
> > > > the
> > > > > > > corresponding AS.
> > > > > >=20
> > > > > > Yes, that's what we understood and agreed,
> > > > however,
> > > > > > that is
> > > > > > not what is presented in the IG: Read the IG
> > > > again:
> > > > > > New text in section 3.11.2:
> > > > > >=20
> > > > > >    In section 4.3.1: ASP/IPSP States, only
> > the
> > > > > > SGP-ASP and the IPSP SE
> > > > > >    scenarios are described. For the IPSP DE
> > > > model,
> > > > > > both IPSPs MUST
> > > > > >    follow the SGP side of the SGP-ASP
> > > > procedures.
> > > > > >=20
> > > > > > That looks pretty plain to me.  According to
> > the
> > > > IG,
> > > > > > DE IPSPs can
> > > > > > send DUPU (they are actually required above
> > to
> > > > do so
> > > > > > when the SGP
> > > > > > procedure dictates).
> > > > > [TOLGA]IMO, the sentence you refer might
> > really be
> > > > > mislaeading, but I don't think it should make
> > the
> > > > SGP
> > > > > send DUPU for the case we are describing.
> > Let's
> > > > say we
> > > > > have IPSP1 and IPSP2. If one user part is down
> > on
> > > > > IPSP2, then it should be ASP part of IPSP2
> > taking
> > > > some
> > > > > action and sending ASPIA, not SGP part of
> > IPSP2.
> > > >=20
> > > > The case I was refering first, was a platform
> > > > hosting only one AS but
> > > > internally made of several services (SI), not
> > always
> > > > available at the same
> > > > time.
> > > > If one SI is not available but the others are
> > still
> > > > "usable", if the AS
> > > > is "Inactivated", all SI will be prohibited! In
> > that
> > > > case the "IPSP" MAY
> > > > (that's my question) send a DUPU for the
> > concerned
> > > > SI but the traffic
> > > > for the others SI is still possible.
> > > > If IPSP cannot send a DUPU (or an UPU encoded),
> > this
> > > > mean that an AS
> > > > identified by a PC only CANNOT corretly play
> > several
> > > > SI.
> > > >=20
> > > > So my question, is the DUPU allowed between two
> > IPSP
> > > > ?=20
> > > >    (or for a same reason between an AS and a
> > SG(p) ?
> > > > )
> > > >=20
> > > > > > --brian
> > > > > >=20
> > > > > > --
> > > > > > Brian F. G. Bidulock
> > > > > > bidulock@openss7.org
> > > > > > http://www.openss7.org/
> > > > > >=20
> > > > > >
> > _______________________________________________
> > > > > > Sigtran mailing list
> > > > > > Sigtran@ietf.org
> > > > > >
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > >=20
> > > > >=20
> > > > > __________________________________
> > > > > Do you Yahoo!?
> > > > > The New Yahoo! Search - Faster. Easier. Bingo.=20
> > > > > http://search.yahoo.com
> > > > >=20
> > > > >
> > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org=20
> https://www1.ietf.org/mailman/listinfo/sigtran
> > > > >=20
> > > >=20
>=20
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > >=20
> > >=20
> > > __________________________________
> > > Do you Yahoo!?
> > > SBC Yahoo! DSL - Now only $29.95 per month! http://sbc.yahoo.com
> > >=20
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > >=20
> >=20
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
>=20
>=20
> __________________________________
> Do you Yahoo!?
> SBC Yahoo! DSL - Now only $29.95 per month! http://sbc.yahoo.com
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>=20

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



From exim@www1.ietf.org  Fri Jul 11 03:01:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA18706
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 03:01:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19artu-0004U4-QU
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 03:01:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B712S1017236
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 03:01:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19artu-0004Ti-2z; Fri, 11 Jul 2003 03:01:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19artF-0004Sm-FI
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 03:00:21 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA18685
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 03:00:17 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19artB-0002yU-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 03:00:17 -0400
Received: from zdemail04.zdem.compaq.com ([161.114.112.28])
	by ietf-mx with esmtp (Exim 4.12)
	id 19artA-0002yO-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 03:00:16 -0400
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail04.zdem.compaq.com (Postfix) with ESMTP
	id 6CC51E3F; Fri, 11 Jul 2003 08:59:47 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Fri, 11 Jul 2003 08:59:04 +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="us-ascii"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] M3UA IG v04
Date: Fri, 11 Jul 2003 08:59:03 +0200
Message-ID: <06CAA4D3753C53408D2CE6340B68444201488AA7@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA IG v04
Thread-Index: AcNHIrKY9MumpL4NRPuzSKhEnMEpUAAVrOQQ
From: "Zannin, Francois" <francois.zannin@hp.com>
To: <bidulock@openss7.org>
Cc: "Tolga Asveren" <tolga_asveren@yahoo.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 11 Jul 2003 06:59:04.0336 (UTC) FILETIME=[EA42B100:01C34779]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Brian,

What about the DUPU between IPSP? Can we make an agreement on how it
MUST be handle when received by an IPSP or a SG?
Do we have to consider it as a protocol error or MUST we use it to
inform local application (or forward an UPU?

Francois.

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]=20
> Sent: Thursday, July 10, 2003 10:34 PM
> To: Zannin, Francois
> Cc: Tolga Asveren; sigtran@ietf.org
> Subject: Re: [Sigtran] M3UA IG v04
>=20
>=20
> Francois,
>=20
> I don't think IPSP is that well defined yet.  We were=20
> supposed to address it in an extension draft.  Some=20
> corrections have been made in the IG, but a lot of work=20
> remains to be done.  We spent some time defining SG-SG=20
> operation which makes more sense that the SE or DE models,=20
> but there wasn't enough interest to advance the work.
>=20
> --brian
>=20
> Zannin, Francois wrote:                                      =20
>                      (Thu, 10 Jul 2003 22:22:57)
> > OK, now, is the DUPU received on a SG or an IPSP considered as a=20
> > protocol error? This has to be clearly stated in the RFC=20
> (or IG) Same=20
> > remark for the SCON received from an IPSP, is it possible=20
> to receive a=20
> > SCON or must the IPSP use the SCTP congestion to handle the=20
> congestion=20
> > on a remote IPSP.
> >=20
> > Francois.
> >=20
> > > -----Original Message-----
> > > From: Tolga Asveren [mailto:tolga_asveren@yahoo.com]
> > > Sent: Thursday, July 10, 2003 10:13 PM
> > > To: sigtran@ietf.org
> > > Subject: RE: [Sigtran] M3UA IG v04
> > >=20
> > >=20
> > > Francois,
> > >=20
> > > IMO, for the scenario you describe, sending ASPIA is
> > > the correct behaviour. Yes, you won't be able to use healthy user=20
> > > parts in such a case, but if you have that level of=20
> granularity, I=20
> > > would think, RKs should be defined accordingly(for=20
> example for your=20
> > > example PC/SI RKs)
> > > The same applies to ASP/SGP case. There is no DUPU
> > > from ASP to SGP direction.
> > > So, IMO, DUPU for IPSP case is not necessary.
> > >=20
> > >      Tolga
> > > --- "Zannin, Francois" <francois.zannin@hp.com> wrote:
> > > > tolga,
> > > >=20
> > > > see inline,
> > > >=20
> > > > Regards,
> > > >=20
> > > > Francois.
> > > >=20
> > > > > -----Original Message-----
> > > > > From: Tolga Asveren
> > > > [mailto:tolga_asveren@yahoo.com]
> > > > > Sent: Thursday, July 10, 2003 9:47 PM
> > > > > To: sigtran@ietf.org
> > > > > Subject: Re: [Sigtran] M3UA IG v04
> > > > >=20
> > > > >=20
> > > > > Brian,
> > > > > [..snip..]
> > > > > > > [TOLGA] Yes, SGPs send DUPU to ASPs but to my=20
> understanding,=20
> > > > > > > Francois's case would
> > > > correspond
> > > > > > > sending DUPU from ASP to SGP.
> > > > > > > As I said, my interpretation of DE IPSP in the
> > > > IG
> > > > > > is a
> > > > > > > combined SGP/ASP. On ASP part of IPSP, the
> > > > user
> > > > > > part
> > > > > > > goes down, then I would expect inactivating
> > > > the
> > > > > > > corresponding AS.
> > > > > >=20
> > > > > > Yes, that's what we understood and agreed,
> > > > however,
> > > > > > that is
> > > > > > not what is presented in the IG: Read the IG
> > > > again:
> > > > > > New text in section 3.11.2:
> > > > > >=20
> > > > > >    In section 4.3.1: ASP/IPSP States, only the=20
> SGP-ASP and the=20
> > > > > > IPSP SE
> > > > > >    scenarios are described. For the IPSP DE
> > > > model,
> > > > > > both IPSPs MUST
> > > > > >    follow the SGP side of the SGP-ASP
> > > > procedures.
> > > > > >=20
> > > > > > That looks pretty plain to me.  According to the
> > > > IG,
> > > > > > DE IPSPs can
> > > > > > send DUPU (they are actually required above to
> > > > do so
> > > > > > when the SGP
> > > > > > procedure dictates).
> > > > > [TOLGA]IMO, the sentence you refer might really be=20
> mislaeading,=20
> > > > > but I don't think it should make the
> > > > SGP
> > > > > send DUPU for the case we are describing. Let's
> > > > say we
> > > > > have IPSP1 and IPSP2. If one user part is down on=20
> IPSP2, then it=20
> > > > > should be ASP part of IPSP2 taking
> > > > some
> > > > > action and sending ASPIA, not SGP part of IPSP2.
> > > >=20
> > > > The case I was refering first, was a platform
> > > > hosting only one AS but
> > > > internally made of several services (SI), not always=20
> available at=20
> > > > the same time.
> > > > If one SI is not available but the others are still
> > > > "usable", if the AS
> > > > is "Inactivated", all SI will be prohibited! In that
> > > > case the "IPSP" MAY
> > > > (that's my question) send a DUPU for the concerned
> > > > SI but the traffic=20
> > > > for the others SI is still possible.
> > > > If IPSP cannot send a DUPU (or an UPU encoded), this
> > > > mean that an AS
> > > > identified by a PC only CANNOT corretly play several
> > > > SI.
> > > >=20
> > > > So my question, is the DUPU allowed between two IPSP
> > > > ?=20
> > > >    (or for a same reason between an AS and a SG(p) ?
> > > > )
> > > >=20
> > > > > > --brian
> > > > > >=20
> > > > > > --
> > > > > > Brian F. G. Bidulock
> > > > > > bidulock@openss7.org
> > > > > > http://www.openss7.org/
> > > > > >=20
> > > > > > _______________________________________________
> > > > > > Sigtran mailing list
> > > > > > Sigtran@ietf.org=20
> > > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > >=20
> > > > >=20
> > > > > __________________________________
> > > > > Do you Yahoo!?
> > > > > The New Yahoo! Search - Faster. Easier. Bingo.=20
> > > > > http://search.yahoo.com
> > > > >=20
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org=20
> https://www1.ietf.org/mailman/listinfo/sigtran
> > > > >=20
> > > >=20
>=20
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > >=20
> > >=20
> > > __________________________________
> > > Do you Yahoo!?
> > > SBC Yahoo! DSL - Now only $29.95 per month! http://sbc.yahoo.com
> > >=20
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > >=20
> >=20
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
>=20
> --=20
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
>=20

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



From exim@www1.ietf.org  Fri Jul 11 03:21:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA19244
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 03:21:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19asDH-0005pv-43
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 03:21:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B7L3Ci022430
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 03:21:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19asDG-0005pR-33; Fri, 11 Jul 2003 03:21:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19asCd-0005oe-Jp
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 03:20:23 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA19231
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 03:20:21 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19asCb-00035y-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 03:20:21 -0400
Received: from penguin-ext.wise.edt.ericsson.se ([193.180.251.47] helo=penguin.al.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19asCa-00035v-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 03:20:20 -0400
Received: from esealnt613.al.sw.ericsson.se ([153.88.254.125])
	by penguin.al.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h6B7KLGJ023686;
	Fri, 11 Jul 2003 09:20:21 +0200 (MEST)
Received: from ESEALNT744.al.sw.ericsson.se ([153.88.251.4]) by esealnt613.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55)
	id N8XJP43W; Fri, 11 Jul 2003 09:20:21 +0200
Received: by ESEALNT744.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <N94MGR14>; Fri, 11 Jul 2003 09:20:24 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF1@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA IG v04
Date: Fri, 11 Jul 2003 09:19:54 +0200
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, 

Pls, see below some comments...

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: jueves, 10 de julio de 2003 14:36
> To: J.Javier Pastor (EE/EEM)
> Cc: 'sigtran@ietf.org'
> Subject: Re: [Sigtran] M3UA IG v04
> 
> 
> Javier,
> 
> 'n' is defined as the number of ASPs that are needed to handle
> the maximum engineered load of the AS.
> 
> This is the total load from the network correct?  As in the
> traffic delivered to the AS, from all SGP, all SG.  Correct?
> 
> What has that to do with the number of ASPs active for particular
> SGP?
> 
We are relying on SGPs to send all the load to the AS. Only when the AS gets to ACTIVE we are sure that it could afford all the load.


> There can be more than on SGP.  Correct?
> 
> There can be more than one SG.  Correct?
> 
> An ASP can be active for any number of SGPs in any number of 
> SGs.  Correct?
> 
> SGPs in an SG should be able to transfer traffic between 
> them.  Correct?

It basically depends on your implementation.
> 
> In general, what has the number of ASPs active for one SGP have
> to do with the traffic handling capacity of an AS?
> 
It has to do with the AS availability, a general rule in telephony nodes.

> 
> Consider the following (n+k == 2+0): (Figure 5).
> Consider first that SGP1 and SGP2 are in the same SG.
> 
> 
>      ________                  ________   
> ____|        |________________|        |
>     |  SGP1  |        ________|  ASP1  |
>     |________|       /        |________|
>      ________       /          ________
>     |        |_____/          |        |
> ____|  SGP2  |________________|  ASP2  |
>     |________|                |________|
> 
> 
Why don't you establish an assoc between SGP1 and ASP2 ?. This network design seems not to be very appropriate for the n+k=2+0

> 3.14 tells me that I SHOULD NOT direct traffic from the
> newtork from SGP1 to ASP1 on startup of ASP1 with SGP1.
> 
It is not recommended, but can always use smooth start. 

> Does that include passing traffic from SGP1 via SGP2?
> 
Nothing is said about that. Therefore it is not included.

> Is this because ASP1 cannot handle the load?
> 
It is because the AS has not sufficient resources to handle the expected traffic from SGP1

> What should SGP1 do?
> 
Wait for more ASPs to become ACTIVE

> Now, consider that SGP1 and SGP2 are in different SGs.
> 
> Should SGP1 tell the network that the AS is inaccessible?
> 
It depends on what you call the network. If you refer to SS7, it is the SG's SPC the visible SPC, so SG will announce un/availability when needed.

> Should SGP1 direct traffic over C-links to SGP2 that will
> ultimately be the same a passing traffic to ASP1?
>
It may, if SGP1 has a route to reach SGP2. 
 
> Why should not SGP1 direct traffic to ASP1?
> 
It could do it when you uses smooth start. But it is not recommended since ASP1 could not handle all the load it receives...

> I don't think that 3.14 is very clear.  'n' has nothing to do
> with the handling capacity from a given SGP to an AS.  It has
> to do with the overall engineered maximum traffic that 'n' ASPs
> can handle for the AS.  In even the more trivial cases the SGP
> cannot follow the recommendation.
> 
It seems to me that the recommendation can be followed. Anyway I may be wrong. So if you can make another proposal that everyone agrees on, pls go ahead.

Cheers // Javier.

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



From exim@www1.ietf.org  Fri Jul 11 03:22:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA19298
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 03:22:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19asED-0005vg-DO
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 03:22:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B7M1nR022781
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 03:22:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19asED-0005vL-8Z; Fri, 11 Jul 2003 03:22:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19asDw-0005sG-6g
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 03:21:44 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA19267
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 03:21:41 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19asDt-00036T-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 03:21:41 -0400
Received: from albatross-ext.wise.edt.ericsson.se ([193.180.251.49] helo=albatross.tn.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19asDs-00036N-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 03:21:40 -0400
Received: from esealnt613.al.sw.ericsson.se ([153.88.254.125])
	by albatross.tn.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h6B7LfIi028259;
	Fri, 11 Jul 2003 09:21:41 +0200 (MEST)
Received: from ESEALNT744.al.sw.ericsson.se ([153.88.251.4]) by esealnt613.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55)
	id N8XJP4ZX; Fri, 11 Jul 2003 09:21:41 +0200
Received: by ESEALNT744.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <N94MGRFL>; Fri, 11 Jul 2003 09:21:44 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF2@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>,
        "'sigtran@ietf.org'"
	 <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA IG v04 AS State Machine
Date: Fri, 11 Jul 2003 09:21:14 +0200
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, 

In the consensus text we have not add or remove anything regarding the AS state scope. Then it remains the same.

Cheers // Javier.

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: jueves, 10 de julio de 2003 15:16
> To: 'sigtran@ietf.org'
> Subject: Re: [Sigtran] M3UA IG v04 AS State Machine
> 
> 
> Javier,
> 
> The AS State Machine illustrated in 3.14 is per-SGP.  Correct?
> 
> SGP can communicate with each other in an SG.  Correct?
> 
> Why should an SGP pend when the last ASP goes ASP-INACTIVE for
> an AS with a given SGP?
> 
> Cannot the SGP redirect traffic via another SGP that has ASPs
> active for the AS?
> 
> Example:
> 
>         _______                    _______
>        |       |______x    x______|       |
>        | SGP1  |         _________| ASP1  |
>        |_______|        /         |_______|
>         _______        /           _______
>        |       |______/           |       |
>        | SGP2  |__________________| ASP2  |
>        |_______|                  |_______|
> 
> 
> When ASP1 goes inactive (or down) w.r.t SGP1, cannot
> SGP1 direct traffic via SGP2?
> 
> What is the state of the AS at SGP1?
> 
> In section 3.14 AC2PN says that:
> 
>   AC2PN: the last ASP in the ASP-ACTIVE state moves to
>   ASP-INACTIVE or ASP-DOWN states, causing the number of ASPs
>   in the ASP-ACTIVE [sic] drop below 1.
> 
> Is that the number of ASPs for SGP1?  for SGP2? for the SG?
> for all (both) SGs?
> 
> In the RFC the AS state machine was per-SGP.  Is the 3.14 state
> machine per-SGP?
> 
>    IA2AC: one ASP moves to [the] ASP-ACTIVE [state], causing
>    [the] number of ASPs in the ASP-ACTIVE state to be n.
> 
> When SGP1 receives an ASP Active and sends an ASP Active Ack
> to ASP1, how many ASPs are active for the AS?   1?  2?
> 
> Please clarify.
> 
> --brian
> 
> 
> _______________________________________________
> 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 exim@www1.ietf.org  Fri Jul 11 03:30:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA19443
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 03:30:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19asM0-0006b8-34
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 03:30:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B7U3QA025275
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 03:30:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19asLy-0006ZN-8p; Fri, 11 Jul 2003 03:30:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19asLP-0006PF-25
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 03:29:27 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA19425
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 03:29:24 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19asLM-00038s-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 03:29:24 -0400
Received: from albatross-ext.wise.edt.ericsson.se ([193.180.251.49] helo=albatross.tn.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19asLL-00038p-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 03:29:23 -0400
Received: from esealnt610.al.sw.ericsson.se ([153.88.254.120])
	by albatross.tn.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h6B7TOIi000127;
	Fri, 11 Jul 2003 09:29:24 +0200 (MEST)
Received: from ESEALNT745.al.sw.ericsson.se ([153.88.251.5]) by esealnt610.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55)
	id 3S6BW6MF; Fri, 11 Jul 2003 09:30:21 +0200
Received: by ESEALNT745.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <N94NNYTB>; Fri, 11 Jul 2003 09:29:02 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF3@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA IG v04 Section 3.14
Date: Fri, 11 Jul 2003 09:28:56 +0200
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, 

Pls, see my comments below...

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: jueves, 10 de julio de 2003 15:21
> To: J.Javier Pastor (EE/EEM)
> Cc: 'sigtran@ietf.org'
> Subject: Re: [Sigtran] M3UA IG v04 Section 3.14
> 
> 
> Javier,
> 
>    ---------
>    New text: (Section 4.3.4.3, for both loadsharing and broadcast)
>    ---------
> 
>    At start-up or re-start phases, an SGP or IPSP, upon 
> reception of an
>    ASP Active message for the first ASP in a Loadshare AS, SHOULD NOT
>    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.
> 
> How can the SGP know when there are sufficient resources?
> 
> 'n'?  How do I count ASPs active for comparison against 'n' at one SGP
> in multiple SGs?
> 
Yes, n is a good measure and it is part of the original text. But no one forbid you using "1" if you want to be more flexible and you are also using the smooth start. 

You can know whether you have "n" counting the ASPs that are active for each AS in your SGP.

> Please provide a definition of, and criteria for, sufficient 
> resources to
> go with the recommendation.

It seems that this is a comment to the RFC, since there is no change included in the IG for that specific sentence. Do you propose any specific wording?

Cheers // Javier.

> 
> --brian
> 
> 

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



From exim@www1.ietf.org  Fri Jul 11 03:53:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA19848
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 03:53:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19asiF-0008Uy-9I
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 03:53:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B7r3Ym032668
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 03:53:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19asiE-0008US-6h; Fri, 11 Jul 2003 03:53:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ashM-0008Nb-Jx
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 03:52:08 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA19795
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 03:52:05 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ashK-0003FG-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 03:52:06 -0400
Received: from albatross-ext.wise.edt.ericsson.se ([193.180.251.49] helo=albatross.tn.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19ashJ-0003F9-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 03:52:05 -0400
Received: from esealnt610.al.sw.ericsson.se ([153.88.254.120])
	by albatross.tn.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h6B7q6Ii005019;
	Fri, 11 Jul 2003 09:52:06 +0200 (MEST)
Received: from ESEALNT745.al.sw.ericsson.se ([153.88.251.5]) by esealnt610.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55)
	id 3S6BW0QC; Fri, 11 Jul 2003 09:53:03 +0200
Received: by ESEALNT745.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <N94NNY8L>; Fri, 11 Jul 2003 09:51:44 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF4@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA IG v04 Section 3.8
Date: Fri, 11 Jul 2003 09:51:40 +0200
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, 

See comments below...

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: jueves, 10 de julio de 2003 15:30
> To: J.Javier Pastor (EE/EEM)
> Cc: 'sigtran@ietf.org'
> Subject: Re: [Sigtran] M3UA IG v04 Section 3.8
> 
> 
> Javier,
> 
> Section 3.8 says:
> 
>    5.4 Auditing examples
> 
>    5.4.1 SG State: Uncongested / Unavailable
> 
>           ASP                          SGP
>           ---                          ---
>            |  -------- DAUD --------->  |
>            |  <------ SCON(0) --------  |
>            |  <------- DUNA ----------  |
> 
> What state of a destination is Uncongested/Unavailable?  There is no
> such state in Q.704.  Uncongested routesets are by definition 
> available.
> Inaccessible routesets have no congestion state by definition.


> 
> Did you think that a routeset could be in an unavailble and 
> congested state?
> 
> What is the use of the SCON(0)?  Why is it being sent?
> 
>    5.4.4 SG state: Uncongested / Unavailable
> 
>           ASP                          SGP
>           ---                          ---
>            |  -------- DAUD --------->  |
>            |  <------- DUNA ----------  |
> 
> What is the difference between 5.4.1 and 5.4.4?  They both say
> Uncongested/Available.
> 
> 
It is a typo. In 5.4.1 it should say: SG state: "Uncongested / Available". And of course a DAVA has to be sent.

> 
>    [...]. Where the SGP maintains the congestion status of the SS7
>    destination, the SGP MUST additionally respond with an SCON message
>    before the DAVA or DRST message.  If the SS7 destination is
> 
> When is the SGP required to maintain congestion status?  

In those networks where the congestion levels are handled... There's no change there regarding the RFC. Do you think the RFC requires some additional explanation in this regard?



> What is the
> criteria for an ASP to determine that an SGP should maintain 
> congestion
> status and is violating this MUST.  Why must the ASP require the SGP
> to send SCON in this circumstance?  What interoperability 
> concern is the
> MUST addressing?
> 
It seems to me that it was not clear whether to send the SCON or not during the bakeoff.

Cheers // Javier. 

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



From exim@www1.ietf.org  Fri Jul 11 03:58:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA19920
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 03:58:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19asn5-0000EQ-KA
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 03:58:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B7w3TU000869
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 03:58:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19asn3-0000Dj-La; Fri, 11 Jul 2003 03:58:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19asmW-0000D7-Lp
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 03:57:28 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA19909
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 03:57:25 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19asmU-0003Hf-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 03:57:26 -0400
Received: from albatross-ext.wise.edt.ericsson.se ([193.180.251.49] helo=albatross.tn.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19asmT-0003Hc-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 03:57:25 -0400
Received: from esealnt612.al.sw.ericsson.se ([153.88.254.118])
	by albatross.tn.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h6B7vQIi006479;
	Fri, 11 Jul 2003 09:57:26 +0200 (MEST)
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 NW4WD4J5; Fri, 11 Jul 2003 09:57:26 +0200
Received: by ESEALNT744.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <N94MGRT9>; Fri, 11 Jul 2003 09:57:29 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF5@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA IG v04 Section 3.18
Date: Fri, 11 Jul 2003 09:56:59 +0200
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, 

Pls, see below...

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: jueves, 10 de julio de 2003 15:34
> To: J.Javier Pastor (EE/EEM)
> Cc: 'sigtran@ietf.org'
> Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> 
> 
> Javier,
> 
> 3.18.3 Solution Description
> 
>    Clarify that the ASP needs to indicate when the congestion 
> level has
>    changed (including abatement).  Further, the ASP peer can maintain
>    a timer, if desired, in case the ASP fails to indicate congestion
>    abatement.
> 
> Where are the procedures for this timer described?  This is 
> the only place
> that the timer is mentioned.  Please describe the timer or 
> remove the sentence
> saying that it has been described...
> 

Do you think is worth include the timer explanation into the new text? Otherwise, we may remove that sentence. 

Thanks // Javier.

 

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



From exim@www1.ietf.org  Fri Jul 11 04:04:34 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA20095
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 04:04:33 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19assw-0000do-N2
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 04:04:06 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B846DD002436
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 04:04:06 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19asst-0000d3-RV; Fri, 11 Jul 2003 04:04:03 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ass5-0000Z6-Dy
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 04:03:13 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA20060
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 04:03:10 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ass2-0003Kn-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 04:03:10 -0400
Received: from albatross-ext.wise.edt.ericsson.se ([193.180.251.49] helo=albatross.tn.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19ass1-0003Kk-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 04:03:10 -0400
Received: from esealnt610.al.sw.ericsson.se ([153.88.254.120])
	by albatross.tn.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h6B83AIi007862;
	Fri, 11 Jul 2003 10:03:10 +0200 (MEST)
Received: from ESEALNT745.al.sw.ericsson.se ([153.88.251.5]) by esealnt610.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55)
	id 3S6BXB3G; Fri, 11 Jul 2003 10:04:07 +0200
Received: by ESEALNT745.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <N94NNZDC>; Fri, 11 Jul 2003 10:02:48 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF6@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA IG v04 Section 3.9
Date: Fri, 11 Jul 2003 10:02:43 +0200
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, 

Pls, see below..

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: jueves, 10 de julio de 2003 15:42
> To: J.Javier Pastor (EE/EEM)
> Cc: 'sigtran@ietf.org'
> Subject: Re: [Sigtran] M3UA IG v04 Section 3.9
> 
> 
> Javier,
> 
> Section 3.9 says:
> 
>           ASP                          SGP
>           ---                          ---
>            |  ------ ASPIA (RC1)----->  |
>            |  <----  ASPIA Ack -------  |
>            |  -----DEREG REQ (RC1)--->  |
>            |  <----DEREG RSP (RC1)----  |
>            |  -------ASPIA (RC1)----->  |
> 
>    What should SG do?
> 
> The new text does not explicity answer this question.
> 
> What does the SGP send in this case?

It is included in the new text for section 4.5.3. Take a look to the second bullet. Although you may be right saying that we should include it also into the "solution description"

> 
> As RC values are per-SGP, it appears to me that it must be
> Error("Invalid Routing Context").
> 
> Correct?

Yes, correct.


Thanks // Javier.

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



From exim@www1.ietf.org  Fri Jul 11 04:28:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA20664
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 04:28:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atG7-0002oQ-0n
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 04:28:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B8S2al010800
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 04:28:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atG5-0002nw-BS; Fri, 11 Jul 2003 04:28:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atFb-0002ic-F4
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 04:27:31 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA20651
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 04:27:27 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19atFY-0003UF-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 04:27:28 -0400
Received: from angelica.ulticom.com ([208.255.120.2] helo=chuckie.dgms.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19atFX-0003U8-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 04:27:27 -0400
Received: from navy.fr.ulticom.com (localhost [127.0.0.1])
	by chuckie.dgms.com (8.9.3/8.9.3) with ESMTP id EAA07963;
	Fri, 11 Jul 2003 04:26:57 -0400 (EDT)
Received: from PCGASTAUDL (pc-gastaud-l.fr.ulticom.com [172.26.32.88])
	by navy.fr.ulticom.com (8.11.6+Sun/8.11.6) with SMTP id h6B8NQE16094;
	Fri, 11 Jul 2003 10:23:27 +0200 (MEST)
From: "Valerie Gastaud" <valerie.gastaud@ulticom.com>
To: <bidulock@openss7.org>, "Tolga Asveren" <tolga_asveren@yahoo.com>
Cc: <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA - RC in SSNM messages
Date: Fri, 11 Jul 2003 10:27:26 +0200
Message-ID: <NDBBIHAOCKGAMIKNLALOEEDHDOAA.valerie.gastaud@ulticom.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.6604 (9.0.2911.0)
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106
Importance: Normal
In-Reply-To: <20030710142654.C24498@openss7.org>
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, Tolga,

so, the conclusion is that when there are multiple AS on the same
association, all the messages (SSNM, DATA and ASPTM) MUST contain the RC
parameter.

if this is agreed by everyone, can we remove the (s) of Application Server
in the following sentence from section 4.3.4.3 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.

and change the RC from optional to conditional in the ASPTM messages as well

and change the SHOULD in section 4.3.4.3 by a MUST:
                    ...  In the case where an ASP wishes to process the
	traffic for more than one Application Server across a common SCTP
	association, the ASP Active message(s) SHOULD contain a list of one
	or more Routing Contexts to indicate for which Application Servers
	the ASP Active message applies. ...

Valerie
-----Original Message-----
From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf Of
Brian F. G. Bidulock
Sent: jeudi 10 juillet 2003 22:27
To: Tolga Asveren
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - RC in SSNM messages


Tolga,


Tolga Asveren wrote:
(Thu, 10 Jul 2003 12:48:14)
> Brian,
> --- "Brian F. G. Bidulock" <bidulock@openss7.org>
> wrote:
> > Tolga,
> >
> > Tolga Asveren wrote:
> >                                 (Thu, 10 Jul 2003
> > 11:10:20)
> > > Brian,
> > >
> > > I was referring to 4.3.4.3 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."
> > > This enables the option to send ASPAC with no RC,
> > even
> > > if multiple AS traffic is carried over one
> > > association. I don't think, this is permitted for
> > > SSNM.
> >
> > It permits a single AS in fitting with the SHOULD I
> > quoted.  Multiple
> > configured AS for the same association would be
> > against the SHOULD.
> > So you are still on your own.
> [TOLGA]Then why does it say "Application Server(s)"
>                                                 ^
> I am pretty sure, what is meant here (In fact even
> just two days ago, there was a qustion related with
> this)

If you recall, the (s) predates the RC in SSNM and DATA
discussion.  The SHOULD was put in after, at the same
time as the RC passages in SSNM and DATA.  If you have
multiple AS on the same association you need to put RC
in all messages.  This was one reason we changed "Optional"
to "Conditional' for RC throughout the document.  The
same condition applies in all cases.

--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 exim@www1.ietf.org  Fri Jul 11 04:37:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA20931
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 04:37:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atOp-0003M0-S6
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 04:37:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B8b3RE012858
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 04:37:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atOm-0003KQ-J3; Fri, 11 Jul 2003 04:37:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atOE-0003JO-La
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 04:36:26 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA20869
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 04:36:22 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19atOB-0003Xu-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 04:36:23 -0400
Received: from albatross-ext.wise.edt.ericsson.se ([193.180.251.49] helo=albatross.tn.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19atOA-0003Xr-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 04:36:22 -0400
Received: from esealnt611.al.sw.ericsson.se ([153.88.254.121])
	by albatross.tn.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h6B8aNIi016588;
	Fri, 11 Jul 2003 10:36:23 +0200 (MEST)
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 3S62JR9S; Fri, 11 Jul 2003 10:39:00 +0200
Received: by ESEALNT744.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <N94MGS2V>; Fri, 11 Jul 2003 10:36:26 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF7@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA IG v04  Section 3.14
Date: Fri, 11 Jul 2003 10:35:57 +0200
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, 

You are right saying that there is currently no way of notifying AS-DOWN. In fact "n" only applies to the traffic case.

Then I see two options:
	1- include AS-DOWN in NTFY
	2- The AS-DOWN state is only got when all the ASPs are in ASP-DOWN state.

I would addopt 2 since it is inline what we are having in the other UALs. 

Thanks // Javier.


> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: jueves, 10 de julio de 2003 17:15
> To: J.Javier Pastor (EE/EEM)
> Cc: 'sigtran@ietf.org'
> Subject: Re: [Sigtran] M3UA IG v04 Section 3.14
> 
> 
> Javier,
> 
> Section 3.14 says:
> 
>    IA2DN: One ASP moves to ASP-DOWN causing that the number of ASPs in
>    either ASP-ACTIVE or ASP-INACTIVE drops below n.
> 
> Am I correct to say that if n=5 and there are 4 ASPs in the 
> ASP-INACTIVE
> and 1 ASP in the ASP-ACTIVE state, and then the 1 active ASP moves to
> the ASP-DOWN state and T(r) expires that the AS is to move to 
> the AS-DOWN
> state?  As follows:
> 
>   If T(r) expires before an ASP becomes ASP-ACTIVE, the SGP MAY stop
>   queuing messages and discards all
>   previously queued messages. The AS will move to the 
> AS-INACTIVE state
>   if at least the number of ASPs in ASP-INACTIVE sum n, otherwise it
>   will move to AS-DOWN state.
> 
> Now, that is an AS state change and we must send a NTFY to 
> all the INACTIVE
> ASPs (the 4 left inactive) of the state change.  But from 
> Section 3.8.2 of
> the RFC, there is no message to notify of an AS-DOWN state:
> 
>          1    reserved
>          2    Application Server Inactive (AS-INACTIVE)
>          3    Application Server Active (AS-ACTIVE)
>          4    Application Server Pending (AS-PENDING)
> 
> You see, we removed it because there was never an instance in which it
> would be sent.  You have inserted an instance again.
> 
> Aren't you going to have to define a Notify message (the 
> Reserved value)
> for "Application Server Down (AS-DOWN)" so that the INACTIVE ASPs can
> be properly notified by the SGP?
> 
> Particularly when the RFC says in Section 4.3.4.5: (Notify Procedures)
> 
>    A Notify message reflecting a change in the AS state MUST 
> be sent to
>    all ASPs in the AS, except those in the ASP-DOWN state, with
>    appropriate Status Information and any ASP Identifier of the failed
>    ASP.  At the ASP, Layer Management is informed with an M-NOTIFY
>    indication primitive.  The Notify message must be sent 
> whether the AS
>    state change was a result of an ASP failure or reception of an ASP
>    State management (ASPSM) / ASP Traffic Management (ASPTM) message.
>    In the second case, the Notify message MUST be sent after 
> any related
>    acknowledgement messages  (e.g., ASP Up Ack, ASP Down Ack, 
> ASP Active
>    Ack, or ASP Inactive Ack).
> 
> The above tells me that I MUST send a Notify message becase of the AS
> state change, but there is no message to inform about AS-DOWN state
> transition.
> 
> What is the purpose of doing this?  What is the protocol 
> difference between
> the AS-DOWN state and the AS-INACTIVE state?  What is it that the SGP
> does or doesn't do in the one state that it doesn't or does 
> do in the other?
> 
> I don't recall this being discussed at all.
> 
> Please clarify.
> 
> -- 
> 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 exim@www1.ietf.org  Fri Jul 11 04:37:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA20932
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 04:37:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atOp-0003M3-SX
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 04:37:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B8b3wf012857
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 04:37:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atOn-0003Kg-4j; Fri, 11 Jul 2003 04:37:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atOf-0003Jx-17
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 04:36:53 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA20899
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 04:36:44 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19atOX-0003YH-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 04:36:45 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[HtRHnagRXi0lA2oq8zEBO/Cfms+hg6uK])
	by ietf-mx with esmtp (Exim 4.12)
	id 19atOW-0003YE-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 04:36:44 -0400
Received: from ns.pigworks.openss7.net (IDENT:i0lXHf5OuuNMJQipruXjnAyYOQ7hdixe@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6B8aiX11359;
	Fri, 11 Jul 2003 02:36:44 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6B8ail02756;
	Fri, 11 Jul 2003 02:36:44 -0600
Date: Fri, 11 Jul 2003 02:36:44 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04
Message-ID: <20030711023644.A1680@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF1@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: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF1@eestqnt105.es.eu.ericsson.se>; from j.javier.pastor@ericsson.com on Fri, Jul 11, 2003 at 09:19:54AM +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>

J.Javier,

J.Javier Pastor (EE/EEM) wrote:                                                   (Fri, 11 Jul 2003 09:19:54)
> 
> Hi Brian, 
> 
> Pls, see below some comments...
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: jueves, 10 de julio de 2003 14:36
> > To: J.Javier Pastor (EE/EEM)
> > Cc: 'sigtran@ietf.org'
> > Subject: Re: [Sigtran] M3UA IG v04
> > 
> > 
> > Javier,
> > 
> > 'n' is defined as the number of ASPs that are needed to handle
> > the maximum engineered load of the AS.
> > 
> > This is the total load from the network correct?  As in the
> > traffic delivered to the AS, from all SGP, all SG.  Correct?
> > 
> > What has that to do with the number of ASPs active for particular
> > SGP?
> > 
> We are relying on SGPs to send all the load to the AS. Only when the AS gets
> to ACTIVE we are sure that it could afford all the load.

What has that to do with the number of ASPs active for a
particular SGP?

'n' is a system value, not a per-SGP value.  The AS state
machine in the IG treats 'n' as a per-SGP value, when it is
defined as a system value.

What has the number of active ASPs for an SGP (the per-SGP
value) have to do with the number of active ASPs for the AS
(system value)?

An SGP cannot tell how many ASPs are active for the AS (system
value), just the number of ASPs that are active to it (per-SGP
value).

How can one count active ASPs for the SGP (per-SGP value) and
compare it to 'n' (the system value)?


> 
> 
> > There can be more than on SGP.  Correct?
> > 
> > There can be more than one SG.  Correct?
> > 
> > An ASP can be active for any number of SGPs in any number of 
> > SGs.  Correct?
> > 
> > SGPs in an SG should be able to transfer traffic between 
> > them.  Correct?
> 
> It basically depends on your implementation.
> > 
> > In general, what has the number of ASPs active for one SGP have
> > to do with the traffic handling capacity of an AS?
> > 
> It has to do with the AS availability, a general rule in telephony nodes.
> 
> > 
> > Consider the following (n+k == 2+0): (Figure 5).
> > Consider first that SGP1 and SGP2 are in the same SG.
> > 
> > 
> >      ________                  ________   
> > ____|        |________________|        |
> >     |  SGP1  |        ________|  ASP1  |
> >     |________|       /        |________|
> >      ________       /          ________
> >     |        |_____/          |        |
> > ____|  SGP2  |________________|  ASP2  |
> >     |________|                |________|
> > 
> > 
> Why don't you establish an assoc between SGP1 and ASP2 ?. This network
> design seems not to be very appropriate for the n+k=2+0

It's a simple example.  The association between SGP1 and ASP2
happens to be down at the moment.

> 
> > 3.14 tells me that I SHOULD NOT direct traffic from the
> > newtork from SGP1 to ASP1 on startup of ASP1 with SGP1.
> > 
> It is not recommended, but can always use smooth start. 

On what grounds can you not recommend directing traffic above?
There are suffcient ASPs active fo the AS (system value) to
handle the maximum engineered load.  'n' of them are active for
the AS (system value), SGP1 has 1 of them active to it (per-SGP)
value.

Please tell me what number you are expecting the SGP to use: the
system value?  or the per-SGP value.

> 
> > Does that include passing traffic from SGP1 via SGP2?
> > 
> Nothing is said about that. Therefore it is not included.

Silence implies acceptance.  If you only want to say that the
SGP should not direct traffic towards the ASP directly over the
association between the SGP and the ASP, please say so.

> 
> > Is this because ASP1 cannot handle the load?
> > 
> It is because the AS has not sufficient resources to handle the expected
> traffic from SGP1

But it does in the diagram.  Two ASPs are active for the AS and
handling traffic for the AS.  There are 'n' ASPs handling
traffic and they can handle the maximum engineerd load.

So there ARE sufficent resources to handle the expected load.
It is just that SGP1 cannot discern that from the number of ASPs
active to it alone.

> 
> > What should SGP1 do?
> > 
> Wait for more ASPs to become ACTIVE

Why? There are 'n' ASPs already handling traffic!

> 
> > Now, consider that SGP1 and SGP2 are in different SGs.
> > 
> > Should SGP1 tell the network that the AS is inaccessible?
> > 
> It depends on what you call the network. If you refer to SS7, it is the SG's
> SPC the visible SPC, so SG will announce un/availability when needed.

> 
> > Should SGP1 direct traffic over C-links to SGP2 that will
> > ultimately be the same a passing traffic to ASP1?
> >
> It may, if SGP1 has a route to reach SGP2. 
>  
> > Why should not SGP1 direct traffic to ASP1?
> > 
> It could do it when you uses smooth start. But it is not recommended since
> ASP1 could not handle all the load it receives...

Look at the diagram.  ASP1 is already exposed to full load via
SGP2.  So is its not true that ASP1 cannot handling the load,
because it is already exposed to it via SGP2.

Also, what do you mean by "could not handle all the load it
receives..."?  An MTP-User (AS) must have overload controls to
address traffic overloads at any time.

Are you advocating that ASP are to be built to not conform to
SS7 specifications in this regard?

Is your ASP not able to handle overload?

Is there anyone on the list whose ASP is not able to handle
overload traffic?

> 
> > I don't think that 3.14 is very clear.  'n' has nothing to do
> > with the handling capacity from a given SGP to an AS.  It has
> > to do with the overall engineered maximum traffic that 'n' ASPs
> > can handle for the AS.  In even the more trivial cases the SGP
> > cannot follow the recommendation.
> > 
> It seems to me that the recommendation can be followed. Anyway I
> may be wrong. So if you can make another proposal that everyone
> agrees on, pls go ahead.

If you mean to count per-SGP APSs and compare the number against
the system-wide engineered value 'n', then every time that the
number of ASPs active for an SGP is different from the number of
ASPs active for the AS (system wide), there will be a problem.

In systems where 'n' are active and 'k' are (cold) backup, every
time that an ASP cannot activate to a given SGP, that SGP is out
of service (as in the above trivial example).

Also, the recommendation is based on the premise that the ASP
cannot handle overload in thes start scenario (where it
obviously is able to handle it in the restart scenario).  This
is a contradiction.  Either the ASP is capable of handling
overload or it is not.

I have asked many times now, and nobody will say that their ASP
cannot handle overload.

The recommendation is technically invalid because of the
following:

1. The number of ASPs active at an SGP (per-SGP) is not the
   number of ASPs active for the AS (system value 'n').

2. ASPs must be able to handle overload, because they must
   conform to SS7 standards.

In the current RFC, it there are ASPs so badly designed that
they cannot handle an overload, they should take care with how
they activate.  They SHOULD coordinate amoung themselves that
'n' are active.  If 'n' ASPs cannot be brought to the ASP-ACTIVE
state for the AS, then none of the ASPs need activate.

Relying on this, more sound, recommendation, the SGP can
activate with the first ASP, knowing that 'n' of them have been
coordinated to become active.

In this way, in the current RFC, the ASP assures that SGP that
'n' are to be active on a system basis (which only the ASPs can
know, and because they have coordinated amoung themselves).  So
the SGP can activate on the first ASP on a per-SGP basis.

The current RFC also allows the SGP to wait further if it has
some additional information about the sufficiency of the number
of active ASPs in addition to the assurance that there will be
'n' active or the first one would not have activated.

The current RFC provides a way around your counting problem, and
even accommodates broken ASPs that can't conform to SS7 overload
controls.  It also leaves the activation of the AS squarely on
the shoulders of the ASPs, where it belongs, and the SGP just
does what it is told.

So, by the RFC, in the above example, when ASP1 activates with
SGP1, SGP1 already has the assurance that ASP1 has coordinated
with ASP2 that both of them will become active.  Thus it can
activate the as immediately.  It also MAY consider that SGP2 has
ASP2 already active as verfication that there are sufficient
ASPs active in the AS, or wait until some SGP has ASP2 active
if, for example, it shares state amoung SGP. to that extent.

So, perhaps not so strangely, the recommendation in 3.14 for the
example above, stops SGP1 from activating the AS when it is
completely reasonable for it to do so.  The RFC, on the other
hand, has SGP1 activating correctly.

I think you need to alter your proposal so that SGP1 can
activate immeidately in the example above, on recommendation
and not option.

You are welcome to go back to the wording in the RFC ("MAY
withhold" instead of "SHOULD NOT direct") to correct this
problem.  Of course, then there would be no need for all the
'n' stuff in the AS state machine.

--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 exim@www1.ietf.org  Fri Jul 11 04:39:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA21081
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 04:39:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atQk-0003o6-5O
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 04:39:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B8d2dV014626
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 04:39:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atQj-0003no-Vp; Fri, 11 Jul 2003 04:39:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atPk-0003gW-Nb
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 04:38:00 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA20979
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 04:37:56 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19atPh-0003Z6-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 04:37:57 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[WFVc4JURMDYgSG38cAwxGZZT068YFB47])
	by ietf-mx with esmtp (Exim 4.12)
	id 19atPg-0003Z3-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 04:37:57 -0400
Received: from ns.pigworks.openss7.net (IDENT:ftwJCzWIg2y7g/cid5seqlJ+XvjsHGok@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6B8btX11376;
	Fri, 11 Jul 2003 02:37:55 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6B8bsM02795;
	Fri, 11 Jul 2003 02:37:54 -0600
Date: Fri, 11 Jul 2003 02:37:54 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04 AS State Machine
Message-ID: <20030711023754.B1680@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF2@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: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF2@eestqnt105.es.eu.ericsson.se>; from j.javier.pastor@ericsson.com on Fri, Jul 11, 2003 at 09:21:14AM +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>

J.Javier,

That doesn't really answer my questions below.  Could you please
clarify the proposal by answering the questions?

--brian

J.Javier Pastor (EE/EEM) wrote:                                                   (Fri, 11 Jul 2003 09:21:14)
> Hi Brian, 
> 
> In the consensus text we have not add or remove anything regarding the AS
> state scope. Then it remains the same.
> 
> Cheers // Javier.
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: jueves, 10 de julio de 2003 15:16
> > To: 'sigtran@ietf.org'
> > Subject: Re: [Sigtran] M3UA IG v04 AS State Machine
> > 
> > 
> > Javier,
> > 
> > The AS State Machine illustrated in 3.14 is per-SGP.  Correct?
> > 
> > SGP can communicate with each other in an SG.  Correct?
> > 
> > Why should an SGP pend when the last ASP goes ASP-INACTIVE for
> > an AS with a given SGP?
> > 
> > Cannot the SGP redirect traffic via another SGP that has ASPs
> > active for the AS?
> > 
> > Example:
> > 
> >         _______                    _______
> >        |       |______x    x______|       |
> >        | SGP1  |         _________| ASP1  |
> >        |_______|        /         |_______|
> >         _______        /           _______
> >        |       |______/           |       |
> >        | SGP2  |__________________| ASP2  |
> >        |_______|                  |_______|
> > 
> > 
> > When ASP1 goes inactive (or down) w.r.t SGP1, cannot
> > SGP1 direct traffic via SGP2?
> > 
> > What is the state of the AS at SGP1?
> > 
> > In section 3.14 AC2PN says that:
> > 
> >   AC2PN: the last ASP in the ASP-ACTIVE state moves to
> >   ASP-INACTIVE or ASP-DOWN states, causing the number of ASPs
> >   in the ASP-ACTIVE [sic] drop below 1.
> > 
> > Is that the number of ASPs for SGP1?  for SGP2? for the SG?
> > for all (both) SGs?
> > 
> > In the RFC the AS state machine was per-SGP.  Is the 3.14 state
> > machine per-SGP?
> > 
> >    IA2AC: one ASP moves to [the] ASP-ACTIVE [state], causing
> >    [the] number of ASPs in the ASP-ACTIVE state to be n.
> > 
> > When SGP1 receives an ASP Active and sends an ASP Active Ack
> > to ASP1, how many ASPs are active for the AS?   1?  2?
> > 
> > Please clarify.
> > 
> > --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 exim@www1.ietf.org  Fri Jul 11 04:48:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA21337
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 04:48:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atZR-0004dA-TT
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 04:48:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B8m10E017799
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 04:48:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atZR-0004cj-1G; Fri, 11 Jul 2003 04:48:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atZ3-0004XO-1U
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 04:47:37 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA21308
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 04:47:31 -0400 (EDT)
From: john.loughney@nokia.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19atYy-0003gd-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 04:47:32 -0400
Received: from mgw-x1.nokia.com ([131.228.20.21])
	by ietf-mx with esmtp (Exim 4.12)
	id 19atYx-0003gZ-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 04:47:31 -0400
Received: from esvir05nok.ntc.nokia.com (esvir05nokt.ntc.nokia.com [172.21.143.37])
	by mgw-x1.nokia.com (Switch-2.2.6/Switch-2.2.6) with ESMTP id h6B8lVa12659
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 11:47:31 +0300 (EET DST)
Received: from esebh001.NOE.Nokia.com (unverified) by esvir05nok.ntc.nokia.com
 (Content Technologies SMTPRS 4.2.5) with ESMTP id <T635d98b366ac158f2549e@esvir05nok.ntc.nokia.com>;
 Fri, 11 Jul 2003 11:47:30 +0300
Received: from esebe014.NOE.Nokia.com ([172.21.138.53]) by esebh001.NOE.Nokia.com with Microsoft SMTPSVC(5.0.2195.6139);
	 Fri, 11 Jul 2003 11:47:30 +0300
Received: from esebe023.NOE.Nokia.com ([172.21.138.115]) by esebe014.NOE.Nokia.com with Microsoft SMTPSVC(5.0.2195.6139);
	 Fri, 11 Jul 2003 11:47:30 +0300
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"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] SUA Question - Trasaction ID based Routing
Date: Fri, 11 Jul 2003 11:47:29 +0300
Message-ID: <DADF50F5EC506B41A0F375ABEB32063658F124@esebe023.ntc.nokia.com>
Thread-Topic: [Sigtran] SUA Question - Trasaction ID based Routing
Thread-Index: AcNHH7v1LigPp2grTqmlOYAqdjIbdQAaOzVw
To: <bidulock@openss7.org>, <JDoss@verisign.com>
Cc: <sigtran@ietf.org>
X-OriginalArrivalTime: 11 Jul 2003 08:47:30.0412 (UTC) FILETIME=[102F1AC0:01C34789]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Brian & James,

Yes, this has a long & twisted history. My original goal was to
try to get SUA finished - something that the IESG has caused
a lot of trouble with. =20

The WG never could come to consensus on how to solve this.  My proposal
to you is to write a short Internet draft on how to solve this,
lets get consensus on it and publish it as a supplemental RFC.

John

> -----Original Message-----
> From: ext Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: 10 July, 2003 23:13
> To: Doss, James
> Cc: sigtran@ietf.org; Loughney John (NRC/Helsinki)
> Subject: Re: [Sigtran] SUA Question - Trasaction ID based Routing
>=20
>=20
> James,
>=20
> Doss, James wrote:                                           =20
>                        (Thu, 10 Jul 2003 12:27:16)
> >=20
> >    Hi John,
> >=20
> >=20
> >=20
> >    In SUA spec draft version 15.
> >=20
> >=20
> >=20
> >    With reference to Page 112 . SCTP Associations Fig 2=20
> Signaling Gateway
> >    Architecture.
>=20
> Page 112 would be the Author's Addresses...
>=20
> >=20
> >=20
> >=20
> >    Will  the  transaction  ID  based  routing  the =20
> responses  to the ASP
> >    originated the transaction.
> >=20
> >=20
> >=20
> >    Ex scenario:
> >=20
> >=20
> >=20
> >    ASP1 from AS3 originated a transaction using SG1 towards=20
> SS7 network.
> >=20
> >=20
> >=20
> >    The  Response  for  this  QUERY  hits  SG2  (Assuming =20
> SG1 and SG2 are
> >    mated).
> >=20
> >=20
> >=20
> >    Question 1:
> >=20
> >=20
> >=20
> >    At this condition will SG2 route this message back to ASP1.
> >=20
> >=20
> >=20
> >    Question 2:
> >=20
> >=20
> >=20
> >    Will  SG1  and  SG2  share  the  information  about =20
> Open Transactions
> >    originated from the ASPs(1..n).
> >=20
> >    Or Else how the Transaction ID based routing (key ) will work.
> >=20
> >=20
> >=20
> >    Please correct me if I understood the routing in a wrong way.
>=20
> The TID Label approach described in 4.7.2 and 4.7.3 and A.3.1=20
> will work for
> simple (and correct) Query/Response pairs in protocol class 0=20
> with operations
> class 3 (no return on error and no rejects or aborts) if you=20
> have sufficient
> control of the TCAP application that you can control the=20
> range of TIDs that
> it assigns per ASP.  Protocol class 0 or 1 and operations=20
> class 1 or 2 (return
> on error, rejects and aborts) will fail because the returned=20
> error does not
> have sufficient information to determine the ASP from TID=20
> (because the ASP's
> TID is not present in the message), or if you cannot strictly=20
> control the way
> that the TCAP layer assigns TIDs at an ASP.
>=20
> So, it works for only the most trivial of applications.  For more
> sophisticated applications, see draft-bidulock-sigtran-tua-01.txt
>=20
> A superior registration procedure was proposed whereby the=20
> ASP could inform
> both SGs of the ASP ownership of a TID on the BEGIN/QWP/QWOP,=20
> but the broken
> TID Label approach slithered its way into SUA.
>=20
> The DRN Label approach is broken worse.  For serious TCAP=20
> applications, use
> TUA.
>=20
> The last thread was "SUA-09: SCCP Connections and TCAP=20
> Transactions" from
> Jan 2002.  (At that time it was too late to make changes--hah!)
>=20
> Before that there is a discussion in "SUA Decision needed -=20
> handlin multiple
> SGs" where it was determined that broken procedures are ok if=20
> those broken
> procedures have a couple of people to nod their heads at each other.
>=20
> See especially "[SUA Issue 14] Changes and addtions required=20
> for decoupled
> operation."  These problems go back to August 2001.
>=20
> Perhaps Verwimp is still around to defend the Label approach,=20
> or perhaps
> not.
>=20
> --brian
>=20
>=20
> --=20
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
>=20

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



From exim@www1.ietf.org  Fri Jul 11 04:55:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA21586
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 04:55:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atgD-00058u-Rf
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 04:55:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B8t1Tw019761
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 04:55:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atgD-00058M-5e; Fri, 11 Jul 2003 04:55:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atfS-00057m-0W
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 04:54:14 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA21554
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 04:54:10 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19atfP-0003lk-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 04:54:11 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[74QXPWXUJ2a7LvQDTf5Yen+sdyNzm9fx])
	by ietf-mx with esmtp (Exim 4.12)
	id 19atfN-0003lh-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 04:54:09 -0400
Received: from ns.pigworks.openss7.net (IDENT:W8Ic22aNJMWM58v7fXz1UO4SrDJOog1Q@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6B8s9X11565;
	Fri, 11 Jul 2003 02:54:09 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6B8s9v03017;
	Fri, 11 Jul 2003 02:54:09 -0600
Date: Fri, 11 Jul 2003 02:54:09 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04 Section 3.14
Message-ID: <20030711025409.C1680@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF3@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: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF3@eestqnt105.es.eu.ericsson.se>; from j.javier.pastor@ericsson.com on Fri, Jul 11, 2003 at 09:28:56AM +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>

J.Javier,

J.Javier Pastor (EE/EEM) wrote:                                                   (Fri, 11 Jul 2003 09:28:56)
> Hi Brian, 
> 
> Pls, see my comments below...
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: jueves, 10 de julio de 2003 15:21
> > To: J.Javier Pastor (EE/EEM)
> > Cc: 'sigtran@ietf.org'
> > Subject: Re: [Sigtran] M3UA IG v04 Section 3.14
> > 
> > 
> > Javier,
> > 
> >    ---------
> >    New text: (Section 4.3.4.3, for both loadsharing and broadcast)
> >    ---------
> > 
> >    At start-up or re-start phases, an SGP or IPSP, upon 
> > reception of an
> >    ASP Active message for the first ASP in a Loadshare AS, SHOULD NOT
> >    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.
> > 
> > How can the SGP know when there are sufficient resources?
> > 
> > 'n'?  How do I count ASPs active for comparison against 'n' at one SGP
> > in multiple SGs?
> > 
> Yes, n is a good measure and it is part of the original text. But no one
> forbid you using "1" if you want to be more flexible and you are also using
> the smooth start. 

You misunderstand.  I was not questioning 'n'.  What do I count
at the SGP to compare against 'n'.  'n' is the system-wide
number of ASPs necessary to handle the maximum engineered load.
I cannot just set it to '1'.  It is determined by the engineered
traffic handling capacity of the ASPs and the maximum engineered
offered load.

The point here is that the SGP does not know how many ASPs are
active for the AS (system wide).  And has nothing to count that
is comparable to 'n'.

On the other hand, the ASPs do have this information.  And they
SHOULD coordinate amoungst themselves that 'n' are active.
Therefore the SGP can be assured that if 1 ASP is active to it,
that 'n' ASPs will be active system-wide.  And it can activate
on the first ASP active to it.

The AS state machine is in conflict with this.  It requires an
SGP to wait when it need not.  The number of ASPs active to a
given SGP cannot be used to determine the number ASPs active for
the AS system-wide.

> 
> You can know whether you have "n" counting the ASPs that are active for each
> AS in your SGP.

It is not necessary that the number of ASPs active for a given
SGP be 'n' for there to be 'n' ASPs active for the AS
systemwide.  Therefore, there can be sufficient ASPs active for
an AS system wide ('n' of them) when there are less than 'n'
ASPs active for a given SGP.

Because the ASPs are required to coordinate amoung themselves
that 'n' are active, the SGP can always be assured that 'n' will
be active and can activate on the first ASP that goes active to
it for the AS.

The recommendation is contrary to that.  It is counting apples
at the SGP an comparing it to oranges at the AS.


> 
> > Please provide a definition of, and criteria for, sufficient 
> > resources to
> > go with the recommendation.
> 
> It seems that this is a comment to the RFC, since there is no change
> included in the IG for that specific sentence. Do you propose any specific
> wording?

Yes, there was a change.  From MAY to SHOULD.  No criteria is
necessary when one says MAY.  When one says SHOULD, a criteria
is necessary.

Originally the SGP has the assurance that the ASPs would
coordinate that 'n' of them were active and it did not have to
worry further about it.  The option (MAY) was provided to allow
the SGP to use additional information for stronger assurances if
it would like.  How it went about doing that was its own
business.

With the change to SHOULD, you are saying the SGP cannot rely on
the assurance offered by the requirement for the ASPs to
coordinate that 'n' of them are active, but must apply some new
criteria of sufficiency.  So please define that criteria.

If you are saying that the SGP should still rely on the fact that
'n' ASPs will coordinate and become active, then the proposed AS
state machine is in conflict, because the SGP should then
activate on the first ASP.

So is the AS state machine in conflict or are you defining a new
criteria?

--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 exim@www1.ietf.org  Fri Jul 11 04:58:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA21656
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 04:58:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atj9-0005GX-0O
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 04:58:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B8w22A020234
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 04:58:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atj7-0005G2-Qb; Fri, 11 Jul 2003 04:58:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atiI-0005FF-Gl
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 04:57:10 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA21622
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 04:57:06 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19atiF-0003my-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 04:57:07 -0400
Received: from auds951.usa.alcatel.com ([143.209.238.80])
	by ietf-mx with esmtp (Exim 4.12)
	id 19atiE-0003md-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 04:57:06 -0400
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6B8uawA002679
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 03:56:36 -0500 (CDT)
Received: from sun2594.ssd.usa.alcatel.com (sun2594.ssd.usa.alcatel.com [143.209.250.252])
	by ssd.usa.alcatel.com (8.12.8/8.12.8) with ESMTP id h6B8uC1C011685
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 03:56:12 -0500 (CDT)
Date: Fri, 11 Jul 2003 03:56:12 -0500 (CDT)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04
In-Reply-To: <20030711023644.A1680@openss7.org>
Message-ID: <Pine.SOL.4.10.10307110353200.25952-100000@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,

See comments inline.

--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, 11 Jul 2003, Brian F. G. Bidulock wrote:

> J.Javier,
> 
> J.Javier Pastor (EE/EEM) wrote:                                                   (Fri, 11 Jul 2003 09:19:54)
> > 
> > Hi Brian, 
> > 
> > Pls, see below some comments...
> > 
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: jueves, 10 de julio de 2003 14:36
> > > To: J.Javier Pastor (EE/EEM)
> > > Cc: 'sigtran@ietf.org'
> > > Subject: Re: [Sigtran] M3UA IG v04
> > > 
> > > 
> > > Javier,
> > > 
> > > 'n' is defined as the number of ASPs that are needed to handle
> > > the maximum engineered load of the AS.
> > > 
Incorrect. RFC says in section 1.4.4.1
    
   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.

> > > This is the total load from the network correct?  As in the
> > > traffic delivered to the AS, from all SGP, all SG.  Correct?
> > > 
> > > What has that to do with the number of ASPs active for particular
> > > SGP?
> > > 
> > We are relying on SGPs to send all the load to the AS. Only when the AS gets
> > to ACTIVE we are sure that it could afford all the load.
> 
> What has that to do with the number of ASPs active for a
> particular SGP?
> 
> 'n' is a system value, not a per-SGP value.  The AS state
> machine in the IG treats 'n' as a per-SGP value, when it is
> defined as a system value.
> 
> What has the number of active ASPs for an SGP (the per-SGP
> value) have to do with the number of active ASPs for the AS
> (system value)?
> 
> An SGP cannot tell how many ASPs are active for the AS (system
> value), just the number of ASPs that are active to it (per-SGP
> value).
> 
> How can one count active ASPs for the SGP (per-SGP value) and
> compare it to 'n' (the system value)?
> 
> 
> > 
> > 
> > > There can be more than on SGP.  Correct?
> > > 
> > > There can be more than one SG.  Correct?
> > > 
> > > An ASP can be active for any number of SGPs in any number of 
> > > SGs.  Correct?
> > > 
> > > SGPs in an SG should be able to transfer traffic between 
> > > them.  Correct?
> > 
> > It basically depends on your implementation.
> > > 
> > > In general, what has the number of ASPs active for one SGP have
> > > to do with the traffic handling capacity of an AS?
> > > 
> > It has to do with the AS availability, a general rule in telephony nodes.
> > 
> > > 
> > > Consider the following (n+k == 2+0): (Figure 5).
> > > Consider first that SGP1 and SGP2 are in the same SG.
> > > 
> > > 
> > >      ________                  ________   
> > > ____|        |________________|        |
> > >     |  SGP1  |        ________|  ASP1  |
> > >     |________|       /        |________|
> > >      ________       /          ________
> > >     |        |_____/          |        |
> > > ____|  SGP2  |________________|  ASP2  |
> > >     |________|                |________|
> > > 
> > > 
> > Why don't you establish an assoc between SGP1 and ASP2 ?. This network
> > design seems not to be very appropriate for the n+k=2+0
> 
> It's a simple example.  The association between SGP1 and ASP2
> happens to be down at the moment.
> 
> > 
> > > 3.14 tells me that I SHOULD NOT direct traffic from the
> > > newtork from SGP1 to ASP1 on startup of ASP1 with SGP1.
> > > 
> > It is not recommended, but can always use smooth start. 
> 
> On what grounds can you not recommend directing traffic above?
> There are suffcient ASPs active fo the AS (system value) to
> handle the maximum engineered load.  'n' of them are active for
> the AS (system value), SGP1 has 1 of them active to it (per-SGP)
> value.
> 
> Please tell me what number you are expecting the SGP to use: the
> system value?  or the per-SGP value.
> 
> > 
> > > Does that include passing traffic from SGP1 via SGP2?
> > > 
> > Nothing is said about that. Therefore it is not included.
> 
> Silence implies acceptance.  If you only want to say that the
> SGP should not direct traffic towards the ASP directly over the
> association between the SGP and the ASP, please say so.
> 
> > 
> > > Is this because ASP1 cannot handle the load?
> > > 
> > It is because the AS has not sufficient resources to handle the expected
> > traffic from SGP1
> 
> But it does in the diagram.  Two ASPs are active for the AS and
> handling traffic for the AS.  There are 'n' ASPs handling
> traffic and they can handle the maximum engineerd load.
> 
> So there ARE sufficent resources to handle the expected load.
> It is just that SGP1 cannot discern that from the number of ASPs
> active to it alone.
> 
> > 
> > > What should SGP1 do?
> > > 
> > Wait for more ASPs to become ACTIVE
> 
> Why? There are 'n' ASPs already handling traffic!
> 
> > 
> > > Now, consider that SGP1 and SGP2 are in different SGs.
> > > 
> > > Should SGP1 tell the network that the AS is inaccessible?
> > > 
> > It depends on what you call the network. If you refer to SS7, it is the SG's
> > SPC the visible SPC, so SG will announce un/availability when needed.
> 
> > 
> > > Should SGP1 direct traffic over C-links to SGP2 that will
> > > ultimately be the same a passing traffic to ASP1?
> > >
> > It may, if SGP1 has a route to reach SGP2. 
> >  
> > > Why should not SGP1 direct traffic to ASP1?
> > > 
> > It could do it when you uses smooth start. But it is not recommended since
> > ASP1 could not handle all the load it receives...
> 
> Look at the diagram.  ASP1 is already exposed to full load via
> SGP2.  So is its not true that ASP1 cannot handling the load,
> because it is already exposed to it via SGP2.
> 
> Also, what do you mean by "could not handle all the load it
> receives..."?  An MTP-User (AS) must have overload controls to
> address traffic overloads at any time.
> 
> Are you advocating that ASP are to be built to not conform to
> SS7 specifications in this regard?
> 
> Is your ASP not able to handle overload?
> 
> Is there anyone on the list whose ASP is not able to handle
> overload traffic?
> 
> > 
> > > I don't think that 3.14 is very clear.  'n' has nothing to do
> > > with the handling capacity from a given SGP to an AS.  It has
> > > to do with the overall engineered maximum traffic that 'n' ASPs
> > > can handle for the AS.  In even the more trivial cases the SGP
> > > cannot follow the recommendation.
> > > 
> > It seems to me that the recommendation can be followed. Anyway I
> > may be wrong. So if you can make another proposal that everyone
> > agrees on, pls go ahead.
> 
> If you mean to count per-SGP APSs and compare the number against
> the system-wide engineered value 'n', then every time that the
> number of ASPs active for an SGP is different from the number of
> ASPs active for the AS (system wide), there will be a problem.
> 
> In systems where 'n' are active and 'k' are (cold) backup, every
> time that an ASP cannot activate to a given SGP, that SGP is out
> of service (as in the above trivial example).
> 
> Also, the recommendation is based on the premise that the ASP
> cannot handle overload in thes start scenario (where it
> obviously is able to handle it in the restart scenario).  This
> is a contradiction.  Either the ASP is capable of handling
> overload or it is not.
> 
> I have asked many times now, and nobody will say that their ASP
> cannot handle overload.
> 
> The recommendation is technically invalid because of the
> following:
> 
> 1. The number of ASPs active at an SGP (per-SGP) is not the
>    number of ASPs active for the AS (system value 'n').
> 
> 2. ASPs must be able to handle overload, because they must
>    conform to SS7 standards.
> 
> In the current RFC, it there are ASPs so badly designed that
> they cannot handle an overload, they should take care with how
> they activate.  They SHOULD coordinate amoung themselves that
> 'n' are active.  If 'n' ASPs cannot be brought to the ASP-ACTIVE
> state for the AS, then none of the ASPs need activate.
> 
> Relying on this, more sound, recommendation, the SGP can
> activate with the first ASP, knowing that 'n' of them have been
> coordinated to become active.
> 
> In this way, in the current RFC, the ASP assures that SGP that
> 'n' are to be active on a system basis (which only the ASPs can
> know, and because they have coordinated amoung themselves).  So
> the SGP can activate on the first ASP on a per-SGP basis.
> 
> The current RFC also allows the SGP to wait further if it has
> some additional information about the sufficiency of the number
> of active ASPs in addition to the assurance that there will be
> 'n' active or the first one would not have activated.
> 
> The current RFC provides a way around your counting problem, and
> even accommodates broken ASPs that can't conform to SS7 overload
> controls.  It also leaves the activation of the AS squarely on
> the shoulders of the ASPs, where it belongs, and the SGP just
> does what it is told.
> 
> So, by the RFC, in the above example, when ASP1 activates with
> SGP1, SGP1 already has the assurance that ASP1 has coordinated
> with ASP2 that both of them will become active.  Thus it can
> activate the as immediately.  It also MAY consider that SGP2 has
> ASP2 already active as verfication that there are sufficient
> ASPs active in the AS, or wait until some SGP has ASP2 active
> if, for example, it shares state amoung SGP. to that extent.
> 
> So, perhaps not so strangely, the recommendation in 3.14 for the
> example above, stops SGP1 from activating the AS when it is
> completely reasonable for it to do so.  The RFC, on the other
> hand, has SGP1 activating correctly.
> 
> I think you need to alter your proposal so that SGP1 can
> activate immeidately in the example above, on recommendation
> and not option.
> 
> You are welcome to go back to the wording in the RFC ("MAY
> withhold" instead of "SHOULD NOT direct") to correct this
> problem.  Of course, then there would be no need for all the
> 'n' stuff in the AS state machine.
> 
> --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 exim@www1.ietf.org  Fri Jul 11 05:07:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA21951
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 05:07:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atrr-00065j-UU
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 05:07:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B973lP023398
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 05:07:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atrq-00064V-2i; Fri, 11 Jul 2003 05:07:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atrU-000640-EA
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 05:06:40 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA21935
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 05:06:36 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19atrR-0003sP-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 05:06:37 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[IFLW2OQH56WLJmZCgiKiTccz3audpIpH])
	by ietf-mx with esmtp (Exim 4.12)
	id 19atrQ-0003sM-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 05:06:36 -0400
Received: from ns.pigworks.openss7.net (IDENT:nZEzKocWIETs8NrFVRcTyQUphQKvoXoQ@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6B96aX11879;
	Fri, 11 Jul 2003 03:06:36 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6B96ZH03516;
	Fri, 11 Jul 2003 03:06:35 -0600
Date: Fri, 11 Jul 2003 03:06:35 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04 Section 3.8
Message-ID: <20030711030635.D1680@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF4@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: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF4@eestqnt105.es.eu.ericsson.se>; from j.javier.pastor@ericsson.com on Fri, Jul 11, 2003 at 09:51:40AM +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>

J.Javier,

J.Javier Pastor (EE/EEM) wrote:                                                   (Fri, 11 Jul 2003 09:51:40)
> Hi Brian, 
> 
> See comments below...
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: jueves, 10 de julio de 2003 15:30
> > To: J.Javier Pastor (EE/EEM)
> > Cc: 'sigtran@ietf.org'
> > Subject: Re: [Sigtran] M3UA IG v04 Section 3.8
> > 
> > 
> > Javier,
> > 
> > Section 3.8 says:
> > 
> >    5.4 Auditing examples
> > 
> >    5.4.1 SG State: Uncongested / Unavailable
> > 
> >           ASP                          SGP
> >           ---                          ---
> >            |  -------- DAUD --------->  |
> >            |  <------ SCON(0) --------  |
> >            |  <------- DUNA ----------  |
> > 
> > What state of a destination is Uncongested/Unavailable?  There is no
> > such state in Q.704.  Uncongested routesets are by definition 
> > available.
> > Inaccessible routesets have no congestion state by definition.
> 

Is the above diagram in error as well?

> 
> > 
> > Did you think that a routeset could be in an unavailble and 
> > congested state?
> > 
> > What is the use of the SCON(0)?  Why is it being sent?
> > 
> >    5.4.4 SG state: Uncongested / Unavailable
> > 
> >           ASP                          SGP
> >           ---                          ---
> >            |  -------- DAUD --------->  |
> >            |  <------- DUNA ----------  |
> > 
> > What is the difference between 5.4.1 and 5.4.4?  They both say
> > Uncongested/Available.
> > 
> > 
> It is a typo. In 5.4.1 it should say: SG state: "Uncongested / Available".
> And of course a DAVA has to be sent.

Why then is SCON(0) sent in 5.4.1 above, and not in 5.4.4.

Perhaps you have the typo reversed?  5.4.1 should be Uncongested/Unavailable
(with DAVA instead of DUNA), and 5.4.4 should be just plain Unavailable.

Like so would be correct:

    5.4.1 SG State: Uncongested / Available
 
           ASP                          SGP
           ---                          ---
            |  -------- DAUD --------->  |
            |  <------ SCON(0) --------  |
            |  <------- DAVA ----------  |

    5.4.4 SG state: Unavailable
 
           ASP                          SGP
           ---                          ---
            |  -------- DAUD --------->  |
            |  <------- DUNA ----------  |
 

> 
> > 
> >    [...]. Where the SGP maintains the congestion status of the SS7
> >    destination, the SGP MUST additionally respond with an SCON message
> >    before the DAVA or DRST message.  If the SS7 destination is
> > 
> > When is the SGP required to maintain congestion status?  
> 
> In those networks where the congestion levels are handled... There's no
> change there regarding the RFC. Do you think the RFC requires some
> additional explanation in this regard?

Well, there is this change:

   Where the SGP does not maintain the congestion status of the SS7
   destination, the response to a DAUD message should always be only a
   DAVA, DRST or DUNA message as appropriate.

When does the SGP not maintain the congestion status of the SS7 destination?
When does it?  The first passage is a MUST and this is a SHOULD.  How is one
to know when an SGP is following the MUST or SHOULD and when it is not?

If it doesn't really matter, why do we say MUST and SHOULD?

> 
> 
> 
> > What is the criteria for an ASP to determine that an SGP should maintain
> > congestion status and is violating this MUST.  Why must the ASP require
> > the SGP to send SCON in this circumstance?  What interoperability concern
> > is the MUST addressing?
> > 
> It seems to me that it was not clear whether to send the SCON or not during
> the bakeoff.

How is it clearer now?  One still doesn't know when an SGP MUST or SHOULD NOT
send SCON because there is no criteria for determining when an SGP maintains
congestion state (a completely undefined concept).

--brian

> 
> 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 exim@www1.ietf.org  Fri Jul 11 05:08:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA21997
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 05:08:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atsn-0006M3-01
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 05:08:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B980Wf024405
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 05:08:00 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atsm-0006LX-RH; Fri, 11 Jul 2003 05:08:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atsK-0006F1-CY
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 05:07:32 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA21949
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 05:07:28 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19atsH-0003sk-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 05:07:29 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[jaxISuEkU39mOJ/wxjzU98Ze94OYORAm])
	by ietf-mx with esmtp (Exim 4.12)
	id 19atsG-0003sh-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 05:07:28 -0400
Received: from ns.pigworks.openss7.net (IDENT:5o2lpYYibpi7qDZI70D4djoD9Xse5VZj@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6B97TX11893;
	Fri, 11 Jul 2003 03:07:29 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6B97Tb03542;
	Fri, 11 Jul 2003 03:07:29 -0600
Date: Fri, 11 Jul 2003 03:07:28 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
Message-ID: <20030711030728.E1680@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF5@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: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF5@eestqnt105.es.eu.ericsson.se>; from j.javier.pastor@ericsson.com on Fri, Jul 11, 2003 at 09:56:59AM +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>

J.Javier,

J.Javier Pastor (EE/EEM) wrote:                                                   (Fri, 11 Jul 2003 09:56:59)
> Hi Brian, 
> 
> Pls, see below...
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: jueves, 10 de julio de 2003 15:34
> > To: J.Javier Pastor (EE/EEM)
> > Cc: 'sigtran@ietf.org'
> > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > 
> > 
> > Javier,
> > 
> > 3.18.3 Solution Description
> > 
> >    Clarify that the ASP needs to indicate when the congestion 
> > level has
> >    changed (including abatement).  Further, the ASP peer can maintain
> >    a timer, if desired, in case the ASP fails to indicate congestion
> >    abatement.
> > 
> > Where are the procedures for this timer described?  This is the only place
> > that the timer is mentioned.  Please describe the timer or remove the
> > sentence saying that it has been described...
> > 
> 
> Do you think is worth include the timer explanation into the new text?
> Otherwise, we may remove that sentence. 

Please remove mention of the timer.

--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 exim@www1.ietf.org  Fri Jul 11 05:12:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA22184
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 05:12:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atwg-0006a8-6G
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 05:12:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B9C2EV025295
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 05:12:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atwf-0006Zk-Rg; Fri, 11 Jul 2003 05:12:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atvi-0006Yz-DK
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 05:11:02 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA22156
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 05:10:58 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19atvf-0003vU-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 05:10:59 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[D+VG89h+CepedNAyqAgLR3r5e1IFkF+D])
	by ietf-mx with esmtp (Exim 4.12)
	id 19atve-0003vR-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 05:10:58 -0400
Received: from ns.pigworks.openss7.net (IDENT:sPaLXAF02L4fNRejdVzwyfUjudUSlY+m@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6B9AwX11937;
	Fri, 11 Jul 2003 03:10:58 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6B9AwY03615;
	Fri, 11 Jul 2003 03:10:58 -0600
Date: Fri, 11 Jul 2003 03:10:57 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04 Section 3.9
Message-ID: <20030711031057.F1680@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF6@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: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF6@eestqnt105.es.eu.ericsson.se>; from j.javier.pastor@ericsson.com on Fri, Jul 11, 2003 at 10:02:43AM +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>

J.Javier,

J.Javier Pastor (EE/EEM) wrote:                                                   (Fri, 11 Jul 2003 10:02:43)
> Hi Brian, 
> 
> Pls, see below..
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: jueves, 10 de julio de 2003 15:42
> > To: J.Javier Pastor (EE/EEM)
> > Cc: 'sigtran@ietf.org'
> > Subject: Re: [Sigtran] M3UA IG v04 Section 3.9
> > 
> > 
> > Javier,
> > 
> > Section 3.9 says:
> > 
> >           ASP                          SGP
> >           ---                          ---
> >            |  ------ ASPIA (RC1)----->  |
> >            |  <----  ASPIA Ack -------  |
> >            |  -----DEREG REQ (RC1)--->  |
> >            |  <----DEREG RSP (RC1)----  |
> >            |  -------ASPIA (RC1)----->  |
> > 
> >    What should SG do?
> > 
> > The new text does not explicity answer this question.
> > 
> > What does the SGP send in this case?
> 
> It is included in the new text for section 4.5.3. Take a look to the second
> bullet. Although you may be right saying that we should include it also into
> the "solution description"

Yes, because you have pulled three bullets in the new text without indicating
which of them addresses the problem.  The new text is fine, the answer to the
question "What should SG do?" is not really answered.  Just place in the
Solution description that the SG should send Error("Invalid Routing Context")
in answer to the original question.

> 
> > 
> > As RC values are per-SGP, it appears to me that it must be
> > Error("Invalid Routing Context").
> > 
> > Correct?
> 
> Yes, correct.
> 
> 
> Thanks // 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 exim@www1.ietf.org  Fri Jul 11 05:14:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA22293
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 05:14:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atyc-0006ii-6t
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 05:14:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B9E2uQ025826
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 05:14:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atyc-0006iR-2z; Fri, 11 Jul 2003 05:14:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19atyR-0006i5-9y
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 05:13:51 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA22254
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 05:13:45 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19atyM-0003wq-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 05:13:46 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[Nxdfcpct5P8Veei98pz+qTlyEDHb+CP0])
	by ietf-mx with esmtp (Exim 4.12)
	id 19atyL-0003wf-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 05:13:45 -0400
Received: from ns.pigworks.openss7.net (IDENT:ZHEm100OXa4/d3Sfb33JFawGC2Vgros/@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6B9DkX11967;
	Fri, 11 Jul 2003 03:13:46 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6B9DkT03645;
	Fri, 11 Jul 2003 03:13:46 -0600
Date: Fri, 11 Jul 2003 03:13:46 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04  Section 3.14
Message-ID: <20030711031346.G1680@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF7@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: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CF7@eestqnt105.es.eu.ericsson.se>; from j.javier.pastor@ericsson.com on Fri, Jul 11, 2003 at 10:35:57AM +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>

J.Javier,

J.Javier Pastor (EE/EEM) wrote:                                                   (Fri, 11 Jul 2003 10:35:57)
> Hi Brian, 
> 
> You are right saying that there is currently no way of notifying AS-DOWN. In
> fact "n" only applies to the traffic case.
> 
> Then I see two options:
> 	1- include AS-DOWN in NTFY
> 	2- The AS-DOWN state is only got when all the ASPs are in ASP-DOWN state.
> 
> I would addopt 2 since it is inline what we are having in the other UALs. 

Yes, please go with 2 as it is also in line with the RFC.

That will also leave only 1 'n' transition in the proposal: the IA2AC
transition, correct?

--brian

> 
> Thanks // Javier.
> 
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: jueves, 10 de julio de 2003 17:15
> > To: J.Javier Pastor (EE/EEM)
> > Cc: 'sigtran@ietf.org'
> > Subject: Re: [Sigtran] M3UA IG v04 Section 3.14
> > 
> > 
> > Javier,
> > 
> > Section 3.14 says:
> > 
> >    IA2DN: One ASP moves to ASP-DOWN causing that the number of ASPs in
> >    either ASP-ACTIVE or ASP-INACTIVE drops below n.
> > 
> > Am I correct to say that if n=5 and there are 4 ASPs in the 
> > ASP-INACTIVE
> > and 1 ASP in the ASP-ACTIVE state, and then the 1 active ASP moves to
> > the ASP-DOWN state and T(r) expires that the AS is to move to 
> > the AS-DOWN
> > state?  As follows:
> > 
> >   If T(r) expires before an ASP becomes ASP-ACTIVE, the SGP MAY stop
> >   queuing messages and discards all
> >   previously queued messages. The AS will move to the 
> > AS-INACTIVE state
> >   if at least the number of ASPs in ASP-INACTIVE sum n, otherwise it
> >   will move to AS-DOWN state.
> > 
> > Now, that is an AS state change and we must send a NTFY to 
> > all the INACTIVE
> > ASPs (the 4 left inactive) of the state change.  But from 
> > Section 3.8.2 of
> > the RFC, there is no message to notify of an AS-DOWN state:
> > 
> >          1    reserved
> >          2    Application Server Inactive (AS-INACTIVE)
> >          3    Application Server Active (AS-ACTIVE)
> >          4    Application Server Pending (AS-PENDING)
> > 
> > You see, we removed it because there was never an instance in which it
> > would be sent.  You have inserted an instance again.
> > 
> > Aren't you going to have to define a Notify message (the 
> > Reserved value)
> > for "Application Server Down (AS-DOWN)" so that the INACTIVE ASPs can
> > be properly notified by the SGP?
> > 
> > Particularly when the RFC says in Section 4.3.4.5: (Notify Procedures)
> > 
> >    A Notify message reflecting a change in the AS state MUST 
> > be sent to
> >    all ASPs in the AS, except those in the ASP-DOWN state, with
> >    appropriate Status Information and any ASP Identifier of the failed
> >    ASP.  At the ASP, Layer Management is informed with an M-NOTIFY
> >    indication primitive.  The Notify message must be sent 
> > whether the AS
> >    state change was a result of an ASP failure or reception of an ASP
> >    State management (ASPSM) / ASP Traffic Management (ASPTM) message.
> >    In the second case, the Notify message MUST be sent after 
> > any related
> >    acknowledgement messages  (e.g., ASP Up Ack, ASP Down Ack, 
> > ASP Active
> >    Ack, or ASP Inactive Ack).
> > 
> > The above tells me that I MUST send a Notify message becase of the AS
> > state change, but there is no message to inform about AS-DOWN state
> > transition.
> > 
> > What is the purpose of doing this?  What is the protocol 
> > difference between
> > the AS-DOWN state and the AS-INACTIVE state?  What is it that the SGP
> > does or doesn't do in the one state that it doesn't or does 
> > do in the other?
> > 
> > I don't recall this being discussed at all.
> > 
> > Please clarify.
> > 
> > -- 
> > 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 exim@www1.ietf.org  Fri Jul 11 05:18:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA22398
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 05:18:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19au2T-0007IE-St
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 05:18:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B9I10p028026
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 05:18:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19au2T-0007Hi-0V; Fri, 11 Jul 2003 05:18:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19au2O-0007HM-TL
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 05:17:56 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA22381
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 05:17:52 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19au2L-0003ya-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 05:17:53 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[MWD+prcbxV5dmIjzt1mBqfMJ8fzyDi+n])
	by ietf-mx with esmtp (Exim 4.12)
	id 19au2L-0003yX-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 05:17:53 -0400
Received: from ns.pigworks.openss7.net (IDENT:t1f79P5JtAb4w9gxXmN8cPa2XHzna3aA@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6B9HrX12011;
	Fri, 11 Jul 2003 03:17:53 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6B9HrH03683;
	Fri, 11 Jul 2003 03:17:53 -0600
Date: Fri, 11 Jul 2003 03:17:53 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04
Message-ID: <20030711031753.H1680@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <20030711023644.A1680@openss7.org> <Pine.SOL.4.10.10307110353200.25952-100000@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.4.10.10307110353200.25952-100000@sun2594.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Fri, Jul 11, 2003 at 03:56:12AM -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>

Samuel,

Samuel Dur D. Jeyaseelan wrote:                                                     (Fri, 11 Jul 2003 03:56:12)
> 
> Brian,
> 
> See comments inline.
> 
> --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, 11 Jul 2003, Brian F. G. Bidulock wrote:
> 
> > J.Javier,
> > 
> > J.Javier Pastor (EE/EEM) wrote:                                                   (Fri, 11 Jul 2003 09:19:54)
> > > 
> > > Hi Brian, 
> > > 
> > > Pls, see below some comments...
> > > 
> > > > -----Original Message-----
> > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > Sent: jueves, 10 de julio de 2003 14:36
> > > > To: J.Javier Pastor (EE/EEM)
> > > > Cc: 'sigtran@ietf.org'
> > > > Subject: Re: [Sigtran] M3UA IG v04
> > > > 
> > > > 
> > > > Javier,
> > > > 
> > > > 'n' is defined as the number of ASPs that are needed to handle
> > > > the maximum engineered load of the AS.
> > > > 
> Incorrect. RFC says in section 1.4.4.1
>     
>    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 was talking about the IG (as in the subject line).  The IG says:

   As "n" ASPs are needed to handle the maximum engineered load of the
   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 exim@www1.ietf.org  Fri Jul 11 05:21:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA22547
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 05:21:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19au5N-0007Ru-0Q
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 05:21:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B9L0Z0028628
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 05:21:00 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19au5M-0007Rd-SX; Fri, 11 Jul 2003 05:21:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19au50-0007Pd-Hf
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 05:20:38 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA22496
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 05:20:34 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19au4x-00040u-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 05:20:35 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[qkkkg6xFxxsPR071O6jPk4G8Po/mXhw2])
	by ietf-mx with esmtp (Exim 4.12)
	id 19au4w-00040r-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 05:20:34 -0400
Received: from ns.pigworks.openss7.net (IDENT:sidbnWNjvGjx4HQErtGMBnZLqyKPoF6a@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6B9KZX12053;
	Fri, 11 Jul 2003 03:20:35 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6B9KYo03738;
	Fri, 11 Jul 2003 03:20:34 -0600
Date: Fri, 11 Jul 2003 03:20:34 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: john.loughney@nokia.com
Cc: JDoss@verisign.com, sigtran@ietf.org
Subject: Re: [Sigtran] SUA Question - Trasaction ID based Routing
Message-ID: <20030711032034.I1680@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: john.loughney@nokia.com, JDoss@verisign.com,
	sigtran@ietf.org
References: <DADF50F5EC506B41A0F375ABEB32063658F124@esebe023.ntc.nokia.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: <DADF50F5EC506B41A0F375ABEB32063658F124@esebe023.ntc.nokia.com>; from john.loughney@nokia.com on Fri, Jul 11, 2003 at 11:47:29AM +0300
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,

Excellent suggestion.

--brian

John Loughney wrote:                                                                  (Fri, 11 Jul 2003 11:47:29)
> Brian & James,
> 
> Yes, this has a long & twisted history. My original goal was to
> try to get SUA finished - something that the IESG has caused
> a lot of trouble with.  
> 
> The WG never could come to consensus on how to solve this.  My proposal
> to you is to write a short Internet draft on how to solve this,
> lets get consensus on it and publish it as a supplemental RFC.
> 
> John
> 
> > -----Original Message-----
> > From: ext Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: 10 July, 2003 23:13
> > To: Doss, James
> > Cc: sigtran@ietf.org; Loughney John (NRC/Helsinki)
> > Subject: Re: [Sigtran] SUA Question - Trasaction ID based Routing
> > 
> > 
> > James,
> > 
> > Doss, James wrote:                                            
> >                        (Thu, 10 Jul 2003 12:27:16)
> > > 
> > >    Hi John,
> > > 
> > > 
> > > 
> > >    In SUA spec draft version 15.
> > > 
> > > 
> > > 
> > >    With reference to Page 112 . SCTP Associations Fig 2 
> > Signaling Gateway
> > >    Architecture.
> > 
> > Page 112 would be the Author's Addresses...
> > 
> > > 
> > > 
> > > 
> > >    Will  the  transaction  ID  based  routing  the  
> > responses  to the ASP
> > >    originated the transaction.
> > > 
> > > 
> > > 
> > >    Ex scenario:
> > > 
> > > 
> > > 
> > >    ASP1 from AS3 originated a transaction using SG1 towards 
> > SS7 network.
> > > 
> > > 
> > > 
> > >    The  Response  for  this  QUERY  hits  SG2  (Assuming  
> > SG1 and SG2 are
> > >    mated).
> > > 
> > > 
> > > 
> > >    Question 1:
> > > 
> > > 
> > > 
> > >    At this condition will SG2 route this message back to ASP1.
> > > 
> > > 
> > > 
> > >    Question 2:
> > > 
> > > 
> > > 
> > >    Will  SG1  and  SG2  share  the  information  about  
> > Open Transactions
> > >    originated from the ASPs(1..n).
> > > 
> > >    Or Else how the Transaction ID based routing (key ) will work.
> > > 
> > > 
> > > 
> > >    Please correct me if I understood the routing in a wrong way.
> > 
> > The TID Label approach described in 4.7.2 and 4.7.3 and A.3.1 
> > will work for
> > simple (and correct) Query/Response pairs in protocol class 0 
> > with operations
> > class 3 (no return on error and no rejects or aborts) if you 
> > have sufficient
> > control of the TCAP application that you can control the 
> > range of TIDs that
> > it assigns per ASP.  Protocol class 0 or 1 and operations 
> > class 1 or 2 (return
> > on error, rejects and aborts) will fail because the returned 
> > error does not
> > have sufficient information to determine the ASP from TID 
> > (because the ASP's
> > TID is not present in the message), or if you cannot strictly 
> > control the way
> > that the TCAP layer assigns TIDs at an ASP.
> > 
> > So, it works for only the most trivial of applications.  For more
> > sophisticated applications, see draft-bidulock-sigtran-tua-01.txt
> > 
> > A superior registration procedure was proposed whereby the 
> > ASP could inform
> > both SGs of the ASP ownership of a TID on the BEGIN/QWP/QWOP, 
> > but the broken
> > TID Label approach slithered its way into SUA.
> > 
> > The DRN Label approach is broken worse.  For serious TCAP 
> > applications, use
> > TUA.
> > 
> > The last thread was "SUA-09: SCCP Connections and TCAP 
> > Transactions" from
> > Jan 2002.  (At that time it was too late to make changes--hah!)
> > 
> > Before that there is a discussion in "SUA Decision needed - 
> > handlin multiple
> > SGs" where it was determined that broken procedures are ok if 
> > those broken
> > procedures have a couple of people to nod their heads at each other.
> > 
> > See especially "[SUA Issue 14] Changes and addtions required 
> > for decoupled
> > operation."  These problems go back to August 2001.
> > 
> > Perhaps Verwimp is still around to defend the Label approach, 
> > or perhaps
> > not.
> > 
> > --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 exim@www1.ietf.org  Fri Jul 11 05:26:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA22651
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 05:26:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19auAE-0007lH-NP
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 05:26:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B9Q2Yq029829
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 05:26:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19auAD-0007kt-Re; Fri, 11 Jul 2003 05:26:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19au9Q-0007he-PJ
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 05:25:13 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA22635
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 05:25:08 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19au9N-00043h-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 05:25:09 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[kmh7BhcMYr+6NoKQPCLWrHIxBjleCDy3])
	by ietf-mx with esmtp (Exim 4.12)
	id 19au9M-00043e-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 05:25:08 -0400
Received: from ns.pigworks.openss7.net (IDENT:AB6BIMcCLplLa2eKV45fziloeoSdeSfB@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6B9P8X12108;
	Fri, 11 Jul 2003 03:25:08 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6B9P7L03790;
	Fri, 11 Jul 2003 03:25:07 -0600
Date: Fri, 11 Jul 2003 03:25:06 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Valerie Gastaud <valerie.gastaud@ulticom.com>
Cc: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - RC in SSNM messages
Message-ID: <20030711032506.J1680@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Valerie Gastaud <valerie.gastaud@ulticom.com>,
	Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <20030710142654.C24498@openss7.org> <NDBBIHAOCKGAMIKNLALOEEDHDOAA.valerie.gastaud@ulticom.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: <NDBBIHAOCKGAMIKNLALOEEDHDOAA.valerie.gastaud@ulticom.com>; from valerie.gastaud@ulticom.com on Fri, Jul 11, 2003 at 10:27:26AM +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>

Valerie,

I would have no problems with that.  Would that firm things up for you?

Does anyone rely on the (s)?

I think that it would be quite easy to redefine 3 configured AS as
a single compound AS at the SG anyway, and still not have to send RC.

--brian

Valerie Gastaud wrote:                                                             (Fri, 11 Jul 2003 10:27:26)
> Brian, Tolga,
> 
> so, the conclusion is that when there are multiple AS on the same
> association, all the messages (SSNM, DATA and ASPTM) MUST contain the RC
> parameter.
> 
> if this is agreed by everyone, can we remove the (s) of Application Server
> in the following sentence from section 4.3.4.3 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.
> 
> and change the RC from optional to conditional in the ASPTM messages as well
> 
> and change the SHOULD in section 4.3.4.3 by a MUST:
>                     ...  In the case where an ASP wishes to process the
> 	traffic for more than one Application Server across a common SCTP
> 	association, the ASP Active message(s) SHOULD contain a list of one
> 	or more Routing Contexts to indicate for which Application Servers
> 	the ASP Active message applies. ...
> 
> Valerie
> -----Original Message-----
> From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf Of
> Brian F. G. Bidulock
> Sent: jeudi 10 juillet 2003 22:27
> To: Tolga Asveren
> Cc: sigtran@ietf.org
> Subject: Re: [Sigtran] M3UA - RC in SSNM messages
> 
> 
> Tolga,
> 
> 
> Tolga Asveren wrote:
> (Thu, 10 Jul 2003 12:48:14)
> > Brian,
> > --- "Brian F. G. Bidulock" <bidulock@openss7.org>
> > wrote:
> > > Tolga,
> > >
> > > Tolga Asveren wrote:
> > >                                 (Thu, 10 Jul 2003
> > > 11:10:20)
> > > > Brian,
> > > >
> > > > I was referring to 4.3.4.3 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."
> > > > This enables the option to send ASPAC with no RC,
> > > even
> > > > if multiple AS traffic is carried over one
> > > > association. I don't think, this is permitted for
> > > > SSNM.
> > >
> > > It permits a single AS in fitting with the SHOULD I
> > > quoted.  Multiple
> > > configured AS for the same association would be
> > > against the SHOULD.
> > > So you are still on your own.
> > [TOLGA]Then why does it say "Application Server(s)"
> >                                                 ^
> > I am pretty sure, what is meant here (In fact even
> > just two days ago, there was a qustion related with
> > this)
> 
> If you recall, the (s) predates the RC in SSNM and DATA
> discussion.  The SHOULD was put in after, at the same
> time as the RC passages in SSNM and DATA.  If you have
> multiple AS on the same association you need to put RC
> in all messages.  This was one reason we changed "Optional"
> to "Conditional' for RC throughout the document.  The
> same condition applies in all cases.
> 
> --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 exim@www1.ietf.org  Fri Jul 11 05:42:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA22996
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 05:42:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19auPi-0000sJ-V4
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 05:42:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6B9g2FP003358
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 05:42:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19auPh-0000ry-Hv; Fri, 11 Jul 2003 05:42:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19auPd-0000re-IS
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 05:41:57 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA22986
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 05:41:53 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19auPa-0004Ap-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 05:41:54 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[BtuJ+MFYSmhXICzsWZdLuYzhz8+QjKX6])
	by ietf-mx with esmtp (Exim 4.12)
	id 19auPY-0004Am-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 05:41:53 -0400
Received: from ns.pigworks.openss7.net (IDENT:99k4C8MwiytlJlWwXzVGid2D2nYy5KRq@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6B9fqX12476;
	Fri, 11 Jul 2003 03:41:52 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6B9fpA04085;
	Fri, 11 Jul 2003 03:41:51 -0600
Date: Fri, 11 Jul 2003 03:41:51 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Zannin, Francois" <francois.zannin@hp.com>
Cc: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M3UA IG v04
Message-ID: <20030711034151.A3995@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Zannin, Francois" <francois.zannin@hp.com>,
	Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
References: <06CAA4D3753C53408D2CE6340B68444201488AA7@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: <06CAA4D3753C53408D2CE6340B68444201488AA7@vbeexc02.emea.cpqcorp.net>; from francois.zannin@hp.com on Fri, Jul 11, 2003 at 08:59:03AM +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>

Francois,

Zannin, Francois wrote:                                                            (Fri, 11 Jul 2003 08:59:03)
> Brian,
> 
> What about the DUPU between IPSP? Can we make an agreement on how it
> MUST be handle when received by an IPSP or a SG?

I would say either Error("Unexpected Message") or
Error("Unsupported Message Type").

> Do we have to consider it as a protocol error or MUST we use it to
> inform local application (or forward an UPU?

I would say it is an error.  The AS should use ASP Inactive at MTP-SAP
granularity and send embedded UPU at PC granularity.  The IPSP should do
the same.

--brian

> 
> Francois.
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] 
> > Sent: Thursday, July 10, 2003 10:34 PM
> > To: Zannin, Francois
> > Cc: Tolga Asveren; sigtran@ietf.org
> > Subject: Re: [Sigtran] M3UA IG v04
> > 
> > 
> > Francois,
> > 
> > I don't think IPSP is that well defined yet.  We were 
> > supposed to address it in an extension draft.  Some 
> > corrections have been made in the IG, but a lot of work 
> > remains to be done.  We spent some time defining SG-SG 
> > operation which makes more sense that the SE or DE models, 
> > but there wasn't enough interest to advance the work.
> > 
> > --brian
> > 
> > Zannin, Francois wrote:                                       
> >                      (Thu, 10 Jul 2003 22:22:57)
> > > OK, now, is the DUPU received on a SG or an IPSP considered as a 
> > > protocol error? This has to be clearly stated in the RFC 
> > (or IG) Same 
> > > remark for the SCON received from an IPSP, is it possible 
> > to receive a 
> > > SCON or must the IPSP use the SCTP congestion to handle the 
> > congestion 
> > > on a remote IPSP.
> > > 
> > > Francois.
> > > 
> > > > -----Original Message-----
> > > > From: Tolga Asveren [mailto:tolga_asveren@yahoo.com]
> > > > Sent: Thursday, July 10, 2003 10:13 PM
> > > > To: sigtran@ietf.org
> > > > Subject: RE: [Sigtran] M3UA IG v04
> > > > 
> > > > 
> > > > Francois,
> > > > 
> > > > IMO, for the scenario you describe, sending ASPIA is
> > > > the correct behaviour. Yes, you won't be able to use healthy user 
> > > > parts in such a case, but if you have that level of 
> > granularity, I 
> > > > would think, RKs should be defined accordingly(for 
> > example for your 
> > > > example PC/SI RKs)
> > > > The same applies to ASP/SGP case. There is no DUPU
> > > > from ASP to SGP direction.
> > > > So, IMO, DUPU for IPSP case is not necessary.
> > > > 
> > > >      Tolga
> > > > --- "Zannin, Francois" <francois.zannin@hp.com> wrote:
> > > > > tolga,
> > > > > 
> > > > > see inline,
> > > > > 
> > > > > Regards,
> > > > > 
> > > > > Francois.
> > > > > 
> > > > > > -----Original Message-----
> > > > > > From: Tolga Asveren
> > > > > [mailto:tolga_asveren@yahoo.com]
> > > > > > Sent: Thursday, July 10, 2003 9:47 PM
> > > > > > To: sigtran@ietf.org
> > > > > > Subject: Re: [Sigtran] M3UA IG v04
> > > > > > 
> > > > > > 
> > > > > > Brian,
> > > > > > [..snip..]
> > > > > > > > [TOLGA] Yes, SGPs send DUPU to ASPs but to my 
> > understanding, 
> > > > > > > > Francois's case would
> > > > > correspond
> > > > > > > > sending DUPU from ASP to SGP.
> > > > > > > > As I said, my interpretation of DE IPSP in the
> > > > > IG
> > > > > > > is a
> > > > > > > > combined SGP/ASP. On ASP part of IPSP, the
> > > > > user
> > > > > > > part
> > > > > > > > goes down, then I would expect inactivating
> > > > > the
> > > > > > > > corresponding AS.
> > > > > > > 
> > > > > > > Yes, that's what we understood and agreed,
> > > > > however,
> > > > > > > that is
> > > > > > > not what is presented in the IG: Read the IG
> > > > > again:
> > > > > > > New text in section 3.11.2:
> > > > > > > 
> > > > > > >    In section 4.3.1: ASP/IPSP States, only the 
> > SGP-ASP and the 
> > > > > > > IPSP SE
> > > > > > >    scenarios are described. For the IPSP DE
> > > > > model,
> > > > > > > both IPSPs MUST
> > > > > > >    follow the SGP side of the SGP-ASP
> > > > > procedures.
> > > > > > > 
> > > > > > > That looks pretty plain to me.  According to the
> > > > > IG,
> > > > > > > DE IPSPs can
> > > > > > > send DUPU (they are actually required above to
> > > > > do so
> > > > > > > when the SGP
> > > > > > > procedure dictates).
> > > > > > [TOLGA]IMO, the sentence you refer might really be 
> > mislaeading, 
> > > > > > but I don't think it should make the
> > > > > SGP
> > > > > > send DUPU for the case we are describing. Let's
> > > > > say we
> > > > > > have IPSP1 and IPSP2. If one user part is down on 
> > IPSP2, then it 
> > > > > > should be ASP part of IPSP2 taking
> > > > > some
> > > > > > action and sending ASPIA, not SGP part of IPSP2.
> > > > > 
> > > > > The case I was refering first, was a platform
> > > > > hosting only one AS but
> > > > > internally made of several services (SI), not always 
> > available at 
> > > > > the same time.
> > > > > If one SI is not available but the others are still
> > > > > "usable", if the AS
> > > > > is "Inactivated", all SI will be prohibited! In that
> > > > > case the "IPSP" MAY
> > > > > (that's my question) send a DUPU for the concerned
> > > > > SI but the traffic 
> > > > > for the others SI is still possible.
> > > > > If IPSP cannot send a DUPU (or an UPU encoded), this
> > > > > mean that an AS
> > > > > identified by a PC only CANNOT corretly play several
> > > > > SI.
> > > > > 
> > > > > So my question, is the DUPU allowed between two IPSP
> > > > > ? 
> > > > >    (or for a same reason between an AS and a SG(p) ?
> > > > > )
> > > > > 
> > > > > > > --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!?
> > > > > > The New Yahoo! Search - Faster. Easier. Bingo. 
> > > > > > http://search.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
> > > > 
> > > > 
> > > > __________________________________
> > > > Do you Yahoo!?
> > > > SBC Yahoo! DSL - Now only $29.95 per month! http://sbc.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
> > 
> > -- 
> > 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 exim@www1.ietf.org  Fri Jul 11 07:42:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA25713
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 07:42:32 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19awHs-000287-BF
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 07:42:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6BBg4eW008175
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 07:42:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19awHp-00027I-Gm; Fri, 11 Jul 2003 07:42:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19awHF-00024r-GW
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 07:41:25 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA25670
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 07:41:23 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19awHE-0004up-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 07:41:24 -0400
Received: from web12809.mail.yahoo.com ([216.136.174.136])
	by ietf-mx with smtp (Exim 4.12)
	id 19awHE-0004um-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 07:41:24 -0400
Message-ID: <20030711114123.53673.qmail@web12809.mail.yahoo.com>
Received: from [208.255.120.2] by web12809.mail.yahoo.com via HTTP; Fri, 11 Jul 2003 04:41:23 PDT
Date: Fri, 11 Jul 2003 04:41:23 -0700 (PDT)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: RE: [Sigtran] M3UA IG v04
To: sigtran@ietf.org
In-Reply-To: <06CAA4D3753C53408D2CE6340B68444201488AA6@vbeexc02.emea.cpqcorp.net>
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>

Francois,

What I had in mind was a new ASPTM to inform about
congestion status of the peers, because I consider
this also part of the traffic management.

     Tolga
--- "Zannin, Francois" <francois.zannin@hp.com> wrote:
> Tolga,
> 
> IN the case of IPSP (but also SGP), if we send an
> ASPTM (INACTIVE I
> guess) would result in the impossibility to exchange
> messages between
> IPSP and then the congestion that can be resorbed by
> exchanges messages
> for already open transaction is eliminated. I don't
> think ASPTM can be
> the best solution to controle the congestion.
> 
> Francois.
> 
> > -----Original Message-----
> > From: Tolga Asveren
> [mailto:tolga_asveren@yahoo.com] 
> > Sent: Thursday, July 10, 2003 10:40 PM
> > To: sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA IG v04
> > 
> > 
> > Francois,
> > --- "Zannin, Francois" <francois.zannin@hp.com>
> wrote:
> > > OK, now, is the DUPU received on a SG or an IPSP
> > > considered as a protocol error?
> > > This has to be clearly stated in the RFC (or IG)
> > > Same remark for the SCON received from an IPSP,
> is
> > > it possible to receive a SCON
> > > or must the IPSP use the SCTP congestion to
> handle
> > > the congestion on a remote IPSP.
> > [TOLGA] Again I would think, SCON is not necessary
> for
> > IPSP communication. 
> > In fact I think, it could be nice to have an ASPTM
> to
> > inform about congestion between peers, because IMO
> it
> > is more elegant than waiting the congestion to
> > propogate to the remote SCTP -maybe using ECN
> might
> > help- but I am representing the minority on this
> issue
> > (a minority of 1 person :) ) 
> > > 
> > > Francois.
> > > 
> > > > -----Original Message-----
> > > > From: Tolga Asveren
> > > [mailto:tolga_asveren@yahoo.com]
> > > > Sent: Thursday, July 10, 2003 10:13 PM
> > > > To: sigtran@ietf.org
> > > > Subject: RE: [Sigtran] M3UA IG v04
> > > > 
> > > > 
> > > > Francois,
> > > > 
> > > > IMO, for the scenario you describe, sending
> ASPIA
> > > is
> > > > the correct behaviour. Yes, you won't be able
> to
> > > use
> > > > healthy user parts in such a case, but if you
> have
> > > > that level of granularity, I would think, RKs
> > > should
> > > > be defined accordingly(for example for your
> > > example
> > > > PC/SI RKs)
> > > > The same applies to ASP/SGP case. There is no
> DUPU
> > > > from ASP to SGP direction.
> > > > So, IMO, DUPU for IPSP case is not necessary.
> > > > 
> > > >      Tolga
> > > > --- "Zannin, Francois"
> <francois.zannin@hp.com>
> > > wrote:
> > > > > tolga,
> > > > > 
> > > > > see inline,
> > > > > 
> > > > > Regards,
> > > > > 
> > > > > Francois.
> > > > > 
> > > > > > -----Original Message-----
> > > > > > From: Tolga Asveren
> > > > > [mailto:tolga_asveren@yahoo.com]
> > > > > > Sent: Thursday, July 10, 2003 9:47 PM
> > > > > > To: sigtran@ietf.org
> > > > > > Subject: Re: [Sigtran] M3UA IG v04
> > > > > > 
> > > > > > 
> > > > > > Brian,
> > > > > > [..snip..]
> > > > > > > > [TOLGA] Yes, SGPs send DUPU to ASPs
> but to
> > > my
> > > > > > > > understanding, Francois's case would
> > > > > correspond
> > > > > > > > sending DUPU from ASP to SGP.
> > > > > > > > As I said, my interpretation of DE
> IPSP in
> > > the
> > > > > IG
> > > > > > > is a
> > > > > > > > combined SGP/ASP. On ASP part of IPSP,
> the
> > > > > user
> > > > > > > part
> > > > > > > > goes down, then I would expect
> > > inactivating
> > > > > the
> > > > > > > > corresponding AS.
> > > > > > > 
> > > > > > > Yes, that's what we understood and
> agreed,
> > > > > however,
> > > > > > > that is
> > > > > > > not what is presented in the IG: Read
> the IG
> > > > > again:
> > > > > > > New text in section 3.11.2:
> > > > > > > 
> > > > > > >    In section 4.3.1: ASP/IPSP States,
> only
> > > the
> > > > > > > SGP-ASP and the IPSP SE
> > > > > > >    scenarios are described. For the IPSP
> DE
> > > > > model,
> > > > > > > both IPSPs MUST
> > > > > > >    follow the SGP side of the SGP-ASP
> > > > > procedures.
> > > > > > > 
> > > > > > > That looks pretty plain to me. 
> According to
> > > the
> > > > > IG,
> > > > > > > DE IPSPs can
> > > > > > > send DUPU (they are actually required
> above
> > > to
> > > > > do so
> > > > > > > when the SGP
> > > > > > > procedure dictates).
> > > > > > [TOLGA]IMO, the sentence you refer might
> > > really be
> > > > > > mislaeading, but I don't think it should
> make
> > > the
> > > > > SGP
> > > > > > send DUPU for the case we are describing.
> > > Let's
> > > > > say we
> > > > > > have IPSP1 and IPSP2. If one user part is
> down
> > > on
> > > > > > IPSP2, then it should be ASP part of IPSP2
> > > taking
> > > > > some
> > > > > > action and sending ASPIA, not SGP part of
> > > IPSP2.
> > > > > 
> > > > > The case I was refering first, was a
> platform
> > > > > hosting only one AS but
> > > > > internally made of several services (SI),
> not
> > > always
> > > > > available at the same
> > > > > time.
> > > > > If one SI is not available but the others
> are
> > > still
> > > > > "usable", if the AS
> > > > > is "Inactivated", all SI will be prohibited!
> In
> > > that
> > > > > case the "IPSP" MAY
> > > > > (that's my question) send a DUPU for the
> > > concerned
> > > > > SI but the traffic
> > > > > for the others SI is still possible.
> > > > > If IPSP cannot send a DUPU (or an UPU
> encoded),
> > > this
> > > > > mean that an AS
> > > > > identified by a PC only CANNOT corretly play
> > > several
> > > > > SI.
> > > > > 
> > > > > So my question, is the DUPU allowed between
> two
> > > IPSP
> > > > > ? 
> 
=== message truncated ===


__________________________________
Do you Yahoo!?
The New Yahoo! Search - Faster. Easier. Bingo.
http://search.yahoo.com

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



From exim@www1.ietf.org  Fri Jul 11 08:31:16 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA25712
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 07:42:32 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19awHs-000286-BB
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 07:42:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6BBg4vn008174
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 07:42:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19awHp-00027Q-Uw; Fri, 11 Jul 2003 07:42:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19awHL-00024w-UR
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 07:41:31 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA25673
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 07:41:29 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19awHL-0004uv-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 07:41:31 -0400
Received: from web12806.mail.yahoo.com ([216.136.174.41])
	by ietf-mx with smtp (Exim 4.12)
	id 19awHK-0004us-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 07:41:30 -0400
Message-ID: <20030711114129.27199.qmail@web12806.mail.yahoo.com>
Received: from [208.255.120.2] by web12806.mail.yahoo.com via HTTP; Fri, 11 Jul 2003 04:41:29 PDT
Date: Fri, 11 Jul 2003 04:41:29 -0700 (PDT)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: RE: [Sigtran] M3UA IG v04
To: sigtran@ietf.org
In-Reply-To: <06CAA4D3753C53408D2CE6340B68444201488AA6@vbeexc02.emea.cpqcorp.net>
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>

Francois,

What I had in mind was a new ASPTM to inform about
congestion status of the peers, because I consider
this also part of the traffic management.

     Tolga
--- "Zannin, Francois" <francois.zannin@hp.com> wrote:
> Tolga,
> 
> IN the case of IPSP (but also SGP), if we send an
> ASPTM (INACTIVE I
> guess) would result in the impossibility to exchange
> messages between
> IPSP and then the congestion that can be resorbed by
> exchanges messages
> for already open transaction is eliminated. I don't
> think ASPTM can be
> the best solution to controle the congestion.
> 
> Francois.
> 
> > -----Original Message-----
> > From: Tolga Asveren
> [mailto:tolga_asveren@yahoo.com] 
> > Sent: Thursday, July 10, 2003 10:40 PM
> > To: sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA IG v04
> > 
> > 
> > Francois,
> > --- "Zannin, Francois" <francois.zannin@hp.com>
> wrote:
> > > OK, now, is the DUPU received on a SG or an IPSP
> > > considered as a protocol error?
> > > This has to be clearly stated in the RFC (or IG)
> > > Same remark for the SCON received from an IPSP,
> is
> > > it possible to receive a SCON
> > > or must the IPSP use the SCTP congestion to
> handle
> > > the congestion on a remote IPSP.
> > [TOLGA] Again I would think, SCON is not necessary
> for
> > IPSP communication. 
> > In fact I think, it could be nice to have an ASPTM
> to
> > inform about congestion between peers, because IMO
> it
> > is more elegant than waiting the congestion to
> > propogate to the remote SCTP -maybe using ECN
> might
> > help- but I am representing the minority on this
> issue
> > (a minority of 1 person :) ) 
> > > 
> > > Francois.
> > > 
> > > > -----Original Message-----
> > > > From: Tolga Asveren
> > > [mailto:tolga_asveren@yahoo.com]
> > > > Sent: Thursday, July 10, 2003 10:13 PM
> > > > To: sigtran@ietf.org
> > > > Subject: RE: [Sigtran] M3UA IG v04
> > > > 
> > > > 
> > > > Francois,
> > > > 
> > > > IMO, for the scenario you describe, sending
> ASPIA
> > > is
> > > > the correct behaviour. Yes, you won't be able
> to
> > > use
> > > > healthy user parts in such a case, but if you
> have
> > > > that level of granularity, I would think, RKs
> > > should
> > > > be defined accordingly(for example for your
> > > example
> > > > PC/SI RKs)
> > > > The same applies to ASP/SGP case. There is no
> DUPU
> > > > from ASP to SGP direction.
> > > > So, IMO, DUPU for IPSP case is not necessary.
> > > > 
> > > >      Tolga
> > > > --- "Zannin, Francois"
> <francois.zannin@hp.com>
> > > wrote:
> > > > > tolga,
> > > > > 
> > > > > see inline,
> > > > > 
> > > > > Regards,
> > > > > 
> > > > > Francois.
> > > > > 
> > > > > > -----Original Message-----
> > > > > > From: Tolga Asveren
> > > > > [mailto:tolga_asveren@yahoo.com]
> > > > > > Sent: Thursday, July 10, 2003 9:47 PM
> > > > > > To: sigtran@ietf.org
> > > > > > Subject: Re: [Sigtran] M3UA IG v04
> > > > > > 
> > > > > > 
> > > > > > Brian,
> > > > > > [..snip..]
> > > > > > > > [TOLGA] Yes, SGPs send DUPU to ASPs
> but to
> > > my
> > > > > > > > understanding, Francois's case would
> > > > > correspond
> > > > > > > > sending DUPU from ASP to SGP.
> > > > > > > > As I said, my interpretation of DE
> IPSP in
> > > the
> > > > > IG
> > > > > > > is a
> > > > > > > > combined SGP/ASP. On ASP part of IPSP,
> the
> > > > > user
> > > > > > > part
> > > > > > > > goes down, then I would expect
> > > inactivating
> > > > > the
> > > > > > > > corresponding AS.
> > > > > > > 
> > > > > > > Yes, that's what we understood and
> agreed,
> > > > > however,
> > > > > > > that is
> > > > > > > not what is presented in the IG: Read
> the IG
> > > > > again:
> > > > > > > New text in section 3.11.2:
> > > > > > > 
> > > > > > >    In section 4.3.1: ASP/IPSP States,
> only
> > > the
> > > > > > > SGP-ASP and the IPSP SE
> > > > > > >    scenarios are described. For the IPSP
> DE
> > > > > model,
> > > > > > > both IPSPs MUST
> > > > > > >    follow the SGP side of the SGP-ASP
> > > > > procedures.
> > > > > > > 
> > > > > > > That looks pretty plain to me. 
> According to
> > > the
> > > > > IG,
> > > > > > > DE IPSPs can
> > > > > > > send DUPU (they are actually required
> above
> > > to
> > > > > do so
> > > > > > > when the SGP
> > > > > > > procedure dictates).
> > > > > > [TOLGA]IMO, the sentence you refer might
> > > really be
> > > > > > mislaeading, but I don't think it should
> make
> > > the
> > > > > SGP
> > > > > > send DUPU for the case we are describing.
> > > Let's
> > > > > say we
> > > > > > have IPSP1 and IPSP2. If one user part is
> down
> > > on
> > > > > > IPSP2, then it should be ASP part of IPSP2
> > > taking
> > > > > some
> > > > > > action and sending ASPIA, not SGP part of
> > > IPSP2.
> > > > > 
> > > > > The case I was refering first, was a
> platform
> > > > > hosting only one AS but
> > > > > internally made of several services (SI),
> not
> > > always
> > > > > available at the same
> > > > > time.
> > > > > If one SI is not available but the others
> are
> > > still
> > > > > "usable", if the AS
> > > > > is "Inactivated", all SI will be prohibited!
> In
> > > that
> > > > > case the "IPSP" MAY
> > > > > (that's my question) send a DUPU for the
> > > concerned
> > > > > SI but the traffic
> > > > > for the others SI is still possible.
> > > > > If IPSP cannot send a DUPU (or an UPU
> encoded),
> > > this
> > > > > mean that an AS
> > > > > identified by a PC only CANNOT corretly play
> > > several
> > > > > SI.
> > > > > 
> > > > > So my question, is the DUPU allowed between
> two
> > > IPSP
> > > > > ? 
> 
=== message truncated ===


__________________________________
Do you Yahoo!?
SBC Yahoo! DSL - Now only $29.95 per month!
http://sbc.yahoo.com

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



From exim@www1.ietf.org  Fri Jul 11 10:41:34 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA01841
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 10:41:34 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19az59-0007eD-Pj
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 10:41:08 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6BEf7YC029397
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 10:41:07 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19az54-0007df-Mq; Fri, 11 Jul 2003 10:41:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19az4q-0007dF-TT
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 10:40:49 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA01822
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 10:40:44 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19az4o-00064U-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 10:40:46 -0400
Received: from web41811.mail.yahoo.com ([66.218.93.145])
	by ietf-mx with smtp (Exim 4.12)
	id 19az4n-00064P-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 10:40:45 -0400
Message-ID: <20030711144012.84572.qmail@web41811.mail.yahoo.com>
Received: from [65.198.40.51] by web41811.mail.yahoo.com via HTTP; Fri, 11 Jul 2003 09:40:12 CDT
Date: Fri, 11 Jul 2003 09:40:12 -0500 (CDT)
From: =?iso-8859-1?q?Pablo=20Salinas?= <pablopesb@yahoo.com>
To: bidulock@openss7.org
Cc: sigtran@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="0-1704390148-1057934412=:83012"
Subject: [Sigtran] Link Key Parameter
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

--0-1704390148-1057934412=:83012
Content-Type: text/plain; charset=iso-8859-1
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id KAA01823
Content-Transfer-Encoding: quoted-printable

Hi Brian,
            Could you tell me what are the Signaling Data Terminal Identi=
fier and the Signaling Data Link Identifier parameters?=20
=20
I've read the RFC's definition, but I can't understand what is their purp=
ose? Why use an SDT and SDL identifiers when the RS and the MGC already u=
se an Interface Identifier?
=20
Or, is it used, because in dynamic registration the only thing the MGC ne=
eds to know about the RS signaling links are the SDT and SDL; while, it i=
s the RS's job to assign a proper Interface Identifier.
=20
Although, if that is the case, the RS and the MGC could agree to use a pr=
e-stablished set of Interface Identifiers. And, not use a pre-stablished =
set of SDT and SDL identifiers.
=20
Another thing, why use two Identifiers (SDT and SDL)?, when they could ag=
ree to use just one of them (SDT or SDL)
=20
                            greetings,
                                            Pablo Salinas



---------------------------------
Do You Yahoo!?
Todo lo que quieres saber de Estados Unidos, Am=E9rica Latina y el resto =
del Mundo.
Vis=EDta Yahoo! Noticias.

--0-1704390148-1057934412=:83012
Content-Type: text/html; charset=iso-8859-1
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id KAA01823
Content-Transfer-Encoding: quoted-printable

<DIV>Hi Brian,</DIV>
<DIV>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; C=
ould you tell me what are the Signaling Data Terminal Identifier and the =
Signaling Data Link Identifier parameters? </DIV>
<DIV>&nbsp;</DIV>
<DIV>I've read the RFC's definition, but I can't understand what is their=
 purpose? Why use an SDT and SDL identifiers when the RS and the MGC alre=
ady use an Interface Identifier?</DIV>
<DIV>&nbsp;</DIV>
<DIV>Or, is it used, because in dynamic registration the only thing the M=
GC needs to know about the RS signaling links are the SDT and SDL; while,=
 it is the RS's job to assign a proper Interface Identifier.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Although, if that is the case, the RS and the MGC could agree to use=
 a pre-stablished set of Interface Identifiers. And, not use a pre-stabli=
shed set of SDT and SDL identifiers.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Another thing, why use two Identifiers (SDT and SDL)?, when they cou=
ld agree to use just one of them (SDT or SDL)</DIV>
<DIV>&nbsp;</DIV>
<DIV>&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; greetings,</DIV>
<DIV>&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; Pablo Salinas</DIV><p><br><h=
r size=3D1 noshade><b>Do You Yahoo!?</b><br>
<a href=3Dhttp://espanol.yahoo.com/mail_tagline/*http://espanol.news.yaho=
o.com><img src=3D"http://us.i1.yimg.com/us.yimg.com/i/not/dailynews.gif" =
width=3D30 height=3D26 border=3D0 align=3Dleft></a>
Todo lo que quieres saber de Estados Unidos, Am=E9rica Latina y el resto =
del Mundo.<br>
Vis=EDta <a href=3Dhttp://espanol.yahoo.com/mail_tagline/*http://espanol.=
news.yahoo.com>Yahoo! Noticias</a>.<br>
--0-1704390148-1057934412=:83012--

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



From exim@www1.ietf.org  Fri Jul 11 12:30:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA05041
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 12:30:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19b0ma-00081v-5o
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 12:30:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6BGU4hm030861
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 12:30:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19b0mX-00081W-BI; Fri, 11 Jul 2003 12:30:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19b0mR-00080m-7j
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 12:29:55 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA05023
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 12:29:50 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19b0mP-0006v3-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 12:29:53 -0400
Received: from angelica.ulticom.com ([208.255.120.2] helo=chuckie.dgms.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19b0mN-0006us-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 12:29:51 -0400
Received: from navy.fr.ulticom.com (localhost [127.0.0.1])
	by chuckie.dgms.com (8.9.3/8.9.3) with ESMTP id MAA01911;
	Fri, 11 Jul 2003 12:29:13 -0400 (EDT)
Received: from PCGASTAUDL (pc-gastaud-l.fr.ulticom.com [172.26.32.88])
	by navy.fr.ulticom.com (8.11.6+Sun/8.11.6) with SMTP id h6BGPgE28993;
	Fri, 11 Jul 2003 18:25:43 +0200 (MEST)
From: "Valerie Gastaud" <valerie.gastaud@ulticom.com>
To: <bidulock@openss7.org>
Cc: "Tolga Asveren" <tolga_asveren@yahoo.com>, <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA - RC in SSNM messages
Date: Fri, 11 Jul 2003 18:29:40 +0200
Message-ID: <NDBBIHAOCKGAMIKNLALOIEDNDOAA.valerie.gastaud@ulticom.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.6604 (9.0.2911.0)
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106
Importance: Normal
In-Reply-To: <20030711032506.J1680@openss7.org>
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,

Yes, this makes it consistent. There is no more difference between ASPTM and
SSNM/DATA messages.

Valerie
-----Original Message-----
From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf Of
Brian F. G. Bidulock
Sent: vendredi 11 juillet 2003 11:25
To: Valerie Gastaud
Cc: Tolga Asveren; sigtran@ietf.org
Subject: Re: [Sigtran] M3UA - RC in SSNM messages


Valerie,

I would have no problems with that.  Would that firm things up for you?

Does anyone rely on the (s)?

I think that it would be quite easy to redefine 3 configured AS as
a single compound AS at the SG anyway, and still not have to send RC.

--brian

Valerie Gastaud wrote:
(Fri, 11 Jul 2003 10:27:26)
> Brian, Tolga,
>
> so, the conclusion is that when there are multiple AS on the same
> association, all the messages (SSNM, DATA and ASPTM) MUST contain the RC
> parameter.
>
> if this is agreed by everyone, can we remove the (s) of Application Server
> in the following sentence from section 4.3.4.3 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.
>
> and change the RC from optional to conditional in the ASPTM messages as
well
>
> and change the SHOULD in section 4.3.4.3 by a MUST:
>                     ...  In the case where an ASP wishes to process the
> 	traffic for more than one Application Server across a common SCTP
> 	association, the ASP Active message(s) SHOULD contain a list of one
> 	or more Routing Contexts to indicate for which Application Servers
> 	the ASP Active message applies. ...
>
> Valerie
> -----Original Message-----
> From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf Of
> Brian F. G. Bidulock
> Sent: jeudi 10 juillet 2003 22:27
> To: Tolga Asveren
> Cc: sigtran@ietf.org
> Subject: Re: [Sigtran] M3UA - RC in SSNM messages
>
>
> Tolga,
>
>
> Tolga Asveren wrote:
> (Thu, 10 Jul 2003 12:48:14)
> > Brian,
> > --- "Brian F. G. Bidulock" <bidulock@openss7.org>
> > wrote:
> > > Tolga,
> > >
> > > Tolga Asveren wrote:
> > >                                 (Thu, 10 Jul 2003
> > > 11:10:20)
> > > > Brian,
> > > >
> > > > I was referring to 4.3.4.3 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."
> > > > This enables the option to send ASPAC with no RC,
> > > even
> > > > if multiple AS traffic is carried over one
> > > > association. I don't think, this is permitted for
> > > > SSNM.
> > >
> > > It permits a single AS in fitting with the SHOULD I
> > > quoted.  Multiple
> > > configured AS for the same association would be
> > > against the SHOULD.
> > > So you are still on your own.
> > [TOLGA]Then why does it say "Application Server(s)"
> >                                                 ^
> > I am pretty sure, what is meant here (In fact even
> > just two days ago, there was a qustion related with
> > this)
>
> If you recall, the (s) predates the RC in SSNM and DATA
> discussion.  The SHOULD was put in after, at the same
> time as the RC passages in SSNM and DATA.  If you have
> multiple AS on the same association you need to put RC
> in all messages.  This was one reason we changed "Optional"
> to "Conditional' for RC throughout the document.  The
> same condition applies in all cases.
>
> --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 exim@www1.ietf.org  Fri Jul 11 15:36:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA13616
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 15:36:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19b3gZ-0004tF-Us
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 15:36:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6BJa3w1018796
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 15:36:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19b3gY-0004sl-8l; Fri, 11 Jul 2003 15:36:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19b3fx-0004rF-O8
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 15:35:25 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA13540
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 15:35:22 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19b3fu-0001IR-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 15:35:22 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[Axq9wuKwMhfEROv2J/FKt8JArcTePB2M])
	by ietf-mx with esmtp (Exim 4.12)
	id 19b3ft-0001IE-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 15:35:22 -0400
Received: from ns.pigworks.openss7.net (IDENT:aHrlxCkGSKWRkPAmtIvgMJ+vqvwYjWYK@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6BJZFX22002;
	Fri, 11 Jul 2003 13:35:15 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6BJZFT13026;
	Fri, 11 Jul 2003 13:35:15 -0600
Date: Fri, 11 Jul 2003 13:35:15 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Pablo Salinas <pablopesb@yahoo.com>
Cc: sigtran@ietf.org
Message-ID: <20030711133515.A12750@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Pablo Salinas <pablopesb@yahoo.com>, sigtran@ietf.org
References: <20030711144012.84572.qmail@web41811.mail.yahoo.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20030711144012.84572.qmail@web41811.mail.yahoo.com>; from pablopesb@yahoo.com on Fri, Jul 11, 2003 at 09:40:12AM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6BJZFX22002
Content-Transfer-Encoding: quoted-printable
Subject: [Sigtran] Re: Link Key Parameter
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Pablo,

One identifies the data link (facility) and the other the data terminal
(electronics).  Link keys are only necessary for automatic allocation
of signalling data links and signalling terminals.  There are two because
in the general case they can be mixed and matched.  That is, one data lin=
k
can be switched onto a different data terminal.  SDLI is an SS7 protocol
value (used in DLC message), normally a CIC.   SDTI just indexes the data
terminal to associate with the data link.

--brian

Pablo Salinas wrote:                                                     =
            (Fri, 11 Jul 2003 09:40:12)
>=20
>    Hi Brian,
>=20
>                 Could  you  tell  me what are the Signaling Data Termin=
al
>    Identifier and the Signaling Data Link Identifier parameters?
>=20
>=20
>=20
>    I've  read  the RFC's definition, but I can't understand what is the=
ir
>    purpose?  Why  use  an SDT and SDL identifiers when the RS and the M=
GC
>    already use an Interface Identifier?
>=20
>=20
>=20
>    Or, is it used, because in dynamic registration the only thing the M=
GC
>    needs to know about the RS signaling links are the SDT and SDL; whil=
e,
>    it is the RS's job to assign a proper Interface Identifier.
>=20
>=20
>=20
>    Although, if that is the case, the RS and the MGC could agree to use=
 a
>    pre-stablished   set   of   Interface  Identifiers.  And,  not  use =
 a
>    pre-stablished set of SDT and SDL identifiers.
>=20
>=20
>=20
>    Another thing, why use two Identifiers (SDT and SDL)?, when they cou=
ld
>    agree to use just one of them (SDT or SDL)
>=20
>=20
>=20
>                                greetings,
>=20
>                                                Pablo Salinas
>      _________________________________________________________________
>=20
>    Do You Yahoo!?
>    [1][dailynews.gif]  Todo  lo  que  quieres  saber  de  Estados Unido=
s,
>    Am=E9rica Latina y el resto del Mundo.
>    Vis=EDta [2]Yahoo! Noticias.
>=20
> References
>=20
>    1. http://espanol.yahoo.com/mail_tagline/*http://espanol.news.yahoo.=
com
>    2. http://espanol.yahoo.com/mail_tagline/*http://espanol.news.yahoo.=
com

--=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



From exim@www1.ietf.org  Fri Jul 11 17:03:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA16922
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 17:03:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19b52l-0003jo-I8
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 17:03:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6BL33og014368
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 17:03:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19b52j-0003jH-0N; Fri, 11 Jul 2003 17:03:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19aumA-0002KJ-CJ
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 06:05:14 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA23464
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 06:05:09 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aum6-0004I0-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 06:05:10 -0400
Received: from zdemail03.zdem.compaq.com ([161.114.112.27])
	by ietf-mx with esmtp (Exim 4.12)
	id 19aum5-0004Hp-00
	for sigtran@ietf.org; Fri, 11 Jul 2003 06:05:09 -0400
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail03.zdem.compaq.com (Postfix) with ESMTP
	id 29F1610E1; Fri, 11 Jul 2003 12:04:35 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Fri, 11 Jul 2003 12:04:35 +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"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Date: Fri, 11 Jul 2003 12:04:34 +0200
Message-ID: <BE462396B4A022449E0EB2DD4753D054342ECA@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Thread-Index: AcNHFtVTJIaZQKR0TPO8xHlLkq/LKgAdr1GQ
From: "Anzile, Christophe" <christophe.anzile@hp.com>
To: <bidulock@openss7.org>
Cc: <john.loughney@nokia.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 11 Jul 2003 10:04:35.0026 (UTC) FILETIME=[D4AB3320:01C34793]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Brian,

You're right. AspId is no use in SGP/ASP exchange.

I'm just getting confused, because SUA draft says in terminology =
chapter:

   IP Server Process (IPSP) - A process instance of an IP-based=20
   application.  An IPSP is essentially the same as an ASP, except that=20
   it uses SUA in a peer-to-peer fashion.

It never says that IPSP could play the role of an SGP while being an =
ASP..
May be it would be nice to add such a note about this..


Christophe.

-----Original Message-----
From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
Sent: Thursday, July 10, 2003 9:09 PM
To: Anzile, Christophe
Cc: john.loughney@nokia.com; sigtran@ietf.org
Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt


Christophe,

IPSP SE is an ASP/SGP relationship and only the ASP side need send
ASP Id.

IPSP DE both send ASPAC and the ASP Id is not needed in ASPUP Ack.

So how would the IPSP use this?

You say to include it in some messages below, but not what the
receiver of the ASPUP Ack is expected (or able) to do with the value.

What would the SE IPSP receiving the ASPUp Ack (remember, acting in ASP
mode) do with it?  There are no procedures defined for an ASP to use
an "SGP Id" if you will...

So, I'm an SE IPSP acting in ASP mode.  I send you an ASPUP and you
send me an ASPUP Ack with an ASP Id in it.  What am I to do with it?
I never send Notify (you do), so what do I do with it?  What use is it?

--brian

Anzile, Christophe wrote:                                                =
          (Thu, 10 Jul 2003 17:35:08)
> My proposal would be :
>=20
> The optional ASP Identifier parameter is specifically useful for IPSP=20
> communication. In that case the IPSP answering the ASP Up message MAY=20
> include its own ASP Identifier value, so that it can give back its ASP
> Identifier to the demanding IPSP in case of a single UP/UP-Ack =
exchange.
> For standard AS-SG communication, this parameter MUST NOT be used.=20
>=20
>=20
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Thursday, July 10, 2003 3:45 PM
> To: john.loughney@nokia.com
> Cc: Anzile, Christophe; sigtran@ietf.org
> Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
>=20
>=20
> John,
>=20
> We discussed this before an nobody had a real problem with it.
>=20
> However, more than just adding the optional parameter to the
> ASP Up Ack, it also needs to be described how one should use it.
> Althoug a lot of uses were discussed, nobody put together a
> coherent proposal for its use.
>=20
> Does someone have a simple proposal for it use?
>=20
> --brian
>=20
> John Loughney wrote:                                                   =
               (Thu, 10 Jul 2003 16:33:51)
> > Hi,
> >=20
> > That could be reasonable - anyone have problems with this?
> >=20
> > John
> >=20
> > > -----Original Message-----
> > > From: ext Anzile, Christophe [mailto:christophe.anzile@hp.com]
> > > Sent: 10 July, 2003 12:26
> > > To: Loughney John (NRC/Helsinki)
> > > Cc: sigtran@ietf.org
> > > Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
> > >=20
> > >=20
> > >=20
> > > Hi John,
> > >=20
> > > section 3.5.2 :=20
> > > it would be judicious to add an optional ASP-ID parameter in=20
> > > the UP-ACK message,
> > > especially for IPSP communication (think of single exchange).
> > >=20
> > > Regards,
> > > Christophe.
> > >=20
> >=20
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
>=20
> --=20
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/

--=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



From exim@www1.ietf.org  Fri Jul 11 17:37:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA18366
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 17:37:32 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19b5Zi-0005dT-Kb
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 17:37:06 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6BLb6Es021654
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 17:37:06 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19b5Zd-0005bR-G0; Fri, 11 Jul 2003 17:37:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19b5Yw-0005TT-PT
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 17:36:18 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA18307
	for <SIGTRAN@IETF.org>; Fri, 11 Jul 2003 17:36:13 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19b5Yu-0002p1-00
	for SIGTRAN@IETF.org; Fri, 11 Jul 2003 17:36:16 -0400
Received: from auds953.usa.alcatel.com ([143.209.238.6])
	by ietf-mx with esmtp (Exim 4.12)
	id 19b5Yt-0002oU-00
	for SIGTRAN@IETF.org; Fri, 11 Jul 2003 17:36:15 -0400
Received: from alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6BLZULX010312;
	Fri, 11 Jul 2003 16:35:30 -0500 (CDT)
Message-ID: <3F0F2DA2.10204@alcatel.com>
Date: Fri, 11 Jul 2003 16:35:30 -0500
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
User-Agent: Mozilla/5.0 (X11; U; SunOS sun4u; en-US; rv:1.0.1) Gecko/20020920 Netscape/7.0
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: SIGTRAN e-mail list <SIGTRAN@ietf.org>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] M2PA: list of comments on M2PAv8 and their resolutions in v9
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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

List of comments on M2PA draft version 8.
Tom George
07/11/2003

Here is my list of comments on M2PA draft version 8 and how they were
incorporated into version 9. For the most part, they are arranged by
the thread from the SIGTRAN list. The date/name/subject identify the
beginning of the thread.

01. 04/24/03 Tolga Asveren "[Sigtran] M2PA: T7 and dropping SCTP
     association after link failure".

     (This was a comment about version 7 that came in after I released
     version 8.)

     Should T7 be required to be per-message? Should M2PA terminate the
     association if the link fails?

     A: These issues have been discussed before. No change is needed.

     Changed wording to not explicitly require many timers.

02. 04/24/03 Jeff Craig "[Sigtran] (M2PA) Comments on M2PA v8".

     Several comments.

     A: Here are the comments for which I made changes.

        T7 and PO: Because LS PO is sent on a different stream, it can
        arrive ahead of a message sent before the LS PO. The sender
        would keep T7 running, but the receiver would buffer and not
        ack the message. This would cause the link to fail due to T7
        expiry.

           To fix this, Delete (c) from 4.1.4 and follow MTP2's
	  behavior. In other words, stop T7 during LPO.

        Typo: In 4.1.5, the typo is corrected - LS PO is changed to LS
        Busy.

        FSN, BSN, PO: 4.2.1 says to ignore the FSN and BSN in non-User
        Data messages. But Q.703 uses the BSN to synchronize sequence
        numbers when exiting PO.

           I deleted the statement about ignoring FSN/BSN in LS
	  messages. I added that LS PR can be used to resynchronize
	  sequence numbers after PO. I added that for Flush, queued
	  incoming messages must be discarded as well as messages in
	  the TB/RTB. Added that Continue means to begin processing
	  the incoming queued messages.

           I did not follow a suggestion to put the LS and User Data
	  messages on the same SCTP stream. This would make the
	  synchronization easier and prevent discarding of messages
	  sent after RPO ends, but it takes away some flexibility in
	  allowing LS messages to go around queued User Data messages.

03. 04/28/03 Rohan "[Sigtran] Proving Message".

     Why is there a proving state and proving messages?

     A: For the same reason as in Q.703 and Q.2140: To verify that the
     link is usable before putting real traffic on it. See 4.1, 1st
     paragraph. No change needed.

04. 04/28/03 Pardeep "[Sigtran] Link States during Intial Alignment in
     M2PA".

     Are all of the MTP2 states during alignment needed in M2PA?

     A: Yes.

05. 04/28/03 Pardeep "[Sigtran] m2pa version control (sec 4.1.9)".

     Does "unsupported version" mean "incompatible old version"?

     A: Yes.

06. 04/28/03 Pardeep "[Sigtran] Need for One More message Type in
     M2PA?".

     Is there a need for an "error" message for incompatible M2PA
     versions?

     A: No. Severe errors like these require manual intervention.

07. 05/02/03 Pardeep "[Sigtran] FSN value".

     A: M2PA should not ignore the FSN in an empty user data message.

        The first FSN should be set to 0, and the first BSN to -1, as
        in Q.703.

        No change to the document is necessary.

08. 05/06/03 HannsJuergen Schwarzbauer "[Sigtran] (M2PA) Comments on
     M2PA v8 - Last Call".

     Several comments.

     A: Here is what I did for the comments.

        Definition of SG: OK as is.

           Changed definition of STP to refer to MTP. Added SP. Added
	  reference for SEP definition.

        RFC 2719: Reference to RFC 2719 is OK.

        XCO/XCA: There is sufficient warning about this.

        FSNC or FSNT: It's FSNC as in Q.703.

        Empty User Data message: OK as is.

        Message Data: Added text to 2.3 to clarify what is in this
        field.

        Filler: OK as is.

        Registered port number: OK as is.

        SLC in 4.1.2: Changed sentence in 4.1.2 to read: "Therefore, it
        is possible that the ***SLC*** values a and b MAY be equal."

09. 05/06/03 Andrew Booth "[Sigtran] M2PA v8".

     Should we include a comment on how to handle T7 after a LS Busy or
     LS Busy Ended is received?

     A: Do what MTP2 does. No change is needed to the draft.

10. 05/07/03 HannsJuergen Schwarzbauer "[Sigtran] (M2PA) Comments on
     M2PA v8 - Last Call".

     Proving_Interval has no value assigned to it.

     A: Proving may be omitted if allowed by the MTP2
     standard. Recommend that Proving_Interval be set to imitate
     expected rate of traffic.

11. 05/07/03 HannsJuergen Schwarzbauer "[Sigtran] (M2PA) Comments on
     M2PA v8 - Last Call".

     Spare bits in the LI field.

     A: Only the 2 priority bits are placed in this field. So there is
     no issue about the number of spare bits.

12. 05/07/03 HannsJuergen Schwarzbauer "[Sigtran] (M2PA) Comments on
     M2PA v8 - Last Call".

     LS messages do not go in retransmission buffer. So why say that
     they are not retrieved?

     A: Because untransmitted messages are sometimes retrieved.

        Clarified in 4.2.1 that LS messages are not placed in
        retransmission buffer. Clarified in 4.2.3 that empty User Data
        messages are not retrieved.

13. 05/21/03 David Lehmann "[Sigtran] M2PA I-D version 8, section
     4.2.1".

        Similar to Jeff Craig comment - item 02, "FSN, BSN, PO".

        Inconsistency in 4.2.1. FSN is set in non-User Data messages,
        but then it says to ignore it.

     A: I deleted the sentence saying to ignore the FSN/BSN in LS
     messages. See item 2 above.


=====================================================================

-- 
* 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.  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 exim@www1.ietf.org  Fri Jul 11 18:23:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA21226
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 18:23:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19b6IA-000093-WA
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 18:23:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6BMN2Yk000536
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 18:23:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19b6I9-00008P-GY; Fri, 11 Jul 2003 18:23:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19b6I4-00008B-21
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 18:22:56 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA21183
	for <SIGTRAN@ietf.org>; Fri, 11 Jul 2003 18:22:50 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19b6I1-0003Tt-00
	for SIGTRAN@ietf.org; Fri, 11 Jul 2003 18:22:53 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19b6Hz-0003TY-00
	for SIGTRAN@ietf.org; Fri, 11 Jul 2003 18:22:52 -0400
Received: from [172.16.1.203] (helo=CCPU00186)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19b6Hm-0004Fk-5R
	for <SIGTRAN@ietf.org>; Fri, 11 Jul 2003 15:22:38 -0700
From: "VIvek Toky" <vivektoky@ccpu.com>
To: "SIGTRAN e-mail list" <SIGTRAN@ietf.org>
Date: Fri, 11 Jul 2003 15:27:14 -0700
Message-ID: <HKEBKFBNPIDALDOKLGKDMEGLCDAA.vivektoky@ccpu.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.6604 (9.0.2911.0)
In-Reply-To: <3F0F2DA2.10204@alcatel.com>
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] M3UA IG V04: Receiving REG for a RK already registered
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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 have one question related to Registration Request handling in M3UA IG
V4. I am pasting few lines from Section 3.6

"If the SGP determines that the received RK was already registered, the SGP
returns a registration Response message to the ASP,containing a Registration
Result "Error - Cannot Support Unique Routing". This error applies whether
the Routing Key is Active or Inactive."

Is there any reason why SGP should send this error back. Why should it not
send the same RC back again ?

thanks,
Vivek.


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



From exim@www1.ietf.org  Fri Jul 11 18:48:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA22291
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 18:48:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19b6gP-0002AZ-Vj
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 18:48:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6BMm5gH008332
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 18:48:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19b6gK-0001zR-WF; Fri, 11 Jul 2003 18:48:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19b6gE-0001zA-Jk
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 18:47:54 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA22229
	for <SIGTRAN@IETF.org>; Fri, 11 Jul 2003 18:47:48 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19b6gB-0003rr-00
	for SIGTRAN@IETF.org; Fri, 11 Jul 2003 18:47:51 -0400
Received: from auds953.usa.alcatel.com ([143.209.238.6])
	by ietf-mx with esmtp (Exim 4.12)
	id 19b6gA-0003rO-00
	for SIGTRAN@IETF.org; Fri, 11 Jul 2003 18:47:50 -0400
Received: from alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6BMlILX016684;
	Fri, 11 Jul 2003 17:47:19 -0500 (CDT)
Message-ID: <3F0F3E76.2060104@alcatel.com>
Date: Fri, 11 Jul 2003 17:47:18 -0500
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
User-Agent: Mozilla/5.0 (X11; U; SunOS sun4u; en-US; rv:1.0.1) Gecko/20020920 Netscape/7.0
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: SIGTRAN e-mail list <SIGTRAN@ietf.org>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] M2PA: List of changes from M2PAv8 to M2PAv9
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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

List of changes in M2PA draft version 9.
Tom George
07/11/2003

This is a more detailed list of changes. It is arranged according to
the order of the document.

This list does not include minor changes to wording, formatting, etc.

General changes:

Changes were made to resolve internal inconsistencies and to make
operation more like MTP2.


Specific changes:

1.2 Terminology - Changed some definitions (STP, SP) to be more
consistent with SS7 standards.

2.3 M2PA Messages - Added paragraph to emphasize contents of "Message
Data".

4.1.3  Link Alignment - Added paragraph "The proving period MAY be
omitted..."

4.1.3  Link Alignment, para "If proving is performed..." - Added
sentence recommending setting for Proving_Interval.

4.1.4 Processor Outage - Added paragraph "When the local processor
outage condition ends..." Mentions that BSN is used to resynchronize
sequence numbers.

4.1.4 Processor Outage, list after "While in a local processor outage
condition:" - Removed (c), which said to leave T7 running for messages
before the PO. Now T7 should be stopped for those messages as well.

4.1.4 Processor Outage - Added 3 paragraphs "If M2PA receives a
Flush..." for Flush and Continue.

4.1.5 Level 2 Flow Control, 1st paragraph - corrected typo by changing
PO to Busy.

4.2.1 Sending and receiving messages - Removed "M2PA SHOULD ignore the
FSN and BSN contained in a Link Status message."

4.2.1 Sending and receiving messages, paragraph "Note that there is no
reason..." - Added LS messages to this statement.

4.2.1 Sending and receiving messages, last paragraph - Changed wording
about T7 so that it does not explicitly require a timer for each
message.

4.2.3 Changeover, paragraph "Note that only..." - Added that empty
User Data messages are not retrieved.

Figures 11 and 12 - Changed T3 and T4 timers and LS Proving messages
to more closely resemble actual procedure.

10 Author's Addresses - Updated author info for Tom George, Ram Dantu.


-- 
* 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.  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 exim@www1.ietf.org  Fri Jul 11 20:55:32 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA28373
	for <sigtran-archive@odin.ietf.org>; Fri, 11 Jul 2003 20:55:32 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19b8fJ-0001Ku-3k
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 20:55:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6C0t5Q6005131
	for sigtran-archive@odin.ietf.org; Fri, 11 Jul 2003 20:55:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19b8fF-0001KO-M2; Fri, 11 Jul 2003 20:55:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19b8eK-0001Jz-TK
	for sigtran@optimus.ietf.org; Fri, 11 Jul 2003 20:54:05 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA28306
	for <SIGTRAN@ietf.org>; Fri, 11 Jul 2003 20:54:01 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19b8eI-0005nV-00
	for SIGTRAN@ietf.org; Fri, 11 Jul 2003 20:54:02 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[ofsS6p5Gp2YuMVlgbhAfHtx07ScNBwkN])
	by ietf-mx with esmtp (Exim 4.12)
	id 19b8eG-0005nO-00
	for SIGTRAN@ietf.org; Fri, 11 Jul 2003 20:54:01 -0400
Received: from ns.pigworks.openss7.net (IDENT:OqCxG8mIMKNxpbI6rmL2cxIzOYF8/mgU@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6C0rsX26653;
	Fri, 11 Jul 2003 18:53:54 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6C0rre17212;
	Fri, 11 Jul 2003 18:53:53 -0600
Date: Fri, 11 Jul 2003 18:53:53 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: VIvek Toky <vivektoky@ccpu.com>
Cc: SIGTRAN e-mail list <SIGTRAN@ietf.org>
Subject: Re: [Sigtran] M3UA IG V04: Receiving REG for a RK already registered
Message-ID: <20030711185353.A17103@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: VIvek Toky <vivektoky@ccpu.com>,
	SIGTRAN e-mail list <SIGTRAN@ietf.org>
References: <3F0F2DA2.10204@alcatel.com> <HKEBKFBNPIDALDOKLGKDMEGLCDAA.vivektoky@ccpu.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <HKEBKFBNPIDALDOKLGKDMEGLCDAA.vivektoky@ccpu.com>; from vivektoky@ccpu.com on Fri, Jul 11, 2003 at 03:27:14PM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6C0rsX26653
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

VIvek,

Because it is an error.  What would be acheived by sending back the alrea=
dy
registered RC instead?

--brian

On Fri, 11 Jul 2003, VIvek Toky wrote:

>   I have one question related to Registration Request handling in M3UA =
IG
> V4. I am pasting few lines from Section 3.6
>=20
> "If the SGP determines that the received RK was already registered, the=
 SGP
> returns a registration Response message to the ASP,containing a Registr=
ation
> Result "Error - Cannot Support Unique Routing". This error applies whet=
her
> the Routing Key is Active or Inactive."
>=20
> Is there any reason why SGP should send this error back. Why should it =
not
> send the same RC back again ?
>=20
> thanks,
> Vivek.
>=20
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Sat Jul 12 00:42:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA07875
	for <sigtran-archive@odin.ietf.org>; Sat, 12 Jul 2003 00:42:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bCCx-0006I0-81
	for sigtran-archive@odin.ietf.org; Sat, 12 Jul 2003 00:42:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6C4g3Ti024170
	for sigtran-archive@odin.ietf.org; Sat, 12 Jul 2003 00:42:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bCCv-0006Hb-HQ; Sat, 12 Jul 2003 00:42:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bCCc-0006HK-Ps
	for sigtran@optimus.ietf.org; Sat, 12 Jul 2003 00:41:42 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA07843
	for <sigtran@ietf.org>; Sat, 12 Jul 2003 00:41:37 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19bCCZ-0000V3-00
	for sigtran@ietf.org; Sat, 12 Jul 2003 00:41:39 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[WAyyo7TVxM08EjLTj+pbpTpH3Q5FTsQp])
	by ietf-mx with esmtp (Exim 4.12)
	id 19bCCY-0000Up-00
	for sigtran@ietf.org; Sat, 12 Jul 2003 00:41:38 -0400
Received: from ns.pigworks.openss7.net (IDENT:2BvaQGXOS76NcIqvzJ+4A4w2nzy1YqZz@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6C4fVX30081;
	Fri, 11 Jul 2003 22:41:31 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6C4fUt20406;
	Fri, 11 Jul 2003 22:41:30 -0600
Date: Fri, 11 Jul 2003 22:41:30 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Ong, Lyndon" <LyOng@Ciena.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Message-ID: <20030711224130.A20278@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Ong, Lyndon" <LyOng@Ciena.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.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: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com>; from LyOng@Ciena.com on Wed, Jul 02, 2003 at 10:21:28AM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Subject: [Sigtran] M2PA Last Call Comment - T1 Editorial
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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 have been updating the M2PA-TEST spec and have noticed a few things in
M2PA-09...

M2PA-09 Says: (top of Page 27):

  The Link Status Ready message replaces the FISU of MTP2 that is sent
  at the end of the proving period. The Link Status Ready message is
  used to verify that both ends have completed proving. When M2PA starts
  timer T1, it SHALL send a Link Status Ready message to its peer in the
  case where MTP2 would send a FISU after proving is complete. M2PA MAY
  send additional Link Status Ready messages while timer T1 is running.

This is all that is mention on T1.

In MTP2 T1 is started when the state machine enters the "Aligned Ready"
or "Aligned Not Ready" state.  Only from the "Aligned Ready" state does
MTP2 send FISU.  From "Aligned Not Ready" state, it sends SIPO.  Also,
if data is pending transmission, MTP2 can enter the "Aligned Ready" state
by sending an MSU rather than a FISU.

The problem is that M2PA-09 says "When M2PA starts timer T1, it SHALL send
a Link Status message...", when in fact if M2PA follows Q.703 could start
T1 and send a DATA message, or could start T1 and send a PROCESSOR-OUTAGE
message.

Perhaps we could say:

  The Link Status Ready message replaces the FISU, the Data message replaces
  the MSU, and the Link Status Processor Outage message replaces the SIPO,
  that would be sent at the end of the proving period.  The Link Status Ready
  Data or Link Status Processor Outage message is used to verify that both
  ends have completed proving.  Whem M2PA starts timer T1, it SHALL send a
  Link Status Ready, Data or Link Status Processor Outage message to is peer
  in the case where MTP2 would send a FISU, MSU or SIPO ater proving is
  complete.  M2PA MAY send additional Link Status Ready, Data or Link Status
  Processor Outage messages while timer T1 is running.

Ok?

--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 exim@www1.ietf.org  Sat Jul 12 01:04:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA08985
	for <sigtran-archive@odin.ietf.org>; Sat, 12 Jul 2003 01:04:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bCYE-0007DY-3D
	for sigtran-archive@odin.ietf.org; Sat, 12 Jul 2003 01:04:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6C542sE027723
	for sigtran-archive@odin.ietf.org; Sat, 12 Jul 2003 01:04:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bCYD-0007D2-6Q; Sat, 12 Jul 2003 01:04:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bCXt-000776-8D
	for sigtran@optimus.ietf.org; Sat, 12 Jul 2003 01:03:41 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA08941
	for <sigtran@ietf.org>; Sat, 12 Jul 2003 01:03:38 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19bCXq-0000oH-00
	for sigtran@ietf.org; Sat, 12 Jul 2003 01:03:38 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[wXbf6EGJqlSdD+wQRmPc35BDurC/bUK3])
	by ietf-mx with esmtp (Exim 4.12)
	id 19bCXp-0000oD-00
	for sigtran@ietf.org; Sat, 12 Jul 2003 01:03:37 -0400
Received: from ns.pigworks.openss7.net (IDENT:SwtiaIxCB/trVrrUv/Vo3EoSVHZwq9wc@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6C53bX30339
	for <sigtran@ietf.org>; Fri, 11 Jul 2003 23:03:37 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6C53ba20808
	for sigtran@ietf.org; Fri, 11 Jul 2003 23:03:37 -0600
Date: Fri, 11 Jul 2003 23:03:37 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Message-ID: <20030711230337.A20466@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "'sigtran@ietf.org'" <sigtran@ietf.org>
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.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: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com>; from LyOng@Ciena.com on Wed, Jul 02, 2003 at 10:21:28AM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Subject: [Sigtran] M2PA Last Call Comment
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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 proving period MAY be omitted if this is allowed by the applicable
  MTP2 standard (e. g., [Q.2210]).

The reference should be Q.2140.

It should also be noted that the same approach MUST be followed on both sides
the link.  Otherwise, one side following Q.703 will have T3 timer expire
waiting for PROVING-NORMAL from the side foregoing proving.

--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 exim@www1.ietf.org  Sat Jul 12 02:30:36 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA26455
	for <sigtran-archive@odin.ietf.org>; Sat, 12 Jul 2003 02:30:36 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bDtZ-000581-Uh
	for sigtran-archive@odin.ietf.org; Sat, 12 Jul 2003 02:30:09 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6C6U9ec019707
	for sigtran-archive@odin.ietf.org; Sat, 12 Jul 2003 02:30:09 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bDtT-00057F-FK; Sat, 12 Jul 2003 02:30:03 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bDss-00051m-91
	for sigtran@optimus.ietf.org; Sat, 12 Jul 2003 02:29:26 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA26355
	for <sigtran@ietf.org>; Sat, 12 Jul 2003 02:29:22 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19bDso-0002BP-00
	for sigtran@ietf.org; Sat, 12 Jul 2003 02:29:22 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[Utngz36v2RgEpqJw21nf5uUpn65xrlDR])
	by ietf-mx with esmtp (Exim 4.12)
	id 19bDsn-0002BM-00
	for sigtran@ietf.org; Sat, 12 Jul 2003 02:29:21 -0400
Received: from ns.pigworks.openss7.net (IDENT:CulW0GyFGIQl4TmQ85c0ND4z9bx+3XaN@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6C6TLX31278
	for <sigtran@ietf.org>; Sat, 12 Jul 2003 00:29:21 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6C6TLI21671
	for sigtran@ietf.org; Sat, 12 Jul 2003 00:29:21 -0600
Date: Sat, 12 Jul 2003 00:29:21 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Message-ID: <20030712002921.A21136@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "'sigtran@ietf.org'" <sigtran@ietf.org>
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.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: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com>; from LyOng@Ciena.com on Wed, Jul 02, 2003 at 10:21:28AM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Subject: [Sigtran] M2PA Last Call Comment - T6
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

M2PA-09 says:

> M2PA SHALL continue transmitting and acknowledging messages while it
> is in receive congestion.

Q.703 says:

| The receiving level 2 also withholds acknowledgement of the message signal
| unit, which triggered off the congestion detection, and of message signal
| units received during the congestion situation; that is fill-in signal units
| or message signal units are sent as usual, but with the backward sequence
| number and backwards indicator bit assigned the values which are contained
| in the last transmitted signal unit before the congestion is recognized.

It is rather important that the busy end not acknowledge messages while it is
in receive congestion.  This is because if the receive congestion results in
link failure, acknowledged messages caught at the congested receiver will be
unavailable for retransmission on an alternate link at the transmitter.

Also, delivering these messages to MTP3 late will aggravate missequencing of
these delayed messages with subsequent messages transmitted on the alternate
link after changeover.

Change M2PA to say:

< M2PA SHALL continue transmitting message while it is in receive congestion
< but MUST NOT acknowledge the message that triggered sending of the Link
< Status Busy message, nor any messages received before the sending of Link
< Status Busy Ended.

M2PA-09 says:

> When the peer M2PA receives the first Link Status Busy message, it
> SHALL start the Remote Congestion timer T6. Additional Link Status
> Busy messages received while T6 is running do not cause T6 to be
> reset.

Q.703 says:

| At the remote end of the signalling link, every reception of a link status
| singal unit containing indication "B" causes the excessive delay of
| acknolwedgement timer T7 to be restarted, if the timer is already running.
| In addition first reception of the link status signal unit containing a
| status indication "B" starts a longer supervision timer T7 (see 12.3) only
| if there are message signal units in the retransmission buffer.  Should
| timer T6 expire, link failure indication is generated.

Change M2PA to say:

< When the peer M2PA receives the first Link Status Busy message, it SHALL
< start the Remote Congestion timer T6 if there are messages in the
< retransmission buffer awaiting acknowledgement (i.e, T7 is running).
< In addition, M2PA will stop the T7 timer if it is running.  Additional Link
< Status Busy messages received while T6 is running do not cause T6 to be
< reset and do not cause T7 to be started.  Also, while T6 is running, T7
< SHALL NOT be started.

< When the peer M2PA receives the Link Status Busy Ended message and T6
< has not expired, it SHALL stop T6 if T6 is running and start T7 if there are
< messages awaiting acknowledgement in the retransmission buffer.

Now, I know that we said that Link Status messages are not used to acknowledge
DATA messages, but it makes sense to permit the Link Status Busy Ended message
to acknowledge outstanding User Data messages, so that upon its receipt, T7
might not need to be started if all messages in the retransmission buffer are
acknolwedged by the Link Status Busy Ended message.

The above also kind of changes the idea of T7 metering the oldest message in
the retransmission buffer.  It would now measure the oldest message in the
retransmission buffer since the receipt of the last Link Status Busy Ended.

M2PA-09 says:

> The peer M2PA SHOULD continue transmitting and acknowledging messages
> while its T6 timer is running, i.e., while the other end is Busy.

But T6 is not necessarily running during the busy condition.

Change M2PA to say:

< The peer M2PA SHOULD continue transmitting and acknowledging messages
< while the other end is Busy: i.e, between receiving Link Status Busy
< and Link Status Busy Ended.

--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 exim@www1.ietf.org  Sat Jul 12 12:59:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA27509
	for <sigtran-archive@odin.ietf.org>; Sat, 12 Jul 2003 12:59:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bNiB-0005OK-Fq
	for sigtran-archive@odin.ietf.org; Sat, 12 Jul 2003 12:59:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6CGx3hZ020723
	for sigtran-archive@odin.ietf.org; Sat, 12 Jul 2003 12:59:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bNi9-0005M9-FP; Sat, 12 Jul 2003 12:59:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bNhU-0005LY-4L
	for sigtran@optimus.ietf.org; Sat, 12 Jul 2003 12:58:20 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA27469
	for <SIGTRAN@ietf.org>; Sat, 12 Jul 2003 12:58:15 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19bNhS-0002HR-00
	for SIGTRAN@ietf.org; Sat, 12 Jul 2003 12:58:18 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19bNhR-0002HL-00
	for SIGTRAN@ietf.org; Sat, 12 Jul 2003 12:58:17 -0400
Received: from [172.16.2.16] (helo=monanumesh)
	by smtp.ccpu.com with esmtp (Exim 4.20 #1 (Debian))
	id 19bNgo-0005AC-P2; Sat, 12 Jul 2003 09:57:38 -0700
Reply-To: <umesh@ccpu.com>
From: "Umesh Vats" <umesh@ccpu.com>
To: <bidulock@openss7.org>, "VIvek Toky" <vivektoky@ccpu.com>
Cc: "SIGTRAN e-mail list" <SIGTRAN@ietf.org>
Subject: RE: [Sigtran] M3UA IG V04: Receiving REG for a RK already registered
Date: Sat, 12 Jul 2003 09:57:27 -0700
Message-ID: <HHEIIBPEJJMEJHNAOKAGGEHGCAAA.umesh@ccpu.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0)
In-Reply-To: <20030711185353.A17103@openss7.org>
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Importance: Normal
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id MAA27470
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Hi  Brian,

This will be useful in the cases where,  REG-RSP from SGP
to ASP is lost. In this case, ASP will retry the reuqest and SGP
can send the same RC back.

Umesh

-----Original Message-----
From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf Of
Brian F. G. Bidulock
Sent: Friday, July 11, 2003 5:54 PM
To: VIvek Toky
Cc: SIGTRAN e-mail list
Subject: Re: [Sigtran] M3UA IG V04: Receiving REG for a RK already
registered


VIvek,

Because it is an error.  What would be acheived by sending back the alrea=
dy
registered RC instead?

--brian

On Fri, 11 Jul 2003, VIvek Toky wrote:

>   I have one question related to Registration Request handling in M3UA =
IG
> V4. I am pasting few lines from Section 3.6
>
> "If the SGP determines that the received RK was already registered, the
SGP
> returns a registration Response message to the ASP,containing a
Registration
> Result "Error - Cannot Support Unique Routing". This error applies whet=
her
> the Routing Key is Active or Inactive."
>
> Is there any reason why SGP should send this error back. Why should it =
not
> send the same RC back again ?
>
> thanks,
> Vivek.
>
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

_______________________________________________
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 exim@www1.ietf.org  Sat Jul 12 15:04:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA03521
	for <sigtran-archive@odin.ietf.org>; Sat, 12 Jul 2003 15:04:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bPf9-00021n-Rs
	for sigtran-archive@odin.ietf.org; Sat, 12 Jul 2003 15:04:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6CJ43DY007789
	for sigtran-archive@odin.ietf.org; Sat, 12 Jul 2003 15:04:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bPf7-00021O-27; Sat, 12 Jul 2003 15:04:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bPeI-0001ug-KR
	for sigtran@optimus.ietf.org; Sat, 12 Jul 2003 15:03:10 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA03355
	for <SIGTRAN@ietf.org>; Sat, 12 Jul 2003 15:03:06 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19bPeF-0003ot-00
	for SIGTRAN@ietf.org; Sat, 12 Jul 2003 15:03:07 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[YK31TQikzcfP/q5/wAf95r4wJgu2h0gI])
	by ietf-mx with esmtp (Exim 4.12)
	id 19bPeE-0003oe-00
	for SIGTRAN@ietf.org; Sat, 12 Jul 2003 15:03:07 -0400
Received: from ns.pigworks.openss7.net (IDENT:aNoKUZH5gaoYscIVPg2ugh6vQO11Ole4@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6CJ2uP10291;
	Sat, 12 Jul 2003 13:02:56 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6CJ2tP09454;
	Sat, 12 Jul 2003 13:02:55 -0600
Date: Sat, 12 Jul 2003 13:02:55 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Umesh Vats <umesh@ccpu.com>
Cc: VIvek Toky <vivektoky@ccpu.com>, SIGTRAN e-mail list <SIGTRAN@ietf.org>
Subject: Re: [Sigtran] M3UA IG V04: Receiving REG for a RK already registered
Message-ID: <20030712130255.A9322@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Umesh Vats <umesh@ccpu.com>,
	VIvek Toky <vivektoky@ccpu.com>,
	SIGTRAN e-mail list <SIGTRAN@ietf.org>
References: <20030711185353.A17103@openss7.org> <HHEIIBPEJJMEJHNAOKAGGEHGCAAA.umesh@ccpu.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <HHEIIBPEJJMEJHNAOKAGGEHGCAAA.umesh@ccpu.com>; from umesh@ccpu.com on Sat, Jul 12, 2003 at 09:57:27AM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6CJ2uP10291
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Umesh,

How can the REG-RSP get lost?  SCTP is reliable.

--brian

On Sat, 12 Jul 2003, Umesh Vats wrote:

> Hi  Brian,
>=20
> This will be useful in the cases where,  REG-RSP from SGP
> to ASP is lost. In this case, ASP will retry the reuqest and SGP
> can send the same RC back.
>=20
> Umesh
>=20
> -----Original Message-----
> From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf O=
f
> Brian F. G. Bidulock
> Sent: Friday, July 11, 2003 5:54 PM
> To: VIvek Toky
> Cc: SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA IG V04: Receiving REG for a RK already
> registered
>=20
>=20
> VIvek,
>=20
> Because it is an error.  What would be acheived by sending back the alr=
eady
> registered RC instead?
>=20
> --brian
>=20
> On Fri, 11 Jul 2003, VIvek Toky wrote:
>=20
> >   I have one question related to Registration Request handling in M3U=
A IG
> > V4. I am pasting few lines from Section 3.6
> >
> > "If the SGP determines that the received RK was already registered, t=
he
> SGP
> > returns a registration Response message to the ASP,containing a
> Registration
> > Result "Error - Cannot Support Unique Routing". This error applies wh=
ether
> > the Routing Key is Active or Inactive."
> >
> > Is there any reason why SGP should send this error back. Why should i=
t not
> > send the same RC back again ?
> >
> > thanks,
> > Vivek.
> >
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
>=20
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Sat Jul 12 19:08:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA18637
	for <sigtran-archive@odin.ietf.org>; Sat, 12 Jul 2003 19:08:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bTTJ-0003YR-R4
	for sigtran-archive@odin.ietf.org; Sat, 12 Jul 2003 19:08:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6CN85cF013657
	for sigtran-archive@odin.ietf.org; Sat, 12 Jul 2003 19:08:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bTTF-0003Y2-IT; Sat, 12 Jul 2003 19:08:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bTT3-0003Sw-BY
	for sigtran@optimus.ietf.org; Sat, 12 Jul 2003 19:07:49 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA18584
	for <sigtran@ietf.org>; Sat, 12 Jul 2003 19:07:43 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19bTSz-00075V-00
	for sigtran@ietf.org; Sat, 12 Jul 2003 19:07:45 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[AeotUxEHWGYnRlq64Dod37lbxi7CQRY9])
	by ietf-mx with esmtp (Exim 4.12)
	id 19bTSy-00075S-00
	for sigtran@ietf.org; Sat, 12 Jul 2003 19:07:45 -0400
Received: from ns.pigworks.openss7.net (IDENT:5asSvG2M1ij49okGMTRHnqG9klvv3irB@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6CN7iP13158
	for <sigtran@ietf.org>; Sat, 12 Jul 2003 17:07:44 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6CN7iP12178
	for sigtran@ietf.org; Sat, 12 Jul 2003 17:07:44 -0600
Date: Sat, 12 Jul 2003 17:07:44 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Message-ID: <20030712170744.A11077@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "'sigtran@ietf.org'" <sigtran@ietf.org>
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.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: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com>; from LyOng@Ciena.com on Wed, Jul 02, 2003 at 10:21:28AM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Subject: [Sigtran] M2PA Last Call Comment - RPO
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

M2PA-09 says:

 While there is a remote processor outage condition:

   (a) Any User Data messages received from the peer after the Link
       Status Processor Outage MUST NOT be acknowledged and MUST be
       buffered.

   (b) M2PA SHOULD continue to acknowledge User Data messages received
       and accepted by MTP3 before the remote processor outage.

In Q.703 the side experiencing PO cannot send MSUs or FISUs without indicating a
recovery from processor outage.  M2PA can send User Data without indicating
processor recovery.  There is no reason for the receiver of User Data not to
process it even after receiving Link Status Processor Outage.

Otherwise, if (a) above is followed, MSUs in the transmission and retransmission
buffers at the time that local processor outage occured will be trapped.  If
User Data is acknowledged by the peer, they will be processed by the receiver.

Change M2PA to say:

 While there is a remote processour outage condition:

   (a) M2PA SHOULD continue to acknowledge User Data messages received
       and accepted by MTP3 regardless of the remote processor outage.

This allows the TB/RTB on the side experiencing processor outage to drain.


M2PA-09 says:

  Note that because the Link Status and User Data messages are sent on
  separate streams, it is possible for a User Data message sent after
  the Link Status Processor Recovered message to arrive before
  it. Therefore, when flushing buffers, M2PA must take care not to flush
  messages sent after the remote processor recovered. M2PA SHOULD NOT
  flush User Data messages with FSN greater than the FSN in the Link
  Status Processor Recovered message. 

MTP3 does not flush the RB after RPO or before RPR (just the RTB), unless it
uses the remote processor outage to local processor outage latching option, and
then it is only because of the LPO condition not the RPO condition.

With the change above, M2PA receives and acknowledges User Data before, during
and after remote processor outage, and comparison to any sequence numbers in the
Link Status Processor Recovered is unnecessary.

M2PA-09 says:

  When the local processor outage condition ends, M2PA SHALL send a Link
  Status Processor Recovered message to its peer. This message is used
  to signal in the end of the processor outage condition, instead of an
  MSU or FISU as in MTP2. The BSN value in this message is used by the
  peer to resynchronize its sequence numbers, if this is required by
  MTP2.

After a flush buffers operation, the side that experienced PO uses the first
received FISU/MSU to resync and discards the FISU/MSU.

ANSI T1.111.3/2000 Section 8 says:

  ... If level 3 requests level 2 [at the side experiencing processor outage] to
  clear its buffers of message signal units, level 2 begins to continuously
  transmit fill-in signal units and discards all message signal units that it
  has received from remote level 2 but not yet delivered to local level 3 and
  all message signal units that it has received from local level 3 but not yet
  transmitted on the link.  The first fill-in signal unit transmitted
  acknowledges all message signal units that have been accepted from remote
  level 2 including any mesage signal units for which acknowledgements were
  previously withheld because of level 2 congestion (see section 9).  Level 2
  then begins accepting message signal units from remote Level 2. When the next
  fill-in or message signal unit is received that passes the acceptance
  procedure of section 4 and sections 5 or 6,

   (1) all message signal units that are available for retransmission are
       discarded [clear RTB].

   (2) the forward sequence number and forward indicator bit of the
       transmitted fill-in signal units are set equal to the backward sequence
       number and backward indicator bit of the received signal unit,

   (3) normal transmission of message signal units and fill-in signal units is
       resumed, and

   (4) level 3 is notified that the buffer clearing has been completed and
       that normal operations has been resumed.

So, you see, the side experiencing the processor outage, after clearing buffers,
resynchronizes its FSN with the BSN of the first received FISU or MSU.  The side
experiencing processor outage could use the first User Data message received
after processor recovery to resyncronize.  However, if the peer has nothing to
send, processor recovery will be delayed.

Note, however, that resynchronization of sequence numbers for the transmit
direction is necessary because MTP2 traps messages in the TB/RTB of the side
experiencing processor outage because it must continuously transmit SIPO during
processor outage.  Permitting the draining of the TB/RTB with the change above
removes the need to flush the TB/RTB on the side experiencing processor outage,
as well as removing the need to resynchronize sequence numbers.

In the other direction, for messages queued to be sent to the side experiencing
processour outage, flushing and resynchronization is still necessary (actions
(1) to (4) per ANSI above).

However, Q.703 permits FISUs/MSUs received by the side experiencing remote
processor outage to discard the first and several subsequent FISU/MSUs when
recovering.  There is even a test case in Q.781 (8.9) that shows an MSU being
discarded (and negative acknowledged) in this fashion.  Because M2PA cannot
discard and expect retransmission of a message, and because sychronization is
not required when the TB/RTB of the side experiencing processor outage is
permitted to drain with the changes above, M2PA cannot permit messages to be
discarded in this fashion.

M2PA-09 says:

  If M2PA receives a Flush command from MTP3, M2PA SHALL discard the
  incoming messages that were queued and unacknowledged during the
  processor outage condition. Furthermore, M2PA SHALL discard messages
  in the transmit and retransmit queues as required by MTP2. 

To handle flushing and resynchronization properly in both directions, change the
above to read as follows:

Change M2PA to say:

  If M2PA receives a Flush command from MTP3, M2PA SHALL discard the
  incoming messages that were queued and unacknowledged during the
  processor outage condition.  Furthermore, the M2PA experiencing
  processor outage SHALL NOT discard message from the transmit and
  retransmit queues and SHALL NOT perform procedures to resynchronize
  sequence numbers, but SHALL continue normal transmission of User Data.
  The M2PA experiencing remote processor outage, SHALL discard messages in
  the transmit and retransmit queues as required by MTP2, and MAY perform
  resynchronization of sequence numbers using the BSN from the received
  Link Status Processor Recovered message.  However, during
  resychronization, M2PA SHALL NOT discard any received User Data
  messages.

As Q.2140 handles processor outage by putting the link out of service, also add:

  When M2PA experiences a local processor outage, it MAY put the link out
  of service by sending a Link Status Out of Service message if allowed by
  the applicable MTP2 standard (e.g, [Q.2140]).

I think that wraps up processor outage problems.

--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 exim@www1.ietf.org  Mon Jul 14 04:33:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA01755
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 04:33:33 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19byld-00039q-Qx
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 04:33:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6E8X5tC012116
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 04:33:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bylZ-00038v-1G; Mon, 14 Jul 2003 04:33:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19byl8-00036T-BP
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 04:32:34 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA01690
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 04:32:31 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19byl5-0002nd-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 04:32:31 -0400
Received: from goliath.siemens.de ([192.35.17.28])
	by ietf-mx with esmtp (Exim 4.12)
	id 19byl4-0002na-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 04:32:30 -0400
Received: from mail3.siemens.de (mail3.siemens.de [139.25.208.14])
	by goliath.siemens.de (8.11.7/8.11.7) with ESMTP id h6E8WTj03751
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 10:32:30 +0200 (MEST)
Received: from blues.mchh.siemens.de (blues.mchh.siemens.de [139.21.204.206])
	by mail3.siemens.de (8.11.7/8.11.7) with ESMTP id h6E8WS026245
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 10:32:28 +0200 (MEST)
Received: from mchh246e.demchh201e.icn.siemens.de (mchh246e.mchh.siemens.de [139.21.200.56])
	by blues.mchh.siemens.de (8.9.3/8.9.1) with ESMTP id KAA28451
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 10:31:22 +0200 (MET DST)
Received: by mchh246e.mchh.siemens.de with Internet Mail Service (5.5.2656.59)
	id <NTVTL70C>; Mon, 14 Jul 2003 10:31:55 +0200
Message-ID: <D73559B38D51D4119CB800508B9527B505AC1965@mchh266e.mchh.siemens.de>
From: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>
To: "List: IETF sigtran (E-Mail)" <sigtran@ietf.org>
Subject: FW: [Sigtran] M2PA Last Call Comment
Date: Mon, 14 Jul 2003 10:31:54 +0200
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2656.59)
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>

resend, because "sigtran list" wasn't included

-----Original Message-----
From: Schwarzbauer Hanns Juergen 
Sent: Monday, July 14, 2003 8:46 AM
To: 'bidulock@openss7.org'
Subject: RE: [Sigtran] M2PA Last Call Comment


Brian,

I don't understand your second point. There is an MTP L3 between the TDM and the IP side, so both sides are decoupled. Thus I can't agree with this point.

greetings

HannsJuergen

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Saturday, July 12, 2003 7:04 AM
> To: 'sigtran@ietf.org'
> Subject: [Sigtran] M2PA Last Call Comment
> 
> 
>   The proving period MAY be omitted if this is allowed by the 
> applicable
>   MTP2 standard (e. g., [Q.2210]).
> 
> The reference should be Q.2140.
> 
> It should also be noted that the same approach MUST be 
> followed on both sides
> the link.  Otherwise, one side following Q.703 will have T3 
> timer expire
> waiting for PROVING-NORMAL from the side foregoing proving.
> 
> --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 exim@www1.ietf.org  Mon Jul 14 05:41:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA03991
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 05:41:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bzpQ-0006XO-Eq
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 05:41:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6E9f4Pa025122
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 05:41:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bzpN-0006Wi-0s; Mon, 14 Jul 2003 05:41:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19bzos-0006W0-5x
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 05:40:30 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA03951
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 05:40:25 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19bzoo-0003ND-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 05:40:26 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[4Lb9rrXy1gf0O7Mp4Gk19iuAePEbUS5C])
	by ietf-mx with esmtp (Exim 4.12)
	id 19bzon-0003N5-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 05:40:25 -0400
Received: from ns.pigworks.openss7.net (IDENT:nTpaYHjTOWVyZRMOmBOy2YQH64KVlmIE@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6E9eFP11555;
	Mon, 14 Jul 2003 03:40:15 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6E9eF913445;
	Mon, 14 Jul 2003 03:40:15 -0600
Date: Mon, 14 Jul 2003 03:40:15 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>
Cc: "List: IETF sigtran (E-Mail)" <sigtran@ietf.org>
Subject: Re: FW: [Sigtran] M2PA Last Call Comment
Message-ID: <20030714034015.A13132@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>,
	"List: IETF sigtran (E-Mail)" <sigtran@ietf.org>
References: <D73559B38D51D4119CB800508B9527B505AC1965@mchh266e.mchh.siemens.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: <D73559B38D51D4119CB800508B9527B505AC1965@mchh266e.mchh.siemens.de>; from hannsjuergen.schwarzbauer@siemens.com on Mon, Jul 14, 2003 at 10:31:54AM +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>

Hanns Juergen,

Consider that one side follows Q.703 and requires proving while the
other side follows Q.2140 and foregoes proving.  The following will
occur:

    ___________________________________________________________________
   |                                                                   |
   |                Q.2140                      Q.703                  |
   |                                                                   |
   |        OUT-OF-SERVICE  ----------------->                         |
   |                        <-----------------  OUT-OF-SERVICE         |
   |                :start                                             |
   |             ALIGNMENT  ----------------->                         |
   |                                            :start                 |
   |                        <-----------------  ALIGNMENT              |
   |                 READY  ----------------->                         |
   |                        <-----------------  PROVING-NORMAL         |
   |                              !                                    |
   |                              !  T3         1.0 <= T3 <= 1.5       |
   |                              !                                    |
   |                        <-----------------  OUT-OF-SERVICE         |
   |                                                                   |
   |___________________________________________________________________|

Note that the Q.703 side must ignore the READY and await a
PROVING-NORMAL or PROVING-EMERGENCY (See Q.781 test case 2.3) and will
time out (See Q.781 test case 1.3).

Therefore unless both sides follow the Q.2140 procedures it will be
impossible to forego proving and successfuly align the link.

The issue has little to do with MTP Level 3.  It is solely an MTP Level
2 compatibility issue, speaks to interoperability, and the keyword MUST
should be used.  So,

  The proving period MAY be ommitted if this is allowed by the
  applicable standard (e.g., [Q.2140]); however, the same standard MUST
  be followed on both sides of the M2PA link.

Does that clear things up for you?

--brian

Schwarzbauer Hanns Juergen wrote:                   (Mon, 14 Jul 2003 10:31:54)
> resend, because "sigtran list" wasn't included
> 
> -----Original Message-----
> From: Schwarzbauer Hanns Juergen 
> Sent: Monday, July 14, 2003 8:46 AM
> To: 'bidulock@openss7.org'
> Subject: RE: [Sigtran] M2PA Last Call Comment
> 
> 
> Brian,
> 
> I don't understand your second point. There is an MTP L3 between the TDM and
> the IP side, so both sides are decoupled. Thus I can't agree with this
> point.
> 
> greetings
> 
> HannsJuergen
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: Saturday, July 12, 2003 7:04 AM
> > To: 'sigtran@ietf.org'
> > Subject: [Sigtran] M2PA Last Call Comment
> > 
> > 
> >   The proving period MAY be omitted if this is allowed by the 
> > applicable
> >   MTP2 standard (e. g., [Q.2210]).
> > 
> > The reference should be Q.2140.
> > 
> > It should also be noted that the same approach MUST be 
> > followed on both sides
> > the link.  Otherwise, one side following Q.703 will have T3 
> > timer expire
> > waiting for PROVING-NORMAL from the side foregoing proving.
> > 
> > --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 exim@www1.ietf.org  Mon Jul 14 06:05:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA04748
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 06:05:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0Ce-0007de-0L
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:05:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EA53oM029341
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:05:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0Cc-0007cY-Cb; Mon, 14 Jul 2003 06:05:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0CB-0007b0-2p
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 06:04:35 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA04714
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 06:04:30 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c0C7-0003cM-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:04:31 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[Zefmcp8Pbi0T6KP/8dKcfkOcrRQWMBO+])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c0C6-0003cB-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:04:30 -0400
Received: from ns.pigworks.openss7.net (IDENT:5DTu3GGY3JA7io7xBccvKEcpuBR7s4Kq@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6EA4UP12193
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 04:04:30 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6EA4UW14153
	for sigtran@ietf.org; Mon, 14 Jul 2003 04:04:30 -0600
Date: Mon, 14 Jul 2003 04:04:30 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Message-ID: <20030714040430.A13583@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "'sigtran@ietf.org'" <sigtran@ietf.org>
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.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: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com>; from LyOng@Ciena.com on Wed, Jul 02, 2003 at 10:21:28AM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Subject: [Sigtran] M2PA Last Call Comment - FSN
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

M2PA-09 says:

  If M2PA receives a User Data message with an FSN that is out of order,
  M2PA SHALL place the link out of service (see section 4.1.6 Link Out
  of Service).

This conflicts with Q.703 that ignores an MSU that contains an FSN that
is other than the next expected FSN (i.e, out of order).  This makes it
difficult to implement (and is probably the last thing standing in the
way of implementing) something like the following (that was requested by
ITU):

                           _________________
                          |                 |
                          |       IWF       |
                          |_________________|
                          |        |        |
                          |  MTP2  |  M2PA  |
                          |________|________|
                               |       |
                           ____|       |___

If M2PA discards the message as Q.703 does, and the other end of the
link was actually broken, the link will be go out of service in due
course anyway.

Change to:

  If M2PA receives a User Data message with an FSN that is out of order,
  M2PA SHALL 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



From exim@www1.ietf.org  Mon Jul 14 06:10:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA04901
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 06:10:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0HR-00087Z-RR
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:10:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EAA1rD031196
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:10:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0HR-000871-65; Mon, 14 Jul 2003 06:10:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0HA-00086d-Dr
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 06:09:44 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA04863
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 06:09:40 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c0H6-0003f6-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:09:40 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[dWIrlDElQrufGkPjgORCEi4ZPjUQ+vTz])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c0H5-0003f3-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:09:40 -0400
Received: from ns.pigworks.openss7.net (IDENT:qw26QR9b9SfbQGi48bYQiPSRGyJK4rF+@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6EA9dP12301
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 04:09:39 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6EA9dj14232
	for sigtran@ietf.org; Mon, 14 Jul 2003 04:09:39 -0600
Date: Mon, 14 Jul 2003 04:09:38 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Message-ID: <20030714040938.B13583@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "'sigtran@ietf.org'" <sigtran@ietf.org>
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.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: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com>; from LyOng@Ciena.com on Wed, Jul 02, 2003 at 10:21:28AM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Subject: [Sigtran] M2PA Last Call Comment - TTC
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

M2PA-09 says:

  Note: For the Japanese version of MTP defined in [JT-Q704], MTP3
  retrieves both unsent and unacknowleged messages for transmission on
  the alternate link(s). In this version of MTP, if M2PA receives a
  Retrieval Request and FSNC request with no FSNC value, or with an
  invalid FSNC, then M2PA SHALL retrieve from its buffers and deliver to
  MTP3 in order:

     (a) any transmitted but unacknowledged User Data messages.

     (b) any untransmitted User Data messages.

JT-Q703 does not have a Retrieval Request and FSNC request.  Only a
Retrieval Request.

Change M2PA to say:

  Note: For the Japanese version of MTP defined in [JT-Q704], MTP3
  retrieves both unsent and unacknowleged messages for transmission on
  the alternate link(s). In this version of MTP, if M2PA receives a
  Retrieval Request then M2PA SHALL retrieve from its buffers and
  deliver to MTP3 in order:

     (a) any transmitted but unacknowledged User Data messages.

     (b) any untransmitted User Data messages.

--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 exim@www1.ietf.org  Mon Jul 14 06:16:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA05128
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 06:16:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0NH-0008Q7-55
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:16:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EAG3j1032366
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:16:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0NG-0008Ps-8a; Mon, 14 Jul 2003 06:16:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0N2-0008PY-Vw
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 06:15:49 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA05107
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 06:15:39 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c0Mp-0003j5-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:15:35 -0400
Received: from goliath.siemens.de ([192.35.17.28])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c0Mo-0003j1-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:15:34 -0400
Received: from mail3.siemens.de (mail3.siemens.de [139.25.208.14])
	by goliath.siemens.de (8.11.7/8.11.7) with ESMTP id h6EAFUo09069;
	Mon, 14 Jul 2003 12:15:30 +0200 (MEST)
Received: from blues.mchh.siemens.de (blues.mchh.siemens.de [139.21.204.206])
	by mail3.siemens.de (8.11.7/8.11.7) with ESMTP id h6EAFU018542;
	Mon, 14 Jul 2003 12:15:30 +0200 (MEST)
Received: from mchh246e.demchh201e.icn.siemens.de (mchh246e.mchh.siemens.de [139.21.200.56])
	by blues.mchh.siemens.de (8.9.3/8.9.1) with ESMTP id MAA11974;
	Mon, 14 Jul 2003 12:14:23 +0200 (MET DST)
Received: by mchh246e.mchh.siemens.de with Internet Mail Service (5.5.2656.59)
	id <NTVTL05H>; Mon, 14 Jul 2003 12:14:56 +0200
Message-ID: <D73559B38D51D4119CB800508B9527B505AC1967@mchh266e.mchh.siemens.de>
From: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>,
        Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>
Cc: "List: IETF sigtran (E-Mail)" <sigtran@ietf.org>
Subject: RE: FW: [Sigtran] M2PA Last Call Comment
Date: Mon, 14 Jul 2003 12:14:56 +0200
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2656.59)
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>

Brian,

sorry, I can't agree. 
My example scenario contains two SS7 nodes, both having only M2PA links between them. I assume that this is still a valid scenario.
Furthermore, if you have an SS7 node with more than one TDM based SS7 link feeding this M2PA link, what are you going to to there.

We are again going to the discussion we had several weeks ago: 

is there an MTP Level 3 coupling the TDM links and the M2PA links and my answer is still YES. Having that in mind, proving on one link is completely independent from proving on another link. 


greetings

HannsJuergen

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Monday, July 14, 2003 11:40 AM
> To: Schwarzbauer Hanns Juergen
> Cc: List: IETF sigtran (E-Mail)
> Subject: Re: FW: [Sigtran] M2PA Last Call Comment
> 
> 
> Hanns Juergen,
> 
> Consider that one side follows Q.703 and requires proving while the
> other side follows Q.2140 and foregoes proving.  The following will
> occur:
> 
>     
> ___________________________________________________________________
>    |                                                          
>          |
>    |                Q.2140                      Q.703         
>          |
>    |                                                          
>          |
>    |        OUT-OF-SERVICE  ----------------->                
>          |
>    |                        <-----------------  
> OUT-OF-SERVICE         |
>    |                :start                                    
>          |
>    |             ALIGNMENT  ----------------->                
>          |
>    |                                            :start        
>          |
>    |                        <-----------------  ALIGNMENT     
>          |
>    |                 READY  ----------------->                
>          |
>    |                        <-----------------  
> PROVING-NORMAL         |
>    |                              !                           
>          |
>    |                              !  T3         1.0 <= T3 <= 
> 1.5       |
>    |                              !                           
>          |
>    |                        <-----------------  
> OUT-OF-SERVICE         |
>    |                                                          
>          |
>    
> |___________________________________________________________________|
> 
> Note that the Q.703 side must ignore the READY and await a
> PROVING-NORMAL or PROVING-EMERGENCY (See Q.781 test case 2.3) and will
> time out (See Q.781 test case 1.3).
> 
> Therefore unless both sides follow the Q.2140 procedures it will be
> impossible to forego proving and successfuly align the link.
> 
> The issue has little to do with MTP Level 3.  It is solely an 
> MTP Level
> 2 compatibility issue, speaks to interoperability, and the 
> keyword MUST
> should be used.  So,
> 
>   The proving period MAY be ommitted if this is allowed by the
>   applicable standard (e.g., [Q.2140]); however, the same 
> standard MUST
>   be followed on both sides of the M2PA link.
> 
> Does that clear things up for you?
> 
> --brian
> 
> Schwarzbauer Hanns Juergen wrote:                   (Mon, 14 
> Jul 2003 10:31:54)
> > resend, because "sigtran list" wasn't included
> > 
> > -----Original Message-----
> > From: Schwarzbauer Hanns Juergen 
> > Sent: Monday, July 14, 2003 8:46 AM
> > To: 'bidulock@openss7.org'
> > Subject: RE: [Sigtran] M2PA Last Call Comment
> > 
> > 
> > Brian,
> > 
> > I don't understand your second point. There is an MTP L3 
> between the TDM and
> > the IP side, so both sides are decoupled. Thus I can't 
> agree with this
> > point.
> > 
> > greetings
> > 
> > HannsJuergen
> > 
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: Saturday, July 12, 2003 7:04 AM
> > > To: 'sigtran@ietf.org'
> > > Subject: [Sigtran] M2PA Last Call Comment
> > > 
> > > 
> > >   The proving period MAY be omitted if this is allowed by the 
> > > applicable
> > >   MTP2 standard (e. g., [Q.2210]).
> > > 
> > > The reference should be Q.2140.
> > > 
> > > It should also be noted that the same approach MUST be 
> > > followed on both sides
> > > the link.  Otherwise, one side following Q.703 will have T3 
> > > timer expire
> > > waiting for PROVING-NORMAL from the side foregoing proving.
> > > 
> > > --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 exim@www1.ietf.org  Mon Jul 14 06:22:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA05320
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 06:22:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0T5-0000DC-9j
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:22:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EAM3Tk000810
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:22:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0T3-0000Ch-W5; Mon, 14 Jul 2003 06:22:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0Sz-0000CS-7l
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 06:21:57 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA05302
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 06:21:52 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c0Sv-0003mg-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:21:53 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[MobT82jggzTzJoOSsgG+HoTWbcLhcMno])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c0Su-0003md-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:21:52 -0400
Received: from ns.pigworks.openss7.net (IDENT:puEKmHeuIXplbQCJmz6eI3Llb9m0YUVM@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6EALqP12441;
	Mon, 14 Jul 2003 04:21:52 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6EALpO14487;
	Mon, 14 Jul 2003 04:21:51 -0600
Date: Mon, 14 Jul 2003 04:21:51 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>
Cc: "List: IETF sigtran (E-Mail)" <sigtran@ietf.org>
Subject: Re: FW: [Sigtran] M2PA Last Call Comment
Message-ID: <20030714042151.C13583@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>,
	"List: IETF sigtran (E-Mail)" <sigtran@ietf.org>
References: <D73559B38D51D4119CB800508B9527B505AC1967@mchh266e.mchh.siemens.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: <D73559B38D51D4119CB800508B9527B505AC1967@mchh266e.mchh.siemens.de>; from hannsjuergen.schwarzbauer@siemens.com on Mon, Jul 14, 2003 at 12:14:56PM +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>

Schwarzbauer,

Huh?  In the diagram I gave, it is only an M2PA link.  One side is
following Q.2140 state machine procedures, the other is following
Q.703 state machine procedures.

This has nothing to do with MTP3.

--brian

Schwarzbauer Hanns Juergen wrote:                                             (Mon, 14 Jul 2003 12:14:56)
> Brian,
> 
> sorry, I can't agree. 
> My example scenario contains two SS7 nodes, both having only M2PA links
> between them. I assume that this is still a valid scenario.  Furthermore, if
> you have an SS7 node with more than one TDM based SS7 link feeding this M2PA
> link, what are you going to to there.
> 
> We are again going to the discussion we had several weeks ago: 
> 
> is there an MTP Level 3 coupling the TDM links and the M2PA links and my
> answer is still YES. Having that in mind, proving on one link is completely
> independent from proving on another link. 
> 
> 
> greetings
> 
> HannsJuergen
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: Monday, July 14, 2003 11:40 AM
> > To: Schwarzbauer Hanns Juergen
> > Cc: List: IETF sigtran (E-Mail)
> > Subject: Re: FW: [Sigtran] M2PA Last Call Comment
> > 
> > 
> > Hanns Juergen,
> > 
> > Consider that one side follows Q.703 and requires proving while the
> > other side follows Q.2140 and foregoes proving.  The following will
> > occur:
> > 
> >     
> > ___________________________________________________________________
> >    |                                                          
> >          |
> >    |                Q.2140                      Q.703         
> >          |
> >    |                                                          
> >          |
> >    |        OUT-OF-SERVICE  ----------------->                
> >          |
> >    |                        <-----------------  
> > OUT-OF-SERVICE         |
> >    |                :start                                    
> >          |
> >    |             ALIGNMENT  ----------------->                
> >          |
> >    |                                            :start        
> >          |
> >    |                        <-----------------  ALIGNMENT     
> >          |
> >    |                 READY  ----------------->                
> >          |
> >    |                        <-----------------  
> > PROVING-NORMAL         |
> >    |                              !                           
> >          |
> >    |                              !  T3         1.0 <= T3 <= 
> > 1.5       |
> >    |                              !                           
> >          |
> >    |                        <-----------------  
> > OUT-OF-SERVICE         |
> >    |                                                          
> >          |
> >    
> > |___________________________________________________________________|
> > 
> > Note that the Q.703 side must ignore the READY and await a
> > PROVING-NORMAL or PROVING-EMERGENCY (See Q.781 test case 2.3) and will
> > time out (See Q.781 test case 1.3).
> > 
> > Therefore unless both sides follow the Q.2140 procedures it will be
> > impossible to forego proving and successfuly align the link.
> > 
> > The issue has little to do with MTP Level 3.  It is solely an 
> > MTP Level
> > 2 compatibility issue, speaks to interoperability, and the 
> > keyword MUST
> > should be used.  So,
> > 
> >   The proving period MAY be ommitted if this is allowed by the
> >   applicable standard (e.g., [Q.2140]); however, the same 
> > standard MUST
> >   be followed on both sides of the M2PA link.
> > 
> > Does that clear things up for you?
> > 
> > --brian
> > 
> > Schwarzbauer Hanns Juergen wrote:                   (Mon, 14 
> > Jul 2003 10:31:54)
> > > resend, because "sigtran list" wasn't included
> > > 
> > > -----Original Message-----
> > > From: Schwarzbauer Hanns Juergen 
> > > Sent: Monday, July 14, 2003 8:46 AM
> > > To: 'bidulock@openss7.org'
> > > Subject: RE: [Sigtran] M2PA Last Call Comment
> > > 
> > > 
> > > Brian,
> > > 
> > > I don't understand your second point. There is an MTP L3 
> > between the TDM and
> > > the IP side, so both sides are decoupled. Thus I can't 
> > agree with this
> > > point.
> > > 
> > > greetings
> > > 
> > > HannsJuergen
> > > 
> > > > -----Original Message-----
> > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > Sent: Saturday, July 12, 2003 7:04 AM
> > > > To: 'sigtran@ietf.org'
> > > > Subject: [Sigtran] M2PA Last Call Comment
> > > > 
> > > > 
> > > >   The proving period MAY be omitted if this is allowed by the 
> > > > applicable
> > > >   MTP2 standard (e. g., [Q.2210]).
> > > > 
> > > > The reference should be Q.2140.
> > > > 
> > > > It should also be noted that the same approach MUST be 
> > > > followed on both sides
> > > > the link.  Otherwise, one side following Q.703 will have T3 
> > > > timer expire
> > > > waiting for PROVING-NORMAL from the side foregoing proving.
> > > > 
> > > > --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 exim@www1.ietf.org  Mon Jul 14 06:26:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA05462
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 06:26:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0Ww-0000Lo-3Y
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:26:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EAQ2Ln001327
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:26:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0Wv-0000L3-7W; Mon, 14 Jul 2003 06:26:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0WH-0000KG-JM
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 06:25:21 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA05433
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 06:25:17 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c0WD-0003p9-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:25:17 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[7wkZR41oPjaArgl0gwEFO7b4WAa4SKUc])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c0WC-0003p6-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:25:16 -0400
Received: from ns.pigworks.openss7.net (IDENT:DDos0TPXv+g2mAABMPI8D2/bvNb3OM8Y@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6EAPGP12488
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 04:25:16 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6EAPG514524
	for sigtran@ietf.org; Mon, 14 Jul 2003 04:25:16 -0600
Date: Mon, 14 Jul 2003 04:25:16 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Message-ID: <20030714042516.D13583@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "'sigtran@ietf.org'" <sigtran@ietf.org>
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.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: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com>; from LyOng@Ciena.com on Wed, Jul 02, 2003 at 10:21:28AM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Subject: [Sigtran] M2PA Last Call Comment - JT-Q703 nit pick
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

M2PA-09 says:

 It is not necessary to put the message length in the LI octet as in
 MTP2. The LI octet is included because the two spare bits in the LI
 octet are used by MTP3 in at least one national version of SS7 to
 carry MTP3 information. For example, the Japanese TTC standard uses
 these spare bits as an MTP3 Message Priority field. See [JT-Q704],
 section 14 "Common Characteristics of message signal unit formats",
 section 14.2 (A) "Priority Indicator (PRI)". For versions of MTP that
 do not use these two bits, the entire octet is spare.

and

   PRI - Priority used only in national MTP defined in [JT-Q704].
         Spare for other MTP versions.

These bits are defined in JT-Q703, not JT-Q704.

Change JT-Q704 to JT-Q703 above and add [JT-Q703] to the references.

--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 exim@www1.ietf.org  Mon Jul 14 06:35:38 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA06219
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 06:35:38 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0fl-0003We-A9
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:35:09 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EAZ9Te013551
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:35:09 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0fd-0003W6-BH; Mon, 14 Jul 2003 06:35:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0em-0003UK-Pv
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 06:34:08 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA06179
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 06:34:04 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c0ei-00041I-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:34:04 -0400
Received: from penguin-ext.wise.edt.ericsson.se ([193.180.251.47] helo=penguin.al.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19c0eh-00041B-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:34:03 -0400
Received: from esealnt610.al.sw.ericsson.se ([153.88.254.120])
	by penguin.al.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h6EAY1GJ028785;
	Mon, 14 Jul 2003 12:34:01 +0200 (MEST)
Received: from ESEALNT745.al.sw.ericsson.se ([153.88.251.5]) by esealnt610.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55)
	id 366BVYYB; Mon, 14 Jul 2003 12:34:03 +0200
Received: by ESEALNT745.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <N94N321P>; Mon, 14 Jul 2003 12:33:35 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CFB@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
To: "'Brian F. G. Bidulock '" <bidulock@openss7.org>
Cc: "''sigtran@ietf.org' '" <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA IG v04 Section 3.8
Date: Mon, 14 Jul 2003 12:33:27 +0200
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 correct text is:

    5.4.1 SG State: Uncongested / Available
 
           ASP                          SGP
           ---                          ---
            |  -------- DAUD --------->  |
            |  <------ SCON(0) --------  |
            |  <------- DAVA ----------  |

    5.4.4 SG state: Unavailable
 
           ASP                          SGP
           ---                          ---
            |  -------- DAUD --------->  |
            |  <------- DUNA ----------  |


(...as you say)


Cheers // Javier.


-----Original Message-----
From: Brian F. G. Bidulock
To: J.Javier Pastor (EE/EEM)
Cc: 'sigtran@ietf.org'
Sent: 11/07/2003 11:06
Subject: Re: [Sigtran] M3UA IG v04 Section 3.8

J.Javier,

J.Javier Pastor (EE/EEM) wrote:
(Fri, 11 Jul 2003 09:51:40)
> Hi Brian, 
> 
> See comments below...
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: jueves, 10 de julio de 2003 15:30
> > To: J.Javier Pastor (EE/EEM)
> > Cc: 'sigtran@ietf.org'
> > Subject: Re: [Sigtran] M3UA IG v04 Section 3.8
> > 
> > 
> > Javier,
> > 
> > Section 3.8 says:
> > 
> >    5.4 Auditing examples
> > 
> >    5.4.1 SG State: Uncongested / Unavailable
> > 
> >           ASP                          SGP
> >           ---                          ---
> >            |  -------- DAUD --------->  |
> >            |  <------ SCON(0) --------  |
> >            |  <------- DUNA ----------  |
> > 
> > What state of a destination is Uncongested/Unavailable?  There is no
> > such state in Q.704.  Uncongested routesets are by definition 
> > available.
> > Inaccessible routesets have no congestion state by definition.
> 

Is the above diagram in error as well?

> 
> > 
> > Did you think that a routeset could be in an unavailble and 
> > congested state?
> > 
> > What is the use of the SCON(0)?  Why is it being sent?
> > 
> >    5.4.4 SG state: Uncongested / Unavailable
> > 
> >           ASP                          SGP
> >           ---                          ---
> >            |  -------- DAUD --------->  |
> >            |  <------- DUNA ----------  |
> > 
> > What is the difference between 5.4.1 and 5.4.4?  They both say
> > Uncongested/Available.
> > 
> > 
> It is a typo. In 5.4.1 it should say: SG state: "Uncongested /
Available".
> And of course a DAVA has to be sent.

Why then is SCON(0) sent in 5.4.1 above, and not in 5.4.4.

Perhaps you have the typo reversed?  5.4.1 should be
Uncongested/Unavailable
(with DAVA instead of DUNA), and 5.4.4 should be just plain Unavailable.

Like so would be correct:

    5.4.1 SG State: Uncongested / Available
 
           ASP                          SGP
           ---                          ---
            |  -------- DAUD --------->  |
            |  <------ SCON(0) --------  |
            |  <------- DAVA ----------  |

    5.4.4 SG state: Unavailable
 
           ASP                          SGP
           ---                          ---
            |  -------- DAUD --------->  |
            |  <------- DUNA ----------  |
 

> 
> > 
> >    [...]. Where the SGP maintains the congestion status of the SS7
> >    destination, the SGP MUST additionally respond with an SCON
message
> >    before the DAVA or DRST message.  If the SS7 destination is
> > 
> > When is the SGP required to maintain congestion status?  
> 
> In those networks where the congestion levels are handled... There's
no
> change there regarding the RFC. Do you think the RFC requires some
> additional explanation in this regard?

Well, there is this change:

   Where the SGP does not maintain the congestion status of the SS7
   destination, the response to a DAUD message should always be only a
   DAVA, DRST or DUNA message as appropriate.

When does the SGP not maintain the congestion status of the SS7
destination?
When does it?  The first passage is a MUST and this is a SHOULD.  How is
one
to know when an SGP is following the MUST or SHOULD and when it is not?

If it doesn't really matter, why do we say MUST and SHOULD?

> 
> 
> 
> > What is the criteria for an ASP to determine that an SGP should
maintain
> > congestion status and is violating this MUST.  Why must the ASP
require
> > the SGP to send SCON in this circumstance?  What interoperability
concern
> > is the MUST addressing?
> > 
> It seems to me that it was not clear whether to send the SCON or not
during
> the bakeoff.

How is it clearer now?  One still doesn't know when an SGP MUST or
SHOULD NOT
send SCON because there is no criteria for determining when an SGP
maintains
congestion state (a completely undefined concept).

--brian

> 
> 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

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



From exim@www1.ietf.org  Mon Jul 14 06:37:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA06399
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 06:37:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0ha-0003j2-1K
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:37:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EAb1fO014298
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:37:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0hZ-0003iW-T2; Mon, 14 Jul 2003 06:37:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0h5-0003gV-8b
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 06:36:31 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA06302
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 06:36:26 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c0h1-00044Z-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:36:27 -0400
Received: from penguin-ext.wise.edt.ericsson.se ([193.180.251.47] helo=penguin.al.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19c0h0-00044P-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:36:26 -0400
Received: from esealnt610.al.sw.ericsson.se ([153.88.254.120])
	by penguin.al.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h6EAaQGJ029142;
	Mon, 14 Jul 2003 12:36:26 +0200 (MEST)
Received: from ESEALNT744.al.sw.ericsson.se ([153.88.251.4]) by esealnt610.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55)
	id 366BVZ26; Mon, 14 Jul 2003 12:36:28 +0200
Received: by ESEALNT744.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <3W410PDT>; Mon, 14 Jul 2003 12:36:26 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CFC@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
To: "'Brian F. G. Bidulock '" <bidulock@openss7.org>
Cc: "''sigtran@ietf.org' '" <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA IG v04 Section 3.18
Date: Mon, 14 Jul 2003 12:35:58 +0200
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, 

We'll remove the mention of the timer, if nobody disagrees...

Cheers // Javier. 

-----Original Message-----
From: Brian F. G. Bidulock
To: J.Javier Pastor (EE/EEM)
Cc: 'sigtran@ietf.org'
Sent: 11/07/2003 11:07
Subject: Re: [Sigtran] M3UA IG v04 Section 3.18

J.Javier,

J.Javier Pastor (EE/EEM) wrote:
(Fri, 11 Jul 2003 09:56:59)
> Hi Brian, 
> 
> Pls, see below...
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: jueves, 10 de julio de 2003 15:34
> > To: J.Javier Pastor (EE/EEM)
> > Cc: 'sigtran@ietf.org'
> > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > 
> > 
> > Javier,
> > 
> > 3.18.3 Solution Description
> > 
> >    Clarify that the ASP needs to indicate when the congestion 
> > level has
> >    changed (including abatement).  Further, the ASP peer can
maintain
> >    a timer, if desired, in case the ASP fails to indicate congestion
> >    abatement.
> > 
> > Where are the procedures for this timer described?  This is the only
place
> > that the timer is mentioned.  Please describe the timer or remove
the
> > sentence saying that it has been described...
> > 
> 
> Do you think is worth include the timer explanation into the new text?
> Otherwise, we may remove that sentence. 

Please remove mention of the timer.

--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 exim@www1.ietf.org  Mon Jul 14 06:37:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA06414
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 06:37:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0ha-0003jl-Lh
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:37:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EAb2Gm014359
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:37:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0ha-0003jU-GY; Mon, 14 Jul 2003 06:37:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0h9-0003gk-ED
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 06:36:35 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA06313
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 06:36:30 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c0h5-00044m-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:36:31 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[HK4ZOT0RptDcsV4ir3CuJp7osHjm5iYB])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c0h3-00044a-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:36:30 -0400
Received: from ns.pigworks.openss7.net (IDENT:teGGwv/XS2OmSdQctiTgsy8MKyVe4Yif@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6EAaSP12719;
	Mon, 14 Jul 2003 04:36:28 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6EAaRJ14692;
	Mon, 14 Jul 2003 04:36:27 -0600
Date: Mon, 14 Jul 2003 04:36:27 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>
Cc: "List: IETF sigtran (E-Mail)" <sigtran@ietf.org>
Subject: Re: FW: [Sigtran] M2PA Last Call Comment
Message-ID: <20030714043627.E13583@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>,
	"List: IETF sigtran (E-Mail)" <sigtran@ietf.org>
References: <D73559B38D51D4119CB800508B9527B505AC1968@mchh266e.mchh.siemens.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: <D73559B38D51D4119CB800508B9527B505AC1968@mchh266e.mchh.siemens.de>; from hannsjuergen.schwarzbauer@siemens.com on Mon, Jul 14, 2003 at 12:27:15PM +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>

Hanns Juergen,

I'll try again.  This has nothing to do with TDM or MTP3.

M2PA-09 says:

  4.  Procedures

  Since M2PA provides MTP3 with an interface and functionality like
  MTP2, its internal functioning is similar to that of MTP2. 

  Except as modified in this document, M2PA SHOULD follow the
  requirements of the applicable MTP2 specification. These may include
  [Q.703] or [T1.111].

and then later:

  The proving period MAY be omitted if this is allowed by the applicable
  MTP2 standard (e. g., [Q.2210]).

So, M2PA permits one side to follow Q.2140 procedures and the other side
to follow Q.703 procedure, with regard to proving.  This is incompatible
as I have shown.

When following the requirements of the applicable MTP2 specification, it
must be noted that the same MTP2 specification must be followed on both
sides, at least with regard to foregoing proving.  Thus the text change.

--brian

Schwarzbauer Hanns Juergen wrote:                                             (Mon, 14 Jul 2003 12:27:15)
> Bidulock,
> 
> your answer is confusing. why should one side of an M2PA link adhere to
> Q.2140 and the other o Q.703 proving procedures? 
>
> You are still assuming a direct and tight coupling between the TDM and the
> IP side.
>
> In turn the throughput of the M2PA link could NEVER exceed that of the
> corresponding TDM link: 64 kbit in case of Q.703 (excluding annex A for the
> moment). That's not what M2PA was designed for!
> 
> greetings
> 
> HannsJuergen
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: Monday, July 14, 2003 12:22 PM
> > To: Schwarzbauer Hanns Juergen
> > Cc: List: IETF sigtran (E-Mail)
> > Subject: Re: FW: [Sigtran] M2PA Last Call Comment
> > 
> > 
> > Schwarzbauer,
> > 
> > Huh?  In the diagram I gave, it is only an M2PA link.  One side is
> > following Q.2140 state machine procedures, the other is following
> > Q.703 state machine procedures.
> > 
> > This has nothing to do with MTP3.
> > 
> > --brian
> > 
> > Schwarzbauer Hanns Juergen wrote:                             
> >                 (Mon, 14 Jul 2003 12:14:56)
> > > Brian,
> > > 
> > > sorry, I can't agree. 
> > > My example scenario contains two SS7 nodes, both having 
> > only M2PA links
> > > between them. I assume that this is still a valid scenario. 
> >  Furthermore, if
> > > you have an SS7 node with more than one TDM based SS7 link 
> > feeding this M2PA
> > > link, what are you going to to there.
> > > 
> > > We are again going to the discussion we had several weeks ago: 
> > > 
> > > is there an MTP Level 3 coupling the TDM links and the M2PA 
> > links and my
> > > answer is still YES. Having that in mind, proving on one 
> > link is completely
> > > independent from proving on another link. 
> > > 
> > > 
> > > greetings
> > > 
> > > HannsJuergen
> > > 
> > > > -----Original Message-----
> > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > Sent: Monday, July 14, 2003 11:40 AM
> > > > To: Schwarzbauer Hanns Juergen
> > > > Cc: List: IETF sigtran (E-Mail)
> > > > Subject: Re: FW: [Sigtran] M2PA Last Call Comment
> > > > 
> > > > 
> > > > Hanns Juergen,
> > > > 
> > > > Consider that one side follows Q.703 and requires proving 
> > while the
> > > > other side follows Q.2140 and foregoes proving.  The 
> > following will
> > > > occur:
> > > > 
> > > >     
> > > > 
> > ___________________________________________________________________
> > > >    |                                                          
> > > >          |
> > > >    |                Q.2140                      Q.703         
> > > >          |
> > > >    |                                                          
> > > >          |
> > > >    |        OUT-OF-SERVICE  ----------------->                
> > > >          |
> > > >    |                        <-----------------  
> > > > OUT-OF-SERVICE         |
> > > >    |                :start                                    
> > > >          |
> > > >    |             ALIGNMENT  ----------------->                
> > > >          |
> > > >    |                                            :start        
> > > >          |
> > > >    |                        <-----------------  ALIGNMENT     
> > > >          |
> > > >    |                 READY  ----------------->                
> > > >          |
> > > >    |                        <-----------------  
> > > > PROVING-NORMAL         |
> > > >    |                              !                           
> > > >          |
> > > >    |                              !  T3         1.0 <= T3 <= 
> > > > 1.5       |
> > > >    |                              !                           
> > > >          |
> > > >    |                        <-----------------  
> > > > OUT-OF-SERVICE         |
> > > >    |                                                          
> > > >          |
> > > >    
> > > > 
> > |___________________________________________________________________|
> > > > 
> > > > Note that the Q.703 side must ignore the READY and await a
> > > > PROVING-NORMAL or PROVING-EMERGENCY (See Q.781 test case 
> > 2.3) and will
> > > > time out (See Q.781 test case 1.3).
> > > > 
> > > > Therefore unless both sides follow the Q.2140 procedures 
> > it will be
> > > > impossible to forego proving and successfuly align the link.
> > > > 
> > > > The issue has little to do with MTP Level 3.  It is solely an 
> > > > MTP Level
> > > > 2 compatibility issue, speaks to interoperability, and the 
> > > > keyword MUST
> > > > should be used.  So,
> > > > 
> > > >   The proving period MAY be ommitted if this is allowed by the
> > > >   applicable standard (e.g., [Q.2140]); however, the same 
> > > > standard MUST
> > > >   be followed on both sides of the M2PA link.
> > > > 
> > > > Does that clear things up for you?
> > > > 
> > > > --brian
> > > > 
> > > > Schwarzbauer Hanns Juergen wrote:                   (Mon, 14 
> > > > Jul 2003 10:31:54)
> > > > > resend, because "sigtran list" wasn't included
> > > > > 
> > > > > -----Original Message-----
> > > > > From: Schwarzbauer Hanns Juergen 
> > > > > Sent: Monday, July 14, 2003 8:46 AM
> > > > > To: 'bidulock@openss7.org'
> > > > > Subject: RE: [Sigtran] M2PA Last Call Comment
> > > > > 
> > > > > 
> > > > > Brian,
> > > > > 
> > > > > I don't understand your second point. There is an MTP L3 
> > > > between the TDM and
> > > > > the IP side, so both sides are decoupled. Thus I can't 
> > > > agree with this
> > > > > point.
> > > > > 
> > > > > greetings
> > > > > 
> > > > > HannsJuergen
> > > > > 
> > > > > > -----Original Message-----
> > > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > > Sent: Saturday, July 12, 2003 7:04 AM
> > > > > > To: 'sigtran@ietf.org'
> > > > > > Subject: [Sigtran] M2PA Last Call Comment
> > > > > > 
> > > > > > 
> > > > > >   The proving period MAY be omitted if this is allowed by the 
> > > > > > applicable
> > > > > >   MTP2 standard (e. g., [Q.2210]).
> > > > > > 
> > > > > > The reference should be Q.2140.
> > > > > > 
> > > > > > It should also be noted that the same approach MUST be 
> > > > > > followed on both sides
> > > > > > the link.  Otherwise, one side following Q.703 will have T3 
> > > > > > timer expire
> > > > > > waiting for PROVING-NORMAL from the side foregoing proving.
> > > > > > 
> > > > > > --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/
> > 

-- 
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 exim@www1.ietf.org  Mon Jul 14 06:41:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA06596
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 06:41:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0lT-0004V3-I6
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:41:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EAf3bT017293
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:41:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0lR-0004Ug-So; Mon, 14 Jul 2003 06:41:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0ku-0004QA-2N
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 06:40:28 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA06569
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 06:40:23 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c0kq-0004AE-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:40:24 -0400
Received: from albatross-ext.wise.edt.ericsson.se ([193.180.251.49] helo=albatross.tn.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19c0ko-0004AB-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:40:23 -0400
Received: from esealnt612.al.sw.ericsson.se ([153.88.254.118])
	by albatross.tn.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h6EAeLIi017496;
	Mon, 14 Jul 2003 12:40:21 +0200 (MEST)
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 NW4WP0KZ; Mon, 14 Jul 2003 12:40:21 +0200
Received: by ESEALNT744.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <3W410PFP>; Mon, 14 Jul 2003 12:40:21 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CFD@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
To: "'Brian F. G. Bidulock '" <bidulock@openss7.org>
Cc: "''sigtran@ietf.org' '" <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA IG v04 Section 3.9
Date: Mon, 14 Jul 2003 12:39:48 +0200
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, 

We'll add it to the Solution Description.

Thanks // Javier.

-----Original Message-----
From: Brian F. G. Bidulock
To: J.Javier Pastor (EE/EEM)
Cc: 'sigtran@ietf.org'
Sent: 11/07/2003 11:10
Subject: Re: [Sigtran] M3UA IG v04 Section 3.9

J.Javier,

J.Javier Pastor (EE/EEM) wrote:
(Fri, 11 Jul 2003 10:02:43)
> Hi Brian, 
> 
> Pls, see below..
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: jueves, 10 de julio de 2003 15:42
> > To: J.Javier Pastor (EE/EEM)
> > Cc: 'sigtran@ietf.org'
> > Subject: Re: [Sigtran] M3UA IG v04 Section 3.9
> > 
> > 
> > Javier,
> > 
> > Section 3.9 says:
> > 
> >           ASP                          SGP
> >           ---                          ---
> >            |  ------ ASPIA (RC1)----->  |
> >            |  <----  ASPIA Ack -------  |
> >            |  -----DEREG REQ (RC1)--->  |
> >            |  <----DEREG RSP (RC1)----  |
> >            |  -------ASPIA (RC1)----->  |
> > 
> >    What should SG do?
> > 
> > The new text does not explicity answer this question.
> > 
> > What does the SGP send in this case?
> 
> It is included in the new text for section 4.5.3. Take a look to the
second
> bullet. Although you may be right saying that we should include it
also into
> the "solution description"

Yes, because you have pulled three bullets in the new text without
indicating
which of them addresses the problem.  The new text is fine, the answer
to the
question "What should SG do?" is not really answered.  Just place in the
Solution description that the SG should send Error("Invalid Routing
Context")
in answer to the original question.

> 
> > 
> > As RC values are per-SGP, it appears to me that it must be
> > Error("Invalid Routing Context").
> > 
> > Correct?
> 
> Yes, correct.
> 
> 
> Thanks // 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 exim@www1.ietf.org  Mon Jul 14 06:44:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA06673
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 06:44:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0oM-0004gE-2k
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:44:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EAi2pm017983
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:44:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0oK-0004fo-LE; Mon, 14 Jul 2003 06:44:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c0o5-0004fD-DD
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 06:43:45 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA06670
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 06:43:40 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c0o1-0004CB-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:43:41 -0400
Received: from albatross-ext.wise.edt.ericsson.se ([193.180.251.49] helo=albatross.tn.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19c0o0-0004C8-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:43:40 -0400
Received: from esealnt610.al.sw.ericsson.se ([153.88.254.120])
	by albatross.tn.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h6EAheIi018182;
	Mon, 14 Jul 2003 12:43:40 +0200 (MEST)
Received: from ESEALNT744.al.sw.ericsson.se ([153.88.251.4]) by esealnt610.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55)
	id 366BV5LK; Mon, 14 Jul 2003 12:43:42 +0200
Received: by ESEALNT744.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <3W410PH6>; Mon, 14 Jul 2003 12:43:40 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CFE@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
To: "'Brian F. G. Bidulock '" <bidulock@openss7.org>
Cc: "''sigtran@ietf.org' '" <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA IG v04  Section 3.14
Date: Mon, 14 Jul 2003 12:43:10 +0200
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, 

What you say it's right, only 1 'n' transition would be kept in the proposal: the IA2AC transition.

I think it makes sense since "n" refers only to the capacity of processing traffic and in the AS-INACTIVE state no traffic is processed.

Does anyone have any objection to this change?


Thanks // Javier.

-----Original Message-----
From: Brian F. G. Bidulock
To: J.Javier Pastor (EE/EEM)
Cc: 'sigtran@ietf.org'
Sent: 11/07/2003 11:13
Subject: Re: [Sigtran] M3UA IG v04  Section 3.14

J.Javier,

J.Javier Pastor (EE/EEM) wrote:
(Fri, 11 Jul 2003 10:35:57)
> Hi Brian, 
> 
> You are right saying that there is currently no way of notifying
AS-DOWN. In
> fact "n" only applies to the traffic case.
> 
> Then I see two options:
> 	1- include AS-DOWN in NTFY
> 	2- The AS-DOWN state is only got when all the ASPs are in
ASP-DOWN state.
> 
> I would addopt 2 since it is inline what we are having in the other
UALs. 

Yes, please go with 2 as it is also in line with the RFC.

That will also leave only 1 'n' transition in the proposal: the IA2AC
transition, correct?

--brian

> 
> Thanks // Javier.
> 
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: jueves, 10 de julio de 2003 17:15
> > To: J.Javier Pastor (EE/EEM)
> > Cc: 'sigtran@ietf.org'
> > Subject: Re: [Sigtran] M3UA IG v04 Section 3.14
> > 
> > 
> > Javier,
> > 
> > Section 3.14 says:
> > 
> >    IA2DN: One ASP moves to ASP-DOWN causing that the number of ASPs
in
> >    either ASP-ACTIVE or ASP-INACTIVE drops below n.
> > 
> > Am I correct to say that if n=5 and there are 4 ASPs in the 
> > ASP-INACTIVE
> > and 1 ASP in the ASP-ACTIVE state, and then the 1 active ASP moves
to
> > the ASP-DOWN state and T(r) expires that the AS is to move to 
> > the AS-DOWN
> > state?  As follows:
> > 
> >   If T(r) expires before an ASP becomes ASP-ACTIVE, the SGP MAY stop
> >   queuing messages and discards all
> >   previously queued messages. The AS will move to the 
> > AS-INACTIVE state
> >   if at least the number of ASPs in ASP-INACTIVE sum n, otherwise it
> >   will move to AS-DOWN state.
> > 
> > Now, that is an AS state change and we must send a NTFY to 
> > all the INACTIVE
> > ASPs (the 4 left inactive) of the state change.  But from 
> > Section 3.8.2 of
> > the RFC, there is no message to notify of an AS-DOWN state:
> > 
> >          1    reserved
> >          2    Application Server Inactive (AS-INACTIVE)
> >          3    Application Server Active (AS-ACTIVE)
> >          4    Application Server Pending (AS-PENDING)
> > 
> > You see, we removed it because there was never an instance in which
it
> > would be sent.  You have inserted an instance again.
> > 
> > Aren't you going to have to define a Notify message (the 
> > Reserved value)
> > for "Application Server Down (AS-DOWN)" so that the INACTIVE ASPs
can
> > be properly notified by the SGP?
> > 
> > Particularly when the RFC says in Section 4.3.4.5: (Notify
Procedures)
> > 
> >    A Notify message reflecting a change in the AS state MUST 
> > be sent to
> >    all ASPs in the AS, except those in the ASP-DOWN state, with
> >    appropriate Status Information and any ASP Identifier of the
failed
> >    ASP.  At the ASP, Layer Management is informed with an M-NOTIFY
> >    indication primitive.  The Notify message must be sent 
> > whether the AS
> >    state change was a result of an ASP failure or reception of an
ASP
> >    State management (ASPSM) / ASP Traffic Management (ASPTM)
message.
> >    In the second case, the Notify message MUST be sent after 
> > any related
> >    acknowledgement messages  (e.g., ASP Up Ack, ASP Down Ack, 
> > ASP Active
> >    Ack, or ASP Inactive Ack).
> > 
> > The above tells me that I MUST send a Notify message becase of the
AS
> > state change, but there is no message to inform about AS-DOWN state
> > transition.
> > 
> > What is the purpose of doing this?  What is the protocol 
> > difference between
> > the AS-DOWN state and the AS-INACTIVE state?  What is it that the
SGP
> > does or doesn't do in the one state that it doesn't or does 
> > do in the other?
> > 
> > I don't recall this being discussed at all.
> > 
> > Please clarify.
> > 
> > -- 
> > 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 exim@www1.ietf.org  Mon Jul 14 06:57:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA07575
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 06:57:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c10w-0005ZJ-8f
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:57:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EAv2Cm021397
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:57:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c10v-0005Yy-Hz; Mon, 14 Jul 2003 06:57:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c10R-0005XF-7N
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 06:56:31 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA07484
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 06:56:26 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c10N-0004PY-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:56:27 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[3f/sY/F/KAIa1fcZQm0z7hl6t/WpXK4u])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c10L-0004PV-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:56:26 -0400
Received: from ns.pigworks.openss7.net (IDENT:M+WOxdY9+EAa/iT7L0LZVfDJ8BFF9n5G@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6EAuPP12956
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 04:56:25 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6EAuPn14963
	for sigtran@ietf.org; Mon, 14 Jul 2003 04:56:25 -0600
Date: Mon, 14 Jul 2003 04:56:24 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Message-ID: <20030714045624.F13583@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "'sigtran@ietf.org'" <sigtran@ietf.org>
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.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: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com>; from LyOng@Ciena.com on Wed, Jul 02, 2003 at 10:21:28AM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Subject: [Sigtran] M2PA Last Call Comment - LI nit pick
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

M2PA-09 says:

  The Data field contains the following fields of the MTP Message Signal
  Unit (MSU):
  
     - Length Indicator (LI), including the two undefined bits between
       the SIO and LI fields.
     - Service Information Octet (SIO)
     - Signaling Information Field (SIF)
  
  The MTP MSU is described in Q.703 [Q.703], section 2.2 "Signal Unit
  Format", and T1.111.3 [T1.111] section 2.2 "Signal Unit Format".
  
  M2PA SHALL NOT add padding to the MTP3 message. 
  
  Note that the Data field SHALL NOT contain other components of the MTP
  MSU format:
  
     - Flag
     - Backward Sequence Number (BSN)
     - Backward Indicator Bit (BIB)
     - Forward Sequence Number (FSN)
     - Forward Indicator Bit (FIB)
     - Check bits (CK) 

  The Data field SHALL be transmitted in the byte order as defined by
  MTP3. 

  It is not necessary to put the message length in the LI octet as in
  MTP2. The LI octet is included because the two spare bits in the LI
  octet are used by MTP3 in at least one national version of SS7 to
  carry MTP3 information. For example, the Japanese TTC standard uses
  these spare bits as an MTP3 Message Priority field. See [JT-Q704],
  section 14 "Common Characteristics of message signal unit formats",
  section 14.2 (A) "Priority Indicator (PRI)". For versions of MTP that
  do not use these two bits, the entire octet is spare.

  Therefore in M2PA the format of the LI octet is:

      0               
      0 1 2 3 4 5 6 7 
     +-+-+-+-+-+-+-+-+
     |PRI|   spare   | (followed by SIO, SIF) 
     +-+-+-+-+-+-+-+-+
  
     PRI - Priority used only in national MTP defined in [JT-Q704].
           Spare for other MTP versions.

  Since the LI octet is not used for a message length, there is no need
  to support the expanded LI field in Q.703 [Q.703] Annex A. Therefore
  the LI field in M2PA is always one octet.

------

LI is not an octet, it is a field.  One that happens to not be included
in the M2PA Data field.

Change to:

------

  The Data field contains the following fields of the MTP Message Signal
  Unit (MSU):
  
     - the Message Priority field (PRI)
     - Service Information Octet (SIO)
     - Signaling Information Field (SIF)
  
  The MTP MSU is described in Q.703 [Q.703], section 2.2 "Signal Unit
  Format", and T1.111.3 [T1.111] section 2.2 "Signal Unit Format".
  The Japanese TTC standard uses the PRI field as an MTP3 Message
  Priority field [JT-Q703].  For versions of MTP that do not use these
  two bits, the entire first octet of the Data field is spare.

  The format of the first octet of the Data field is:

      0               
      0 1 2 3 4 5 6 7 
     +-+-+-+-+-+-+-+-+
     |PRI|   spare   | (followed by SIO, SIF) 
     +-+-+-+-+-+-+-+-+
  
     PRI - Priority used only in national MTP defined in [JT-Q703].
           Spare for other MTP versions.

  Note that the Data field SHALL NOT contain other components of the MTP
  MSU format:
  
     - Flag
     - Backward Sequence Number (BSN)
     - Backward Indicator Bit (BIB)
     - Forward Sequence Number (FSN)
     - Forward Indicator Bit (FIB)
     - Length Indicator (LI)
     - Check bits (CK) 

  The Data field SHALL be transmitted in the byte order as defined by
  MTP3. 

  M2PA SHALL NOT add padding to the MTP3 message. 


------

This could be less confusing that saying on the onset that the LI field
is part of the Data field and then going on to explain how it isn't.

--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 exim@www1.ietf.org  Mon Jul 14 06:58:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA07686
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 06:58:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c11t-0005gP-GQ
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:58:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EAw1ex021837
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 06:58:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c11t-0005g7-CU; Mon, 14 Jul 2003 06:58:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c11p-0005fs-Io
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 06:57:57 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA07618
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 06:57:52 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c11l-0004Ru-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:57:53 -0400
Received: from thoth.sbs.de ([192.35.17.2])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c11k-0004Rq-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 06:57:52 -0400
Received: from mail1.siemens.de (mail1.siemens.de [139.23.33.14])
	by thoth.sbs.de (8.11.7/8.11.7) with ESMTP id h6EAvnI29321;
	Mon, 14 Jul 2003 12:57:49 +0200 (MEST)
Received: from blues.mchh.siemens.de (blues.mchh.siemens.de [139.21.204.206])
	by mail1.siemens.de (8.11.7/8.11.7) with ESMTP id h6EAvna14374;
	Mon, 14 Jul 2003 12:57:49 +0200 (MEST)
Received: from mchh246e.demchh201e.icn.siemens.de (mchh246e.mchh.siemens.de [139.21.200.56])
	by blues.mchh.siemens.de (8.9.3/8.9.1) with ESMTP id MAA16980;
	Mon, 14 Jul 2003 12:56:42 +0200 (MET DST)
Received: by mchh246e.mchh.siemens.de with Internet Mail Service (5.5.2656.59)
	id <NTVTMAZ1>; Mon, 14 Jul 2003 12:57:15 +0200
Message-ID: <D73559B38D51D4119CB800508B9527B505AC196A@mchh266e.mchh.siemens.de>
From: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>,
        Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>
Cc: "List: IETF sigtran (E-Mail)" <sigtran@ietf.org>
Subject: RE: FW: [Sigtran] M2PA Last Call Comment
Date: Mon, 14 Jul 2003 12:57:14 +0200
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2656.59)
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>

Brian,

I can't fallow your arguements regarding "applicable MTP2 specification". 

In my picture of M2PA there is NO applicable MTP2 specification a certain side of M2PA has to adhere.

There is only one issue, if one side wants to perform proving for whatever reasons, the other side has to support that. 
Therefore actively starting proving could be flagged as an optional feature, whereas supporting the peer node in performing the proving is mandatory.

greetings

HannsJuergen

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Monday, July 14, 2003 12:36 PM
> To: Schwarzbauer Hanns Juergen
> Cc: List: IETF sigtran (E-Mail)
> Subject: Re: FW: [Sigtran] M2PA Last Call Comment
> 
> 
> Hanns Juergen,
> 
> I'll try again.  This has nothing to do with TDM or MTP3.
> 
> M2PA-09 says:
> 
>   4.  Procedures
> 
>   Since M2PA provides MTP3 with an interface and functionality like
>   MTP2, its internal functioning is similar to that of MTP2. 
> 
>   Except as modified in this document, M2PA SHOULD follow the
>   requirements of the applicable MTP2 specification. These may include
>   [Q.703] or [T1.111].
> 
> and then later:
> 
>   The proving period MAY be omitted if this is allowed by the 
> applicable
>   MTP2 standard (e. g., [Q.2210]).
> 
> So, M2PA permits one side to follow Q.2140 procedures and the 
> other side
> to follow Q.703 procedure, with regard to proving.  This is 
> incompatible
> as I have shown.
> 
> When following the requirements of the applicable MTP2 
> specification, it
> must be noted that the same MTP2 specification must be 
> followed on both
> sides, at least with regard to foregoing proving.  Thus the 
> text change.
> 
> --brian
> 
> Schwarzbauer Hanns Juergen wrote:                             
>                 (Mon, 14 Jul 2003 12:27:15)
> > Bidulock,
> > 
> > your answer is confusing. why should one side of an M2PA 
> link adhere to
> > Q.2140 and the other o Q.703 proving procedures? 
> >
> > You are still assuming a direct and tight coupling between 
> the TDM and the
> > IP side.
> >
> > In turn the throughput of the M2PA link could NEVER exceed 
> that of the
> > corresponding TDM link: 64 kbit in case of Q.703 (excluding 
> annex A for the
> > moment). That's not what M2PA was designed for!
> > 
> > greetings
> > 
> > HannsJuergen
> > 
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: Monday, July 14, 2003 12:22 PM
> > > To: Schwarzbauer Hanns Juergen
> > > Cc: List: IETF sigtran (E-Mail)
> > > Subject: Re: FW: [Sigtran] M2PA Last Call Comment
> > > 
> > > 
> > > Schwarzbauer,
> > > 
> > > Huh?  In the diagram I gave, it is only an M2PA link.  One side is
> > > following Q.2140 state machine procedures, the other is following
> > > Q.703 state machine procedures.
> > > 
> > > This has nothing to do with MTP3.
> > > 
> > > --brian
> > > 
> > > Schwarzbauer Hanns Juergen wrote:                             
> > >                 (Mon, 14 Jul 2003 12:14:56)
> > > > Brian,
> > > > 
> > > > sorry, I can't agree. 
> > > > My example scenario contains two SS7 nodes, both having 
> > > only M2PA links
> > > > between them. I assume that this is still a valid scenario. 
> > >  Furthermore, if
> > > > you have an SS7 node with more than one TDM based SS7 link 
> > > feeding this M2PA
> > > > link, what are you going to to there.
> > > > 
> > > > We are again going to the discussion we had several weeks ago: 
> > > > 
> > > > is there an MTP Level 3 coupling the TDM links and the M2PA 
> > > links and my
> > > > answer is still YES. Having that in mind, proving on one 
> > > link is completely
> > > > independent from proving on another link. 
> > > > 
> > > > 
> > > > greetings
> > > > 
> > > > HannsJuergen
> > > > 
> > > > > -----Original Message-----
> > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > Sent: Monday, July 14, 2003 11:40 AM
> > > > > To: Schwarzbauer Hanns Juergen
> > > > > Cc: List: IETF sigtran (E-Mail)
> > > > > Subject: Re: FW: [Sigtran] M2PA Last Call Comment
> > > > > 
> > > > > 
> > > > > Hanns Juergen,
> > > > > 
> > > > > Consider that one side follows Q.703 and requires proving 
> > > while the
> > > > > other side follows Q.2140 and foregoes proving.  The 
> > > following will
> > > > > occur:
> > > > > 
> > > > >     
> > > > > 
> > > 
> ___________________________________________________________________
> > > > >    |                                                          
> > > > >          |
> > > > >    |                Q.2140                      Q.703         
> > > > >          |
> > > > >    |                                                          
> > > > >          |
> > > > >    |        OUT-OF-SERVICE  ----------------->                
> > > > >          |
> > > > >    |                        <-----------------  
> > > > > OUT-OF-SERVICE         |
> > > > >    |                :start                                    
> > > > >          |
> > > > >    |             ALIGNMENT  ----------------->                
> > > > >          |
> > > > >    |                                            :start        
> > > > >          |
> > > > >    |                        <-----------------  ALIGNMENT     
> > > > >          |
> > > > >    |                 READY  ----------------->                
> > > > >          |
> > > > >    |                        <-----------------  
> > > > > PROVING-NORMAL         |
> > > > >    |                              !                           
> > > > >          |
> > > > >    |                              !  T3         1.0 <= T3 <= 
> > > > > 1.5       |
> > > > >    |                              !                           
> > > > >          |
> > > > >    |                        <-----------------  
> > > > > OUT-OF-SERVICE         |
> > > > >    |                                                          
> > > > >          |
> > > > >    
> > > > > 
> > > 
> |___________________________________________________________________|
> > > > > 
> > > > > Note that the Q.703 side must ignore the READY and await a
> > > > > PROVING-NORMAL or PROVING-EMERGENCY (See Q.781 test case 
> > > 2.3) and will
> > > > > time out (See Q.781 test case 1.3).
> > > > > 
> > > > > Therefore unless both sides follow the Q.2140 procedures 
> > > it will be
> > > > > impossible to forego proving and successfuly align the link.
> > > > > 
> > > > > The issue has little to do with MTP Level 3.  It is solely an 
> > > > > MTP Level
> > > > > 2 compatibility issue, speaks to interoperability, and the 
> > > > > keyword MUST
> > > > > should be used.  So,
> > > > > 
> > > > >   The proving period MAY be ommitted if this is allowed by the
> > > > >   applicable standard (e.g., [Q.2140]); however, the same 
> > > > > standard MUST
> > > > >   be followed on both sides of the M2PA link.
> > > > > 
> > > > > Does that clear things up for you?
> > > > > 
> > > > > --brian
> > > > > 
> > > > > Schwarzbauer Hanns Juergen wrote:                   (Mon, 14 
> > > > > Jul 2003 10:31:54)
> > > > > > resend, because "sigtran list" wasn't included
> > > > > > 
> > > > > > -----Original Message-----
> > > > > > From: Schwarzbauer Hanns Juergen 
> > > > > > Sent: Monday, July 14, 2003 8:46 AM
> > > > > > To: 'bidulock@openss7.org'
> > > > > > Subject: RE: [Sigtran] M2PA Last Call Comment
> > > > > > 
> > > > > > 
> > > > > > Brian,
> > > > > > 
> > > > > > I don't understand your second point. There is an MTP L3 
> > > > > between the TDM and
> > > > > > the IP side, so both sides are decoupled. Thus I can't 
> > > > > agree with this
> > > > > > point.
> > > > > > 
> > > > > > greetings
> > > > > > 
> > > > > > HannsJuergen
> > > > > > 
> > > > > > > -----Original Message-----
> > > > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > > > Sent: Saturday, July 12, 2003 7:04 AM
> > > > > > > To: 'sigtran@ietf.org'
> > > > > > > Subject: [Sigtran] M2PA Last Call Comment
> > > > > > > 
> > > > > > > 
> > > > > > >   The proving period MAY be omitted if this is 
> allowed by the 
> > > > > > > applicable
> > > > > > >   MTP2 standard (e. g., [Q.2210]).
> > > > > > > 
> > > > > > > The reference should be Q.2140.
> > > > > > > 
> > > > > > > It should also be noted that the same approach MUST be 
> > > > > > > followed on both sides
> > > > > > > the link.  Otherwise, one side following Q.703 
> will have T3 
> > > > > > > timer expire
> > > > > > > waiting for PROVING-NORMAL from the side 
> foregoing proving.
> > > > > > > 
> > > > > > > --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/
> > > 
> 
> -- 
> 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 exim@www1.ietf.org  Mon Jul 14 07:09:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA08478
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 07:09:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c1CY-0006YI-Dl
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 07:09:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EB92gU025177
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 07:09:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c1CX-0006Xx-Bb; Mon, 14 Jul 2003 07:09:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c1CJ-0006Xd-Ei
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 07:08:47 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA08423
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 07:08:42 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c1CE-0004gO-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 07:08:42 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[GMAhj/UDEToztMd3mXGNCa+9iTMgpn0r])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c1CD-0004gD-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 07:08:41 -0400
Received: from ns.pigworks.openss7.net (IDENT:PNIB6F2M6X7xdE9LvDIzgXWOPo1fmLNT@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6EB8fP13124;
	Mon, 14 Jul 2003 05:08:41 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6EB8e415293;
	Mon, 14 Jul 2003 05:08:40 -0600
Date: Mon, 14 Jul 2003 05:08:40 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>
Cc: "List: IETF sigtran (E-Mail)" <sigtran@ietf.org>
Subject: Re: FW: [Sigtran] M2PA Last Call Comment
Message-ID: <20030714050840.G13583@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>,
	"List: IETF sigtran (E-Mail)" <sigtran@ietf.org>
References: <D73559B38D51D4119CB800508B9527B505AC196A@mchh266e.mchh.siemens.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: <D73559B38D51D4119CB800508B9527B505AC196A@mchh266e.mchh.siemens.de>; from hannsjuergen.schwarzbauer@siemens.com on Mon, Jul 14, 2003 at 12:57:14PM +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>

Hanss Juergen,

> >   Except as modified in this document, M2PA SHOULD follow the
> >   requirements of the applicable MTP2 specification. These may include
> >   [Q.703] or [T1.111].

These "applicable MTP2 specifications" that M2PA-09 says that
I SHOULD follow.

--brian

Schwarzbauer Hanns Juergen wrote:                                             (Mon, 14 Jul 2003 12:57:14)
> Brian,
> 
> I can't fallow your arguements regarding "applicable MTP2 specification". 
> 
> In my picture of M2PA there is NO applicable MTP2 specification a certain
> side of M2PA has to adhere.
> 
> There is only one issue, if one side wants to perform proving for whatever
> reasons, the other side has to support that. 
>
> Therefore actively starting proving could be flagged as an optional feature,
> whereas supporting the peer node in performing the proving is mandatory.
> 
> greetings
> 
> HannsJuergen
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: Monday, July 14, 2003 12:36 PM
> > To: Schwarzbauer Hanns Juergen
> > Cc: List: IETF sigtran (E-Mail)
> > Subject: Re: FW: [Sigtran] M2PA Last Call Comment
> > 
> > 
> > Hanns Juergen,
> > 
> > I'll try again.  This has nothing to do with TDM or MTP3.
> > 
> > M2PA-09 says:
> > 
> >   4.  Procedures
> > 
> >   Since M2PA provides MTP3 with an interface and functionality like
> >   MTP2, its internal functioning is similar to that of MTP2. 
> > 
> >   Except as modified in this document, M2PA SHOULD follow the
> >   requirements of the applicable MTP2 specification. These may include
> >   [Q.703] or [T1.111].
> > 
> > and then later:
> > 
> >   The proving period MAY be omitted if this is allowed by the 
> > applicable
> >   MTP2 standard (e. g., [Q.2210]).
> > 
> > So, M2PA permits one side to follow Q.2140 procedures and the 
> > other side
> > to follow Q.703 procedure, with regard to proving.  This is 
> > incompatible
> > as I have shown.
> > 
> > When following the requirements of the applicable MTP2 
> > specification, it
> > must be noted that the same MTP2 specification must be 
> > followed on both
> > sides, at least with regard to foregoing proving.  Thus the 
> > text change.
> > 
> > --brian
> > 
> > Schwarzbauer Hanns Juergen wrote:                             
> >                 (Mon, 14 Jul 2003 12:27:15)
> > > Bidulock,
> > > 
> > > your answer is confusing. why should one side of an M2PA 
> > link adhere to
> > > Q.2140 and the other o Q.703 proving procedures? 
> > >
> > > You are still assuming a direct and tight coupling between 
> > the TDM and the
> > > IP side.
> > >
> > > In turn the throughput of the M2PA link could NEVER exceed 
> > that of the
> > > corresponding TDM link: 64 kbit in case of Q.703 (excluding 
> > annex A for the
> > > moment). That's not what M2PA was designed for!
> > > 
> > > greetings
> > > 
> > > HannsJuergen
> > > 
> > > > -----Original Message-----
> > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > Sent: Monday, July 14, 2003 12:22 PM
> > > > To: Schwarzbauer Hanns Juergen
> > > > Cc: List: IETF sigtran (E-Mail)
> > > > Subject: Re: FW: [Sigtran] M2PA Last Call Comment
> > > > 
> > > > 
> > > > Schwarzbauer,
> > > > 
> > > > Huh?  In the diagram I gave, it is only an M2PA link.  One side is
> > > > following Q.2140 state machine procedures, the other is following
> > > > Q.703 state machine procedures.
> > > > 
> > > > This has nothing to do with MTP3.
> > > > 
> > > > --brian
> > > > 
> > > > Schwarzbauer Hanns Juergen wrote:                             
> > > >                 (Mon, 14 Jul 2003 12:14:56)
> > > > > Brian,
> > > > > 
> > > > > sorry, I can't agree. 
> > > > > My example scenario contains two SS7 nodes, both having 
> > > > only M2PA links
> > > > > between them. I assume that this is still a valid scenario. 
> > > >  Furthermore, if
> > > > > you have an SS7 node with more than one TDM based SS7 link 
> > > > feeding this M2PA
> > > > > link, what are you going to to there.
> > > > > 
> > > > > We are again going to the discussion we had several weeks ago: 
> > > > > 
> > > > > is there an MTP Level 3 coupling the TDM links and the M2PA 
> > > > links and my
> > > > > answer is still YES. Having that in mind, proving on one 
> > > > link is completely
> > > > > independent from proving on another link. 
> > > > > 
> > > > > 
> > > > > greetings
> > > > > 
> > > > > HannsJuergen
> > > > > 
> > > > > > -----Original Message-----
> > > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > > Sent: Monday, July 14, 2003 11:40 AM
> > > > > > To: Schwarzbauer Hanns Juergen
> > > > > > Cc: List: IETF sigtran (E-Mail)
> > > > > > Subject: Re: FW: [Sigtran] M2PA Last Call Comment
> > > > > > 
> > > > > > 
> > > > > > Hanns Juergen,
> > > > > > 
> > > > > > Consider that one side follows Q.703 and requires proving 
> > > > while the
> > > > > > other side follows Q.2140 and foregoes proving.  The 
> > > > following will
> > > > > > occur:
> > > > > > 
> > > > > >     
> > > > > > 
> > > > 
> > ___________________________________________________________________
> > > > > >    |                                                          
> > > > > >          |
> > > > > >    |                Q.2140                      Q.703         
> > > > > >          |
> > > > > >    |                                                          
> > > > > >          |
> > > > > >    |        OUT-OF-SERVICE  ----------------->                
> > > > > >          |
> > > > > >    |                        <-----------------  
> > > > > > OUT-OF-SERVICE         |
> > > > > >    |                :start                                    
> > > > > >          |
> > > > > >    |             ALIGNMENT  ----------------->                
> > > > > >          |
> > > > > >    |                                            :start        
> > > > > >          |
> > > > > >    |                        <-----------------  ALIGNMENT     
> > > > > >          |
> > > > > >    |                 READY  ----------------->                
> > > > > >          |
> > > > > >    |                        <-----------------  
> > > > > > PROVING-NORMAL         |
> > > > > >    |                              !                           
> > > > > >          |
> > > > > >    |                              !  T3         1.0 <= T3 <= 
> > > > > > 1.5       |
> > > > > >    |                              !                           
> > > > > >          |
> > > > > >    |                        <-----------------  
> > > > > > OUT-OF-SERVICE         |
> > > > > >    |                                                          
> > > > > >          |
> > > > > >    
> > > > > > 
> > > > 
> > |___________________________________________________________________|
> > > > > > 
> > > > > > Note that the Q.703 side must ignore the READY and await a
> > > > > > PROVING-NORMAL or PROVING-EMERGENCY (See Q.781 test case 
> > > > 2.3) and will
> > > > > > time out (See Q.781 test case 1.3).
> > > > > > 
> > > > > > Therefore unless both sides follow the Q.2140 procedures 
> > > > it will be
> > > > > > impossible to forego proving and successfuly align the link.
> > > > > > 
> > > > > > The issue has little to do with MTP Level 3.  It is solely an 
> > > > > > MTP Level
> > > > > > 2 compatibility issue, speaks to interoperability, and the 
> > > > > > keyword MUST
> > > > > > should be used.  So,
> > > > > > 
> > > > > >   The proving period MAY be ommitted if this is allowed by the
> > > > > >   applicable standard (e.g., [Q.2140]); however, the same 
> > > > > > standard MUST
> > > > > >   be followed on both sides of the M2PA link.
> > > > > > 
> > > > > > Does that clear things up for you?
> > > > > > 
> > > > > > --brian
> > > > > > 
> > > > > > Schwarzbauer Hanns Juergen wrote:                   (Mon, 14 
> > > > > > Jul 2003 10:31:54)
> > > > > > > resend, because "sigtran list" wasn't included
> > > > > > > 
> > > > > > > -----Original Message-----
> > > > > > > From: Schwarzbauer Hanns Juergen 
> > > > > > > Sent: Monday, July 14, 2003 8:46 AM
> > > > > > > To: 'bidulock@openss7.org'
> > > > > > > Subject: RE: [Sigtran] M2PA Last Call Comment
> > > > > > > 
> > > > > > > 
> > > > > > > Brian,
> > > > > > > 
> > > > > > > I don't understand your second point. There is an MTP L3 
> > > > > > between the TDM and
> > > > > > > the IP side, so both sides are decoupled. Thus I can't 
> > > > > > agree with this
> > > > > > > point.
> > > > > > > 
> > > > > > > greetings
> > > > > > > 
> > > > > > > HannsJuergen
> > > > > > > 
> > > > > > > > -----Original Message-----
> > > > > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > > > > Sent: Saturday, July 12, 2003 7:04 AM
> > > > > > > > To: 'sigtran@ietf.org'
> > > > > > > > Subject: [Sigtran] M2PA Last Call Comment
> > > > > > > > 
> > > > > > > > 
> > > > > > > >   The proving period MAY be omitted if this is 
> > allowed by the 
> > > > > > > > applicable
> > > > > > > >   MTP2 standard (e. g., [Q.2210]).
> > > > > > > > 
> > > > > > > > The reference should be Q.2140.
> > > > > > > > 
> > > > > > > > It should also be noted that the same approach MUST be 
> > > > > > > > followed on both sides
> > > > > > > > the link.  Otherwise, one side following Q.703 
> > will have T3 
> > > > > > > > timer expire
> > > > > > > > waiting for PROVING-NORMAL from the side 
> > foregoing proving.
> > > > > > > > 
> > > > > > > > --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/
> > > > 
> > 
> > -- 
> > 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 exim@www1.ietf.org  Mon Jul 14 07:43:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA11952
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 07:43:33 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c1jU-0008MR-FH
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 07:43:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EBh4fW032138
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 07:43:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c1jR-0008M3-Cw; Mon, 14 Jul 2003 07:43:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c1jH-0008Li-Vj
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 07:42:52 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA11861
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 07:42:49 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c1jG-0005b8-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 07:42:50 -0400
Received: from goliath.siemens.de ([192.35.17.28])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c1jF-0005av-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 07:42:49 -0400
Received: from mail1.siemens.de (mail1.siemens.de [139.23.33.14])
	by goliath.siemens.de (8.11.7/8.11.7) with ESMTP id h6EBggo23928;
	Mon, 14 Jul 2003 13:42:42 +0200 (MEST)
Received: from blues.mchh.siemens.de (blues.mchh.siemens.de [139.21.204.206])
	by mail1.siemens.de (8.11.7/8.11.7) with ESMTP id h6EBgda08200;
	Mon, 14 Jul 2003 13:42:39 +0200 (MEST)
Received: from mchh248e.mchh.siemens.de (mchh248e.mchh.siemens.de [139.21.200.58])
	by blues.mchh.siemens.de (8.9.3/8.9.1) with ESMTP id NAA22285;
	Mon, 14 Jul 2003 13:41:26 +0200 (MET DST)
Received: by mchh248e.mchh.siemens.de with Internet Mail Service (5.5.2653.19)
	id <NTWN6BAF>; Mon, 14 Jul 2003 13:41:58 +0200
Message-ID: <D73559B38D51D4119CB800508B9527B505AC196C@mchh266e.mchh.siemens.de>
From: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>
Cc: "List: IETF sigtran (E-Mail)" <sigtran@ietf.org>
Subject: RE: FW: [Sigtran] M2PA Last Call Comment
Date: Mon, 14 Jul 2003 13:41:57 +0200
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>

Brian,

sorry for insisting, looking at the figure below, I see no "applicable MTP2 specification" neither for the STP nor the SEP:
In case of STP is it Q.2140 or Q.703?
in case of SEP there are no TDM links at all?

greetings

HannsJuergen

  

64kbitlink#1+---------+                     +---------+
    --------+         |                     |         |
64kbitlink#n|SS7 Node |     M2PA link       |SS7 Node |
    --------+-e.g. STP+---------------------+ e.g. SEP|
HSLlink #1  |         |                     |         |
    ........+         |                     |         |
HSLlink #m  |         |                     |         |
    ........+         |                     |         |
            +---------+                     +---------+

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Monday, July 14, 2003 1:09 PM
> To: Schwarzbauer Hanns Juergen
> Cc: List: IETF sigtran (E-Mail)
> Subject: Re: FW: [Sigtran] M2PA Last Call Comment
> 
> 
> Hanss Juergen,
> 
> > >   Except as modified in this document, M2PA SHOULD follow the
> > >   requirements of the applicable MTP2 specification. 
> These may include
> > >   [Q.703] or [T1.111].
> 
> These "applicable MTP2 specifications" that M2PA-09 says that
> I SHOULD follow.
> 
> --brian
> 
> Schwarzbauer Hanns Juergen wrote:                             
>                 (Mon, 14 Jul 2003 12:57:14)
> > Brian,
> > 
> > I can't fallow your arguements regarding "applicable MTP2 
> specification". 
> > 
> > In my picture of M2PA there is NO applicable MTP2 
> specification a certain
> > side of M2PA has to adhere.
> > 
> > There is only one issue, if one side wants to perform 
> proving for whatever
> > reasons, the other side has to support that. 
> >
> > Therefore actively starting proving could be flagged as an 
> optional feature,
> > whereas supporting the peer node in performing the proving 
> is mandatory.
> > 
> > greetings
> > 
> > HannsJuergen
> > 
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: Monday, July 14, 2003 12:36 PM
> > > To: Schwarzbauer Hanns Juergen
> > > Cc: List: IETF sigtran (E-Mail)
> > > Subject: Re: FW: [Sigtran] M2PA Last Call Comment
> > > 
> > > 
> > > Hanns Juergen,
> > > 
> > > I'll try again.  This has nothing to do with TDM or MTP3.
> > > 
> > > M2PA-09 says:
> > > 
> > >   4.  Procedures
> > > 
> > >   Since M2PA provides MTP3 with an interface and 
> functionality like
> > >   MTP2, its internal functioning is similar to that of MTP2. 
> > > 
> > >   Except as modified in this document, M2PA SHOULD follow the
> > >   requirements of the applicable MTP2 specification. 
> These may include
> > >   [Q.703] or [T1.111].
> > > 
> > > and then later:
> > > 
> > >   The proving period MAY be omitted if this is allowed by the 
> > > applicable
> > >   MTP2 standard (e. g., [Q.2210]).
> > > 
> > > So, M2PA permits one side to follow Q.2140 procedures and the 
> > > other side
> > > to follow Q.703 procedure, with regard to proving.  This is 
> > > incompatible
> > > as I have shown.
> > > 
> > > When following the requirements of the applicable MTP2 
> > > specification, it
> > > must be noted that the same MTP2 specification must be 
> > > followed on both
> > > sides, at least with regard to foregoing proving.  Thus the 
> > > text change.
> > > 
> > > --brian
> > > 
> > > Schwarzbauer Hanns Juergen wrote:                             
> > >                 (Mon, 14 Jul 2003 12:27:15)
> > > > Bidulock,
> > > > 
> > > > your answer is confusing. why should one side of an M2PA 
> > > link adhere to
> > > > Q.2140 and the other o Q.703 proving procedures? 
> > > >
> > > > You are still assuming a direct and tight coupling between 
> > > the TDM and the
> > > > IP side.
> > > >
> > > > In turn the throughput of the M2PA link could NEVER exceed 
> > > that of the
> > > > corresponding TDM link: 64 kbit in case of Q.703 (excluding 
> > > annex A for the
> > > > moment). That's not what M2PA was designed for!
> > > > 
> > > > greetings
> > > > 
> > > > HannsJuergen
> > > > 
> > > > > -----Original Message-----
> > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > Sent: Monday, July 14, 2003 12:22 PM
> > > > > To: Schwarzbauer Hanns Juergen
> > > > > Cc: List: IETF sigtran (E-Mail)
> > > > > Subject: Re: FW: [Sigtran] M2PA Last Call Comment
> > > > > 
> > > > > 
> > > > > Schwarzbauer,
> > > > > 
> > > > > Huh?  In the diagram I gave, it is only an M2PA link. 
>  One side is
> > > > > following Q.2140 state machine procedures, the other 
> is following
> > > > > Q.703 state machine procedures.
> > > > > 
> > > > > This has nothing to do with MTP3.
> > > > > 
> > > > > --brian
> > > > > 
> > > > > Schwarzbauer Hanns Juergen wrote:                             
> > > > >                 (Mon, 14 Jul 2003 12:14:56)
> > > > > > Brian,
> > > > > > 
> > > > > > sorry, I can't agree. 
> > > > > > My example scenario contains two SS7 nodes, both having 
> > > > > only M2PA links
> > > > > > between them. I assume that this is still a valid scenario. 
> > > > >  Furthermore, if
> > > > > > you have an SS7 node with more than one TDM based SS7 link 
> > > > > feeding this M2PA
> > > > > > link, what are you going to to there.
> > > > > > 
> > > > > > We are again going to the discussion we had several 
> weeks ago: 
> > > > > > 
> > > > > > is there an MTP Level 3 coupling the TDM links and the M2PA 
> > > > > links and my
> > > > > > answer is still YES. Having that in mind, proving on one 
> > > > > link is completely
> > > > > > independent from proving on another link. 
> > > > > > 
> > > > > > 
> > > > > > greetings
> > > > > > 
> > > > > > HannsJuergen
> > > > > > 
> > > > > > > -----Original Message-----
> > > > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > > > Sent: Monday, July 14, 2003 11:40 AM
> > > > > > > To: Schwarzbauer Hanns Juergen
> > > > > > > Cc: List: IETF sigtran (E-Mail)
> > > > > > > Subject: Re: FW: [Sigtran] M2PA Last Call Comment
> > > > > > > 
> > > > > > > 
> > > > > > > Hanns Juergen,
> > > > > > > 
> > > > > > > Consider that one side follows Q.703 and requires proving 
> > > > > while the
> > > > > > > other side follows Q.2140 and foregoes proving.  The 
> > > > > following will
> > > > > > > occur:
> > > > > > > 
> > > > > > >     
> > > > > > > 
> > > > > 
> > > 
> ___________________________________________________________________
> > > > > > >    |                                              
>             
> > > > > > >          |
> > > > > > >    |                Q.2140                      
> Q.703         
> > > > > > >          |
> > > > > > >    |                                              
>             
> > > > > > >          |
> > > > > > >    |        OUT-OF-SERVICE  ----------------->    
>             
> > > > > > >          |
> > > > > > >    |                        <-----------------  
> > > > > > > OUT-OF-SERVICE         |
> > > > > > >    |                :start                        
>             
> > > > > > >          |
> > > > > > >    |             ALIGNMENT  ----------------->    
>             
> > > > > > >          |
> > > > > > >    |                                            
> :start        
> > > > > > >          |
> > > > > > >    |                        <-----------------  
> ALIGNMENT     
> > > > > > >          |
> > > > > > >    |                 READY  ----------------->    
>             
> > > > > > >          |
> > > > > > >    |                        <-----------------  
> > > > > > > PROVING-NORMAL         |
> > > > > > >    |                              !               
>             
> > > > > > >          |
> > > > > > >    |                              !  T3         
> 1.0 <= T3 <= 
> > > > > > > 1.5       |
> > > > > > >    |                              !               
>             
> > > > > > >          |
> > > > > > >    |                        <-----------------  
> > > > > > > OUT-OF-SERVICE         |
> > > > > > >    |                                              
>             
> > > > > > >          |
> > > > > > >    
> > > > > > > 
> > > > > 
> > > 
> |___________________________________________________________________|
> > > > > > > 
> > > > > > > Note that the Q.703 side must ignore the READY and await a
> > > > > > > PROVING-NORMAL or PROVING-EMERGENCY (See Q.781 test case 
> > > > > 2.3) and will
> > > > > > > time out (See Q.781 test case 1.3).
> > > > > > > 
> > > > > > > Therefore unless both sides follow the Q.2140 procedures 
> > > > > it will be
> > > > > > > impossible to forego proving and successfuly 
> align the link.
> > > > > > > 
> > > > > > > The issue has little to do with MTP Level 3.  It 
> is solely an 
> > > > > > > MTP Level
> > > > > > > 2 compatibility issue, speaks to 
> interoperability, and the 
> > > > > > > keyword MUST
> > > > > > > should be used.  So,
> > > > > > > 
> > > > > > >   The proving period MAY be ommitted if this is 
> allowed by the
> > > > > > >   applicable standard (e.g., [Q.2140]); however, the same 
> > > > > > > standard MUST
> > > > > > >   be followed on both sides of the M2PA link.
> > > > > > > 
> > > > > > > Does that clear things up for you?
> > > > > > > 
> > > > > > > --brian
> > > > > > > 
> > > > > > > Schwarzbauer Hanns Juergen wrote:                 
>   (Mon, 14 
> > > > > > > Jul 2003 10:31:54)
> > > > > > > > resend, because "sigtran list" wasn't included
> > > > > > > > 
> > > > > > > > -----Original Message-----
> > > > > > > > From: Schwarzbauer Hanns Juergen 
> > > > > > > > Sent: Monday, July 14, 2003 8:46 AM
> > > > > > > > To: 'bidulock@openss7.org'
> > > > > > > > Subject: RE: [Sigtran] M2PA Last Call Comment
> > > > > > > > 
> > > > > > > > 
> > > > > > > > Brian,
> > > > > > > > 
> > > > > > > > I don't understand your second point. There is 
> an MTP L3 
> > > > > > > between the TDM and
> > > > > > > > the IP side, so both sides are decoupled. Thus I can't 
> > > > > > > agree with this
> > > > > > > > point.
> > > > > > > > 
> > > > > > > > greetings
> > > > > > > > 
> > > > > > > > HannsJuergen
> > > > > > > > 
> > > > > > > > > -----Original Message-----
> > > > > > > > > From: Brian F. G. Bidulock 
> [mailto:bidulock@openss7.org]
> > > > > > > > > Sent: Saturday, July 12, 2003 7:04 AM
> > > > > > > > > To: 'sigtran@ietf.org'
> > > > > > > > > Subject: [Sigtran] M2PA Last Call Comment
> > > > > > > > > 
> > > > > > > > > 
> > > > > > > > >   The proving period MAY be omitted if this is 
> > > allowed by the 
> > > > > > > > > applicable
> > > > > > > > >   MTP2 standard (e. g., [Q.2210]).
> > > > > > > > > 
> > > > > > > > > The reference should be Q.2140.
> > > > > > > > > 
> > > > > > > > > It should also be noted that the same 
> approach MUST be 
> > > > > > > > > followed on both sides
> > > > > > > > > the link.  Otherwise, one side following Q.703 
> > > will have T3 
> > > > > > > > > timer expire
> > > > > > > > > waiting for PROVING-NORMAL from the side 
> > > foregoing proving.
> > > > > > > > > 
> > > > > > > > > --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/
> > > > > 
> > > 
> > > -- 
> > > 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 exim@www1.ietf.org  Mon Jul 14 10:08:32 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA22196
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 10:08:32 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c3zo-00079W-SG
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 10:08:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EE84va027488
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 10:08:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c3zm-00078q-8s; Mon, 14 Jul 2003 10:08:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c3zj-000789-Fl
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 10:07:59 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA22103
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 10:07:55 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c3zh-0000AX-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 10:07:57 -0400
Received: from auds953.usa.alcatel.com ([143.209.238.6])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c3zg-00009p-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 10:07:56 -0400
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6EE7BLZ023591;
	Mon, 14 Jul 2003 09:07:11 -0500 (CDT)
Received: from sun2685.ssd.usa.alcatel.com (sun2685.ssd.usa.alcatel.com [143.209.151.167])
	by ssd.usa.alcatel.com (8.12.8/8.12.8) with ESMTP id h6EE6a1C021014;
	Mon, 14 Jul 2003 09:06:36 -0500 (CDT)
Date: Mon, 14 Jul 2003 09:06:36 -0500 (CDT)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
cc: "'Brian F. G. Bidulock '" <bidulock@openss7.org>,
        "''sigtran@ietf.org' '" <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA IG v04  Section 3.14
In-Reply-To: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CFE@eestqnt105.es.eu.ericsson.se>
Message-ID: <Pine.SOL.4.10.10307140703090.21400-100000@sun2685.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,Brian,


 (i was bit late to enter into this conversation)

 Including a new status type 'AS-DOWN' in NTFY messages is a good idea,
 but section 4.3.4.5 in RFC says an ASP in DOWN state will not process
 NTFY message....?


 Let me take an example of n=5, SGP needs 5 ASPes in ACTIVE state
 to declare the AS as ACTIVE(true-only for start case).

 Start-case
 ----------

 ACTIVE ASPes    INACTIVE ASPes  DOWN ASPes    AS state        case-#
 --------------------------------------------------------------------
    0             0               5            DOWN            case-1
    0             1               4            DOWN            case-2
    0             2               3            DOWN            case-3
    .
    .
    .
    0             5               0            INACTIVE        case-4
    1             4               0            INACTIVE        case-5
    .
    .
    .
    5             0               0            ACTIVE          case-6

   (assume traffic flows)

    Non-start case
    --------------
    4             1               0            ACTIVE          case-7
    3             2               0            ACTIVE          case-8
    .
    .
    .
    1             4               0            ACTIVE          case-8
    0             5               0            INACTIVE        case-9
    0             4               1            still INACTIVE  case-10
    0             3               2            still INACTIVE  case-11
    .
    .
    .
    0             1               4            still INACTIVE  case-12
    0             0               5            DOWN            case-13


 I think we should define the following state-conversion definition.

 for Start-case
 --------------

 DN2IA : One ASP moves to ASP-INACTIVE causing that n ASPs are in
         ASP-INACTIVE states.

 IA2DN : One ASP moves to ASP-DOWN causing that n ASPs are in
         ASP-DOWN state.

 IA2AC : one ASP moves to ASP-ACTIVE, causing number of ASPs are in 
         ASP-ACTIVE state.


Waiting for comments to go ahead.


--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 Mon, 14 Jul 2003, J.Javier Pastor (EE/EEM) wrote:

> Brian, 
> 
> What you say it's right, only 1 'n' transition would be kept in the proposal: the IA2AC transition.
> 
> I think it makes sense since "n" refers only to the capacity of processing traffic and in the AS-INACTIVE state no traffic is processed.
> 
> Does anyone have any objection to this change?
> 
> 
> Thanks // Javier.
> 
> -----Original Message-----
> From: Brian F. G. Bidulock
> To: J.Javier Pastor (EE/EEM)
> Cc: 'sigtran@ietf.org'
> Sent: 11/07/2003 11:13
> Subject: Re: [Sigtran] M3UA IG v04  Section 3.14
> 
> J.Javier,
> 
> J.Javier Pastor (EE/EEM) wrote:
> (Fri, 11 Jul 2003 10:35:57)
> > Hi Brian, 
> > 
> > You are right saying that there is currently no way of notifying
> AS-DOWN. In
> > fact "n" only applies to the traffic case.
> > 
> > Then I see two options:
> > 	1- include AS-DOWN in NTFY
> > 	2- The AS-DOWN state is only got when all the ASPs are in
> ASP-DOWN state.
> > 
> > I would addopt 2 since it is inline what we are having in the other
> UALs. 
> 
> Yes, please go with 2 as it is also in line with the RFC.
> 
> That will also leave only 1 'n' transition in the proposal: the IA2AC
> transition, correct?
> 
> --brian
> 
> > 
> > Thanks // Javier.
> > 
> > 
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: jueves, 10 de julio de 2003 17:15
> > > To: J.Javier Pastor (EE/EEM)
> > > Cc: 'sigtran@ietf.org'
> > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.14
> > > 
> > > 
> > > Javier,
> > > 
> > > Section 3.14 says:
> > > 
> > >    IA2DN: One ASP moves to ASP-DOWN causing that the number of ASPs
> in
> > >    either ASP-ACTIVE or ASP-INACTIVE drops below n.
> > > 
> > > Am I correct to say that if n=5 and there are 4 ASPs in the 
> > > ASP-INACTIVE
> > > and 1 ASP in the ASP-ACTIVE state, and then the 1 active ASP moves
> to
> > > the ASP-DOWN state and T(r) expires that the AS is to move to 
> > > the AS-DOWN
> > > state?  As follows:
> > > 
> > >   If T(r) expires before an ASP becomes ASP-ACTIVE, the SGP MAY stop
> > >   queuing messages and discards all
> > >   previously queued messages. The AS will move to the 
> > > AS-INACTIVE state
> > >   if at least the number of ASPs in ASP-INACTIVE sum n, otherwise it
> > >   will move to AS-DOWN state.
> > > 
> > > Now, that is an AS state change and we must send a NTFY to 
> > > all the INACTIVE
> > > ASPs (the 4 left inactive) of the state change.  But from 
> > > Section 3.8.2 of
> > > the RFC, there is no message to notify of an AS-DOWN state:
> > > 
> > >          1    reserved
> > >          2    Application Server Inactive (AS-INACTIVE)
> > >          3    Application Server Active (AS-ACTIVE)
> > >          4    Application Server Pending (AS-PENDING)
> > > 
> > > You see, we removed it because there was never an instance in which
> it
> > > would be sent.  You have inserted an instance again.
> > > 
> > > Aren't you going to have to define a Notify message (the 
> > > Reserved value)
> > > for "Application Server Down (AS-DOWN)" so that the INACTIVE ASPs
> can
> > > be properly notified by the SGP?
> > > 
> > > Particularly when the RFC says in Section 4.3.4.5: (Notify
> Procedures)
> > > 
> > >    A Notify message reflecting a change in the AS state MUST 
> > > be sent to
> > >    all ASPs in the AS, except those in the ASP-DOWN state, with
> > >    appropriate Status Information and any ASP Identifier of the
> failed
> > >    ASP.  At the ASP, Layer Management is informed with an M-NOTIFY
> > >    indication primitive.  The Notify message must be sent 
> > > whether the AS
> > >    state change was a result of an ASP failure or reception of an
> ASP
> > >    State management (ASPSM) / ASP Traffic Management (ASPTM)
> message.
> > >    In the second case, the Notify message MUST be sent after 
> > > any related
> > >    acknowledgement messages  (e.g., ASP Up Ack, ASP Down Ack, 
> > > ASP Active
> > >    Ack, or ASP Inactive Ack).
> > > 
> > > The above tells me that I MUST send a Notify message becase of the
> AS
> > > state change, but there is no message to inform about AS-DOWN state
> > > transition.
> > > 
> > > What is the purpose of doing this?  What is the protocol 
> > > difference between
> > > the AS-DOWN state and the AS-INACTIVE state?  What is it that the
> SGP
> > > does or doesn't do in the one state that it doesn't or does 
> > > do in the other?
> > > 
> > > I don't recall this being discussed at all.
> > > 
> > > Please clarify.
> > > 
> > > -- 
> > > 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 exim@www1.ietf.org  Mon Jul 14 13:03:32 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA05003
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 13:03:32 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c6jA-0001cL-Tc
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 13:03:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EH34wb006199
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 13:03:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c6j7-0001bj-Bx; Mon, 14 Jul 2003 13:03:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c6iX-0001a2-71
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 13:02:25 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA04841
	for <SIGTRAN@ietf.org>; Mon, 14 Jul 2003 13:02:20 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c6iV-00034x-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 13:02:23 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19c6iT-00034r-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 13:02:21 -0400
Received: from [172.16.1.40] (helo=ccpu00251)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19c6iK-0001ha-Oe
	for <SIGTRAN@ietf.org>; Mon, 14 Jul 2003 10:02:12 -0700
Reply-To: <umesh@ccpu.com>
From: "Umesh Vats" <umesh@ccpu.com>
To: "SIGTRAN e-mail list" <SIGTRAN@ietf.org>
Date: Mon, 14 Jul 2003 10:07:44 -0700
Message-ID: <ACEMIJONPBBPDLOKCABHMEDFCBAA.umesh@ccpu.com>
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 IMO, Build 9.0.2416 (9.0.2911.0)
In-Reply-To: <20030711185353.A17103@openss7.org>
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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

Right now, if ASP sends REG-REQ for an already registered routing key,
SGP sends unsuccesful REG-RSP. 
But in some scenarios, succesful REG-RSP from SGP to ASP may be lost 
due to network problems/local node problems. In these cases, ASP may
retry REG-REQ. 
In such scenarios (duplicate REG-REQ), as per M3UA IG V03 SGP sends REG-RSP
with error as "Error - Cannot Support Unique Routing"

Is it possible for SGP to include the Routing Context (RC assigned by SGP
for the local rk id received in the first REG-REQ from ASP) in addition 
to the "Local_Routing Key Identifier" in the unsuccessful REG-RSP message
mentioned above. This will allow ASP to recover the RC value.

Thanks
Umesh

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003


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



From exim@www1.ietf.org  Mon Jul 14 13:22:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA06346
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 13:22:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c71V-0002xD-E1
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 13:22:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EHM1mu011340
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 13:22:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c71U-0002wn-F5; Mon, 14 Jul 2003 13:22:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c70w-0002wR-4y
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 13:21:26 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA06291
	for <SIGTRAN@ietf.org>; Mon, 14 Jul 2003 13:21:21 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c70u-0003SX-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 13:21:24 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19c70t-0003SU-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 13:21:23 -0400
Received: from [172.16.1.203] (helo=CCPU00186)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19c70g-0003GS-Vd; Mon, 14 Jul 2003 10:21:10 -0700
From: "VIvek Toky" <vivektoky@ccpu.com>
To: <umesh@ccpu.com>, "SIGTRAN e-mail list" <SIGTRAN@ietf.org>
Subject: RE: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Date: Mon, 14 Jul 2003 10:25:50 -0700
Message-ID: <HKEBKFBNPIDALDOKLGKDAEHCCDAA.vivektoky@ccpu.com>
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 IMO, Build 9.0.6604 (9.0.2911.0)
In-Reply-To: <ACEMIJONPBBPDLOKCABHMEDFCBAA.umesh@ccpu.com>
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
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 Umesh is right. If REG Response is lost ASP will never be able to
register the same RK again. As it doesn't know the RC so it would not be
able to de-register it and it can also not register it again.

-Vivek.

-----Original Message-----
From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf Of
Umesh Vats
Sent: Monday, July 14, 2003 10:08 AM
To: SIGTRAN e-mail list
Subject: [Sigtran] M3UA : RC in unsuccesful REG-RSP


Right now, if ASP sends REG-REQ for an already registered routing key,
SGP sends unsuccesful REG-RSP.
But in some scenarios, succesful REG-RSP from SGP to ASP may be lost
due to network problems/local node problems. In these cases, ASP may
retry REG-REQ.
In such scenarios (duplicate REG-REQ), as per M3UA IG V03 SGP sends REG-RSP
with error as "Error - Cannot Support Unique Routing"

Is it possible for SGP to include the Routing Context (RC assigned by SGP
for the local rk id received in the first REG-REQ from ASP) in addition
to the "Local_Routing Key Identifier" in the unsuccessful REG-RSP message
mentioned above. This will allow ASP to recover the RC value.

Thanks
Umesh

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003


_______________________________________________
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 exim@www1.ietf.org  Mon Jul 14 14:20:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA10992
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 14:20:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c7vf-0007PK-AY
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 14:20:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EIK3dM028465
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 14:20:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c7ve-0007Oi-8o; Mon, 14 Jul 2003 14:20:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c7vL-0007LM-8S
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 14:19:43 -0400
Received: from gw.openss7.com (IDENT:wLLL58lceLRgkUrwHZgUvldyFNIk0Mmz@gw.openss7.com [142.179.199.224])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA10870
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 14:19:38 -0400 (EDT)
Received: from ns.pigworks.openss7.net (IDENT:NS8uZoLZCbeK6fEoXTaS/w39WI6KoweS@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6EIJcP21264;
	Mon, 14 Jul 2003 12:19:38 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6EIJbe23043;
	Mon, 14 Jul 2003 12:19:37 -0600
Date: Mon, 14 Jul 2003 12:19:37 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
        "''sigtran@ietf.org' '" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04  Section 3.14
Message-ID: <20030714121937.A22972@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
	"''sigtran@ietf.org' '" <sigtran@ietf.org>
References: <1AB3D30B989BF141BBD5C70057B2EF7C01B65CFE@eestqnt105.es.eu.ericsson.se> <Pine.SOL.4.10.10307140703090.21400-100000@sun2685.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.4.10.10307140703090.21400-100000@sun2685.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Mon, Jul 14, 2003 at 09:06:36AM -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>

Samuel,

To what end?

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Mon, 14 Jul 2003 09:06:36)
> 
> Javier,Brian,
> 
> 
>  (i was bit late to enter into this conversation)
> 
>  Including a new status type 'AS-DOWN' in NTFY messages is a good idea,
>  but section 4.3.4.5 in RFC says an ASP in DOWN state will not process
>  NTFY message....?
> 
> 
>  Let me take an example of n=5, SGP needs 5 ASPes in ACTIVE state
>  to declare the AS as ACTIVE(true-only for start case).
> 
>  Start-case
>  ----------
> 
>  ACTIVE ASPes    INACTIVE ASPes  DOWN ASPes    AS state        case-#
>  --------------------------------------------------------------------
>     0             0               5            DOWN            case-1
>     0             1               4            DOWN            case-2
>     0             2               3            DOWN            case-3
>     .
>     .
>     .
>     0             5               0            INACTIVE        case-4
>     1             4               0            INACTIVE        case-5
>     .
>     .
>     .
>     5             0               0            ACTIVE          case-6
> 
>    (assume traffic flows)
> 
>     Non-start case
>     --------------
>     4             1               0            ACTIVE          case-7
>     3             2               0            ACTIVE          case-8
>     .
>     .
>     .
>     1             4               0            ACTIVE          case-8
>     0             5               0            INACTIVE        case-9
>     0             4               1            still INACTIVE  case-10
>     0             3               2            still INACTIVE  case-11
>     .
>     .
>     .
>     0             1               4            still INACTIVE  case-12
>     0             0               5            DOWN            case-13
> 
> 
>  I think we should define the following state-conversion definition.
> 
>  for Start-case
>  --------------
> 
>  DN2IA : One ASP moves to ASP-INACTIVE causing that n ASPs are in
>          ASP-INACTIVE states.
> 
>  IA2DN : One ASP moves to ASP-DOWN causing that n ASPs are in
>          ASP-DOWN state.
> 
>  IA2AC : one ASP moves to ASP-ACTIVE, causing number of ASPs are in 
>          ASP-ACTIVE state.
> 
> 
> Waiting for comments to go ahead.
> 
> 
> --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 Mon, 14 Jul 2003, J.Javier Pastor (EE/EEM) wrote:
> 
> > Brian, 
> > 
> > What you say it's right, only 1 'n' transition would be kept in the proposal: the IA2AC transition.
> > 
> > I think it makes sense since "n" refers only to the capacity of processing traffic and in the AS-INACTIVE state no traffic is processed.
> > 
> > Does anyone have any objection to this change?
> > 
> > 
> > Thanks // Javier.
> > 
> > -----Original Message-----
> > From: Brian F. G. Bidulock
> > To: J.Javier Pastor (EE/EEM)
> > Cc: 'sigtran@ietf.org'
> > Sent: 11/07/2003 11:13
> > Subject: Re: [Sigtran] M3UA IG v04  Section 3.14
> > 
> > J.Javier,
> > 
> > J.Javier Pastor (EE/EEM) wrote:
> > (Fri, 11 Jul 2003 10:35:57)
> > > Hi Brian, 
> > > 
> > > You are right saying that there is currently no way of notifying
> > AS-DOWN. In
> > > fact "n" only applies to the traffic case.
> > > 
> > > Then I see two options:
> > > 	1- include AS-DOWN in NTFY
> > > 	2- The AS-DOWN state is only got when all the ASPs are in
> > ASP-DOWN state.
> > > 
> > > I would addopt 2 since it is inline what we are having in the other
> > UALs. 
> > 
> > Yes, please go with 2 as it is also in line with the RFC.
> > 
> > That will also leave only 1 'n' transition in the proposal: the IA2AC
> > transition, correct?
> > 
> > --brian
> > 
> > > 
> > > Thanks // Javier.
> > > 
> > > 
> > > > -----Original Message-----
> > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > Sent: jueves, 10 de julio de 2003 17:15
> > > > To: J.Javier Pastor (EE/EEM)
> > > > Cc: 'sigtran@ietf.org'
> > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.14
> > > > 
> > > > 
> > > > Javier,
> > > > 
> > > > Section 3.14 says:
> > > > 
> > > >    IA2DN: One ASP moves to ASP-DOWN causing that the number of ASPs
> > in
> > > >    either ASP-ACTIVE or ASP-INACTIVE drops below n.
> > > > 
> > > > Am I correct to say that if n=5 and there are 4 ASPs in the 
> > > > ASP-INACTIVE
> > > > and 1 ASP in the ASP-ACTIVE state, and then the 1 active ASP moves
> > to
> > > > the ASP-DOWN state and T(r) expires that the AS is to move to 
> > > > the AS-DOWN
> > > > state?  As follows:
> > > > 
> > > >   If T(r) expires before an ASP becomes ASP-ACTIVE, the SGP MAY stop
> > > >   queuing messages and discards all
> > > >   previously queued messages. The AS will move to the 
> > > > AS-INACTIVE state
> > > >   if at least the number of ASPs in ASP-INACTIVE sum n, otherwise it
> > > >   will move to AS-DOWN state.
> > > > 
> > > > Now, that is an AS state change and we must send a NTFY to 
> > > > all the INACTIVE
> > > > ASPs (the 4 left inactive) of the state change.  But from 
> > > > Section 3.8.2 of
> > > > the RFC, there is no message to notify of an AS-DOWN state:
> > > > 
> > > >          1    reserved
> > > >          2    Application Server Inactive (AS-INACTIVE)
> > > >          3    Application Server Active (AS-ACTIVE)
> > > >          4    Application Server Pending (AS-PENDING)
> > > > 
> > > > You see, we removed it because there was never an instance in which
> > it
> > > > would be sent.  You have inserted an instance again.
> > > > 
> > > > Aren't you going to have to define a Notify message (the 
> > > > Reserved value)
> > > > for "Application Server Down (AS-DOWN)" so that the INACTIVE ASPs
> > can
> > > > be properly notified by the SGP?
> > > > 
> > > > Particularly when the RFC says in Section 4.3.4.5: (Notify
> > Procedures)
> > > > 
> > > >    A Notify message reflecting a change in the AS state MUST 
> > > > be sent to
> > > >    all ASPs in the AS, except those in the ASP-DOWN state, with
> > > >    appropriate Status Information and any ASP Identifier of the
> > failed
> > > >    ASP.  At the ASP, Layer Management is informed with an M-NOTIFY
> > > >    indication primitive.  The Notify message must be sent 
> > > > whether the AS
> > > >    state change was a result of an ASP failure or reception of an
> > ASP
> > > >    State management (ASPSM) / ASP Traffic Management (ASPTM)
> > message.
> > > >    In the second case, the Notify message MUST be sent after 
> > > > any related
> > > >    acknowledgement messages  (e.g., ASP Up Ack, ASP Down Ack, 
> > > > ASP Active
> > > >    Ack, or ASP Inactive Ack).
> > > > 
> > > > The above tells me that I MUST send a Notify message becase of the
> > AS
> > > > state change, but there is no message to inform about AS-DOWN state
> > > > transition.
> > > > 
> > > > What is the purpose of doing this?  What is the protocol 
> > > > difference between
> > > > the AS-DOWN state and the AS-INACTIVE state?  What is it that the
> > SGP
> > > > does or doesn't do in the one state that it doesn't or does 
> > > > do in the other?
> > > > 
> > > > I don't recall this being discussed at all.
> > > > 
> > > > Please clarify.
> > > > 
> > > > -- 
> > > > 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 exim@www1.ietf.org  Mon Jul 14 14:22:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA11171
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 14:22:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c7xa-0007hE-88
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 14:22:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EIM2Up029581
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 14:22:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c7xZ-0007gr-FI; Mon, 14 Jul 2003 14:22:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c7x2-0007eF-81
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 14:21:28 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA11073
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 14:21:25 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c7wz-0004gI-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 14:21:25 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[yePha4SBqgV0WWLv4yg4TMU/zylUGZhn])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c7wx-0004g1-00
	for sigtran@ietf.org; Mon, 14 Jul 2003 14:21:23 -0400
Received: from ns.pigworks.openss7.net (IDENT:0hTOuI2n+tSbnx648g+ic8hdzgXW+yDi@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6EILMP21296;
	Mon, 14 Jul 2003 12:21:22 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6EILLp23078;
	Mon, 14 Jul 2003 12:21:21 -0600
Date: Mon, 14 Jul 2003 12:21:21 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>
Cc: "List: IETF sigtran (E-Mail)" <sigtran@ietf.org>
Subject: Re: FW: [Sigtran] M2PA Last Call Comment
Message-ID: <20030714122121.B22972@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>,
	"List: IETF sigtran (E-Mail)" <sigtran@ietf.org>
References: <D73559B38D51D4119CB800508B9527B505AC196C@mchh266e.mchh.siemens.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: <D73559B38D51D4119CB800508B9527B505AC196C@mchh266e.mchh.siemens.de>; from hannsjuergen.schwarzbauer@siemens.com on Mon, Jul 14, 2003 at 01:41:57PM +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>

Hanns Juergen,

Sorry for insisting but READ THIS PARAGRAPH FROM M2PA-09!

> > > >   Except as modified in this document, M2PA SHOULD follow the
> > > >   requirements of the applicable MTP2 specification.  These may include
> > > >   [Q.703] or [T1.111].

--brian

Schwarzbauer Hanns Juergen wrote:                                             (Mon, 14 Jul 2003 13:41:57)
> Brian,
> 
> sorry for insisting, looking at the figure below, I see no "applicable MTP2 specification" neither for the STP nor the SEP:
> In case of STP is it Q.2140 or Q.703?
> in case of SEP there are no TDM links at all?
> 
> greetings
> 
> HannsJuergen
> 
>   
> 
> 64kbitlink#1+---------+                     +---------+
>     --------+         |                     |         |
> 64kbitlink#n|SS7 Node |     M2PA link       |SS7 Node |
>     --------+-e.g. STP+---------------------+ e.g. SEP|
> HSLlink #1  |         |                     |         |
>     ........+         |                     |         |
> HSLlink #m  |         |                     |         |
>     ........+         |                     |         |
>             +---------+                     +---------+
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: Monday, July 14, 2003 1:09 PM
> > To: Schwarzbauer Hanns Juergen
> > Cc: List: IETF sigtran (E-Mail)
> > Subject: Re: FW: [Sigtran] M2PA Last Call Comment
> > 
> > 
> > Hanss Juergen,
> > 
> > > >   Except as modified in this document, M2PA SHOULD follow the
> > > >   requirements of the applicable MTP2 specification. 
> > These may include
> > > >   [Q.703] or [T1.111].
> > 
> > These "applicable MTP2 specifications" that M2PA-09 says that
> > I SHOULD follow.
> > 
> > --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 exim@www1.ietf.org  Mon Jul 14 14:23:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA11298
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 14:23:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c7yX-0007rv-Hj
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 14:23:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EIN1mm030233
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 14:23:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c7yX-0007rO-E1; Mon, 14 Jul 2003 14:23:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19c7yB-0007lj-Sb
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 14:22:39 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA11193
	for <SIGTRAN@ietf.org>; Mon, 14 Jul 2003 14:22:36 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c7y9-0004hu-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 14:22:37 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[lM2lsYGjYjW9V009szOuqvEvo+1IKbS7])
	by ietf-mx with esmtp (Exim 4.12)
	id 19c7y8-0004hk-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 14:22:36 -0400
Received: from ns.pigworks.openss7.net (IDENT:i0gRF6WOO+PyaIs9JS2DYfg2pMXIMXWg@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6EIMZP21313;
	Mon, 14 Jul 2003 12:22:35 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6EIMZP23117;
	Mon, 14 Jul 2003 12:22:35 -0600
Date: Mon, 14 Jul 2003 12:22:34 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: VIvek Toky <vivektoky@ccpu.com>
Cc: umesh@ccpu.com, SIGTRAN e-mail list <SIGTRAN@ietf.org>
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Message-ID: <20030714122234.C22972@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: VIvek Toky <vivektoky@ccpu.com>, umesh@ccpu.com,
	SIGTRAN e-mail list <SIGTRAN@ietf.org>
References: <ACEMIJONPBBPDLOKCABHMEDFCBAA.umesh@ccpu.com> <HKEBKFBNPIDALDOKLGKDAEHCCDAA.vivektoky@ccpu.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <HKEBKFBNPIDALDOKLGKDAEHCCDAA.vivektoky@ccpu.com>; from vivektoky@ccpu.com on Mon, Jul 14, 2003 at 10:25:50AM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6EIMZP21313
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

VIvek,

REG response cannot be lost.

--brian

On Mon, 14 Jul 2003, VIvek Toky wrote:

> I think Umesh is right. If REG Response is lost ASP will never be able =
to
> register the same RK again. As it doesn't know the RC so it would not b=
e
> able to de-register it and it can also not register it again.
>=20
> -Vivek.
>=20
> -----Original Message-----
> From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf O=
f
> Umesh Vats
> Sent: Monday, July 14, 2003 10:08 AM
> To: SIGTRAN e-mail list
> Subject: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>=20
>=20
> Right now, if ASP sends REG-REQ for an already registered routing key,
> SGP sends unsuccesful REG-RSP.
> But in some scenarios, succesful REG-RSP from SGP to ASP may be lost
> due to network problems/local node problems. In these cases, ASP may
> retry REG-REQ.
> In such scenarios (duplicate REG-REQ), as per M3UA IG V03 SGP sends REG=
-RSP
> with error as "Error - Cannot Support Unique Routing"
>=20
> Is it possible for SGP to include the Routing Context (RC assigned by S=
GP
> for the local rk id received in the first REG-REQ from ASP) in addition
> to the "Local_Routing Key Identifier" in the unsuccessful REG-RSP messa=
ge
> mentioned above. This will allow ASP to recover the RC value.
>=20
> Thanks
> Umesh
>=20
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>=20
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>=20
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Mon Jul 14 17:21:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA26465
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 17:21:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cAkq-0003aQ-95
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 17:21:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ELL4VT013782
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 17:21:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cAko-0003a3-5V; Mon, 14 Jul 2003 17:21:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cAkl-0003Zi-QD
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 17:20:59 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA26418
	for <SIGTRAN@ietf.org>; Mon, 14 Jul 2003 17:20:56 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cAkj-0000KL-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 17:20:57 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19cAkg-0000K7-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 17:20:55 -0400
Received: from [172.16.1.40] (helo=ccpu00251)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19cAkU-0002ZM-50; Mon, 14 Jul 2003 14:20:42 -0700
Reply-To: <umesh@ccpu.com>
From: "Umesh Vats" <umesh@ccpu.com>
To: <bidulock@openss7.org>, "VIvek Toky" <vivektoky@ccpu.com>
Cc: "SIGTRAN e-mail list" <SIGTRAN@ietf.org>
Subject: RE: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Date: Mon, 14 Jul 2003 14:26:05 -0700
Message-ID: <ACEMIJONPBBPDLOKCABHAEDHCBAA.umesh@ccpu.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0)
In-Reply-To: <20030714122234.C22972@openss7.org>
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id RAA26419
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Brian,

REG-RSP may be lost at the local ASP node itself. And anyway
REG-RSP is being generated by SG with local Rk Id. It can put
RC also so that ASP can correlate RC and RK in such cases.

Umesh

-----Original Message-----
From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf Of
Brian F. G. Bidulock
Sent: Monday, July 14, 2003 11:23 AM
To: VIvek Toky
Cc: umesh@ccpu.com; SIGTRAN e-mail list
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP


VIvek,

REG response cannot be lost.

--brian

On Mon, 14 Jul 2003, VIvek Toky wrote:

> I think Umesh is right. If REG Response is lost ASP will never be able =
to
> register the same RK again. As it doesn't know the RC so it would not b=
e
> able to de-register it and it can also not register it again.
>
> -Vivek.
>
> -----Original Message-----
> From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf O=
f
> Umesh Vats
> Sent: Monday, July 14, 2003 10:08 AM
> To: SIGTRAN e-mail list
> Subject: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>
>
> Right now, if ASP sends REG-REQ for an already registered routing key,
> SGP sends unsuccesful REG-RSP.
> But in some scenarios, succesful REG-RSP from SGP to ASP may be lost
> due to network problems/local node problems. In these cases, ASP may
> retry REG-REQ.
> In such scenarios (duplicate REG-REQ), as per M3UA IG V03 SGP sends
REG-RSP
> with error as "Error - Cannot Support Unique Routing"
>
> Is it possible for SGP to include the Routing Context (RC assigned by S=
GP
> for the local rk id received in the first REG-REQ from ASP) in addition
> to the "Local_Routing Key Identifier" in the unsuccessful REG-RSP messa=
ge
> mentioned above. This will allow ASP to recover the RC value.
>
> Thanks
> Umesh
>
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>
>
> _______________________________________________
> 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    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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

---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003


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



From exim@www1.ietf.org  Mon Jul 14 17:31:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA27373
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 17:31:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cAuV-000403-Ew
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 17:31:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ELV304015361
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 17:31:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cAuU-0003ze-9W; Mon, 14 Jul 2003 17:31:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cAtk-0003yC-E4
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 17:30:16 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA27244
	for <SIGTRAN@ietf.org>; Mon, 14 Jul 2003 17:30:12 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cAti-0000WU-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 17:30:14 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[e7LeZllC5KJedrdcsdHLD6+qPsRVdXs0])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cAth-0000WR-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 17:30:13 -0400
Received: from ns.pigworks.openss7.net (IDENT:pPwzB7O58u8dCCBZKsSsFkqZxa8Vgl0e@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6ELU8P23866;
	Mon, 14 Jul 2003 15:30:08 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6ELU7l27318;
	Mon, 14 Jul 2003 15:30:07 -0600
Date: Mon, 14 Jul 2003 15:30:07 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Umesh Vats <umesh@ccpu.com>
Cc: VIvek Toky <vivektoky@ccpu.com>, SIGTRAN e-mail list <SIGTRAN@ietf.org>
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Message-ID: <20030714153007.A27293@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Umesh Vats <umesh@ccpu.com>,
	VIvek Toky <vivektoky@ccpu.com>,
	SIGTRAN e-mail list <SIGTRAN@ietf.org>
References: <20030714122234.C22972@openss7.org> <ACEMIJONPBBPDLOKCABHAEDHCBAA.umesh@ccpu.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <ACEMIJONPBBPDLOKCABHAEDHCBAA.umesh@ccpu.com>; from umesh@ccpu.com on Mon, Jul 14, 2003 at 02:26:05PM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6ELU8P23866
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Umesh,

An implementation that is discarding messages is broken.

--brian

On Mon, 14 Jul 2003, Umesh Vats wrote:

> Brian,
>=20
> REG-RSP may be lost at the local ASP node itself. And anyway
> REG-RSP is being generated by SG with local Rk Id. It can put
> RC also so that ASP can correlate RC and RK in such cases.
>=20
> Umesh
>=20
> -----Original Message-----
> From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf O=
f
> Brian F. G. Bidulock
> Sent: Monday, July 14, 2003 11:23 AM
> To: VIvek Toky
> Cc: umesh@ccpu.com; SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>=20
>=20
> VIvek,
>=20
> REG response cannot be lost.
>=20
> --brian
>=20
> On Mon, 14 Jul 2003, VIvek Toky wrote:
>=20
> > I think Umesh is right. If REG Response is lost ASP will never be abl=
e to
> > register the same RK again. As it doesn't know the RC so it would not=
 be
> > able to de-register it and it can also not register it again.
> >
> > -Vivek.
> >
> > -----Original Message-----
> > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf=
 Of
> > Umesh Vats
> > Sent: Monday, July 14, 2003 10:08 AM
> > To: SIGTRAN e-mail list
> > Subject: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> >
> >
> > Right now, if ASP sends REG-REQ for an already registered routing key=
,
> > SGP sends unsuccesful REG-RSP.
> > But in some scenarios, succesful REG-RSP from SGP to ASP may be lost
> > due to network problems/local node problems. In these cases, ASP may
> > retry REG-REQ.
> > In such scenarios (duplicate REG-REQ), as per M3UA IG V03 SGP sends
> REG-RSP
> > with error as "Error - Cannot Support Unique Routing"
> >
> > Is it possible for SGP to include the Routing Context (RC assigned by=
 SGP
> > for the local rk id received in the first REG-REQ from ASP) in additi=
on
> > to the "Local_Routing Key Identifier" in the unsuccessful REG-RSP mes=
sage
> > mentioned above. This will allow ASP to recover the RC value.
> >
> > Thanks
> > Umesh
> >
> > ---
> > Outgoing mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> >
> > _______________________________________________
> > 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
>=20
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>=20
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>=20
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Mon Jul 14 17:41:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA28152
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 17:41:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cB4A-00058p-7s
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 17:41:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ELf21P019757
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 17:41:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cB49-00058Y-BV; Mon, 14 Jul 2003 17:41:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cB3r-00058J-Of
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 17:40:43 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA28089
	for <SIGTRAN@ietf.org>; Mon, 14 Jul 2003 17:40:39 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cB3p-0000kU-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 17:40:41 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19cB3m-0000kI-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 17:40:39 -0400
Received: from [172.16.1.40] (helo=ccpu00251)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19cB3a-00041i-N3; Mon, 14 Jul 2003 14:40:26 -0700
Reply-To: <umesh@ccpu.com>
From: "Umesh Vats" <umesh@ccpu.com>
To: <bidulock@openss7.org>
Cc: "VIvek Toky" <vivektoky@ccpu.com>,
        "SIGTRAN e-mail list" <SIGTRAN@ietf.org>
Subject: RE: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Date: Mon, 14 Jul 2003 14:46:00 -0700
Message-ID: <ACEMIJONPBBPDLOKCABHOEDHCBAA.umesh@ccpu.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0)
In-Reply-To: <20030714153007.A27293@openss7.org>
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id RAA28090
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

It is a very reasonable scenario. On the nodes, which work
in highly-available fault tolerant modes, some message
losses are expected when active/standby switchovers happen.

But there should be mechanism for recovery with minimum impact.

In the specific case I mentioned, if the ASP receives RC in
REG-RSP, it can easily correlate RC with RK. Then, if everything
else is OK, it can keep using same RK. Or it can send request to
SGP for deregistering RK (using RC receivd from SGP).

thanks
Umesh


-----Original Message-----
From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
Sent: Monday, July 14, 2003 2:30 PM
To: Umesh Vats
Cc: VIvek Toky; SIGTRAN e-mail list
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP


Umesh,

An implementation that is discarding messages is broken.

--brian

On Mon, 14 Jul 2003, Umesh Vats wrote:

> Brian,
>
> REG-RSP may be lost at the local ASP node itself. And anyway
> REG-RSP is being generated by SG with local Rk Id. It can put
> RC also so that ASP can correlate RC and RK in such cases.
>
> Umesh
>
> -----Original Message-----
> From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf O=
f
> Brian F. G. Bidulock
> Sent: Monday, July 14, 2003 11:23 AM
> To: VIvek Toky
> Cc: umesh@ccpu.com; SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>
>
> VIvek,
>
> REG response cannot be lost.
>
> --brian
>
> On Mon, 14 Jul 2003, VIvek Toky wrote:
>
> > I think Umesh is right. If REG Response is lost ASP will never be abl=
e
to
> > register the same RK again. As it doesn't know the RC so it would not=
 be
> > able to de-register it and it can also not register it again.
> >
> > -Vivek.
> >
> > -----Original Message-----
> > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf=
 Of
> > Umesh Vats
> > Sent: Monday, July 14, 2003 10:08 AM
> > To: SIGTRAN e-mail list
> > Subject: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> >
> >
> > Right now, if ASP sends REG-REQ for an already registered routing key=
,
> > SGP sends unsuccesful REG-RSP.
> > But in some scenarios, succesful REG-RSP from SGP to ASP may be lost
> > due to network problems/local node problems. In these cases, ASP may
> > retry REG-REQ.
> > In such scenarios (duplicate REG-REQ), as per M3UA IG V03 SGP sends
> REG-RSP
> > with error as "Error - Cannot Support Unique Routing"
> >
> > Is it possible for SGP to include the Routing Context (RC assigned by
SGP
> > for the local rk id received in the first REG-REQ from ASP) in additi=
on
> > to the "Local_Routing Key Identifier" in the unsuccessful REG-RSP
message
> > mentioned above. This will allow ASP to recover the RC value.
> >
> > Thanks
> > Umesh
> >
> > ---
> > Outgoing mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> >
> > _______________________________________________
> > 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    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

--
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003


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



From exim@www1.ietf.org  Mon Jul 14 18:11:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA00172
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 18:11:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cBXE-0006me-Ks
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 18:11:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EMB42H026069
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 18:11:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cBXC-0006mE-Fz; Mon, 14 Jul 2003 18:11:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cBWr-0006lV-LT
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 18:10:42 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA00072
	for <SIGTRAN@ietf.org>; Mon, 14 Jul 2003 18:10:37 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cBWo-00015j-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 18:10:38 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[hjJ+hKmQ2q3+38mgZhafo8SakLoXxfNz])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cBWn-00015g-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 18:10:37 -0400
Received: from ns.pigworks.openss7.net (IDENT:DAB9CGOIWYjHd5HYWWVvaq1XuG0HF0oU@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6EMAbP24437;
	Mon, 14 Jul 2003 16:10:37 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6EMAa828133;
	Mon, 14 Jul 2003 16:10:36 -0600
Date: Mon, 14 Jul 2003 16:10:36 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Umesh Vats <umesh@ccpu.com>
Cc: VIvek Toky <vivektoky@ccpu.com>, SIGTRAN e-mail list <SIGTRAN@ietf.org>
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Message-ID: <20030714161036.A28088@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Umesh Vats <umesh@ccpu.com>,
	VIvek Toky <vivektoky@ccpu.com>,
	SIGTRAN e-mail list <SIGTRAN@ietf.org>
References: <20030714153007.A27293@openss7.org> <ACEMIJONPBBPDLOKCABHOEDHCBAA.umesh@ccpu.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <ACEMIJONPBBPDLOKCABHOEDHCBAA.umesh@ccpu.com>; from umesh@ccpu.com on Mon, Jul 14, 2003 at 02:46:00PM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6EMAbP24437
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Umesh,

Active/standby switchovers occur BETWEEN SCTP associations.  REG RSP is
only relevant WITHIN an SCTP association.  So the scenario you describe
does not apply.

Why does your ASP lose REG RSP messages?  Is is so poorly designed?

The only scenario that we considered for the RFC where the ASP may lose
messages is if the ASP crashes and then come back up again.  Then,
however, if the SCTP association reamins open, it is the responsibility
of the system to send ASP DOWN.  Or, in any case, when the ASP comes
back up, it sends ASP UP, that clears all registrations anyway.

I'm sorry, but you have not presented a reasonable situation yet.
Please try again.

--brian

On Mon, 14 Jul 2003, Umesh Vats wrote:

> It is a very reasonable scenario. On the nodes, which work
> in highly-available fault tolerant modes, some message
> losses are expected when active/standby switchovers happen.
>=20
> But there should be mechanism for recovery with minimum impact.
>=20
> In the specific case I mentioned, if the ASP receives RC in
> REG-RSP, it can easily correlate RC with RK. Then, if everything
> else is OK, it can keep using same RK. Or it can send request to
> SGP for deregistering RK (using RC receivd from SGP).
>=20
> thanks
> Umesh
>=20
>=20
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Monday, July 14, 2003 2:30 PM
> To: Umesh Vats
> Cc: VIvek Toky; SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>=20
>=20
> Umesh,
>=20
> An implementation that is discarding messages is broken.
>=20
> --brian
>=20
> On Mon, 14 Jul 2003, Umesh Vats wrote:
>=20
> > Brian,
> >
> > REG-RSP may be lost at the local ASP node itself. And anyway
> > REG-RSP is being generated by SG with local Rk Id. It can put
> > RC also so that ASP can correlate RC and RK in such cases.
> >
> > Umesh
> >
> > -----Original Message-----
> > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf=
 Of
> > Brian F. G. Bidulock
> > Sent: Monday, July 14, 2003 11:23 AM
> > To: VIvek Toky
> > Cc: umesh@ccpu.com; SIGTRAN e-mail list
> > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> >
> >
> > VIvek,
> >
> > REG response cannot be lost.
> >
> > --brian
> >
> > On Mon, 14 Jul 2003, VIvek Toky wrote:
> >
> > > I think Umesh is right. If REG Response is lost ASP will never be a=
ble
> to
> > > register the same RK again. As it doesn't know the RC so it would n=
ot be
> > > able to de-register it and it can also not register it again.
> > >
> > > -Vivek.
> > >
> > > -----Original Message-----
> > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Beha=
lf Of
> > > Umesh Vats
> > > Sent: Monday, July 14, 2003 10:08 AM
> > > To: SIGTRAN e-mail list
> > > Subject: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > >
> > >
> > > Right now, if ASP sends REG-REQ for an already registered routing k=
ey,
> > > SGP sends unsuccesful REG-RSP.
> > > But in some scenarios, succesful REG-RSP from SGP to ASP may be los=
t
> > > due to network problems/local node problems. In these cases, ASP ma=
y
> > > retry REG-REQ.
> > > In such scenarios (duplicate REG-REQ), as per M3UA IG V03 SGP sends
> > REG-RSP
> > > with error as "Error - Cannot Support Unique Routing"
> > >
> > > Is it possible for SGP to include the Routing Context (RC assigned =
by
> SGP
> > > for the local rk id received in the first REG-REQ from ASP) in addi=
tion
> > > to the "Local_Routing Key Identifier" in the unsuccessful REG-RSP
> message
> > > mentioned above. This will allow ASP to recover the RC value.
> > >
> > > Thanks
> > > Umesh
> > >
> > > ---
> > > Outgoing mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > >
> > > _______________________________________________
> > > 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    =A6 The reasonable man adapts himself to the =
=A6
> > bidulock@openss7.org    =A6 world; the unreasonable one persists in  =
=A6
> > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =
=A6
> >                         =A6 Therefore  all  progress  depends on the =
=A6
> >                         =A6 unreasonable man. -- George Bernard Shaw =
=A6
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> > ---
> > Incoming mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > ---
> > Outgoing mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>=20
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>=20
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>=20
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Mon Jul 14 18:29:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA01616
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 18:29:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cBoc-0007vu-UI
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 18:29:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EMT2BE030490
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 18:29:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cBob-0007vU-9x; Mon, 14 Jul 2003 18:29:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cBns-0007v1-Ea
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 18:28:16 -0400
Received: from smtp.ccpu.com (defender.ccpu.com [216.54.134.34])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA01542
	for <SIGTRAN@ietf.org>; Mon, 14 Jul 2003 18:28:11 -0400 (EDT)
Received: from [172.16.1.40] (helo=ccpu00251)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19cBnl-0007Yo-JO; Mon, 14 Jul 2003 15:28:09 -0700
Reply-To: <umesh@ccpu.com>
From: "Umesh Vats" <umesh@ccpu.com>
To: <bidulock@openss7.org>
Cc: "VIvek Toky" <vivektoky@ccpu.com>,
        "SIGTRAN e-mail list" <SIGTRAN@ietf.org>
Subject: RE: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Date: Mon, 14 Jul 2003 15:33:42 -0700
Message-ID: <ACEMIJONPBBPDLOKCABHMEDICBAA.umesh@ccpu.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0)
In-Reply-To: <20030714161036.A28088@openss7.org>
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id SAA01544
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Brian,

I am talking about node level fault-tolerance.
Consider the following scenario :

      host1
    ASP1-Active---|
        |         |    host3
    Common Data   |---SCTP-------association--------SGP
        |         |
    ASP1-Standby--|
       host2

Here, I am considering node level fault-tolerance. Two,
ASPs are operating in active/standby mode. Both the copies
represent single ASP to SGP (via single SCTP association)
where SCTP is running on host other than those of ASPs.
Also, both the ASPs refer to same data(AS, ASP, Association states,
routing contexts).

Suppose ASP1-Active sends REG-REQ for RK1 to SGP. SGP responds back
with REG-RSP to ASP1. Now, ASP1-Active receives the REG-RSP
but before it could update the common data, it goes down. In
this case, switchover happens and ASP1-Standby become active.
Now, ASP1-Standby sees that RK1 is not registered and sends REG-REQ
again to SGP. The problem comes here.

The other method you discussed (sending ASP-DOWN) will bring down
remaing application servers also which may be already actively receiving
traffic.

Thanks
Umesh

-----Original Message-----
From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf Of
Brian F. G. Bidulock
Sent: Monday, July 14, 2003 3:11 PM
To: Umesh Vats
Cc: VIvek Toky; SIGTRAN e-mail list
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP


Umesh,

Active/standby switchovers occur BETWEEN SCTP associations.  REG RSP is
only relevant WITHIN an SCTP association.  So the scenario you describe
does not apply.

Why does your ASP lose REG RSP messages?  Is is so poorly designed?

The only scenario that we considered for the RFC where the ASP may lose
messages is if the ASP crashes and then come back up again.  Then,
however, if the SCTP association reamins open, it is the responsibility
of the system to send ASP DOWN.  Or, in any case, when the ASP comes
back up, it sends ASP UP, that clears all registrations anyway.

I'm sorry, but you have not presented a reasonable situation yet.
Please try again.

--brian

On Mon, 14 Jul 2003, Umesh Vats wrote:

> It is a very reasonable scenario. On the nodes, which work
> in highly-available fault tolerant modes, some message
> losses are expected when active/standby switchovers happen.
>
> But there should be mechanism for recovery with minimum impact.
>
> In the specific case I mentioned, if the ASP receives RC in
> REG-RSP, it can easily correlate RC with RK. Then, if everything
> else is OK, it can keep using same RK. Or it can send request to
> SGP for deregistering RK (using RC receivd from SGP).
>
> thanks
> Umesh
>
>
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Monday, July 14, 2003 2:30 PM
> To: Umesh Vats
> Cc: VIvek Toky; SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>
>
> Umesh,
>
> An implementation that is discarding messages is broken.
>
> --brian
>
> On Mon, 14 Jul 2003, Umesh Vats wrote:
>
> > Brian,
> >
> > REG-RSP may be lost at the local ASP node itself. And anyway
> > REG-RSP is being generated by SG with local Rk Id. It can put
> > RC also so that ASP can correlate RC and RK in such cases.
> >
> > Umesh
> >
> > -----Original Message-----
> > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf=
 Of
> > Brian F. G. Bidulock
> > Sent: Monday, July 14, 2003 11:23 AM
> > To: VIvek Toky
> > Cc: umesh@ccpu.com; SIGTRAN e-mail list
> > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> >
> >
> > VIvek,
> >
> > REG response cannot be lost.
> >
> > --brian
> >
> > On Mon, 14 Jul 2003, VIvek Toky wrote:
> >
> > > I think Umesh is right. If REG Response is lost ASP will never be a=
ble
> to
> > > register the same RK again. As it doesn't know the RC so it would n=
ot
be
> > > able to de-register it and it can also not register it again.
> > >
> > > -Vivek.
> > >
> > > -----Original Message-----
> > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Beha=
lf
Of
> > > Umesh Vats
> > > Sent: Monday, July 14, 2003 10:08 AM
> > > To: SIGTRAN e-mail list
> > > Subject: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > >
> > >
> > > Right now, if ASP sends REG-REQ for an already registered routing k=
ey,
> > > SGP sends unsuccesful REG-RSP.
> > > But in some scenarios, succesful REG-RSP from SGP to ASP may be los=
t
> > > due to network problems/local node problems. In these cases, ASP ma=
y
> > > retry REG-REQ.
> > > In such scenarios (duplicate REG-REQ), as per M3UA IG V03 SGP sends
> > REG-RSP
> > > with error as "Error - Cannot Support Unique Routing"
> > >
> > > Is it possible for SGP to include the Routing Context (RC assigned =
by
> SGP
> > > for the local rk id received in the first REG-REQ from ASP) in
addition
> > > to the "Local_Routing Key Identifier" in the unsuccessful REG-RSP
> message
> > > mentioned above. This will allow ASP to recover the RC value.
> > >
> > > Thanks
> > > Umesh
> > >
> > > ---
> > > Outgoing mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > >
> > > _______________________________________________
> > > 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    =A6 The reasonable man adapts himself to the =
=A6
> > bidulock@openss7.org    =A6 world; the unreasonable one persists in  =
=A6
> > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =
=A6
> >                         =A6 Therefore  all  progress  depends on the =
=A6
> >                         =A6 unreasonable man. -- George Bernard Shaw =
=A6
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> > ---
> > Incoming mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > ---
> > Outgoing mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

--
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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

---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003


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



From exim@www1.ietf.org  Mon Jul 14 18:36:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA02327
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 18:36:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cBvO-0008NL-MI
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 18:36:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EMa2bB032191
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 18:36:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cBvN-0008Mv-Gs; Mon, 14 Jul 2003 18:36:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cBua-0008CN-IR
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 18:35:12 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA02159
	for <SIGTRAN@ietf.org>; Mon, 14 Jul 2003 18:35:07 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cBuX-0001Rp-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 18:35:09 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[ivp3mzZNzywbusH7TICu3VMMXKdDzh4B])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cBuW-0001Ri-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 18:35:08 -0400
Received: from ns.pigworks.openss7.net (IDENT:jXGYQDBueZ4ruxg7fWjh2AGjTf0XX5HX@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6EMZ7P24783;
	Mon, 14 Jul 2003 16:35:07 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6EMZ6N28348;
	Mon, 14 Jul 2003 16:35:06 -0600
Date: Mon, 14 Jul 2003 16:35:05 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Umesh Vats <umesh@ccpu.com>
Cc: VIvek Toky <vivektoky@ccpu.com>, SIGTRAN e-mail list <SIGTRAN@ietf.org>
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Message-ID: <20030714163505.A28262@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Umesh Vats <umesh@ccpu.com>,
	VIvek Toky <vivektoky@ccpu.com>,
	SIGTRAN e-mail list <SIGTRAN@ietf.org>
References: <20030714161036.A28088@openss7.org> <ACEMIJONPBBPDLOKCABHMEDICBAA.umesh@ccpu.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <ACEMIJONPBBPDLOKCABHMEDICBAA.umesh@ccpu.com>; from umesh@ccpu.com on Mon, Jul 14, 2003 at 03:33:42PM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6EMZ7P24783
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Umesh,

M3UA only supports 1 ASP per SCTP association.  You configuration
is not supported.

By your argument, any M3UA message could be lost between SCTP
and M3UA.  This is not provided for the the specification.  Short
of having M3UA acknowledge and retransmit ALL messages, there will
not be support in M3UA for this arrangement.

You need a better design.

--brian

On Mon, 14 Jul 2003, Umesh Vats wrote:

> Brian,
>=20
> I am talking about node level fault-tolerance.
> Consider the following scenario :
>=20
>       host1
>     ASP1-Active---|
>         |         |    host3
>     Common Data   |---SCTP-------association--------SGP
>         |         |
>     ASP1-Standby--|
>        host2
>=20
> Here, I am considering node level fault-tolerance. Two,
> ASPs are operating in active/standby mode. Both the copies
> represent single ASP to SGP (via single SCTP association)
> where SCTP is running on host other than those of ASPs.
> Also, both the ASPs refer to same data(AS, ASP, Association states,
> routing contexts).
>=20
> Suppose ASP1-Active sends REG-REQ for RK1 to SGP. SGP responds back
> with REG-RSP to ASP1. Now, ASP1-Active receives the REG-RSP
> but before it could update the common data, it goes down. In
> this case, switchover happens and ASP1-Standby become active.
> Now, ASP1-Standby sees that RK1 is not registered and sends REG-REQ
> again to SGP. The problem comes here.
>=20
> The other method you discussed (sending ASP-DOWN) will bring down
> remaing application servers also which may be already actively receivin=
g
> traffic.
>=20
> Thanks
> Umesh
>=20
> -----Original Message-----
> From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf O=
f
> Brian F. G. Bidulock
> Sent: Monday, July 14, 2003 3:11 PM
> To: Umesh Vats
> Cc: VIvek Toky; SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>=20
>=20
> Umesh,
>=20
> Active/standby switchovers occur BETWEEN SCTP associations.  REG RSP is
> only relevant WITHIN an SCTP association.  So the scenario you describe
> does not apply.
>=20
> Why does your ASP lose REG RSP messages?  Is is so poorly designed?
>=20
> The only scenario that we considered for the RFC where the ASP may lose
> messages is if the ASP crashes and then come back up again.  Then,
> however, if the SCTP association reamins open, it is the responsibility
> of the system to send ASP DOWN.  Or, in any case, when the ASP comes
> back up, it sends ASP UP, that clears all registrations anyway.
>=20
> I'm sorry, but you have not presented a reasonable situation yet.
> Please try again.
>=20
> --brian
>=20
> On Mon, 14 Jul 2003, Umesh Vats wrote:
>=20
> > It is a very reasonable scenario. On the nodes, which work
> > in highly-available fault tolerant modes, some message
> > losses are expected when active/standby switchovers happen.
> >
> > But there should be mechanism for recovery with minimum impact.
> >
> > In the specific case I mentioned, if the ASP receives RC in
> > REG-RSP, it can easily correlate RC with RK. Then, if everything
> > else is OK, it can keep using same RK. Or it can send request to
> > SGP for deregistering RK (using RC receivd from SGP).
> >
> > thanks
> > Umesh
> >
> >
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: Monday, July 14, 2003 2:30 PM
> > To: Umesh Vats
> > Cc: VIvek Toky; SIGTRAN e-mail list
> > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> >
> >
> > Umesh,
> >
> > An implementation that is discarding messages is broken.
> >
> > --brian
> >
> > On Mon, 14 Jul 2003, Umesh Vats wrote:
> >
> > > Brian,
> > >
> > > REG-RSP may be lost at the local ASP node itself. And anyway
> > > REG-RSP is being generated by SG with local Rk Id. It can put
> > > RC also so that ASP can correlate RC and RK in such cases.
> > >
> > > Umesh
> > >
> > > -----Original Message-----
> > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Beha=
lf Of
> > > Brian F. G. Bidulock
> > > Sent: Monday, July 14, 2003 11:23 AM
> > > To: VIvek Toky
> > > Cc: umesh@ccpu.com; SIGTRAN e-mail list
> > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > >
> > >
> > > VIvek,
> > >
> > > REG response cannot be lost.
> > >
> > > --brian
> > >
> > > On Mon, 14 Jul 2003, VIvek Toky wrote:
> > >
> > > > I think Umesh is right. If REG Response is lost ASP will never be=
 able
> > to
> > > > register the same RK again. As it doesn't know the RC so it would=
 not
> be
> > > > able to de-register it and it can also not register it again.
> > > >
> > > > -Vivek.
> > > >
> > > > -----Original Message-----
> > > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Be=
half
> Of
> > > > Umesh Vats
> > > > Sent: Monday, July 14, 2003 10:08 AM
> > > > To: SIGTRAN e-mail list
> > > > Subject: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > > >
> > > >
> > > > Right now, if ASP sends REG-REQ for an already registered routing=
 key,
> > > > SGP sends unsuccesful REG-RSP.
> > > > But in some scenarios, succesful REG-RSP from SGP to ASP may be l=
ost
> > > > due to network problems/local node problems. In these cases, ASP =
may
> > > > retry REG-REQ.
> > > > In such scenarios (duplicate REG-REQ), as per M3UA IG V03 SGP sen=
ds
> > > REG-RSP
> > > > with error as "Error - Cannot Support Unique Routing"
> > > >
> > > > Is it possible for SGP to include the Routing Context (RC assigne=
d by
> > SGP
> > > > for the local rk id received in the first REG-REQ from ASP) in
> addition
> > > > to the "Local_Routing Key Identifier" in the unsuccessful REG-RSP
> > message
> > > > mentioned above. This will allow ASP to recover the RC value.
> > > >
> > > > Thanks
> > > > Umesh
> > > >
> > > > ---
> > > > Outgoing mail is certified Virus Free.
> > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > > >
> > > >
> > > > _______________________________________________
> > > > 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    =A6 The reasonable man adapts himself to th=
e =A6
> > > bidulock@openss7.org    =A6 world; the unreasonable one persists in=
  =A6
> > > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself=
. =A6
> > >                         =A6 Therefore  all  progress  depends on th=
e =A6
> > >                         =A6 unreasonable man. -- George Bernard Sha=
w =A6
> > >
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >
> > > ---
> > > Incoming mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > > ---
> > > Outgoing mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > --
> > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =
=A6
> > bidulock@openss7.org    =A6 world; the unreasonable one persists in  =
=A6
> > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =
=A6
> >                         =A6 Therefore  all  progress  depends on the =
=A6
> >                         =A6 unreasonable man. -- George Bernard Shaw =
=A6
> >
> > ---
> > Incoming mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > ---
> > Outgoing mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>=20
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>=20
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>=20
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Mon Jul 14 18:40:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA02719
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 18:40:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cBzF-0000Hq-Qt
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 18:40:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EMe1Pr001095
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 18:40:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cBzF-0000HJ-4f; Mon, 14 Jul 2003 18:40:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cByJ-0000GV-Ii
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 18:39:03 -0400
Received: from smtp.ccpu.com (defender.ccpu.com [216.54.134.34])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA02568
	for <SIGTRAN@ietf.org>; Mon, 14 Jul 2003 18:38:55 -0400 (EDT)
Received: from [172.16.1.203] (helo=CCPU00186)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19cBy8-0008Hj-I8; Mon, 14 Jul 2003 15:38:52 -0700
From: "VIvek Toky" <vivektoky@ccpu.com>
To: <bidulock@openss7.org>, "Umesh Vats" <umesh@ccpu.com>
Cc: "SIGTRAN e-mail list" <SIGTRAN@ietf.org>
Subject: RE: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Date: Mon, 14 Jul 2003 15:43:32 -0700
Message-ID: <HKEBKFBNPIDALDOKLGKDCEHGCDAA.vivektoky@ccpu.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.6604 (9.0.2911.0)
In-Reply-To: <20030714163505.A28262@openss7.org>
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id SAA02573
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Brian,
   Probably your thinking is limilted to your implementation only. Some
implmentions may internally be fault tolerant or distributed. Which means
that multiple M3UA instances can be designed to use same SCTP assocation =
and
to represent single ASP. Now, explaining you the advantage  of that may b=
e
far beyond the topic but I don't see any harm in including RC in the
messages, as some implementaions may use it to their benefit.

  I am not sure why are you objecting to this proposal , do you mean to s=
ay
that its difficult for your implemntaion to find an RC for already
registered RK ?

regards,
Vivek.

-----Original Message-----
From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf Of
Brian F. G. Bidulock
Sent: Monday, July 14, 2003 3:35 PM
To: Umesh Vats
Cc: VIvek Toky; SIGTRAN e-mail list
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP


Umesh,

M3UA only supports 1 ASP per SCTP association.  You configuration
is not supported.

By your argument, any M3UA message could be lost between SCTP
and M3UA.  This is not provided for the the specification.  Short
of having M3UA acknowledge and retransmit ALL messages, there will
not be support in M3UA for this arrangement.

You need a better design.

--brian

On Mon, 14 Jul 2003, Umesh Vats wrote:

> Brian,
>
> I am talking about node level fault-tolerance.
> Consider the following scenario :
>
>       host1
>     ASP1-Active---|
>         |         |    host3
>     Common Data   |---SCTP-------association--------SGP
>         |         |
>     ASP1-Standby--|
>        host2
>
> Here, I am considering node level fault-tolerance. Two,
> ASPs are operating in active/standby mode. Both the copies
> represent single ASP to SGP (via single SCTP association)
> where SCTP is running on host other than those of ASPs.
> Also, both the ASPs refer to same data(AS, ASP, Association states,
> routing contexts).
>
> Suppose ASP1-Active sends REG-REQ for RK1 to SGP. SGP responds back
> with REG-RSP to ASP1. Now, ASP1-Active receives the REG-RSP
> but before it could update the common data, it goes down. In
> this case, switchover happens and ASP1-Standby become active.
> Now, ASP1-Standby sees that RK1 is not registered and sends REG-REQ
> again to SGP. The problem comes here.
>
> The other method you discussed (sending ASP-DOWN) will bring down
> remaing application servers also which may be already actively receivin=
g
> traffic.
>
> Thanks
> Umesh
>
> -----Original Message-----
> From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf O=
f
> Brian F. G. Bidulock
> Sent: Monday, July 14, 2003 3:11 PM
> To: Umesh Vats
> Cc: VIvek Toky; SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>
>
> Umesh,
>
> Active/standby switchovers occur BETWEEN SCTP associations.  REG RSP is
> only relevant WITHIN an SCTP association.  So the scenario you describe
> does not apply.
>
> Why does your ASP lose REG RSP messages?  Is is so poorly designed?
>
> The only scenario that we considered for the RFC where the ASP may lose
> messages is if the ASP crashes and then come back up again.  Then,
> however, if the SCTP association reamins open, it is the responsibility
> of the system to send ASP DOWN.  Or, in any case, when the ASP comes
> back up, it sends ASP UP, that clears all registrations anyway.
>
> I'm sorry, but you have not presented a reasonable situation yet.
> Please try again.
>
> --brian
>
> On Mon, 14 Jul 2003, Umesh Vats wrote:
>
> > It is a very reasonable scenario. On the nodes, which work
> > in highly-available fault tolerant modes, some message
> > losses are expected when active/standby switchovers happen.
> >
> > But there should be mechanism for recovery with minimum impact.
> >
> > In the specific case I mentioned, if the ASP receives RC in
> > REG-RSP, it can easily correlate RC with RK. Then, if everything
> > else is OK, it can keep using same RK. Or it can send request to
> > SGP for deregistering RK (using RC receivd from SGP).
> >
> > thanks
> > Umesh
> >
> >
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: Monday, July 14, 2003 2:30 PM
> > To: Umesh Vats
> > Cc: VIvek Toky; SIGTRAN e-mail list
> > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> >
> >
> > Umesh,
> >
> > An implementation that is discarding messages is broken.
> >
> > --brian
> >
> > On Mon, 14 Jul 2003, Umesh Vats wrote:
> >
> > > Brian,
> > >
> > > REG-RSP may be lost at the local ASP node itself. And anyway
> > > REG-RSP is being generated by SG with local Rk Id. It can put
> > > RC also so that ASP can correlate RC and RK in such cases.
> > >
> > > Umesh
> > >
> > > -----Original Message-----
> > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Beha=
lf
Of
> > > Brian F. G. Bidulock
> > > Sent: Monday, July 14, 2003 11:23 AM
> > > To: VIvek Toky
> > > Cc: umesh@ccpu.com; SIGTRAN e-mail list
> > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > >
> > >
> > > VIvek,
> > >
> > > REG response cannot be lost.
> > >
> > > --brian
> > >
> > > On Mon, 14 Jul 2003, VIvek Toky wrote:
> > >
> > > > I think Umesh is right. If REG Response is lost ASP will never be
able
> > to
> > > > register the same RK again. As it doesn't know the RC so it would
not
> be
> > > > able to de-register it and it can also not register it again.
> > > >
> > > > -Vivek.
> > > >
> > > > -----Original Message-----
> > > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On
Behalf
> Of
> > > > Umesh Vats
> > > > Sent: Monday, July 14, 2003 10:08 AM
> > > > To: SIGTRAN e-mail list
> > > > Subject: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > > >
> > > >
> > > > Right now, if ASP sends REG-REQ for an already registered routing
key,
> > > > SGP sends unsuccesful REG-RSP.
> > > > But in some scenarios, succesful REG-RSP from SGP to ASP may be l=
ost
> > > > due to network problems/local node problems. In these cases, ASP =
may
> > > > retry REG-REQ.
> > > > In such scenarios (duplicate REG-REQ), as per M3UA IG V03 SGP sen=
ds
> > > REG-RSP
> > > > with error as "Error - Cannot Support Unique Routing"
> > > >
> > > > Is it possible for SGP to include the Routing Context (RC assigne=
d
by
> > SGP
> > > > for the local rk id received in the first REG-REQ from ASP) in
> addition
> > > > to the "Local_Routing Key Identifier" in the unsuccessful REG-RSP
> > message
> > > > mentioned above. This will allow ASP to recover the RC value.
> > > >
> > > > Thanks
> > > > Umesh
> > > >
> > > > ---
> > > > Outgoing mail is certified Virus Free.
> > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > > >
> > > >
> > > > _______________________________________________
> > > > 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    =A6 The reasonable man adapts himself to th=
e =A6
> > > bidulock@openss7.org    =A6 world; the unreasonable one persists in=
  =A6
> > > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself=
. =A6
> > >                         =A6 Therefore  all  progress  depends on th=
e =A6
> > >                         =A6 unreasonable man. -- George Bernard Sha=
w =A6
> > >
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >
> > > ---
> > > Incoming mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > > ---
> > > Outgoing mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > --
> > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =
=A6
> > bidulock@openss7.org    =A6 world; the unreasonable one persists in  =
=A6
> > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =
=A6
> >                         =A6 Therefore  all  progress  depends on the =
=A6
> >                         =A6 unreasonable man. -- George Bernard Shaw =
=A6
> >
> > ---
> > Incoming mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > ---
> > Outgoing mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

--
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

_______________________________________________
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 exim@www1.ietf.org  Mon Jul 14 18:47:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA03466
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 18:47:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cC62-0000hG-BB
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 18:47:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EMl2HE002664
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 18:47:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cC61-0000gm-Am; Mon, 14 Jul 2003 18:47:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cC5P-0000g9-EU
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 18:46:23 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA03320
	for <SIGTRAN@ietf.org>; Mon, 14 Jul 2003 18:46:18 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cC5M-0001kB-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 18:46:20 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[K7S4kLtdTYKuEuyeWUMb7jF/iCuK1KNg])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cC5K-0001k8-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 18:46:18 -0400
Received: from ns.pigworks.openss7.net (IDENT:xDy4JxQi6Bn4WvaVniIWG4AUWHX6oTJa@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6EMkIP24901;
	Mon, 14 Jul 2003 16:46:18 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6EMkHc28451;
	Mon, 14 Jul 2003 16:46:17 -0600
Date: Mon, 14 Jul 2003 16:46:17 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: VIvek Toky <vivektoky@ccpu.com>
Cc: Umesh Vats <umesh@ccpu.com>, SIGTRAN e-mail list <SIGTRAN@ietf.org>
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Message-ID: <20030714164617.A28376@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: VIvek Toky <vivektoky@ccpu.com>,
	Umesh Vats <umesh@ccpu.com>, SIGTRAN e-mail list <SIGTRAN@ietf.org>
References: <20030714163505.A28262@openss7.org> <HKEBKFBNPIDALDOKLGKDCEHGCDAA.vivektoky@ccpu.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <HKEBKFBNPIDALDOKLGKDCEHGCDAA.vivektoky@ccpu.com>; from vivektoky@ccpu.com on Mon, Jul 14, 2003 at 03:43:32PM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6EMkIP24901
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

VIvek,

Because you would have to handle ANY messages being lost to
accomodate this approach.  REG RSP would not be enough.  I
don't see the point in putting a half-bandaid on something
that isn't broken.

Make you distributed system either not lose messages, or
perform ASP Down/Up procedures between switchover.

I do have a distributed implementation, but it does not lose
M3UA messages without an ASP going down.

In the diagram below, where is M3UA?  If it is not riding
directly on top of SCTP (it is after all an SCTP Adaptation
Layer) then it is not provided for in the M3UA specification.
If you have M3UA running on host3, there is no problem, it
will not lose messages.

If you are extending multiple instances of M3UA running over a
single SCTP association, this is not supported.  (1 ASP, 1
SCTP Association).

Am I clear?  Don't expect M3UA to accomodate your unsupported
and poor distributed design.  Make a better design.

The ASP that loses M3UA messages is either ASP-DOWN or broken.

--brian

On Mon, 14 Jul 2003, VIvek Toky wrote:

> Brian,
>    Probably your thinking is limilted to your implementation only. Some
> implmentions may internally be fault tolerant or distributed. Which mea=
ns
> that multiple M3UA instances can be designed to use same SCTP assocatio=
n and
> to represent single ASP. Now, explaining you the advantage  of that may=
 be
> far beyond the topic but I don't see any harm in including RC in the
> messages, as some implementaions may use it to their benefit.
>=20
>   I am not sure why are you objecting to this proposal , do you mean to=
 say
> that its difficult for your implemntaion to find an RC for already
> registered RK ?
>=20
> regards,
> Vivek.
>=20
> -----Original Message-----
> From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf O=
f
> Brian F. G. Bidulock
> Sent: Monday, July 14, 2003 3:35 PM
> To: Umesh Vats
> Cc: VIvek Toky; SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>=20
>=20
> Umesh,
>=20
> M3UA only supports 1 ASP per SCTP association.  You configuration
> is not supported.
>=20
> By your argument, any M3UA message could be lost between SCTP
> and M3UA.  This is not provided for the the specification.  Short
> of having M3UA acknowledge and retransmit ALL messages, there will
> not be support in M3UA for this arrangement.
>=20
> You need a better design.
>=20
> --brian
>=20
> On Mon, 14 Jul 2003, Umesh Vats wrote:
>=20
> > Brian,
> >
> > I am talking about node level fault-tolerance.
> > Consider the following scenario :
> >
> >       host1
> >     ASP1-Active---|
> >         |         |    host3
> >     Common Data   |---SCTP-------association--------SGP
> >         |         |
> >     ASP1-Standby--|
> >        host2
> >
> > Here, I am considering node level fault-tolerance. Two,
> > ASPs are operating in active/standby mode. Both the copies
> > represent single ASP to SGP (via single SCTP association)
> > where SCTP is running on host other than those of ASPs.
> > Also, both the ASPs refer to same data(AS, ASP, Association states,
> > routing contexts).
> >
> > Suppose ASP1-Active sends REG-REQ for RK1 to SGP. SGP responds back
> > with REG-RSP to ASP1. Now, ASP1-Active receives the REG-RSP
> > but before it could update the common data, it goes down. In
> > this case, switchover happens and ASP1-Standby become active.
> > Now, ASP1-Standby sees that RK1 is not registered and sends REG-REQ
> > again to SGP. The problem comes here.
> >
> > The other method you discussed (sending ASP-DOWN) will bring down
> > remaing application servers also which may be already actively receiv=
ing
> > traffic.
> >
> > Thanks
> > Umesh
> >
> > -----Original Message-----
> > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf=
 Of
> > Brian F. G. Bidulock
> > Sent: Monday, July 14, 2003 3:11 PM
> > To: Umesh Vats
> > Cc: VIvek Toky; SIGTRAN e-mail list
> > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> >
> >
> > Umesh,
> >
> > Active/standby switchovers occur BETWEEN SCTP associations.  REG RSP =
is
> > only relevant WITHIN an SCTP association.  So the scenario you descri=
be
> > does not apply.
> >
> > Why does your ASP lose REG RSP messages?  Is is so poorly designed?
> >
> > The only scenario that we considered for the RFC where the ASP may lo=
se
> > messages is if the ASP crashes and then come back up again.  Then,
> > however, if the SCTP association reamins open, it is the responsibili=
ty
> > of the system to send ASP DOWN.  Or, in any case, when the ASP comes
> > back up, it sends ASP UP, that clears all registrations anyway.
> >
> > I'm sorry, but you have not presented a reasonable situation yet.
> > Please try again.
> >
> > --brian
> >
> > On Mon, 14 Jul 2003, Umesh Vats wrote:
> >
> > > It is a very reasonable scenario. On the nodes, which work
> > > in highly-available fault tolerant modes, some message
> > > losses are expected when active/standby switchovers happen.
> > >
> > > But there should be mechanism for recovery with minimum impact.
> > >
> > > In the specific case I mentioned, if the ASP receives RC in
> > > REG-RSP, it can easily correlate RC with RK. Then, if everything
> > > else is OK, it can keep using same RK. Or it can send request to
> > > SGP for deregistering RK (using RC receivd from SGP).
> > >
> > > thanks
> > > Umesh
> > >
> > >
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: Monday, July 14, 2003 2:30 PM
> > > To: Umesh Vats
> > > Cc: VIvek Toky; SIGTRAN e-mail list
> > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > >
> > >
> > > Umesh,
> > >
> > > An implementation that is discarding messages is broken.
> > >
> > > --brian
> > >
> > > On Mon, 14 Jul 2003, Umesh Vats wrote:
> > >
> > > > Brian,
> > > >
> > > > REG-RSP may be lost at the local ASP node itself. And anyway
> > > > REG-RSP is being generated by SG with local Rk Id. It can put
> > > > RC also so that ASP can correlate RC and RK in such cases.
> > > >
> > > > Umesh
> > > >
> > > > -----Original Message-----
> > > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Be=
half
> Of
> > > > Brian F. G. Bidulock
> > > > Sent: Monday, July 14, 2003 11:23 AM
> > > > To: VIvek Toky
> > > > Cc: umesh@ccpu.com; SIGTRAN e-mail list
> > > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > > >
> > > >
> > > > VIvek,
> > > >
> > > > REG response cannot be lost.
> > > >
> > > > --brian
> > > >
> > > > On Mon, 14 Jul 2003, VIvek Toky wrote:
> > > >
> > > > > I think Umesh is right. If REG Response is lost ASP will never =
be
> able
> > > to
> > > > > register the same RK again. As it doesn't know the RC so it wou=
ld
> not
> > be
> > > > > able to de-register it and it can also not register it again.
> > > > >
> > > > > -Vivek.
> > > > >
> > > > > -----Original Message-----
> > > > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On
> Behalf
> > Of
> > > > > Umesh Vats
> > > > > Sent: Monday, July 14, 2003 10:08 AM
> > > > > To: SIGTRAN e-mail list
> > > > > Subject: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > > > >
> > > > >
> > > > > Right now, if ASP sends REG-REQ for an already registered routi=
ng
> key,
> > > > > SGP sends unsuccesful REG-RSP.
> > > > > But in some scenarios, succesful REG-RSP from SGP to ASP may be=
 lost
> > > > > due to network problems/local node problems. In these cases, AS=
P may
> > > > > retry REG-REQ.
> > > > > In such scenarios (duplicate REG-REQ), as per M3UA IG V03 SGP s=
ends
> > > > REG-RSP
> > > > > with error as "Error - Cannot Support Unique Routing"
> > > > >
> > > > > Is it possible for SGP to include the Routing Context (RC assig=
ned
> by
> > > SGP
> > > > > for the local rk id received in the first REG-REQ from ASP) in
> > addition
> > > > > to the "Local_Routing Key Identifier" in the unsuccessful REG-R=
SP
> > > message
> > > > > mentioned above. This will allow ASP to recover the RC value.
> > > > >
> > > > > Thanks
> > > > > Umesh
> > > > >
> > > > > ---
> > > > > Outgoing mail is certified Virus Free.
> > > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/200=
3
> > > > >
> > > > >
> > > > > _______________________________________________
> > > > > 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    =A6 The reasonable man adapts himself to =
the =A6
> > > > bidulock@openss7.org    =A6 world; the unreasonable one persists =
in  =A6
> > > > http://www.openss7.org/ =A6 trying  to adapt the  world  to himse=
lf. =A6
> > > >                         =A6 Therefore  all  progress  depends on =
the =A6
> > > >                         =A6 unreasonable man. -- George Bernard S=
haw =A6
> > > >
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > >
> > > > ---
> > > > Incoming mail is certified Virus Free.
> > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > > >
> > > > ---
> > > > Outgoing mail is certified Virus Free.
> > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > > --
> > > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to th=
e =A6
> > > bidulock@openss7.org    =A6 world; the unreasonable one persists in=
  =A6
> > > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself=
. =A6
> > >                         =A6 Therefore  all  progress  depends on th=
e =A6
> > >                         =A6 unreasonable man. -- George Bernard Sha=
w =A6
> > >
> > > ---
> > > Incoming mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > > ---
> > > Outgoing mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > --
> > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =
=A6
> > bidulock@openss7.org    =A6 world; the unreasonable one persists in  =
=A6
> > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =
=A6
> >                         =A6 Therefore  all  progress  depends on the =
=A6
> >                         =A6 unreasonable man. -- George Bernard Shaw =
=A6
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> > ---
> > Incoming mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > ---
> > Outgoing mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>=20
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Mon Jul 14 18:57:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA04042
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 18:57:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cCFi-0001fX-N4
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 18:57:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EMv2cp006406
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 18:57:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cCFh-0001fD-N8; Mon, 14 Jul 2003 18:57:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cCEp-0001em-J1
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 18:56:07 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA04024
	for <SIGTRAN@ietf.org>; Mon, 14 Jul 2003 18:56:02 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cCEm-0001vc-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 18:56:04 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19cCEl-0001vZ-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 18:56:03 -0400
Received: from [172.16.1.40] (helo=ccpu00251)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19cCEc-0000pL-EG; Mon, 14 Jul 2003 15:55:54 -0700
Reply-To: <umesh@ccpu.com>
From: "Umesh Vats" <umesh@ccpu.com>
To: "SIGTRAN e-mail list" <SIGTRAN@ietf.org>
Cc: "Vivek Toky" <vivektoky@ccpu.com>
Subject: RE: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Date: Mon, 14 Jul 2003 16:01:27 -0700
Message-ID: <ACEMIJONPBBPDLOKCABHOEDJCBAA.umesh@ccpu.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0)
In-Reply-To: <20030714164617.A28376@openss7.org>
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id SAA04025
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

What do others think about this issue ?
I think, in a practical scenarios its very musch possible
that message loss may happen between ASP and SGP.

Currently SGP sends REG-RSP in case duplicate REG-REQ for
same RK is received. If we assume ideal scenario where no message loss
can ever happen, SGP should simply dump REG-REQ and not send REG-RSP.

But if it is sending REG-RSP with error (as proposed in M3UA IG-V04),
then sufficient information should be included so that ASP may come
to know the correlation between RC and RK.

thanks
Umesh




-----Original Message-----
From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
Sent: Monday, July 14, 2003 3:46 PM
To: VIvek Toky
Cc: Umesh Vats; SIGTRAN e-mail list
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP


VIvek,

Because you would have to handle ANY messages being lost to
accomodate this approach.  REG RSP would not be enough.  I
don't see the point in putting a half-bandaid on something
that isn't broken.

Make you distributed system either not lose messages, or
perform ASP Down/Up procedures between switchover.

I do have a distributed implementation, but it does not lose
M3UA messages without an ASP going down.

In the diagram below, where is M3UA?  If it is not riding
directly on top of SCTP (it is after all an SCTP Adaptation
Layer) then it is not provided for in the M3UA specification.
If you have M3UA running on host3, there is no problem, it
will not lose messages.

If you are extending multiple instances of M3UA running over a
single SCTP association, this is not supported.  (1 ASP, 1
SCTP Association).

Am I clear?  Don't expect M3UA to accomodate your unsupported
and poor distributed design.  Make a better design.

The ASP that loses M3UA messages is either ASP-DOWN or broken.

--brian

On Mon, 14 Jul 2003, VIvek Toky wrote:

> Brian,
>    Probably your thinking is limilted to your implementation only. Some
> implmentions may internally be fault tolerant or distributed. Which mea=
ns
> that multiple M3UA instances can be designed to use same SCTP assocatio=
n
and
> to represent single ASP. Now, explaining you the advantage  of that may=
 be
> far beyond the topic but I don't see any harm in including RC in the
> messages, as some implementaions may use it to their benefit.
>
>   I am not sure why are you objecting to this proposal , do you mean to
say
> that its difficult for your implemntaion to find an RC for already
> registered RK ?
>
> regards,
> Vivek.
>
> -----Original Message-----
> From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf O=
f
> Brian F. G. Bidulock
> Sent: Monday, July 14, 2003 3:35 PM
> To: Umesh Vats
> Cc: VIvek Toky; SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>
>
> Umesh,
>
> M3UA only supports 1 ASP per SCTP association.  You configuration
> is not supported.
>
> By your argument, any M3UA message could be lost between SCTP
> and M3UA.  This is not provided for the the specification.  Short
> of having M3UA acknowledge and retransmit ALL messages, there will
> not be support in M3UA for this arrangement.
>
> You need a better design.
>
> --brian
>
> On Mon, 14 Jul 2003, Umesh Vats wrote:
>
> > Brian,
> >
> > I am talking about node level fault-tolerance.
> > Consider the following scenario :
> >
> >       host1
> >     ASP1-Active---|
> >         |         |    host3
> >     Common Data   |---SCTP-------association--------SGP
> >         |         |
> >     ASP1-Standby--|
> >        host2
> >
> > Here, I am considering node level fault-tolerance. Two,
> > ASPs are operating in active/standby mode. Both the copies
> > represent single ASP to SGP (via single SCTP association)
> > where SCTP is running on host other than those of ASPs.
> > Also, both the ASPs refer to same data(AS, ASP, Association states,
> > routing contexts).
> >
> > Suppose ASP1-Active sends REG-REQ for RK1 to SGP. SGP responds back
> > with REG-RSP to ASP1. Now, ASP1-Active receives the REG-RSP
> > but before it could update the common data, it goes down. In
> > this case, switchover happens and ASP1-Standby become active.
> > Now, ASP1-Standby sees that RK1 is not registered and sends REG-REQ
> > again to SGP. The problem comes here.
> >
> > The other method you discussed (sending ASP-DOWN) will bring down
> > remaing application servers also which may be already actively receiv=
ing
> > traffic.
> >
> > Thanks
> > Umesh
> >
> > -----Original Message-----
> > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf=
 Of
> > Brian F. G. Bidulock
> > Sent: Monday, July 14, 2003 3:11 PM
> > To: Umesh Vats
> > Cc: VIvek Toky; SIGTRAN e-mail list
> > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> >
> >
> > Umesh,
> >
> > Active/standby switchovers occur BETWEEN SCTP associations.  REG RSP =
is
> > only relevant WITHIN an SCTP association.  So the scenario you descri=
be
> > does not apply.
> >
> > Why does your ASP lose REG RSP messages?  Is is so poorly designed?
> >
> > The only scenario that we considered for the RFC where the ASP may lo=
se
> > messages is if the ASP crashes and then come back up again.  Then,
> > however, if the SCTP association reamins open, it is the responsibili=
ty
> > of the system to send ASP DOWN.  Or, in any case, when the ASP comes
> > back up, it sends ASP UP, that clears all registrations anyway.
> >
> > I'm sorry, but you have not presented a reasonable situation yet.
> > Please try again.
> >
> > --brian
> >
> > On Mon, 14 Jul 2003, Umesh Vats wrote:
> >
> > > It is a very reasonable scenario. On the nodes, which work
> > > in highly-available fault tolerant modes, some message
> > > losses are expected when active/standby switchovers happen.
> > >
> > > But there should be mechanism for recovery with minimum impact.
> > >
> > > In the specific case I mentioned, if the ASP receives RC in
> > > REG-RSP, it can easily correlate RC with RK. Then, if everything
> > > else is OK, it can keep using same RK. Or it can send request to
> > > SGP for deregistering RK (using RC receivd from SGP).
> > >
> > > thanks
> > > Umesh
> > >
> > >
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: Monday, July 14, 2003 2:30 PM
> > > To: Umesh Vats
> > > Cc: VIvek Toky; SIGTRAN e-mail list
> > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > >
> > >
> > > Umesh,
> > >
> > > An implementation that is discarding messages is broken.
> > >
> > > --brian
> > >
> > > On Mon, 14 Jul 2003, Umesh Vats wrote:
> > >
> > > > Brian,
> > > >
> > > > REG-RSP may be lost at the local ASP node itself. And anyway
> > > > REG-RSP is being generated by SG with local Rk Id. It can put
> > > > RC also so that ASP can correlate RC and RK in such cases.
> > > >
> > > > Umesh
> > > >
> > > > -----Original Message-----
> > > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On
Behalf
> Of
> > > > Brian F. G. Bidulock
> > > > Sent: Monday, July 14, 2003 11:23 AM
> > > > To: VIvek Toky
> > > > Cc: umesh@ccpu.com; SIGTRAN e-mail list
> > > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > > >
> > > >
> > > > VIvek,
> > > >
> > > > REG response cannot be lost.
> > > >
> > > > --brian
> > > >
> > > > On Mon, 14 Jul 2003, VIvek Toky wrote:
> > > >
> > > > > I think Umesh is right. If REG Response is lost ASP will never =
be
> able
> > > to
> > > > > register the same RK again. As it doesn't know the RC so it wou=
ld
> not
> > be
> > > > > able to de-register it and it can also not register it again.
> > > > >
> > > > > -Vivek.
> > > > >
> > > > > -----Original Message-----
> > > > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On
> Behalf
> > Of
> > > > > Umesh Vats
> > > > > Sent: Monday, July 14, 2003 10:08 AM
> > > > > To: SIGTRAN e-mail list
> > > > > Subject: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > > > >
> > > > >
> > > > > Right now, if ASP sends REG-REQ for an already registered routi=
ng
> key,
> > > > > SGP sends unsuccesful REG-RSP.
> > > > > But in some scenarios, succesful REG-RSP from SGP to ASP may be
lost
> > > > > due to network problems/local node problems. In these cases, AS=
P
may
> > > > > retry REG-REQ.
> > > > > In such scenarios (duplicate REG-REQ), as per M3UA IG V03 SGP
sends
> > > > REG-RSP
> > > > > with error as "Error - Cannot Support Unique Routing"
> > > > >
> > > > > Is it possible for SGP to include the Routing Context (RC assig=
ned
> by
> > > SGP
> > > > > for the local rk id received in the first REG-REQ from ASP) in
> > addition
> > > > > to the "Local_Routing Key Identifier" in the unsuccessful REG-R=
SP
> > > message
> > > > > mentioned above. This will allow ASP to recover the RC value.
> > > > >
> > > > > Thanks
> > > > > Umesh
> > > > >
> > > > > ---
> > > > > Outgoing mail is certified Virus Free.
> > > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/200=
3
> > > > >
> > > > >
> > > > > _______________________________________________
> > > > > 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    =A6 The reasonable man adapts himself to =
the =A6
> > > > bidulock@openss7.org    =A6 world; the unreasonable one persists =
in  =A6
> > > > http://www.openss7.org/ =A6 trying  to adapt the  world  to himse=
lf. =A6
> > > >                         =A6 Therefore  all  progress  depends on =
the =A6
> > > >                         =A6 unreasonable man. -- George Bernard S=
haw =A6
> > > >
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > >
> > > > ---
> > > > Incoming mail is certified Virus Free.
> > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > > >
> > > > ---
> > > > Outgoing mail is certified Virus Free.
> > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > > --
> > > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to th=
e =A6
> > > bidulock@openss7.org    =A6 world; the unreasonable one persists in=
  =A6
> > > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself=
. =A6
> > >                         =A6 Therefore  all  progress  depends on th=
e =A6
> > >                         =A6 unreasonable man. -- George Bernard Sha=
w =A6
> > >
> > > ---
> > > Incoming mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > > ---
> > > Outgoing mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > --
> > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =
=A6
> > bidulock@openss7.org    =A6 world; the unreasonable one persists in  =
=A6
> > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =
=A6
> >                         =A6 Therefore  all  progress  depends on the =
=A6
> >                         =A6 unreasonable man. -- George Bernard Shaw =
=A6
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> > ---
> > Incoming mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > ---
> > Outgoing mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003


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



From exim@www1.ietf.org  Mon Jul 14 19:01:26 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA04163
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 19:01:26 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cCJZ-0001st-Cv
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 19:01:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6EN11NY007231
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 19:01:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cCJZ-0001sN-0r; Mon, 14 Jul 2003 19:01:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cCJD-0001rs-4C
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 19:00:39 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA04145
	for <SIGTRAN@ietf.org>; Mon, 14 Jul 2003 19:00:34 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cCJ9-0001xw-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 19:00:35 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[wbSy8ZWxHdwM6snVTshGg5NylkimOYQl])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cCJ8-0001xt-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 19:00:34 -0400
Received: from ns.pigworks.openss7.net (IDENT:tVwacw7yFridK6WsvzTwDLvOx1AMc39f@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6EN0SP25141;
	Mon, 14 Jul 2003 17:00:28 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6EN0RP28609;
	Mon, 14 Jul 2003 17:00:27 -0600
Date: Mon, 14 Jul 2003 17:00:27 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Umesh Vats <umesh@ccpu.com>
Cc: SIGTRAN e-mail list <SIGTRAN@ietf.org>, Vivek Toky <vivektoky@ccpu.com>
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Message-ID: <20030714170027.A28531@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Umesh Vats <umesh@ccpu.com>,
	SIGTRAN e-mail list <SIGTRAN@ietf.org>,
	Vivek Toky <vivektoky@ccpu.com>
References: <20030714164617.A28376@openss7.org> <ACEMIJONPBBPDLOKCABHOEDJCBAA.umesh@ccpu.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <ACEMIJONPBBPDLOKCABHOEDJCBAA.umesh@ccpu.com>; from umesh@ccpu.com on Mon, Jul 14, 2003 at 04:01:27PM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6EN0SP25141
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Umesh,

When host1 sends REG REQ have is save local-routing-key-id in
common data.  When host3 sends messages to host1 and 2 have it
save a local copy until acknowledged by the host to which it was
sent.  If host1 goes down between receiving and REG RSP and before
acknowledging it, host3 will send REG RSP to host2.  host2 uses
local-routing-key-id in REG RSP to correlate REG REQ from common
data.

Problem solved.

There is no need for such a design to lose messages, particularly
REG RSP.

Please don't expect the protocol to change to accomodate poor design.

--brian

On Mon, 14 Jul 2003, Umesh Vats wrote:

> What do others think about this issue ?
> I think, in a practical scenarios its very musch possible
> that message loss may happen between ASP and SGP.
>=20
> Currently SGP sends REG-RSP in case duplicate REG-REQ for
> same RK is received. If we assume ideal scenario where no message loss
> can ever happen, SGP should simply dump REG-REQ and not send REG-RSP.
>=20
> But if it is sending REG-RSP with error (as proposed in M3UA IG-V04),
> then sufficient information should be included so that ASP may come
> to know the correlation between RC and RK.
>=20
> thanks
> Umesh
>=20
>=20
>=20
>=20
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Monday, July 14, 2003 3:46 PM
> To: VIvek Toky
> Cc: Umesh Vats; SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>=20
>=20
> VIvek,
>=20
> Because you would have to handle ANY messages being lost to
> accomodate this approach.  REG RSP would not be enough.  I
> don't see the point in putting a half-bandaid on something
> that isn't broken.
>=20
> Make you distributed system either not lose messages, or
> perform ASP Down/Up procedures between switchover.
>=20
> I do have a distributed implementation, but it does not lose
> M3UA messages without an ASP going down.
>=20
> In the diagram below, where is M3UA?  If it is not riding
> directly on top of SCTP (it is after all an SCTP Adaptation
> Layer) then it is not provided for in the M3UA specification.
> If you have M3UA running on host3, there is no problem, it
> will not lose messages.
>=20
> If you are extending multiple instances of M3UA running over a
> single SCTP association, this is not supported.  (1 ASP, 1
> SCTP Association).
>=20
> Am I clear?  Don't expect M3UA to accomodate your unsupported
> and poor distributed design.  Make a better design.
>=20
> The ASP that loses M3UA messages is either ASP-DOWN or broken.
>=20
> --brian
>=20
> On Mon, 14 Jul 2003, VIvek Toky wrote:
>=20
> > Brian,
> >    Probably your thinking is limilted to your implementation only. So=
me
> > implmentions may internally be fault tolerant or distributed. Which m=
eans
> > that multiple M3UA instances can be designed to use same SCTP assocat=
ion
> and
> > to represent single ASP. Now, explaining you the advantage  of that m=
ay be
> > far beyond the topic but I don't see any harm in including RC in the
> > messages, as some implementaions may use it to their benefit.
> >
> >   I am not sure why are you objecting to this proposal , do you mean =
to
> say
> > that its difficult for your implemntaion to find an RC for already
> > registered RK ?
> >
> > regards,
> > Vivek.
> >
> > -----Original Message-----
> > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf=
 Of
> > Brian F. G. Bidulock
> > Sent: Monday, July 14, 2003 3:35 PM
> > To: Umesh Vats
> > Cc: VIvek Toky; SIGTRAN e-mail list
> > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> >
> >
> > Umesh,
> >
> > M3UA only supports 1 ASP per SCTP association.  You configuration
> > is not supported.
> >
> > By your argument, any M3UA message could be lost between SCTP
> > and M3UA.  This is not provided for the the specification.  Short
> > of having M3UA acknowledge and retransmit ALL messages, there will
> > not be support in M3UA for this arrangement.
> >
> > You need a better design.
> >
> > --brian
> >
> > On Mon, 14 Jul 2003, Umesh Vats wrote:
> >
> > > Brian,
> > >
> > > I am talking about node level fault-tolerance.
> > > Consider the following scenario :
> > >
> > >       host1
> > >     ASP1-Active---|
> > >         |         |    host3
> > >     Common Data   |---SCTP-------association--------SGP
> > >         |         |
> > >     ASP1-Standby--|
> > >        host2
> > >
> > > Here, I am considering node level fault-tolerance. Two,
> > > ASPs are operating in active/standby mode. Both the copies
> > > represent single ASP to SGP (via single SCTP association)
> > > where SCTP is running on host other than those of ASPs.
> > > Also, both the ASPs refer to same data(AS, ASP, Association states,
> > > routing contexts).
> > >
> > > Suppose ASP1-Active sends REG-REQ for RK1 to SGP. SGP responds back
> > > with REG-RSP to ASP1. Now, ASP1-Active receives the REG-RSP
> > > but before it could update the common data, it goes down. In
> > > this case, switchover happens and ASP1-Standby become active.
> > > Now, ASP1-Standby sees that RK1 is not registered and sends REG-REQ
> > > again to SGP. The problem comes here.
> > >
> > > The other method you discussed (sending ASP-DOWN) will bring down
> > > remaing application servers also which may be already actively rece=
iving
> > > traffic.
> > >
> > > Thanks
> > > Umesh
> > >
> > > -----Original Message-----
> > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Beha=
lf Of
> > > Brian F. G. Bidulock
> > > Sent: Monday, July 14, 2003 3:11 PM
> > > To: Umesh Vats
> > > Cc: VIvek Toky; SIGTRAN e-mail list
> > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > >
> > >
> > > Umesh,
> > >
> > > Active/standby switchovers occur BETWEEN SCTP associations.  REG RS=
P is
> > > only relevant WITHIN an SCTP association.  So the scenario you desc=
ribe
> > > does not apply.
> > >
> > > Why does your ASP lose REG RSP messages?  Is is so poorly designed?
> > >
> > > The only scenario that we considered for the RFC where the ASP may =
lose
> > > messages is if the ASP crashes and then come back up again.  Then,
> > > however, if the SCTP association reamins open, it is the responsibi=
lity
> > > of the system to send ASP DOWN.  Or, in any case, when the ASP come=
s
> > > back up, it sends ASP UP, that clears all registrations anyway.
> > >
> > > I'm sorry, but you have not presented a reasonable situation yet.
> > > Please try again.
> > >
> > > --brian
> > >
> > > On Mon, 14 Jul 2003, Umesh Vats wrote:
> > >
> > > > It is a very reasonable scenario. On the nodes, which work
> > > > in highly-available fault tolerant modes, some message
> > > > losses are expected when active/standby switchovers happen.
> > > >
> > > > But there should be mechanism for recovery with minimum impact.
> > > >
> > > > In the specific case I mentioned, if the ASP receives RC in
> > > > REG-RSP, it can easily correlate RC with RK. Then, if everything
> > > > else is OK, it can keep using same RK. Or it can send request to
> > > > SGP for deregistering RK (using RC receivd from SGP).
> > > >
> > > > thanks
> > > > Umesh
> > > >
> > > >
> > > > -----Original Message-----
> > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > Sent: Monday, July 14, 2003 2:30 PM
> > > > To: Umesh Vats
> > > > Cc: VIvek Toky; SIGTRAN e-mail list
> > > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > > >
> > > >
> > > > Umesh,
> > > >
> > > > An implementation that is discarding messages is broken.
> > > >
> > > > --brian
> > > >
> > > > On Mon, 14 Jul 2003, Umesh Vats wrote:
> > > >
> > > > > Brian,
> > > > >
> > > > > REG-RSP may be lost at the local ASP node itself. And anyway
> > > > > REG-RSP is being generated by SG with local Rk Id. It can put
> > > > > RC also so that ASP can correlate RC and RK in such cases.
> > > > >
> > > > > Umesh
> > > > >
> > > > > -----Original Message-----
> > > > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On
> Behalf
> > Of
> > > > > Brian F. G. Bidulock
> > > > > Sent: Monday, July 14, 2003 11:23 AM
> > > > > To: VIvek Toky
> > > > > Cc: umesh@ccpu.com; SIGTRAN e-mail list
> > > > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > > > >
> > > > >
> > > > > VIvek,
> > > > >
> > > > > REG response cannot be lost.
> > > > >
> > > > > --brian
> > > > >
> > > > > On Mon, 14 Jul 2003, VIvek Toky wrote:
> > > > >
> > > > > > I think Umesh is right. If REG Response is lost ASP will neve=
r be
> > able
> > > > to
> > > > > > register the same RK again. As it doesn't know the RC so it w=
ould
> > not
> > > be
> > > > > > able to de-register it and it can also not register it again.
> > > > > >
> > > > > > -Vivek.
> > > > > >
> > > > > > -----Original Message-----
> > > > > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]O=
n
> > Behalf
> > > Of
> > > > > > Umesh Vats
> > > > > > Sent: Monday, July 14, 2003 10:08 AM
> > > > > > To: SIGTRAN e-mail list
> > > > > > Subject: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > > > > >
> > > > > >
> > > > > > Right now, if ASP sends REG-REQ for an already registered rou=
ting
> > key,
> > > > > > SGP sends unsuccesful REG-RSP.
> > > > > > But in some scenarios, succesful REG-RSP from SGP to ASP may =
be
> lost
> > > > > > due to network problems/local node problems. In these cases, =
ASP
> may
> > > > > > retry REG-REQ.
> > > > > > In such scenarios (duplicate REG-REQ), as per M3UA IG V03 SGP
> sends
> > > > > REG-RSP
> > > > > > with error as "Error - Cannot Support Unique Routing"
> > > > > >
> > > > > > Is it possible for SGP to include the Routing Context (RC ass=
igned
> > by
> > > > SGP
> > > > > > for the local rk id received in the first REG-REQ from ASP) i=
n
> > > addition
> > > > > > to the "Local_Routing Key Identifier" in the unsuccessful REG=
-RSP
> > > > message
> > > > > > mentioned above. This will allow ASP to recover the RC value.
> > > > > >
> > > > > > Thanks
> > > > > > Umesh
> > > > > >
> > > > > > ---
> > > > > > Outgoing mail is certified Virus Free.
> > > > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2=
003
> > > > > >
> > > > > >
> > > > > > _______________________________________________
> > > > > > 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    =A6 The reasonable man adapts himself t=
o the =A6
> > > > > bidulock@openss7.org    =A6 world; the unreasonable one persist=
s in  =A6
> > > > > http://www.openss7.org/ =A6 trying  to adapt the  world  to him=
self. =A6
> > > > >                         =A6 Therefore  all  progress  depends o=
n the =A6
> > > > >                         =A6 unreasonable man. -- George Bernard=
 Shaw =A6
> > > > >
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > >
> > > > > ---
> > > > > Incoming mail is certified Virus Free.
> > > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/200=
3
> > > > >
> > > > > ---
> > > > > Outgoing mail is certified Virus Free.
> > > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/200=
3
> > > >
> > > > --
> > > > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to =
the =A6
> > > > bidulock@openss7.org    =A6 world; the unreasonable one persists =
in  =A6
> > > > http://www.openss7.org/ =A6 trying  to adapt the  world  to himse=
lf. =A6
> > > >                         =A6 Therefore  all  progress  depends on =
the =A6
> > > >                         =A6 unreasonable man. -- George Bernard S=
haw =A6
> > > >
> > > > ---
> > > > Incoming mail is certified Virus Free.
> > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > > >
> > > > ---
> > > > Outgoing mail is certified Virus Free.
> > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > > --
> > > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to th=
e =A6
> > > bidulock@openss7.org    =A6 world; the unreasonable one persists in=
  =A6
> > > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself=
. =A6
> > >                         =A6 Therefore  all  progress  depends on th=
e =A6
> > >                         =A6 unreasonable man. -- George Bernard Sha=
w =A6
> > >
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >
> > > ---
> > > Incoming mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > > ---
> > > Outgoing mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > --
> > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =
=A6
> > bidulock@openss7.org    =A6 world; the unreasonable one persists in  =
=A6
> > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =
=A6
> >                         =A6 Therefore  all  progress  depends on the =
=A6
> >                         =A6 unreasonable man. -- George Bernard Shaw =
=A6
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
>=20
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>=20
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>=20
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>=20
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Mon Jul 14 19:15:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA04434
	for <sigtran-archive@odin.ietf.org>; Mon, 14 Jul 2003 19:15:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cCX7-0002so-Q8
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 19:15:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ENF1mj011081
	for sigtran-archive@odin.ietf.org; Mon, 14 Jul 2003 19:15:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cCX6-0002sU-RT; Mon, 14 Jul 2003 19:15:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cCWp-0002s4-8n
	for sigtran@optimus.ietf.org; Mon, 14 Jul 2003 19:14:43 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA04425
	for <SIGTRAN@ietf.org>; Mon, 14 Jul 2003 19:14:39 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cCWn-00021l-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 19:14:41 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19cCWl-00021i-00
	for SIGTRAN@ietf.org; Mon, 14 Jul 2003 19:14:39 -0400
Received: from [172.16.1.40] (helo=ccpu00251)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19cCWX-00025N-3z; Mon, 14 Jul 2003 16:14:25 -0700
Reply-To: <umesh@ccpu.com>
From: "Umesh Vats" <umesh@ccpu.com>
To: <bidulock@openss7.org>
Cc: "SIGTRAN e-mail list" <SIGTRAN@ietf.org>,
        "Vivek Toky" <vivektoky@ccpu.com>
Subject: RE: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Date: Mon, 14 Jul 2003 16:19:58 -0700
Message-ID: <ACEMIJONPBBPDLOKCABHIEDKCBAA.umesh@ccpu.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0)
In-Reply-To: <20030714170027.A28531@openss7.org>
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id TAA04426
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Brian,

I am sure you have sufficient background of highly-available/distributed
systems as your implementation is 100% fault tolerant with
nil (0%) message loss.

In the given scenario, what if host1 receives REG-RSP from host3 but befo=
re
updating
the RC received in message to common data, host1 goes down.

In this case, host2 doesn't know the RC and RK relation.
Performing ASP-DOWN procedure will make ASP DOWN for all the Application
Servers
for which traffic exchange might be happening.

Infact, your implementation can use this idea to further decrease message
loss
from 0% to -1% percentage or less :)

Thanks
Umesh

-----Original Message-----
From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
Sent: Monday, July 14, 2003 4:00 PM
To: Umesh Vats
Cc: SIGTRAN e-mail list; Vivek Toky
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP


Umesh,

When host1 sends REG REQ have is save local-routing-key-id in
common data.  When host3 sends messages to host1 and 2 have it
save a local copy until acknowledged by the host to which it was
sent.  If host1 goes down between receiving and REG RSP and before
acknowledging it, host3 will send REG RSP to host2.  host2 uses
local-routing-key-id in REG RSP to correlate REG REQ from common
data.

Problem solved.

There is no need for such a design to lose messages, particularly
REG RSP.

Please don't expect the protocol to change to accomodate poor design.

--brian

On Mon, 14 Jul 2003, Umesh Vats wrote:

> What do others think about this issue ?
> I think, in a practical scenarios its very musch possible
> that message loss may happen between ASP and SGP.
>
> Currently SGP sends REG-RSP in case duplicate REG-REQ for
> same RK is received. If we assume ideal scenario where no message loss
> can ever happen, SGP should simply dump REG-REQ and not send REG-RSP.
>
> But if it is sending REG-RSP with error (as proposed in M3UA IG-V04),
> then sufficient information should be included so that ASP may come
> to know the correlation between RC and RK.
>
> thanks
> Umesh
>
>
>
>
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Monday, July 14, 2003 3:46 PM
> To: VIvek Toky
> Cc: Umesh Vats; SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>
>
> VIvek,
>
> Because you would have to handle ANY messages being lost to
> accomodate this approach.  REG RSP would not be enough.  I
> don't see the point in putting a half-bandaid on something
> that isn't broken.
>
> Make you distributed system either not lose messages, or
> perform ASP Down/Up procedures between switchover.
>
> I do have a distributed implementation, but it does not lose
> M3UA messages without an ASP going down.
>
> In the diagram below, where is M3UA?  If it is not riding
> directly on top of SCTP (it is after all an SCTP Adaptation
> Layer) then it is not provided for in the M3UA specification.
> If you have M3UA running on host3, there is no problem, it
> will not lose messages.
>
> If you are extending multiple instances of M3UA running over a
> single SCTP association, this is not supported.  (1 ASP, 1
> SCTP Association).
>
> Am I clear?  Don't expect M3UA to accomodate your unsupported
> and poor distributed design.  Make a better design.
>
> The ASP that loses M3UA messages is either ASP-DOWN or broken.
>
> --brian
>
> On Mon, 14 Jul 2003, VIvek Toky wrote:
>
> > Brian,
> >    Probably your thinking is limilted to your implementation only. So=
me
> > implmentions may internally be fault tolerant or distributed. Which
means
> > that multiple M3UA instances can be designed to use same SCTP assocat=
ion
> and
> > to represent single ASP. Now, explaining you the advantage  of that m=
ay
be
> > far beyond the topic but I don't see any harm in including RC in the
> > messages, as some implementaions may use it to their benefit.
> >
> >   I am not sure why are you objecting to this proposal , do you mean =
to
> say
> > that its difficult for your implemntaion to find an RC for already
> > registered RK ?
> >
> > regards,
> > Vivek.
> >
> > -----Original Message-----
> > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf=
 Of
> > Brian F. G. Bidulock
> > Sent: Monday, July 14, 2003 3:35 PM
> > To: Umesh Vats
> > Cc: VIvek Toky; SIGTRAN e-mail list
> > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> >
> >
> > Umesh,
> >
> > M3UA only supports 1 ASP per SCTP association.  You configuration
> > is not supported.
> >
> > By your argument, any M3UA message could be lost between SCTP
> > and M3UA.  This is not provided for the the specification.  Short
> > of having M3UA acknowledge and retransmit ALL messages, there will
> > not be support in M3UA for this arrangement.
> >
> > You need a better design.
> >
> > --brian
> >
> > On Mon, 14 Jul 2003, Umesh Vats wrote:
> >
> > > Brian,
> > >
> > > I am talking about node level fault-tolerance.
> > > Consider the following scenario :
> > >
> > >       host1
> > >     ASP1-Active---|
> > >         |         |    host3
> > >     Common Data   |---SCTP-------association--------SGP
> > >         |         |
> > >     ASP1-Standby--|
> > >        host2
> > >
> > > Here, I am considering node level fault-tolerance. Two,
> > > ASPs are operating in active/standby mode. Both the copies
> > > represent single ASP to SGP (via single SCTP association)
> > > where SCTP is running on host other than those of ASPs.
> > > Also, both the ASPs refer to same data(AS, ASP, Association states,
> > > routing contexts).
> > >
> > > Suppose ASP1-Active sends REG-REQ for RK1 to SGP. SGP responds back
> > > with REG-RSP to ASP1. Now, ASP1-Active receives the REG-RSP
> > > but before it could update the common data, it goes down. In
> > > this case, switchover happens and ASP1-Standby become active.
> > > Now, ASP1-Standby sees that RK1 is not registered and sends REG-REQ
> > > again to SGP. The problem comes here.
> > >
> > > The other method you discussed (sending ASP-DOWN) will bring down
> > > remaing application servers also which may be already actively
receiving
> > > traffic.
> > >
> > > Thanks
> > > Umesh
> > >
> > > -----Original Message-----
> > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Beha=
lf
Of
> > > Brian F. G. Bidulock
> > > Sent: Monday, July 14, 2003 3:11 PM
> > > To: Umesh Vats
> > > Cc: VIvek Toky; SIGTRAN e-mail list
> > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > >
> > >
> > > Umesh,
> > >
> > > Active/standby switchovers occur BETWEEN SCTP associations.  REG RS=
P
is
> > > only relevant WITHIN an SCTP association.  So the scenario you
describe
> > > does not apply.
> > >
> > > Why does your ASP lose REG RSP messages?  Is is so poorly designed?
> > >
> > > The only scenario that we considered for the RFC where the ASP may
lose
> > > messages is if the ASP crashes and then come back up again.  Then,
> > > however, if the SCTP association reamins open, it is the
responsibility
> > > of the system to send ASP DOWN.  Or, in any case, when the ASP come=
s
> > > back up, it sends ASP UP, that clears all registrations anyway.
> > >
> > > I'm sorry, but you have not presented a reasonable situation yet.
> > > Please try again.
> > >
> > > --brian
> > >
> > > On Mon, 14 Jul 2003, Umesh Vats wrote:
> > >
> > > > It is a very reasonable scenario. On the nodes, which work
> > > > in highly-available fault tolerant modes, some message
> > > > losses are expected when active/standby switchovers happen.
> > > >
> > > > But there should be mechanism for recovery with minimum impact.
> > > >
> > > > In the specific case I mentioned, if the ASP receives RC in
> > > > REG-RSP, it can easily correlate RC with RK. Then, if everything
> > > > else is OK, it can keep using same RK. Or it can send request to
> > > > SGP for deregistering RK (using RC receivd from SGP).
> > > >
> > > > thanks
> > > > Umesh
> > > >
> > > >
> > > > -----Original Message-----
> > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > Sent: Monday, July 14, 2003 2:30 PM
> > > > To: Umesh Vats
> > > > Cc: VIvek Toky; SIGTRAN e-mail list
> > > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > > >
> > > >
> > > > Umesh,
> > > >
> > > > An implementation that is discarding messages is broken.
> > > >
> > > > --brian
> > > >
> > > > On Mon, 14 Jul 2003, Umesh Vats wrote:
> > > >
> > > > > Brian,
> > > > >
> > > > > REG-RSP may be lost at the local ASP node itself. And anyway
> > > > > REG-RSP is being generated by SG with local Rk Id. It can put
> > > > > RC also so that ASP can correlate RC and RK in such cases.
> > > > >
> > > > > Umesh
> > > > >
> > > > > -----Original Message-----
> > > > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On
> Behalf
> > Of
> > > > > Brian F. G. Bidulock
> > > > > Sent: Monday, July 14, 2003 11:23 AM
> > > > > To: VIvek Toky
> > > > > Cc: umesh@ccpu.com; SIGTRAN e-mail list
> > > > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > > > >
> > > > >
> > > > > VIvek,
> > > > >
> > > > > REG response cannot be lost.
> > > > >
> > > > > --brian
> > > > >
> > > > > On Mon, 14 Jul 2003, VIvek Toky wrote:
> > > > >
> > > > > > I think Umesh is right. If REG Response is lost ASP will neve=
r
be
> > able
> > > > to
> > > > > > register the same RK again. As it doesn't know the RC so it
would
> > not
> > > be
> > > > > > able to de-register it and it can also not register it again.
> > > > > >
> > > > > > -Vivek.
> > > > > >
> > > > > > -----Original Message-----
> > > > > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]O=
n
> > Behalf
> > > Of
> > > > > > Umesh Vats
> > > > > > Sent: Monday, July 14, 2003 10:08 AM
> > > > > > To: SIGTRAN e-mail list
> > > > > > Subject: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > > > > >
> > > > > >
> > > > > > Right now, if ASP sends REG-REQ for an already registered
routing
> > key,
> > > > > > SGP sends unsuccesful REG-RSP.
> > > > > > But in some scenarios, succesful REG-RSP from SGP to ASP may =
be
> lost
> > > > > > due to network problems/local node problems. In these cases, =
ASP
> may
> > > > > > retry REG-REQ.
> > > > > > In such scenarios (duplicate REG-REQ), as per M3UA IG V03 SGP
> sends
> > > > > REG-RSP
> > > > > > with error as "Error - Cannot Support Unique Routing"
> > > > > >
> > > > > > Is it possible for SGP to include the Routing Context (RC
assigned
> > by
> > > > SGP
> > > > > > for the local rk id received in the first REG-REQ from ASP) i=
n
> > > addition
> > > > > > to the "Local_Routing Key Identifier" in the unsuccessful
REG-RSP
> > > > message
> > > > > > mentioned above. This will allow ASP to recover the RC value.
> > > > > >
> > > > > > Thanks
> > > > > > Umesh
> > > > > >
> > > > > > ---
> > > > > > Outgoing mail is certified Virus Free.
> > > > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2=
003
> > > > > >
> > > > > >
> > > > > > _______________________________________________
> > > > > > 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    =A6 The reasonable man adapts himself t=
o the
=A6
> > > > > bidulock@openss7.org    =A6 world; the unreasonable one persist=
s in
=A6
> > > > > http://www.openss7.org/ =A6 trying  to adapt the  world  to him=
self.
=A6
> > > > >                         =A6 Therefore  all  progress  depends o=
n the
=A6
> > > > >                         =A6 unreasonable man. -- George Bernard=
 Shaw
=A6
> > > > >
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > >
> > > > > ---
> > > > > Incoming mail is certified Virus Free.
> > > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/200=
3
> > > > >
> > > > > ---
> > > > > Outgoing mail is certified Virus Free.
> > > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/200=
3
> > > >
> > > > --
> > > > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to =
the =A6
> > > > bidulock@openss7.org    =A6 world; the unreasonable one persists =
in  =A6
> > > > http://www.openss7.org/ =A6 trying  to adapt the  world  to himse=
lf. =A6
> > > >                         =A6 Therefore  all  progress  depends on =
the =A6
> > > >                         =A6 unreasonable man. -- George Bernard S=
haw =A6
> > > >
> > > > ---
> > > > Incoming mail is certified Virus Free.
> > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > > >
> > > > ---
> > > > Outgoing mail is certified Virus Free.
> > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > > --
> > > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to th=
e =A6
> > > bidulock@openss7.org    =A6 world; the unreasonable one persists in=
  =A6
> > > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself=
. =A6
> > >                         =A6 Therefore  all  progress  depends on th=
e =A6
> > >                         =A6 unreasonable man. -- George Bernard Sha=
w =A6
> > >
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >
> > > ---
> > > Incoming mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > > ---
> > > Outgoing mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > --
> > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =
=A6
> > bidulock@openss7.org    =A6 world; the unreasonable one persists in  =
=A6
> > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =
=A6
> >                         =A6 Therefore  all  progress  depends on the =
=A6
> >                         =A6 unreasonable man. -- George Bernard Shaw =
=A6
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
>
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003


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



From exim@www1.ietf.org  Tue Jul 15 00:08:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA28673
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 00:08:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cH6i-0004Nw-IJ
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 00:08:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6F484lQ016850
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 00:08:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cH6e-0004NK-3i; Tue, 15 Jul 2003 00:08:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cH5h-00041o-0V
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 00:07:01 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA28420
	for <sigtran@ietf.org>; Tue, 15 Jul 2003 00:06:57 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cH5e-0000PY-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 00:06:58 -0400
Received: from auds952.usa.alcatel.com ([143.209.238.7])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cH5d-0000Nu-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 00:06:57 -0400
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6F46NB7025850
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 23:06:25 -0500 (CDT)
Received: from sun2685.ssd.usa.alcatel.com (sun2685.ssd.usa.alcatel.com [143.209.151.167])
	by ssd.usa.alcatel.com (8.12.8/8.12.8) with ESMTP id h6F45c1C006571
	for <sigtran@ietf.org>; Mon, 14 Jul 2003 23:05:38 -0500 (CDT)
Date: Mon, 14 Jul 2003 23:05:38 -0500 (CDT)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "''sigtran@ietf.org' '" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04  Section 3.14
In-Reply-To: <20030714121937.A22972@openss7.org>
Message-ID: <Pine.SOL.4.10.10307142254250.21400-100000@sun2685.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,

 In n=5(example) case, if 4 ASPes are in DOWN state and 1 ASP is in ACTIVE
 state and which handles the traffic.Suppose ACTIVE ASP moves to INACTIVE 
 or DOWN state so SGP tries sending NTFY(AS-DOWN) to other ASPes.

 How does an ASP process NTFY message in DOWN state?


--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 Mon, 14 Jul 2003, Brian F. G. Bidulock wrote:

> Samuel,
> 
> To what end?
> 
> --brian
> 
> Samuel Dur D. Jeyaseelan wrote:                                                     (Mon, 14 Jul 2003 09:06:36)
> > 
> > Javier,Brian,
> > 
> > 
> >  (i was bit late to enter into this conversation)
> > 
> >  Including a new status type 'AS-DOWN' in NTFY messages is a good idea,
> >  but section 4.3.4.5 in RFC says an ASP in DOWN state will not process
> >  NTFY message....?
> > 
> > 
> >  Let me take an example of n=5, SGP needs 5 ASPes in ACTIVE state
> >  to declare the AS as ACTIVE(true-only for start case).
> > 
> >  Start-case
> >  ----------
> > 
> >  ACTIVE ASPes    INACTIVE ASPes  DOWN ASPes    AS state        case-#
> >  --------------------------------------------------------------------
> >     0             0               5            DOWN            case-1
> >     0             1               4            DOWN            case-2
> >     0             2               3            DOWN            case-3
> >     .
> >     .
> >     .
> >     0             5               0            INACTIVE        case-4
> >     1             4               0            INACTIVE        case-5
> >     .
> >     .
> >     .
> >     5             0               0            ACTIVE          case-6
> > 
> >    (assume traffic flows)
> > 
> >     Non-start case
> >     --------------
> >     4             1               0            ACTIVE          case-7
> >     3             2               0            ACTIVE          case-8
> >     .
> >     .
> >     .
> >     1             4               0            ACTIVE          case-8
> >     0             5               0            INACTIVE        case-9
> >     0             4               1            still INACTIVE  case-10
> >     0             3               2            still INACTIVE  case-11
> >     .
> >     .
> >     .
> >     0             1               4            still INACTIVE  case-12
> >     0             0               5            DOWN            case-13
> > 
> > 
> >  I think we should define the following state-conversion definition.
> > 
> >  for Start-case
> >  --------------
> > 
> >  DN2IA : One ASP moves to ASP-INACTIVE causing that n ASPs are in
> >          ASP-INACTIVE states.
> > 
> >  IA2DN : One ASP moves to ASP-DOWN causing that n ASPs are in
> >          ASP-DOWN state.
> > 
> >  IA2AC : one ASP moves to ASP-ACTIVE, causing number of ASPs are in 
> >          ASP-ACTIVE state.
> > 
> > 
> > Waiting for comments to go ahead.
> > 
> > 
> > --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 Mon, 14 Jul 2003, J.Javier Pastor (EE/EEM) wrote:
> > 
> > > Brian, 
> > > 
> > > What you say it's right, only 1 'n' transition would be kept in the proposal: the IA2AC transition.
> > > 
> > > I think it makes sense since "n" refers only to the capacity of processing traffic and in the AS-INACTIVE state no traffic is processed.
> > > 
> > > Does anyone have any objection to this change?
> > > 
> > > 
> > > Thanks // Javier.
> > > 
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock
> > > To: J.Javier Pastor (EE/EEM)
> > > Cc: 'sigtran@ietf.org'
> > > Sent: 11/07/2003 11:13
> > > Subject: Re: [Sigtran] M3UA IG v04  Section 3.14
> > > 
> > > J.Javier,
> > > 
> > > J.Javier Pastor (EE/EEM) wrote:
> > > (Fri, 11 Jul 2003 10:35:57)
> > > > Hi Brian, 
> > > > 
> > > > You are right saying that there is currently no way of notifying
> > > AS-DOWN. In
> > > > fact "n" only applies to the traffic case.
> > > > 
> > > > Then I see two options:
> > > > 	1- include AS-DOWN in NTFY
> > > > 	2- The AS-DOWN state is only got when all the ASPs are in
> > > ASP-DOWN state.
> > > > 
> > > > I would addopt 2 since it is inline what we are having in the other
> > > UALs. 
> > > 
> > > Yes, please go with 2 as it is also in line with the RFC.
> > > 
> > > That will also leave only 1 'n' transition in the proposal: the IA2AC
> > > transition, correct?
> > > 
> > > --brian
> > > 
> > > > 
> > > > Thanks // Javier.
> > > > 
> > > > 
> > > > > -----Original Message-----
> > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > Sent: jueves, 10 de julio de 2003 17:15
> > > > > To: J.Javier Pastor (EE/EEM)
> > > > > Cc: 'sigtran@ietf.org'
> > > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.14
> > > > > 
> > > > > 
> > > > > Javier,
> > > > > 
> > > > > Section 3.14 says:
> > > > > 
> > > > >    IA2DN: One ASP moves to ASP-DOWN causing that the number of ASPs
> > > in
> > > > >    either ASP-ACTIVE or ASP-INACTIVE drops below n.
> > > > > 
> > > > > Am I correct to say that if n=5 and there are 4 ASPs in the 
> > > > > ASP-INACTIVE
> > > > > and 1 ASP in the ASP-ACTIVE state, and then the 1 active ASP moves
> > > to
> > > > > the ASP-DOWN state and T(r) expires that the AS is to move to 
> > > > > the AS-DOWN
> > > > > state?  As follows:
> > > > > 
> > > > >   If T(r) expires before an ASP becomes ASP-ACTIVE, the SGP MAY stop
> > > > >   queuing messages and discards all
> > > > >   previously queued messages. The AS will move to the 
> > > > > AS-INACTIVE state
> > > > >   if at least the number of ASPs in ASP-INACTIVE sum n, otherwise it
> > > > >   will move to AS-DOWN state.
> > > > > 
> > > > > Now, that is an AS state change and we must send a NTFY to 
> > > > > all the INACTIVE
> > > > > ASPs (the 4 left inactive) of the state change.  But from 
> > > > > Section 3.8.2 of
> > > > > the RFC, there is no message to notify of an AS-DOWN state:
> > > > > 
> > > > >          1    reserved
> > > > >          2    Application Server Inactive (AS-INACTIVE)
> > > > >          3    Application Server Active (AS-ACTIVE)
> > > > >          4    Application Server Pending (AS-PENDING)
> > > > > 
> > > > > You see, we removed it because there was never an instance in which
> > > it
> > > > > would be sent.  You have inserted an instance again.
> > > > > 
> > > > > Aren't you going to have to define a Notify message (the 
> > > > > Reserved value)
> > > > > for "Application Server Down (AS-DOWN)" so that the INACTIVE ASPs
> > > can
> > > > > be properly notified by the SGP?
> > > > > 
> > > > > Particularly when the RFC says in Section 4.3.4.5: (Notify
> > > Procedures)
> > > > > 
> > > > >    A Notify message reflecting a change in the AS state MUST 
> > > > > be sent to
> > > > >    all ASPs in the AS, except those in the ASP-DOWN state, with
> > > > >    appropriate Status Information and any ASP Identifier of the
> > > failed
> > > > >    ASP.  At the ASP, Layer Management is informed with an M-NOTIFY
> > > > >    indication primitive.  The Notify message must be sent 
> > > > > whether the AS
> > > > >    state change was a result of an ASP failure or reception of an
> > > ASP
> > > > >    State management (ASPSM) / ASP Traffic Management (ASPTM)
> > > message.
> > > > >    In the second case, the Notify message MUST be sent after 
> > > > > any related
> > > > >    acknowledgement messages  (e.g., ASP Up Ack, ASP Down Ack, 
> > > > > ASP Active
> > > > >    Ack, or ASP Inactive Ack).
> > > > > 
> > > > > The above tells me that I MUST send a Notify message becase of the
> > > AS
> > > > > state change, but there is no message to inform about AS-DOWN state
> > > > > transition.
> > > > > 
> > > > > What is the purpose of doing this?  What is the protocol 
> > > > > difference between
> > > > > the AS-DOWN state and the AS-INACTIVE state?  What is it that the
> > > SGP
> > > > > does or doesn't do in the one state that it doesn't or does 
> > > > > do in the other?
> > > > > 
> > > > > I don't recall this being discussed at all.
> > > > > 
> > > > > Please clarify.
> > > > > 
> > > > > -- 
> > > > > 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 exim@www1.ietf.org  Tue Jul 15 00:11:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA28946
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 00:11:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cH9a-0004V7-8J
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 00:11:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6F4B2LH017287
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 00:11:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cH9a-0004Ui-4q; Tue, 15 Jul 2003 00:11:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cH9B-0004T2-6u
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 00:10:37 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA28861
	for <sigtran@ietf.org>; Tue, 15 Jul 2003 00:10:33 -0400 (EDT)
From: kakaul@hss.hns.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cH98-0000Vb-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 00:10:34 -0400
Received: from hindon.hss.co.in ([202.54.26.202])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cH97-0000Uj-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 00:10:33 -0400
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h6F47Cs24739;
	Tue, 15 Jul 2003 09:37:12 +0530 (IST)
Received: from dlfmail1.hss.hns.com (localhost [127.0.0.1])
	by ultra.hss.co.in (8.10.0/8.10.0) with ESMTP id h6F4AuS09684;
	Tue, 15 Jul 2003 09:40:57 +0530 (IST)
Subject: RE: [Sigtran] M3UA IG v04 Section 3.18
To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
Cc: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.5  September 22, 2000
Message-ID: <OF321198D1.857EA3E7-ON65256D64.0015BCC7@hss.hns.com>
Date: Tue, 15 Jul 2003 09:35:27 +0530
X-MIMETrack: Serialize by Router on Dlfmail1/HSS(Release 6.0|September 26, 2002) at 07/15/2003
 09:33:09 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>





Hi Javier,
                    This timer may be used by some implementations to
supervise the congestion status of point code if SCON messages send by SGP
are lost in transit to ASP.It was discussed earlier and was agreed upon by
all to keep this timer.I disagree with removing the mention of the timer.

Regards
Kamesh Kaul
www.hssworld.com
Hughes Software Systems




"J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>@ietf.org on
07/14/2003 04:05:58 PM

Sent by:    sigtran-admin@ietf.org


To:    "'Brian F. G. Bidulock '" <bidulock@openss7.org>
cc:    "''sigtran@ietf.org' '" <sigtran@ietf.org>

Subject:    RE: [Sigtran] M3UA IG v04 Section 3.18


Brian,

We'll remove the mention of the timer, if nobody disagrees...

Cheers // Javier.

-----Original Message-----
From: Brian F. G. Bidulock
To: J.Javier Pastor (EE/EEM)
Cc: 'sigtran@ietf.org'
Sent: 11/07/2003 11:07
Subject: Re: [Sigtran] M3UA IG v04 Section 3.18

J.Javier,

J.Javier Pastor (EE/EEM) wrote:
(Fri, 11 Jul 2003 09:56:59)
> Hi Brian,
>
> Pls, see below...
>
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: jueves, 10 de julio de 2003 15:34
> > To: J.Javier Pastor (EE/EEM)
> > Cc: 'sigtran@ietf.org'
> > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> >
> >
> > Javier,
> >
> > 3.18.3 Solution Description
> >
> >    Clarify that the ASP needs to indicate when the congestion
> > level has
> >    changed (including abatement).  Further, the ASP peer can
maintain
> >    a timer, if desired, in case the ASP fails to indicate congestion
> >    abatement.
> >
> > Where are the procedures for this timer described?  This is the only
place
> > that the timer is mentioned.  Please describe the timer or remove
the
> > sentence saying that it has been described...
> >
>
> Do you think is worth include the timer explanation into the new text?
> Otherwise, we may remove that sentence.

Please remove mention of the timer.

--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 exim@www1.ietf.org  Tue Jul 15 00:58:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA02914
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 00:58:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cHt5-0007NR-K4
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 00:58:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6F4w3vJ028358
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 00:58:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cHt3-0007Mj-9Z; Tue, 15 Jul 2003 00:58:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cHsS-0007ME-N5
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 00:57:24 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA02845
	for <sigtran@ietf.org>; Tue, 15 Jul 2003 00:57:20 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cHsP-0001TO-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 00:57:21 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[EXlG4vH7u1W0dM/ekCxcfthHR2ZMqhTb])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cHsO-0001T7-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 00:57:20 -0400
Received: from ns.pigworks.openss7.net (IDENT:eVk83fVaU/G0e37pFR7zQNySg11WJJQc@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6F4vHP29266;
	Mon, 14 Jul 2003 22:57:17 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6F4vGN31967;
	Mon, 14 Jul 2003 22:57:16 -0600
Date: Mon, 14 Jul 2003 22:57:16 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: "''sigtran@ietf.org' '" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04  Section 3.14
Message-ID: <20030714225716.A31916@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"''sigtran@ietf.org' '" <sigtran@ietf.org>
References: <20030714121937.A22972@openss7.org> <Pine.SOL.4.10.10307142254250.21400-100000@sun2685.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.4.10.10307142254250.21400-100000@sun2685.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Mon, Jul 14, 2003 at 11:05:38PM -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>

Samuel,

As the RFC stands the AS is only DOWN when the last ASP is down.  There
is no ASP to send the notification to.

Just what do you think that introductin an AS-DOWN state with ASPs in
the ASP-INACTIVE state would accomplish?  (A: Nothing.  So leave the
RFC alone.)

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Mon, 14 Jul 2003 23:05:38)
> 
> Brian,
> 
>  In n=5(example) case, if 4 ASPes are in DOWN state and 1 ASP is in ACTIVE
>  state and which handles the traffic.Suppose ACTIVE ASP moves to INACTIVE 
>  or DOWN state so SGP tries sending NTFY(AS-DOWN) to other ASPes.
> 
>  How does an ASP process NTFY message in DOWN state?
> 
> 
> --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 Mon, 14 Jul 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > To what end?
> > 
> > --brian
> > 
> > Samuel Dur D. Jeyaseelan wrote:                                                     (Mon, 14 Jul 2003 09:06:36)
> > > 
> > > Javier,Brian,
> > > 
> > > 
> > >  (i was bit late to enter into this conversation)
> > > 
> > >  Including a new status type 'AS-DOWN' in NTFY messages is a good idea,
> > >  but section 4.3.4.5 in RFC says an ASP in DOWN state will not process
> > >  NTFY message....?
> > > 
> > > 
> > >  Let me take an example of n=5, SGP needs 5 ASPes in ACTIVE state
> > >  to declare the AS as ACTIVE(true-only for start case).
> > > 
> > >  Start-case
> > >  ----------
> > > 
> > >  ACTIVE ASPes    INACTIVE ASPes  DOWN ASPes    AS state        case-#
> > >  --------------------------------------------------------------------
> > >     0             0               5            DOWN            case-1
> > >     0             1               4            DOWN            case-2
> > >     0             2               3            DOWN            case-3
> > >     .
> > >     .
> > >     .
> > >     0             5               0            INACTIVE        case-4
> > >     1             4               0            INACTIVE        case-5
> > >     .
> > >     .
> > >     .
> > >     5             0               0            ACTIVE          case-6
> > > 
> > >    (assume traffic flows)
> > > 
> > >     Non-start case
> > >     --------------
> > >     4             1               0            ACTIVE          case-7
> > >     3             2               0            ACTIVE          case-8
> > >     .
> > >     .
> > >     .
> > >     1             4               0            ACTIVE          case-8
> > >     0             5               0            INACTIVE        case-9
> > >     0             4               1            still INACTIVE  case-10
> > >     0             3               2            still INACTIVE  case-11
> > >     .
> > >     .
> > >     .
> > >     0             1               4            still INACTIVE  case-12
> > >     0             0               5            DOWN            case-13
> > > 
> > > 
> > >  I think we should define the following state-conversion definition.
> > > 
> > >  for Start-case
> > >  --------------
> > > 
> > >  DN2IA : One ASP moves to ASP-INACTIVE causing that n ASPs are in
> > >          ASP-INACTIVE states.
> > > 
> > >  IA2DN : One ASP moves to ASP-DOWN causing that n ASPs are in
> > >          ASP-DOWN state.
> > > 
> > >  IA2AC : one ASP moves to ASP-ACTIVE, causing number of ASPs are in 
> > >          ASP-ACTIVE state.
> > > 
> > > 
> > > Waiting for comments to go ahead.
> > > 
> > > 
> > > --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 Mon, 14 Jul 2003, J.Javier Pastor (EE/EEM) wrote:
> > > 
> > > > Brian, 
> > > > 
> > > > What you say it's right, only 1 'n' transition would be kept in the proposal: the IA2AC transition.
> > > > 
> > > > I think it makes sense since "n" refers only to the capacity of processing traffic and in the AS-INACTIVE state no traffic is processed.
> > > > 
> > > > Does anyone have any objection to this change?
> > > > 
> > > > 
> > > > Thanks // Javier.
> > > > 
> > > > -----Original Message-----
> > > > From: Brian F. G. Bidulock
> > > > To: J.Javier Pastor (EE/EEM)
> > > > Cc: 'sigtran@ietf.org'
> > > > Sent: 11/07/2003 11:13
> > > > Subject: Re: [Sigtran] M3UA IG v04  Section 3.14
> > > > 
> > > > J.Javier,
> > > > 
> > > > J.Javier Pastor (EE/EEM) wrote:
> > > > (Fri, 11 Jul 2003 10:35:57)
> > > > > Hi Brian, 
> > > > > 
> > > > > You are right saying that there is currently no way of notifying
> > > > AS-DOWN. In
> > > > > fact "n" only applies to the traffic case.
> > > > > 
> > > > > Then I see two options:
> > > > > 	1- include AS-DOWN in NTFY
> > > > > 	2- The AS-DOWN state is only got when all the ASPs are in
> > > > ASP-DOWN state.
> > > > > 
> > > > > I would addopt 2 since it is inline what we are having in the other
> > > > UALs. 
> > > > 
> > > > Yes, please go with 2 as it is also in line with the RFC.
> > > > 
> > > > That will also leave only 1 'n' transition in the proposal: the IA2AC
> > > > transition, correct?
> > > > 
> > > > --brian
> > > > 
> > > > > 
> > > > > Thanks // Javier.
> > > > > 
> > > > > 
> > > > > > -----Original Message-----
> > > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > > Sent: jueves, 10 de julio de 2003 17:15
> > > > > > To: J.Javier Pastor (EE/EEM)
> > > > > > Cc: 'sigtran@ietf.org'
> > > > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.14
> > > > > > 
> > > > > > 
> > > > > > Javier,
> > > > > > 
> > > > > > Section 3.14 says:
> > > > > > 
> > > > > >    IA2DN: One ASP moves to ASP-DOWN causing that the number of ASPs
> > > > in
> > > > > >    either ASP-ACTIVE or ASP-INACTIVE drops below n.
> > > > > > 
> > > > > > Am I correct to say that if n=5 and there are 4 ASPs in the 
> > > > > > ASP-INACTIVE
> > > > > > and 1 ASP in the ASP-ACTIVE state, and then the 1 active ASP moves
> > > > to
> > > > > > the ASP-DOWN state and T(r) expires that the AS is to move to 
> > > > > > the AS-DOWN
> > > > > > state?  As follows:
> > > > > > 
> > > > > >   If T(r) expires before an ASP becomes ASP-ACTIVE, the SGP MAY stop
> > > > > >   queuing messages and discards all
> > > > > >   previously queued messages. The AS will move to the 
> > > > > > AS-INACTIVE state
> > > > > >   if at least the number of ASPs in ASP-INACTIVE sum n, otherwise it
> > > > > >   will move to AS-DOWN state.
> > > > > > 
> > > > > > Now, that is an AS state change and we must send a NTFY to 
> > > > > > all the INACTIVE
> > > > > > ASPs (the 4 left inactive) of the state change.  But from 
> > > > > > Section 3.8.2 of
> > > > > > the RFC, there is no message to notify of an AS-DOWN state:
> > > > > > 
> > > > > >          1    reserved
> > > > > >          2    Application Server Inactive (AS-INACTIVE)
> > > > > >          3    Application Server Active (AS-ACTIVE)
> > > > > >          4    Application Server Pending (AS-PENDING)
> > > > > > 
> > > > > > You see, we removed it because there was never an instance in which
> > > > it
> > > > > > would be sent.  You have inserted an instance again.
> > > > > > 
> > > > > > Aren't you going to have to define a Notify message (the 
> > > > > > Reserved value)
> > > > > > for "Application Server Down (AS-DOWN)" so that the INACTIVE ASPs
> > > > can
> > > > > > be properly notified by the SGP?
> > > > > > 
> > > > > > Particularly when the RFC says in Section 4.3.4.5: (Notify
> > > > Procedures)
> > > > > > 
> > > > > >    A Notify message reflecting a change in the AS state MUST 
> > > > > > be sent to
> > > > > >    all ASPs in the AS, except those in the ASP-DOWN state, with
> > > > > >    appropriate Status Information and any ASP Identifier of the
> > > > failed
> > > > > >    ASP.  At the ASP, Layer Management is informed with an M-NOTIFY
> > > > > >    indication primitive.  The Notify message must be sent 
> > > > > > whether the AS
> > > > > >    state change was a result of an ASP failure or reception of an
> > > > ASP
> > > > > >    State management (ASPSM) / ASP Traffic Management (ASPTM)
> > > > message.
> > > > > >    In the second case, the Notify message MUST be sent after 
> > > > > > any related
> > > > > >    acknowledgement messages  (e.g., ASP Up Ack, ASP Down Ack, 
> > > > > > ASP Active
> > > > > >    Ack, or ASP Inactive Ack).
> > > > > > 
> > > > > > The above tells me that I MUST send a Notify message becase of the
> > > > AS
> > > > > > state change, but there is no message to inform about AS-DOWN state
> > > > > > transition.
> > > > > > 
> > > > > > What is the purpose of doing this?  What is the protocol 
> > > > > > difference between
> > > > > > the AS-DOWN state and the AS-INACTIVE state?  What is it that the
> > > > SGP
> > > > > > does or doesn't do in the one state that it doesn't or does 
> > > > > > do in the other?
> > > > > > 
> > > > > > I don't recall this being discussed at all.
> > > > > > 
> > > > > > Please clarify.
> > > > > > 
> > > > > > -- 
> > > > > > 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

-- 
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 exim@www1.ietf.org  Tue Jul 15 00:59:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA03031
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 00:59:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cHu1-0007Vf-CB
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 00:59:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6F4x1d6028861
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 00:59:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cHu1-0007VL-73; Tue, 15 Jul 2003 00:59:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cHtm-0007Ul-7B
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 00:58:46 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA02934
	for <sigtran@ietf.org>; Tue, 15 Jul 2003 00:58:41 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cHtj-0001Uq-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 00:58:43 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[T7luigXifcSaC3knGWiguxsoyXEvhjQL])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cHti-0001Ub-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 00:58:42 -0400
Received: from ns.pigworks.openss7.net (IDENT:cAbOz8jarWcXahcRRZZVOAVUBY5dYrzR@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6F4wfP29283;
	Mon, 14 Jul 2003 22:58:41 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6F4wfR32006;
	Mon, 14 Jul 2003 22:58:41 -0600
Date: Mon, 14 Jul 2003 22:58:40 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: kakaul@hss.hns.com
Cc: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
        sigtran@ietf.org
Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
Message-ID: <20030714225840.B31916@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: kakaul@hss.hns.com,
	"J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
	sigtran@ietf.org
References: <OF321198D1.857EA3E7-ON65256D64.0015BCC7@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: <OF321198D1.857EA3E7-ON65256D64.0015BCC7@hss.hns.com>; from kakaul@hss.hns.com on Tue, Jul 15, 2003 at 09:35:27AM +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>

kakaul,

Then add text describing how to use the timer in an interoperable way.
Also, please describe why the peer needs to know about the existence
of this timer to remain interoperable.  Otherwise, there would not be
any need to describe it.

--brian


kakaul@hss.hns.com wrote:                                                           (Tue, 15 Jul 2003 09:35:27)
> 
> 
> 
> 
> Hi Javier,
>                     This timer may be used by some implementations to
> supervise the congestion status of point code if SCON messages send by SGP
> are lost in transit to ASP.It was discussed earlier and was agreed upon by
> all to keep this timer.I disagree with removing the mention of the timer.
> 
> Regards
> Kamesh Kaul
> www.hssworld.com
> Hughes Software Systems
> 
> 
> 
> 
> "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>@ietf.org on
> 07/14/2003 04:05:58 PM
> 
> Sent by:    sigtran-admin@ietf.org
> 
> 
> To:    "'Brian F. G. Bidulock '" <bidulock@openss7.org>
> cc:    "''sigtran@ietf.org' '" <sigtran@ietf.org>
> 
> Subject:    RE: [Sigtran] M3UA IG v04 Section 3.18
> 
> 
> Brian,
> 
> We'll remove the mention of the timer, if nobody disagrees...
> 
> Cheers // Javier.
> 
> -----Original Message-----
> From: Brian F. G. Bidulock
> To: J.Javier Pastor (EE/EEM)
> Cc: 'sigtran@ietf.org'
> Sent: 11/07/2003 11:07
> Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> 
> J.Javier,
> 
> J.Javier Pastor (EE/EEM) wrote:
> (Fri, 11 Jul 2003 09:56:59)
> > Hi Brian,
> >
> > Pls, see below...
> >
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: jueves, 10 de julio de 2003 15:34
> > > To: J.Javier Pastor (EE/EEM)
> > > Cc: 'sigtran@ietf.org'
> > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > >
> > >
> > > Javier,
> > >
> > > 3.18.3 Solution Description
> > >
> > >    Clarify that the ASP needs to indicate when the congestion
> > > level has
> > >    changed (including abatement).  Further, the ASP peer can
> maintain
> > >    a timer, if desired, in case the ASP fails to indicate congestion
> > >    abatement.
> > >
> > > Where are the procedures for this timer described?  This is the only
> place
> > > that the timer is mentioned.  Please describe the timer or remove
> the
> > > sentence saying that it has been described...
> > >
> >
> > Do you think is worth include the timer explanation into the new text?
> > Otherwise, we may remove that sentence.
> 
> Please remove mention of the timer.
> 
> --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 exim@www1.ietf.org  Tue Jul 15 01:14:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA04452
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 01:14:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cI8Y-0000ak-G5
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 01:14:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6F5E2J0002268
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 01:14:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cI8X-0000aI-Qi; Tue, 15 Jul 2003 01:14:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cI7j-0000Zu-Jr
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 01:13:11 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA04326
	for <SIGTRAN@ietf.org>; Tue, 15 Jul 2003 01:13:07 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cI7g-0001pq-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 01:13:08 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[qCkBw2TSzN/mhhPd9LqOpWnAY8BF13LZ])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cI7f-0001pm-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 01:13:07 -0400
Received: from ns.pigworks.openss7.net (IDENT:iLn/rgaCiwn5VSMataTHFJWUMHZKuLE9@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6F5D7P29477;
	Mon, 14 Jul 2003 23:13:07 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6F5D7U32296;
	Mon, 14 Jul 2003 23:13:07 -0600
Date: Mon, 14 Jul 2003 23:13:06 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Umesh Vats <umesh@ccpu.com>
Cc: SIGTRAN e-mail list <SIGTRAN@ietf.org>, Vivek Toky <vivektoky@ccpu.com>
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Message-ID: <20030714231306.C31916@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Umesh Vats <umesh@ccpu.com>,
	SIGTRAN e-mail list <SIGTRAN@ietf.org>,
	Vivek Toky <vivektoky@ccpu.com>
References: <20030714170027.A28531@openss7.org> <ACEMIJONPBBPDLOKCABHIEDKCBAA.umesh@ccpu.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <ACEMIJONPBBPDLOKCABHIEDKCBAA.umesh@ccpu.com>; from umesh@ccpu.com on Mon, Jul 14, 2003 at 04:19:58PM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6F5D7P29477
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Umesh,

On Mon, 14 Jul 2003, Umesh Vats wrote:

> Brian,
>=20
> I am sure you have sufficient background of highly-available/distribute=
d
> systems as your implementation is 100% fault tolerant with
> nil (0%) message loss.
>=20
> In the given scenario, what if host1 receives REG-RSP from host3 but be=
fore
> updating
> the RC received in message to common data, host1 goes down.

An host1 does not acknowledge the message to host3, so when host2 comes
up, host3 sends the REG RSP again to host2.

>=20
> In this case, host2 doesn't know the RC and RK relation.

The RC is in the REG RSP that host3 resends to host2.  The RK is in commo=
n
data along with the Local-Routing-Key-Identifier that is also contained i=
n
the REG RSP.  host2 will have no problem correlating the REG RSP with the
REG REQ, as would host1 (on the LRK-Id).

> Performing ASP-DOWN procedure will make ASP DOWN for all the Applicatio=
n
> Servers for which traffic exchange might be happening.

So, don't.  Have host3 send the REG RSP message not acknowledged by host1
to host2 when host2 takes over.

>=20
> Infact, your implementation can use this idea to further decrease messa=
ge
> loss from 0% to -1% percentage or less :)

Your implementation does not need to lose REG RSP unless it has been
designed to do so.  Don't do that.  Use techniques such as above so that
you do not lose messages.  Do not expect M3UA to repeat messages to handl=
e
the idiosynchrasies of your own internal architecture.

The reason for sending "Cannot support unique routing" was because we spe=
nt
some great pains removing RK depedency from the ASP.  The ASP can be desi=
gned
completely RK neutral and treat the RK as opaque.  Only the AS need know =
the
internals of the RK.  One side effect of this is that mutliple AS at the =
ASP
may attempt to activate for the same RK.  The ASP cannot detect this an r=
emain
RK neutral.  So, when a second AS attemtps to register the same RK (and o=
btain
its own unique RC), it is the SGP that refuses the registration with "Can=
not
support unique routing", instead of requiring the ASP to maintain RKs and
police AS/RC/RK uniqueness.  The RFC places the obligation at the SGP (M3=
UA,
NIF or MTP) to enforce the 1:1:1 relationship of AS:RC:RK, because it is =
the
one that requires this relationship so that there is "unique routing" in =
the
SGP routing table.

This is rather important to interoperability, because there is no mechani=
sm
in the M3UA ASP specification to enforce this relationship, only the mech=
anism
at the SGP.  Altering the behviour of the SGP would break this
interoperability mechanism.

There are probably three or four other mechanism within your architecture
that will solve the problem for you without changing M3UA and breaking th=
is
more important interoperability behavior.  So I cannot agree with the cha=
nge.

--brian

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Tue Jul 15 01:17:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA04873
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 01:17:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cIBR-0000td-Bw
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 01:17:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6F5H1uo003419
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 01:17:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cIBR-0000t3-8H; Tue, 15 Jul 2003 01:17:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cIAd-0000sb-PM
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 01:16:11 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA04688
	for <sigtran@ietf.org>; Tue, 15 Jul 2003 01:16:07 -0400 (EDT)
From: ygahlaut@hss.hns.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cIAa-0001w9-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 01:16:08 -0400
Received: from hindon.hss.co.in ([202.54.26.202])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cIAY-0001uL-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 01:16:07 -0400
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h6F5Ccs03500
	for <sigtran@ietf.org>; Tue, 15 Jul 2003 10:42:38 +0530 (IST)
Received: from dlfmail1.hss.hns.com (localhost [127.0.0.1])
	by ultra.hss.co.in (8.10.0/8.10.0) with ESMTP id h6F5GNf17573
	for <sigtran@ietf.org>; Tue, 15 Jul 2003 10:46:23 +0530 (IST)
To: sigtran@ietf.org
Date: Tue, 15 Jul 2003 10:41:00 +0530
Message-ID: <OFB687C5D7.A5D7600F-ON65256D64.001A3D55@hss.hns.com>
X-MIMETrack: Serialize by Router on Dlfmail1/HSS(Release 6.0|September 26, 2002) at 07/15/2003
 10:38:35 AM
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Subject: [Sigtran] SUA v15 - Confusion about Reg Req
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>






 At page 100 of SUA v15  says
   " An ASP MAY modify an existing Routing Key by including a Routing
    Context parameter in the REG REQ.  If the SGP determines that the
    Routing Context applies to an existing Routing Key, the SG MAY adjust
    the existing Routing Key to match the new information provided in the
    Routing Key parameter.  A Registration Response "Routing Key Change
    Refused" is returned if the SGP does not accept the modification of
    the Routing Key. "

 but in the section  3.8.1 on pages 51 - 52 the REG REQ message do not have
 RC Parameter.
 And the above mentioned Error Code "Routing Key Change Refused" exits in
 the parameter Registration Status (section 3.9.22  page 65)

 Which one is right ?

 Regards
 Yogesh


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



From exim@www1.ietf.org  Tue Jul 15 01:22:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA05472
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 01:22:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cIGI-000137-2v
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 01:22:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6F5M26Y004033
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 01:22:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cIGG-00012l-Rk; Tue, 15 Jul 2003 01:22:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cIFM-00012B-MS
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 01:21:04 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA05326
	for <sigtran@ietf.org>; Tue, 15 Jul 2003 01:21:00 -0400 (EDT)
From: kakaul@hss.hns.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cIFJ-000250-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 01:21:01 -0400
Received: from hindon.hss.co.in ([202.54.26.202])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cIFI-000240-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 01:21:00 -0400
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h6F5Hhs04148;
	Tue, 15 Jul 2003 10:47:43 +0530 (IST)
Received: from dlfmail1.hss.hns.com (localhost [127.0.0.1])
	by ultra.hss.co.in (8.10.0/8.10.0) with ESMTP id h6F5LSs18209;
	Tue, 15 Jul 2003 10:51:28 +0530 (IST)
Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
To: bidulock@openss7.org
Cc: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.5  September 22, 2000
Message-ID: <OFEEBA05EC.BCEADFA4-ON65256D64.001B6F0C@hss.hns.com>
Date: Tue, 15 Jul 2003 10:46:03 +0530
X-MIMETrack: Serialize by Router on Dlfmail1/HSS(Release 6.0|September 26, 2002) at 07/15/2003
 10:43:40 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,
   There is no interoperability issue involved in it and also i dont see
any need of peer knowing the existence of this timer.It is local to ASP and
is used by ASP to supervise the congestion status of the affected point
code reachable through a given SG.ASP after receiving an SCON message with
congestion status greater than 0 can start this timer and if no SCON
message is received from peer till the expiry of congestion timer ASP can
decrement affected point code  congestion status by 1 and start the timer
provisioning of congestion again if congestion status of affected point
code has not reached 0 or no congestion.

Regards
Kamesh Kaul
www.hssworld.com
Hughes Software Systems




"Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/15/2003
10:28:40 AM

Please respond to bidulock@openss7.org

Sent by:    sigtran-admin@ietf.org


To:    Kamesh Kaul/HSS@HSS
cc:    "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
       sigtran@ietf.org

Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18


kakaul,

Then add text describing how to use the timer in an interoperable way.
Also, please describe why the peer needs to know about the existence
of this timer to remain interoperable.  Otherwise, there would not be
any need to describe it.

--brian


kakaul@hss.hns.com wrote:
(Tue, 15 Jul 2003 09:35:27)
>
>
>
>
> Hi Javier,
>                     This timer may be used by some implementations to
> supervise the congestion status of point code if SCON messages send by
SGP
> are lost in transit to ASP.It was discussed earlier and was agreed upon
by
> all to keep this timer.I disagree with removing the mention of the timer.
>
> Regards
> Kamesh Kaul
> www.hssworld.com
> Hughes Software Systems
>
>
>
>
> "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>@ietf.org on
> 07/14/2003 04:05:58 PM
>
> Sent by:    sigtran-admin@ietf.org
>
>
> To:    "'Brian F. G. Bidulock '" <bidulock@openss7.org>
> cc:    "''sigtran@ietf.org' '" <sigtran@ietf.org>
>
> Subject:    RE: [Sigtran] M3UA IG v04 Section 3.18
>
>
> Brian,
>
> We'll remove the mention of the timer, if nobody disagrees...
>
> Cheers // Javier.
>
> -----Original Message-----
> From: Brian F. G. Bidulock
> To: J.Javier Pastor (EE/EEM)
> Cc: 'sigtran@ietf.org'
> Sent: 11/07/2003 11:07
> Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
>
> J.Javier,
>
> J.Javier Pastor (EE/EEM) wrote:
> (Fri, 11 Jul 2003 09:56:59)
> > Hi Brian,
> >
> > Pls, see below...
> >
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: jueves, 10 de julio de 2003 15:34
> > > To: J.Javier Pastor (EE/EEM)
> > > Cc: 'sigtran@ietf.org'
> > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > >
> > >
> > > Javier,
> > >
> > > 3.18.3 Solution Description
> > >
> > >    Clarify that the ASP needs to indicate when the congestion
> > > level has
> > >    changed (including abatement).  Further, the ASP peer can
> maintain
> > >    a timer, if desired, in case the ASP fails to indicate congestion
> > >    abatement.
> > >
> > > Where are the procedures for this timer described?  This is the only
> place
> > > that the timer is mentioned.  Please describe the timer or remove
> the
> > > sentence saying that it has been described...
> > >
> >
> > Do you think is worth include the timer explanation into the new text?
> > Otherwise, we may remove that sentence.
>
> Please remove mention of the timer.
>
> --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



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



From exim@www1.ietf.org  Tue Jul 15 02:47:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA25910
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 02:47:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cJaZ-0008QS-49
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 02:47:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6F6l3ec032379
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 02:47:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cJaW-0008Pi-L8; Tue, 15 Jul 2003 02:47:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cJaF-0008Mp-HS
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 02:46:43 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA25783
	for <sigtran@ietf.org>; Tue, 15 Jul 2003 02:46:38 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cJaB-0004GH-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 02:46:39 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[K9fLnnt55XnTOLAJYhWkZiNfttyFYcKK])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cJaA-0004G9-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 02:46:38 -0400
Received: from ns.pigworks.openss7.net (IDENT:9at8Gysj+sVnwWa4yEiQWIW34nA4RecH@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6F6kdP30710;
	Tue, 15 Jul 2003 00:46:39 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6F6kct01166;
	Tue, 15 Jul 2003 00:46:38 -0600
Date: Tue, 15 Jul 2003 00:46:38 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: kakaul@hss.hns.com
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
Message-ID: <20030715004638.A1007@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: kakaul@hss.hns.com, sigtran@ietf.org
References: <OFEEBA05EC.BCEADFA4-ON65256D64.001B6F0C@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: <OFEEBA05EC.BCEADFA4-ON65256D64.001B6F0C@hss.hns.com>; from kakaul@hss.hns.com on Tue, Jul 15, 2003 at 10:46:03AM +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>

kakaul,

If there is no interoperablity issue then it is implementation dependent.
A description of a procedure for the timer could be added as an IMPLEMENTATION
NOTE making a suggestion to the implementor of one way to address local issues
at the ASP, but it cannot be required.

But looking at your description, this mechanism would be a duplication of the
mechanism used by SCCP, ISUP, BICC, which use a dual-timer approach to
detecting congestion abatement.  Why would it be necessary for the ASP to do
this as well?

Also, you should not that SCON received from one SG is no different than SCON
received from another.  The SCON represents a statistical sampling of
congestion in the SS7 network between the ASP and the destination, and is not
referenced to an SG.  Therefore, the ASP should never perform this procedure
on a per-SG basis.  The SCON received from either SG will generate an
MTP-STATUS primitive to the AS, and the AS will run suitable timers at the
MTP-User (where the responsibility lies to avoid sending messages of a
priority less that the congestion state).

Understand that SCON is not an M3UA congestion indication: it is an SS7
network congestion indication that is directed at the MTP-User.  The ASP need
only translate SCON messages to MTP-STATUS primitives to the AS to which they
correspond.  A AS that conforms to MTP-User standards specifications with deal
with the MTP-STATUS primitives appropriately.

It is also important to understand that the ASP should not change the routing
of messages on the basis of a received SCON message.  Where applicable and for
applicable SS7 standards specifications, the SG will send DRST under very
specific and standardized congestion situations.  The ASP is already expected
to perform some functions of MTP routing in response to DRST.

So, now I don't understand why you expect the ASP to run this timer at all.

Please explain.

--brian

kakaul@hss.hns.com wrote:                                                           (Tue, 15 Jul 2003 10:46:03)
> 
> 
> 
> 
> Brian,
>    There is no interoperability issue involved in it and also i dont see
> any need of peer knowing the existence of this timer.It is local to ASP and
> is used by ASP to supervise the congestion status of the affected point
> code reachable through a given SG.ASP after receiving an SCON message with
> congestion status greater than 0 can start this timer and if no SCON
> message is received from peer till the expiry of congestion timer ASP can
> decrement affected point code  congestion status by 1 and start the timer
> provisioning of congestion again if congestion status of affected point
> code has not reached 0 or no congestion.
> 
> Regards
> Kamesh Kaul
> www.hssworld.com
> Hughes Software Systems
> 
> 
> 
> 
> "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/15/2003
> 10:28:40 AM
> 
> Please respond to bidulock@openss7.org
> 
> Sent by:    sigtran-admin@ietf.org
> 
> 
> To:    Kamesh Kaul/HSS@HSS
> cc:    "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
>        sigtran@ietf.org
> 
> Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> 
> 
> kakaul,
> 
> Then add text describing how to use the timer in an interoperable way.
> Also, please describe why the peer needs to know about the existence
> of this timer to remain interoperable.  Otherwise, there would not be
> any need to describe it.
> 
> --brian
> 
> 
> kakaul@hss.hns.com wrote:
> (Tue, 15 Jul 2003 09:35:27)
> >
> >
> >
> >
> > Hi Javier,
> >                     This timer may be used by some implementations to
> > supervise the congestion status of point code if SCON messages send by
> SGP
> > are lost in transit to ASP.It was discussed earlier and was agreed upon
> by
> > all to keep this timer.I disagree with removing the mention of the timer.
> >
> > Regards
> > Kamesh Kaul
> > www.hssworld.com
> > Hughes Software Systems
> >
> >
> >
> >
> > "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>@ietf.org on
> > 07/14/2003 04:05:58 PM
> >
> > Sent by:    sigtran-admin@ietf.org
> >
> >
> > To:    "'Brian F. G. Bidulock '" <bidulock@openss7.org>
> > cc:    "''sigtran@ietf.org' '" <sigtran@ietf.org>
> >
> > Subject:    RE: [Sigtran] M3UA IG v04 Section 3.18
> >
> >
> > Brian,
> >
> > We'll remove the mention of the timer, if nobody disagrees...
> >
> > Cheers // Javier.
> >
> > -----Original Message-----
> > From: Brian F. G. Bidulock
> > To: J.Javier Pastor (EE/EEM)
> > Cc: 'sigtran@ietf.org'
> > Sent: 11/07/2003 11:07
> > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> >
> > J.Javier,
> >
> > J.Javier Pastor (EE/EEM) wrote:
> > (Fri, 11 Jul 2003 09:56:59)
> > > Hi Brian,
> > >
> > > Pls, see below...
> > >
> > > > -----Original Message-----
> > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > Sent: jueves, 10 de julio de 2003 15:34
> > > > To: J.Javier Pastor (EE/EEM)
> > > > Cc: 'sigtran@ietf.org'
> > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > > >
> > > >
> > > > Javier,
> > > >
> > > > 3.18.3 Solution Description
> > > >
> > > >    Clarify that the ASP needs to indicate when the congestion
> > > > level has
> > > >    changed (including abatement).  Further, the ASP peer can
> > maintain
> > > >    a timer, if desired, in case the ASP fails to indicate congestion
> > > >    abatement.
> > > >
> > > > Where are the procedures for this timer described?  This is the only
> > place
> > > > that the timer is mentioned.  Please describe the timer or remove
> > the
> > > > sentence saying that it has been described...
> > > >
> > >
> > > Do you think is worth include the timer explanation into the new text?
> > > Otherwise, we may remove that sentence.
> >
> > Please remove mention of the timer.
> >
> > --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
> 

-- 
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 exim@www1.ietf.org  Tue Jul 15 03:00:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA26857
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 03:00:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cJnA-00014P-Pi
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 03:00:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6F704xd004104
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 03:00:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cJn7-00013v-2h; Tue, 15 Jul 2003 03:00:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cJn2-00013Q-15
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 02:59:56 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA26771
	for <sigtran@ietf.org>; Tue, 15 Jul 2003 02:59:50 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cJmx-0004Wr-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 02:59:51 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[6p5o3q+pmQC4a5FFZk3V0HztxrrKroiR])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cJmw-0004Wo-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 02:59:50 -0400
Received: from ns.pigworks.openss7.net (IDENT:3NV0HvRCjDzlg4hkwwrd2jEKyU5sz5y7@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6F6xpP30850;
	Tue, 15 Jul 2003 00:59:51 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6F6xod01354;
	Tue, 15 Jul 2003 00:59:50 -0600
Date: Tue, 15 Jul 2003 00:59:50 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: ygahlaut@hss.hns.com
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] SUA v15 - Confusion about Reg Req
Message-ID: <20030715005950.B1007@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: ygahlaut@hss.hns.com, sigtran@ietf.org
References: <OFB687C5D7.A5D7600F-ON65256D64.001A3D55@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: <OFB687C5D7.A5D7600F-ON65256D64.001A3D55@hss.hns.com>; from ygahlaut@hss.hns.com on Tue, Jul 15, 2003 at 10:41:00AM +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>

ygahlaut,

The RC parameter was missed from the REG REQ.  We will include it in the final.
The RC parameter will be included in the Routing Key (because an RC parameter
is specific to each RK in the REG REQ) as an optional parameter.  In addition
Network Appearance will not be required when RC is present in the Routing Key
(see Section 3.10.14.)

Changes as follows:

-----------
Old text
-----------

3.10.14 Routing Key 
 
      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 
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 
     |         Tag = 0x010E          |             Length            | 
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 
     |         Tag = 0x0018          |        Length = 8             |  
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+  
     |                  Local Routing Key Identifier                 |  
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+    
     \                         Key parameter(s)                      \ 
     /                                                               / 
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 
    
   Local Routing Key Identifier field: 32-bits (unsigned integer) 
      
        Key field: variable 
    
   The Key field contains the following parameters: 
    
        Parameter 
           Traffic Mode Type          Optional 
           Network Appearance         Optional *1 
           Source Address             Optional 
           Destination Address        Optional 
           Address Range              Optional 
    
   Note 1: The Network Appearance parameter must be included in the 
           Routing Key when the ASP is able to register in                                                                     
           multiple SS7 Network contexts. 
    
-----------
New text
-----------

3.10.14 Routing Key 
 
      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 
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 
     |         Tag = 0x010E          |             Length            | 
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 
     |         Tag = 0x0018          |        Length = 8             |  
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+  
     |                  Local Routing Key Identifier                 |  
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+    
     \                         Key parameter(s)                      \ 
     /                                                               / 
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 
    
   Local Routing Key Identifier field: 32-bits (unsigned integer) 
      
        Key field: variable 
    
   The Key field contains the following parameters: 
    
        Parameter 
>          Routing Context            Optional *2
           Traffic Mode Type          Optional 
           Network Appearance         Optional *1 
           Source Address             Optional 
           Destination Address        Optional 
           Address Range              Optional 
    
   Note 1: The Network Appearance parameter must be included in the 
           Routing Key when the ASP is able to register in                                                                     
>          multiple SS7 Network contexts and the Routing Context
>          parameter is not included.
>
>  Note 2: The Routing Context parameter is included in the Routing
>          Key when the ASP wishes to modify an existing Routing Key
>          that has already been registered with the given Routing
>          Context.  When the Routing Context parameter is included
>          in the Routing Key, it SHOULD be the first parameter in
>          the Routing Key.

How's that?

--brian

ygahlaut@hss.hns.com wrote:                                                       (Tue, 15 Jul 2003 10:41:00)
> 
> 
> 
> 
> 
>  At page 100 of SUA v15  says
>    " An ASP MAY modify an existing Routing Key by including a Routing
>     Context parameter in the REG REQ.  If the SGP determines that the
>     Routing Context applies to an existing Routing Key, the SG MAY adjust
>     the existing Routing Key to match the new information provided in the
>     Routing Key parameter.  A Registration Response "Routing Key Change
>     Refused" is returned if the SGP does not accept the modification of
>     the Routing Key. "
> 
>  but in the section  3.8.1 on pages 51 - 52 the REG REQ message do not have
>  RC Parameter.
>  And the above mentioned Error Code "Routing Key Change Refused" exits in
>  the parameter Registration Status (section 3.9.22  page 65)
> 
>  Which one is right ?
> 
>  Regards
>  Yogesh
> 
> 
> _______________________________________________
> 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 exim@www1.ietf.org  Tue Jul 15 04:26:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA03880
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 04:26:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cL8O-0006Ja-Nh
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 04:26:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6F8Q4Sw024271
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 04:26:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cL8N-0006Ip-O0; Tue, 15 Jul 2003 04:26:03 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cL7Y-0006Hb-0q
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 04:25:14 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA03776
	for <SIGTRAN@ietf.org>; Tue, 15 Jul 2003 04:25:07 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cL7V-0006G3-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 04:25:09 -0400
Received: from zdemail03.zdem.compaq.com ([161.114.112.27])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cL7T-0006Ep-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 04:25:08 -0400
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail03.zdem.compaq.com (Postfix) with ESMTP
	id ABB4010C4; Tue, 15 Jul 2003 10:24:32 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Tue, 15 Jul 2003 10:24:32 +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="us-ascii"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] M3UA IG V04: Receiving REG for a RK already registered
Date: Tue, 15 Jul 2003 10:24:31 +0200
Message-ID: <06CAA4D3753C53408D2CE6340B68444214C46D@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA IG V04: Receiving REG for a RK already registered
Thread-Index: AcNIqHZ6S45uJXEHRheiLAxJoNVmQACAHhLg
From: "Zannin, Francois" <francois.zannin@hp.com>
To: <bidulock@openss7.org>, "Umesh Vats" <umesh@ccpu.com>
Cc: "VIvek Toky" <vivektoky@ccpu.com>,
        "SIGTRAN e-mail list" <SIGTRAN@ietf.org>
X-OriginalArrivalTime: 15 Jul 2003 08:24:32.0485 (UTC) FILETIME=[84872150:01C34AAA]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Brian,

This is true that SCTP is a very reliable transport layer but it may
happen that a M3UA on top of SCTP not "bug free" can lost a received
message (maybe due to local congestion?). In that case, if the ASP has
lost the reg_resp, than it cannot deregister the lost RC (because it is
not known), the guilty ASP is then unable to recover the missing RC
unless maybe to send an ASP-DOWN/ASP-UP in order to allows again the AS
registration. Then, all already registered AS on the same associations
are lost!

If the Routing KEY is the same as the one already registered, isn't it
possible for the SG to repeat the same RC?

If the AS (RC) was already ACTIVE on this association, yes it is an
error, then the AS mUST return to the INACTIVE state.

Francois.

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]=20
> Sent: Saturday, July 12, 2003 9:03 PM
> To: Umesh Vats
> Cc: VIvek Toky; SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA IG V04: Receiving REG for a RK=20
> already registered
>=20
>=20
> Umesh,
>=20
> How can the REG-RSP get lost?  SCTP is reliable.
>=20
> --brian
>=20
> On Sat, 12 Jul 2003, Umesh Vats wrote:
>=20
> > Hi  Brian,
> >=20
> > This will be useful in the cases where,  REG-RSP from SGP
> > to ASP is lost. In this case, ASP will retry the reuqest=20
> and SGP can=20
> > send the same RC back.
> >=20
> > Umesh
> >=20
> > -----Original Message-----
> > From: sigtran-admin@ietf.org=20
> [mailto:sigtran-admin@ietf.org]On Behalf=20
> > Of Brian F. G. Bidulock
> > Sent: Friday, July 11, 2003 5:54 PM
> > To: VIvek Toky
> > Cc: SIGTRAN e-mail list
> > Subject: Re: [Sigtran] M3UA IG V04: Receiving REG for a RK already=20
> > registered
> >=20
> >=20
> > VIvek,
> >=20
> > Because it is an error.  What would be acheived by sending back the=20
> > already registered RC instead?
> >=20
> > --brian
> >=20
> > On Fri, 11 Jul 2003, VIvek Toky wrote:
> >=20
> > >   I have one question related to Registration Request handling in=20
> > > M3UA IG V4. I am pasting few lines from Section 3.6
> > >
> > > "If the SGP determines that the received RK was already=20
> registered,=20
> > > the
> > SGP
> > > returns a registration Response message to the ASP,containing a
> > Registration
> > > Result "Error - Cannot Support Unique Routing". This=20
> error applies=20
> > > whether the Routing Key is Active or Inactive."
> > >
> > > Is there any reason why SGP should send this error back.=20
> Why should=20
> > > it not send the same RC back again ?
> > >
> > > thanks,
> > > Vivek.
> > >
> > >
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> >=20
> > --
> > 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 |
> >=20
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
>=20
> --=20
> 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 |
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>=20

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



From exim@www1.ietf.org  Tue Jul 15 04:30:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA04257
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 04:30:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cLCG-00070l-I2
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 04:30:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6F8U4CJ026945
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 04:30:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cLCF-00070M-Gn; Tue, 15 Jul 2003 04:30:03 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cLC3-0006wU-3y
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 04:29:51 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA04146
	for <sigtran@ietf.org>; Tue, 15 Jul 2003 04:29:46 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cLC0-0006MD-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 04:29:48 -0400
Received: from email10.etsi.org ([212.234.161.112] helo=email10.etsihq.org)
	by ietf-mx with esmtp (Exim 4.12)
	id 19cLBz-0006Ku-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 04:29:47 -0400
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"
Content-Transfer-Encoding: quoted-printable
Date: Tue, 15 Jul 2003 10:29:17 +0200
Message-ID: <4091553999CBE4409CC2B562152B5A33019C866B@email10.etsihq.org>
Thread-Topic: [Sigtran] SUA v15 - Confusion about Reg Req
Thread-Index: AcNKnr06ssZqe3GMSkeMOLRCfEgLVgAB3Wug
From: =?us-ascii?Q?Sebastien_Flaux?= <Sebastien.Flaux@etsi.org>
To: <sigtran@ietf.org>
Content-Transfer-Encoding: quoted-printable
Subject: [Sigtran] Sigtran Plugtest at ETSI 1-5, Sep 2003
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Dear all,

It is a reminder concerning the Sigtran Plugtests back off organised at
ETSI.

We noticed today that only two companies registered for the event.

Please don't forget to register !

Best Regards,

Sebastien Flaux.

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



From exim@www1.ietf.org  Tue Jul 15 06:40:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA13344
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 06:40:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cNE3-0006Yl-OI
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 06:40:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6FAe3rg025179
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 06:40:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cNE2-0006XZ-1z; Tue, 15 Jul 2003 06:40:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cNDz-0006X0-Nr
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 06:39:59 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA13329
	for <sigtran@ietf.org>; Tue, 15 Jul 2003 06:39:53 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cNDq-0000VK-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 06:39:50 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[HkNH0DLLvIyuS/Ha5CgjlSWFPXHhu4FQ])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cNDL-0000V9-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 06:39:19 -0400
Received: from ns.pigworks.openss7.net (IDENT:yyjbcIOdBJ8yh6IC4SnUFCibh039Ahh0@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6FAbwP01579;
	Tue, 15 Jul 2003 04:37:58 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6FAbvj04562;
	Tue, 15 Jul 2003 04:37:57 -0600
Date: Tue, 15 Jul 2003 04:37:57 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Sebastien Flaux <Sebastien.Flaux@etsi.org>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Sigtran Plugtest at ETSI 1-5, Sep 2003
Message-ID: <20030715043757.D1007@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Sebastien Flaux <Sebastien.Flaux@etsi.org>,
	sigtran@ietf.org
References: <4091553999CBE4409CC2B562152B5A33019C866B@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: <4091553999CBE4409CC2B562152B5A33019C866B@email10.etsihq.org>; from Sebastien.Flaux@etsi.org on Tue, Jul 15, 2003 at 10:29:17AM +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>

Sebastien,

What's a "back off"?  ;)

--brian

Sebastien Flaux wrote:                                                           (Tue, 15 Jul 2003 10:29:17)
> Dear all,
> 
> It is a reminder concerning the Sigtran Plugtests back off organised at
> ETSI.
> 
> We noticed today that only two companies registered for the event.
> 
> Please don't forget to register !
> 
> Best Regards,
> 
> Sebastien Flaux.
> 
> _______________________________________________
> 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 exim@www1.ietf.org  Tue Jul 15 07:26:53 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA13346
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 06:40:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cNE3-0006Yj-ON
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 06:40:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6FAe3ZM025181
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 06:40:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cNE2-0006Xp-Mx; Tue, 15 Jul 2003 06:40:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cNDz-0006Wz-LC
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 06:39:59 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA13330
	for <SIGTRAN@ietf.org>; Tue, 15 Jul 2003 06:39:53 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cNDq-0000VL-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 06:39:50 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[q8O8nDQ4QHWER6plIbVUlm3R8KpmJvMp])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cNDL-0000Ul-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 06:39:19 -0400
Received: from ns.pigworks.openss7.net (IDENT:+JVFH+lWmK5phD6Lkf03BXWkDZeEfHl1@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6FAapP01568;
	Tue, 15 Jul 2003 04:36:51 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6FAapD04555;
	Tue, 15 Jul 2003 04:36:51 -0600
Date: Tue, 15 Jul 2003 04:36:51 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Zannin, Francois" <francois.zannin@hp.com>
Cc: Umesh Vats <umesh@ccpu.com>, VIvek Toky <vivektoky@ccpu.com>,
        SIGTRAN e-mail list <SIGTRAN@ietf.org>
Subject: Re: [Sigtran] M3UA IG V04: Receiving REG for a RK already registered
Message-ID: <20030715043651.C1007@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Zannin, Francois" <francois.zannin@hp.com>,
	Umesh Vats <umesh@ccpu.com>, VIvek Toky <vivektoky@ccpu.com>,
	SIGTRAN e-mail list <SIGTRAN@ietf.org>
References: <06CAA4D3753C53408D2CE6340B68444214C46D@vbeexc02.emea.cpqcorp.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <06CAA4D3753C53408D2CE6340B68444214C46D@vbeexc02.emea.cpqcorp.net>; from francois.zannin@hp.com on Tue, Jul 15, 2003 at 10:24:31AM +0200
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6FAapP01568
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Francois,

Well, not quite.  Even the buggy M3UA should be able to retain the
Local-Routing-Key-Identifier from the original REG REQ.
All that the ASP need do is reissue If the M3UA cannot maintain the
state of the original REG REQ (i.e, retain the
Local-Routing-Key-Identifier used) then it cannot correlate the REG RSP
anyway.  the REG RSP with the same Local-Routing-Key-Identifier.  The
SGP uses this to determine that the REG REQ is a repeat attempt an
replies with a REG RSP.

REG RSP("Cannot support unique routing") is intended as a reply to a REG
REQ from a different AS at the ASP.  This could be identified by the
presence of a different Local-Routing-Key-Identifier than the original.

I know that this not mentioned in the IG 04, but we discussed it.

Example: an ASP sends a REG REQ and waits.  It gives up and sends
another REG REQ.  It then receives two REG RSP.  We discussed that if an
ASP were to send two REG REQ for the same registration that they should
contain the same LRK-Id.

In the case where a REG REQ contains the same LRK-Id for the same RK,
the response to the second could be the same as the response to the
first. (Successful).

Now, the objectionable lines are in IG 04:

   --------
   New text: (Section 4.4.1)
   ---------

   If the SGP determines that the received RK was already registered,
   the SGP returns a Registration Response message to the ASP,
   containing a Registration Result "Error - Cannot Support Unique
   Routing". This error applies whether the Routing Key is Active or
   Inactive.

But, this (strangely) does not change the pertinent passages in the RFC
which say:

   If the SGP determines that a unique Routing Key cannot be created,
   the SGP returns a Registration Response message to the ASP, with a
   Registration Status of "Error - "Cannot Support Unique Routing"  An
   incoming signalling message received at an SGP should not match
   against more than one Routing Key.

This error was really intended to preclude "overlapping" routing keys
which are verbodden in M3UA.  The error was never intended for attempts
at re-registration by the same AS.

How does the following change sound?

   --------
   New text: (Section 4.4.1)
   ---------

   If the SGP determines that the received RK was already registered,
   and the Local-Routing-Key-Identifier in the RK is the same as the
   Local-Routing-Key-Identifier that was used to register the RK, then
   the SGP treats the REG REQ as a duplicate and returns a Registration
   Response message to the ASP, containing a Registration Result
   "Successful" and the Routing Context associated with the RK.

   If the SGP determines that the received RK was already registered,
   and the Local-Routing-Key-Identifier in the RK does not match the
   Local-Routing-Key-Identifier that was used to register the RK, the
   the SGP treats the REG REQ as an overlapping RK request and returns
   a Registration Response message to the ASP, containing a Registration
   Result "Error - Cannot Support Unique Routing". This error applies
   whether the Routing Key is Active or Inactive.

This would only push the SGP into maintain, along with the fact that the
ASP is registered against the RK, the Local-Routing-Key-Id that was used
by the ASP to register.  4 bytes.  Not too much.

Then an AS can send as many REG REQ as it likes, as long as it uses the
same LRK-Id, it will get the same response.

Sound good?

--brian

On Tue, 15 Jul 2003, Zannin, Francois wrote:

> Brian,
>=20
> This is true that SCTP is a very reliable transport layer but it may
> happen that a M3UA on top of SCTP not "bug free" can lost a received
> message (maybe due to local congestion?). In that case, if the ASP has
> lost the reg_resp, than it cannot deregister the lost RC (because it is
> not known), the guilty ASP is then unable to recover the missing RC
> unless maybe to send an ASP-DOWN/ASP-UP in order to allows again the AS
> registration. Then, all already registered AS on the same associations
> are lost!
>=20
> If the Routing KEY is the same as the one already registered, isn't it
> possible for the SG to repeat the same RC?
>=20
> If the AS (RC) was already ACTIVE on this association, yes it is an
> error, then the AS mUST return to the INACTIVE state.
>=20
> Francois.
>=20

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Tue Jul 15 07:34:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA14942
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 07:34:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cO4I-0000Iy-NN
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 07:34:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6FBY2fK001165
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 07:34:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cO4H-0000I8-KP; Tue, 15 Jul 2003 07:34:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cO3r-0000Hd-6e
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 07:33:35 -0400
Received: from zdemail04.zdem.compaq.com (zdemail04.zdem.compaq.com [161.114.112.28])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA14891
	for <SIGTRAN@ietf.org>; Tue, 15 Jul 2003 07:33:31 -0400 (EDT)
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail04.zdem.compaq.com (Postfix) with ESMTP
	id D91ABFC1; Tue, 15 Jul 2003 13:33:02 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Tue, 15 Jul 2003 13:33:02 +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="us-ascii"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] M3UA IG V04: Receiving REG for a RK already registered
Date: Tue, 15 Jul 2003 13:32:57 +0200
Message-ID: <06CAA4D3753C53408D2CE6340B68444214C46E@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA IG V04: Receiving REG for a RK already registered
Thread-Index: AcNKvXqv4ukcH12ZSVO74oeW+VRmBAABonqg
From: "Zannin, Francois" <francois.zannin@hp.com>
To: <bidulock@openss7.org>
Cc: "Umesh Vats" <umesh@ccpu.com>, "VIvek Toky" <vivektoky@ccpu.com>,
        "SIGTRAN e-mail list" <SIGTRAN@ietf.org>
X-OriginalArrivalTime: 15 Jul 2003 11:33:02.0715 (UTC) FILETIME=[D9F344B0:01C34AC4]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

It seem to resolve the problem and keep the "overlapping" control on an
ASP valid.
Do not forget to include this modification in next IG ;-)

Francois.

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]=20
> Sent: Tuesday, July 15, 2003 12:37 PM
> To: Zannin, Francois
> Cc: Umesh Vats; VIvek Toky; SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA IG V04: Receiving REG for a RK=20
> already registered
>=20
>=20
> Francois,
>=20
> Well, not quite.  Even the buggy M3UA should be able to=20
> retain the Local-Routing-Key-Identifier from the original REG=20
> REQ. All that the ASP need do is reissue If the M3UA cannot=20
> maintain the state of the original REG REQ (i.e, retain the=20
> Local-Routing-Key-Identifier used) then it cannot correlate=20
> the REG RSP anyway.  the REG RSP with the same=20
> Local-Routing-Key-Identifier.  The SGP uses this to determine=20
> that the REG REQ is a repeat attempt an replies with a REG RSP.
>=20
> REG RSP("Cannot support unique routing") is intended as a=20
> reply to a REG REQ from a different AS at the ASP.  This=20
> could be identified by the presence of a different=20
> Local-Routing-Key-Identifier than the original.
>=20
> I know that this not mentioned in the IG 04, but we discussed it.
>=20
> Example: an ASP sends a REG REQ and waits.  It gives up and=20
> sends another REG REQ.  It then receives two REG RSP.  We=20
> discussed that if an ASP were to send two REG REQ for the=20
> same registration that they should contain the same LRK-Id.
>=20
> In the case where a REG REQ contains the same LRK-Id for the=20
> same RK, the response to the second could be the same as the=20
> response to the first. (Successful).
>=20
> Now, the objectionable lines are in IG 04:
>=20
>    --------
>    New text: (Section 4.4.1)
>    ---------
>=20
>    If the SGP determines that the received RK was already registered,
>    the SGP returns a Registration Response message to the ASP,
>    containing a Registration Result "Error - Cannot Support Unique
>    Routing". This error applies whether the Routing Key is Active or
>    Inactive.
>=20
> But, this (strangely) does not change the pertinent passages=20
> in the RFC which say:
>=20
>    If the SGP determines that a unique Routing Key cannot be created,
>    the SGP returns a Registration Response message to the ASP, with a
>    Registration Status of "Error - "Cannot Support Unique Routing"  An
>    incoming signalling message received at an SGP should not match
>    against more than one Routing Key.
>=20
> This error was really intended to preclude "overlapping"=20
> routing keys which are verbodden in M3UA.  The error was=20
> never intended for attempts at re-registration by the same AS.
>=20
> How does the following change sound?
>=20
>    --------
>    New text: (Section 4.4.1)
>    ---------
>=20
>    If the SGP determines that the received RK was already registered,
>    and the Local-Routing-Key-Identifier in the RK is the same as the
>    Local-Routing-Key-Identifier that was used to register the RK, then
>    the SGP treats the REG REQ as a duplicate and returns a=20
> Registration
>    Response message to the ASP, containing a Registration Result
>    "Successful" and the Routing Context associated with the RK.
>=20
>    If the SGP determines that the received RK was already registered,
>    and the Local-Routing-Key-Identifier in the RK does not match the
>    Local-Routing-Key-Identifier that was used to register the RK, the
>    the SGP treats the REG REQ as an overlapping RK request and returns
>    a Registration Response message to the ASP, containing a=20
> Registration
>    Result "Error - Cannot Support Unique Routing". This error applies
>    whether the Routing Key is Active or Inactive.
>=20
> This would only push the SGP into maintain, along with the=20
> fact that the ASP is registered against the RK, the=20
> Local-Routing-Key-Id that was used by the ASP to register.  4=20
> bytes.  Not too much.
>=20
> Then an AS can send as many REG REQ as it likes, as long as=20
> it uses the same LRK-Id, it will get the same response.
>=20
> Sound good?
>=20
> --brian
>=20
> On Tue, 15 Jul 2003, Zannin, Francois wrote:
>=20
> > Brian,
> >=20
> > This is true that SCTP is a very reliable transport layer but it may

> > happen that a M3UA on top of SCTP not "bug free" can lost a received

> > message (maybe due to local congestion?). In that case, if the ASP
has=20
> > lost the reg_resp, than it cannot deregister the lost RC (because it

> > is not known), the guilty ASP is then unable to recover the missing
RC=20
> > unless maybe to send an ASP-DOWN/ASP-UP in order to allows again the

> > AS registration. Then, all already registered AS on the same=20
> > associations are lost!
> >=20
> > If the Routing KEY is the same as the one already registered, isn't
it=20
> > possible for the SG to repeat the same RC?
> >=20
> > If the AS (RC) was already ACTIVE on this association, yes it is an=20
> > error, then the AS mUST return to the INACTIVE state.
> >=20
> > Francois.
> >=20
>=20
> --=20
> 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 |
>=20

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



From exim@www1.ietf.org  Tue Jul 15 08:26:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA16200
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 08:26:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cOse-0002eY-8A
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 08:26:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6FCQ4Zw010187
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 08:26:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cOsb-0002eC-M0; Tue, 15 Jul 2003 08:26:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cOsK-0002dv-8c
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 08:25:44 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA16193
	for <sigtran@ietf.org>; Tue, 15 Jul 2003 08:25:40 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cOsE-0001PX-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 08:25:38 -0400
Received: from email10.etsi.org ([212.234.161.112] helo=email10.etsihq.org)
	by ietf-mx with esmtp (Exim 4.12)
	id 19cOrn-0001PA-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 08:25:11 -0400
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"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] Sigtran Plugtest at ETSI 1-5, Sep 2003
Date: Tue, 15 Jul 2003 14:23:56 +0200
Message-ID: <4091553999CBE4409CC2B562152B5A3301EBCA44@email10.etsihq.org>
Thread-Topic: [Sigtran] Sigtran Plugtest at ETSI 1-5, Sep 2003
Thread-Index: AcNKvVtk48fSxHUHS+ilTlEvXFnkOQADcQkg
From: =?us-ascii?Q?Sebastien_Flaux?= <Sebastien.Flaux@etsi.org>
To: <bidulock@openss7.org>
Cc: <sigtran@ietf.org>
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Sorry for the mistake:
The SIGTRAN bake off.

Best Regards,

Sebastien Flaux

-----Original Message-----
From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
Sent: 15 July 2003 12:38
To: Sebastien Flaux
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] Sigtran Plugtest at ETSI 1-5, Sep 2003


Sebastien,

What's a "back off"?  ;)

--brian

Sebastien Flaux wrote:
(Tue, 15 Jul 2003 10:29:17)
> Dear all,
>=20
> It is a reminder concerning the Sigtran Plugtests back off organised
at
> ETSI.
>=20
> We noticed today that only two companies registered for the event.
>=20
> Please don't forget to register !
>=20
> Best Regards,
>=20
> Sebastien Flaux.
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--=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



From exim@www1.ietf.org  Tue Jul 15 09:35:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA18071
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 09:35:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cPxP-0005tg-Kv
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 09:35:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6FDZ3LP022667
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 09:35:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cPxM-0005tM-MF; Tue, 15 Jul 2003 09:35:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cPxG-0005t4-4d
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 09:34:54 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA18053
	for <sigtran@ietf.org>; Tue, 15 Jul 2003 09:34:49 -0400 (EDT)
From: kakaul@hss.hns.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cPxE-0001xW-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 09:34:52 -0400
Received: from hindon.hss.co.in ([202.54.26.202])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cPx2-0001xK-00
	for sigtran@ietf.org; Tue, 15 Jul 2003 09:34:41 -0400
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h6FDVHs00824;
	Tue, 15 Jul 2003 19:01:17 +0530 (IST)
Received: from dlfmail1.hss.hns.com (localhost [127.0.0.1])
	by ultra.hss.co.in (8.10.0/8.10.0) with ESMTP id h6FDZ1F12867;
	Tue, 15 Jul 2003 19:05:02 +0530 (IST)
Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
To: bidulock@openss7.org
Cc: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.5  September 22, 2000
Message-ID: <OF62FF9733.E9CC03AC-ON65256D64.004838B3@hss.hns.com>
Date: Tue, 15 Jul 2003 18:59:37 +0530
X-MIMETrack: Serialize by Router on Dlfmail1/HSS(Release 6.0|September 26, 2002) at 07/15/2003
 06:57:12 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
   some comments inline




"Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/15/2003
12:16:38 PM

Please respond to bidulock@openss7.org

Sent by:    sigtran-admin@ietf.org


To:    Kamesh Kaul/HSS@HSS
cc:    sigtran@ietf.org

Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18


kakaul,

If there is no interoperablity issue then it is implementation dependent.
A description of a procedure for the timer could be added as an
IMPLEMENTATION
NOTE making a suggestion to the implementor of one way to address local
issues
at the ASP, but it cannot be required.

But looking at your description, this mechanism would be a duplication of
the
mechanism used by SCCP, ISUP, BICC, which use a dual-timer approach to
detecting congestion abatement.  Why would it be necessary for the ASP to
do
this as well?
>> It is also done in MTP3 for nation options with priority do you mean to
say that RCT mechanism is not required in MTP3.

Also, you should not that SCON received from one SG is no different than
SCON
received from another.  The SCON represents a statistical sampling of
congestion in the SS7 network between the ASP and the destination, and is
not
referenced to an SG.  Therefore, the ASP should never perform this
procedure
on a per-SG basis.  The SCON received from either SG will generate an
MTP-STATUS primitive to the AS, and the AS will run suitable timers at the
MTP-User (where the responsibility lies to avoid sending messages of a
priority less that the congestion state).
>> In national option with priority MTP3 user is given MTP-STATUS
indication only if point code (routeset )congestion status
changes.Similarly at ASP MTP-STATUS indication shall be given to the user
part only if derived status of affected point code changes as shown in
figure below :


                              SCON (1)+--+----++
                             <------- |        |
                          +-----------+  SG1   +-----+
                          |           |        |     |
                          |           +--------+     |      SS7 NODE
           +--+----++     |                          |    +----------+
           |        |     |                          |    |   PC     |
           |  ASP   +-----+                          +----+  X-Y-Z   |
           |        |     |                          |    |          |
           +--------+     |                          |    +----------+
                          |          +--+----++      |
             AS:A-B-C     |          |        |      |
                          +----------+  SG2   +------+
                            <------- |        |
                             SCON(3) +--------+

MTP-User will be given only MTP-STATUS indication with congestion status 3
if SCON(3) reaches before SCON(1).So for this you have to maintain
congestion status of affected point code through each SG and ultimately
given the Derived congestion status to MTP-User.

Understand that SCON is not an M3UA congestion indication: it is an SS7
network congestion indication that is directed at the MTP-User.  The ASP
need
only translate SCON messages to MTP-STATUS primitives to the AS to which
they
correspond.  A AS that conforms to MTP-User standards specifications with
deal
with the MTP-STATUS primitives appropriately.
>>It is only true in international and national networks without priority
option.

It is also important to understand that the ASP should not change the
routing
of messages on the basis of a received SCON message.  Where applicable and
for
applicable SS7 standards specifications, the SG will send DRST under very
specific and standardized congestion situations.  The ASP is already
expected
to perform some functions of MTP routing in response to DRST.
>> Here i definitely agree with you that rerouting of messages should not
be done at ASP to avoid missequencing.

So, now I don't understand why you expect the ASP to run this timer at all.
>> Some implementations may want to provision congestion of affected point
code through supervisory congestion timer, they should be allowed to do so.

Please explain.

--brian

kakaul@hss.hns.com wrote:
(Tue, 15 Jul 2003 10:46:03)
>
>
>
>
> Brian,
>    There is no interoperability issue involved in it and also i dont see
> any need of peer knowing the existence of this timer.It is local to ASP
and
> is used by ASP to supervise the congestion status of the affected point
> code reachable through a given SG.ASP after receiving an SCON message
with
> congestion status greater than 0 can start this timer and if no SCON
> message is received from peer till the expiry of congestion timer ASP can
> decrement affected point code  congestion status by 1 and start the timer
> provisioning of congestion again if congestion status of affected point
> code has not reached 0 or no congestion.
>
> Regards
> Kamesh Kaul
> www.hssworld.com
> Hughes Software Systems
>
>
>
>
> "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/15/2003
> 10:28:40 AM
>
> Please respond to bidulock@openss7.org
>
> Sent by:    sigtran-admin@ietf.org
>
>
> To:    Kamesh Kaul/HSS@HSS
> cc:    "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
>        sigtran@ietf.org
>
> Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
>
>
> kakaul,
>
> Then add text describing how to use the timer in an interoperable way.
> Also, please describe why the peer needs to know about the existence
> of this timer to remain interoperable.  Otherwise, there would not be
> any need to describe it.
>
> --brian
>
>
> kakaul@hss.hns.com wrote:
> (Tue, 15 Jul 2003 09:35:27)
> >
> >
> >
> >
> > Hi Javier,
> >                     This timer may be used by some implementations to
> > supervise the congestion status of point code if SCON messages send by
> SGP
> > are lost in transit to ASP.It was discussed earlier and was agreed upon
> by
> > all to keep this timer.I disagree with removing the mention of the
timer.
> >
> > Regards
> > Kamesh Kaul
> > www.hssworld.com
> > Hughes Software Systems
> >
> >
> >
> >
> > "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>@ietf.org on
> > 07/14/2003 04:05:58 PM
> >
> > Sent by:    sigtran-admin@ietf.org
> >
> >
> > To:    "'Brian F. G. Bidulock '" <bidulock@openss7.org>
> > cc:    "''sigtran@ietf.org' '" <sigtran@ietf.org>
> >
> > Subject:    RE: [Sigtran] M3UA IG v04 Section 3.18
> >
> >
> > Brian,
> >
> > We'll remove the mention of the timer, if nobody disagrees...
> >
> > Cheers // Javier.
> >
> > -----Original Message-----
> > From: Brian F. G. Bidulock
> > To: J.Javier Pastor (EE/EEM)
> > Cc: 'sigtran@ietf.org'
> > Sent: 11/07/2003 11:07
> > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> >
> > J.Javier,
> >
> > J.Javier Pastor (EE/EEM) wrote:
> > (Fri, 11 Jul 2003 09:56:59)
> > > Hi Brian,
> > >
> > > Pls, see below...
> > >
> > > > -----Original Message-----
> > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > Sent: jueves, 10 de julio de 2003 15:34
> > > > To: J.Javier Pastor (EE/EEM)
> > > > Cc: 'sigtran@ietf.org'
> > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > > >
> > > >
> > > > Javier,
> > > >
> > > > 3.18.3 Solution Description
> > > >
> > > >    Clarify that the ASP needs to indicate when the congestion
> > > > level has
> > > >    changed (including abatement).  Further, the ASP peer can
> > maintain
> > > >    a timer, if desired, in case the ASP fails to indicate
congestion
> > > >    abatement.
> > > >
> > > > Where are the procedures for this timer described?  This is the
only
> > place
> > > > that the timer is mentioned.  Please describe the timer or remove
> > the
> > > > sentence saying that it has been described...
> > > >
> > >
> > > Do you think is worth include the timer explanation into the new
text?
> > > Otherwise, we may remove that sentence.
> >
> > Please remove mention of the timer.
> >
> > --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
>

--
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 exim@www1.ietf.org  Tue Jul 15 15:24:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA01151
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 15:24:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cVP8-0005qk-VX
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 15:24:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6FJO2Qc022479
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 15:24:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cVP6-0005qK-LN; Tue, 15 Jul 2003 15:24:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cVOG-0005l0-VC
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 15:23:09 -0400
Received: from smtp.ccpu.com (defender.ccpu.com [216.54.134.34])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA01117
	for <SIGTRAN@ietf.org>; Tue, 15 Jul 2003 15:23:04 -0400 (EDT)
Received: from [172.16.1.40] (helo=ccpu00251)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19cVOB-0003pW-FE; Tue, 15 Jul 2003 12:23:03 -0700
Reply-To: <umesh@ccpu.com>
From: "Umesh Vats" <umesh@ccpu.com>
To: <bidulock@openss7.org>
Cc: "SIGTRAN e-mail list" <SIGTRAN@ietf.org>,
        "Vivek Toky" <vivektoky@ccpu.com>
Subject: RE: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Date: Tue, 15 Jul 2003 12:28:39 -0700
Message-ID: <ACEMIJONPBBPDLOKCABHEEEACBAA.umesh@ccpu.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0)
In-Reply-To: <20030714231306.C31916@openss7.org>
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id PAA01118
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

REG-RSP from SGP to ASP ,may be lost at SG itself before it reaches SCTP
due to local problems at SG. In this case, what should ASP do ?

I don't understand why sending of RC from SG poses any problem.
SG already knows about the RK and RC. It just has to return the RC.

Umesh


-----Original Message-----
From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf Of
Brian F. G. Bidulock
Sent: Monday, July 14, 2003 10:13 PM
To: Umesh Vats
Cc: SIGTRAN e-mail list; Vivek Toky
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP


Umesh,

On Mon, 14 Jul 2003, Umesh Vats wrote:

> Brian,
>
> I am sure you have sufficient background of highly-available/distribute=
d
> systems as your implementation is 100% fault tolerant with
> nil (0%) message loss.
>
> In the given scenario, what if host1 receives REG-RSP from host3 but
before
> updating
> the RC received in message to common data, host1 goes down.

An host1 does not acknowledge the message to host3, so when host2 comes
up, host3 sends the REG RSP again to host2.

>
> In this case, host2 doesn't know the RC and RK relation.

The RC is in the REG RSP that host3 resends to host2.  The RK is in commo=
n
data along with the Local-Routing-Key-Identifier that is also contained i=
n
the REG RSP.  host2 will have no problem correlating the REG RSP with the
REG REQ, as would host1 (on the LRK-Id).

> Performing ASP-DOWN procedure will make ASP DOWN for all the Applicatio=
n
> Servers for which traffic exchange might be happening.

So, don't.  Have host3 send the REG RSP message not acknowledged by host1
to host2 when host2 takes over.

>
> Infact, your implementation can use this idea to further decrease messa=
ge
> loss from 0% to -1% percentage or less :)

Your implementation does not need to lose REG RSP unless it has been
designed to do so.  Don't do that.  Use techniques such as above so that
you do not lose messages.  Do not expect M3UA to repeat messages to handl=
e
the idiosynchrasies of your own internal architecture.

The reason for sending "Cannot support unique routing" was because we spe=
nt
some great pains removing RK depedency from the ASP.  The ASP can be
designed
completely RK neutral and treat the RK as opaque.  Only the AS need know =
the
internals of the RK.  One side effect of this is that mutliple AS at the =
ASP
may attempt to activate for the same RK.  The ASP cannot detect this an
remain
RK neutral.  So, when a second AS attemtps to register the same RK (and
obtain
its own unique RC), it is the SGP that refuses the registration with "Can=
not
support unique routing", instead of requiring the ASP to maintain RKs and
police AS/RC/RK uniqueness.  The RFC places the obligation at the SGP (M3=
UA,
NIF or MTP) to enforce the 1:1:1 relationship of AS:RC:RK, because it is =
the
one that requires this relationship so that there is "unique routing" in =
the
SGP routing table.

This is rather important to interoperability, because there is no mechani=
sm
in the M3UA ASP specification to enforce this relationship, only the
mechanism
at the SGP.  Altering the behviour of the SGP would break this
interoperability mechanism.

There are probably three or four other mechanism within your architecture
that will solve the problem for you without changing M3UA and breaking th=
is
more important interoperability behavior.  So I cannot agree with the
change.

--brian

--
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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

---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003


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



From exim@www1.ietf.org  Tue Jul 15 15:26:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA01232
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 15:26:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cVR3-0005wH-Q1
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 15:26:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6FJQ1eX022816
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 15:26:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cVR3-0005vl-Fy; Tue, 15 Jul 2003 15:26:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cVQP-0005v1-Bo
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 15:25:21 -0400
Received: from smtp.ccpu.com (defender.ccpu.com [216.54.134.34])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA01190
	for <SIGTRAN@ietf.org>; Tue, 15 Jul 2003 15:25:15 -0400 (EDT)
Received: from [172.16.1.40] (helo=ccpu00251)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19cVQH-0003zc-0H; Tue, 15 Jul 2003 12:25:13 -0700
Reply-To: <umesh@ccpu.com>
From: "Umesh Vats" <umesh@ccpu.com>
To: <bidulock@openss7.org>, "Zannin, Francois" <francois.zannin@hp.com>
Cc: "VIvek Toky" <vivektoky@ccpu.com>,
        "SIGTRAN e-mail list" <SIGTRAN@ietf.org>
Subject: RE: [Sigtran] M3UA IG V04: Receiving REG for a RK already registered
Date: Tue, 15 Jul 2003 12:30:48 -0700
Message-ID: <ACEMIJONPBBPDLOKCABHKEEACBAA.umesh@ccpu.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0)
In-Reply-To: <20030715043651.C1007@openss7.org>
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id PAA01191
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Whay does SG need to store the Local RK-Id ? Local RK-Id
is just like transaction id for correlating REG-REQ and REG-RSP for
one exchange of routing key registration procedure.

To find out whether the RK request from an ASP is duplicate or not,
SG can just compare the RK's already registered by the corresponding ASP.

Umesh

-----Original Message-----
From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf Of
Brian F. G. Bidulock
Sent: Tuesday, July 15, 2003 3:37 AM
To: Zannin, Francois
Cc: Umesh Vats; VIvek Toky; SIGTRAN e-mail list
Subject: Re: [Sigtran] M3UA IG V04: Receiving REG for a RK already
registered


Francois,

Well, not quite.  Even the buggy M3UA should be able to retain the
Local-Routing-Key-Identifier from the original REG REQ.
All that the ASP need do is reissue If the M3UA cannot maintain the
state of the original REG REQ (i.e, retain the
Local-Routing-Key-Identifier used) then it cannot correlate the REG RSP
anyway.  the REG RSP with the same Local-Routing-Key-Identifier.  The
SGP uses this to determine that the REG REQ is a repeat attempt an
replies with a REG RSP.

REG RSP("Cannot support unique routing") is intended as a reply to a REG
REQ from a different AS at the ASP.  This could be identified by the
presence of a different Local-Routing-Key-Identifier than the original.

I know that this not mentioned in the IG 04, but we discussed it.

Example: an ASP sends a REG REQ and waits.  It gives up and sends
another REG REQ.  It then receives two REG RSP.  We discussed that if an
ASP were to send two REG REQ for the same registration that they should
contain the same LRK-Id.

In the case where a REG REQ contains the same LRK-Id for the same RK,
the response to the second could be the same as the response to the
first. (Successful).

Now, the objectionable lines are in IG 04:

   --------
   New text: (Section 4.4.1)
   ---------

   If the SGP determines that the received RK was already registered,
   the SGP returns a Registration Response message to the ASP,
   containing a Registration Result "Error - Cannot Support Unique
   Routing". This error applies whether the Routing Key is Active or
   Inactive.

But, this (strangely) does not change the pertinent passages in the RFC
which say:

   If the SGP determines that a unique Routing Key cannot be created,
   the SGP returns a Registration Response message to the ASP, with a
   Registration Status of "Error - "Cannot Support Unique Routing"  An
   incoming signalling message received at an SGP should not match
   against more than one Routing Key.

This error was really intended to preclude "overlapping" routing keys
which are verbodden in M3UA.  The error was never intended for attempts
at re-registration by the same AS.

How does the following change sound?

   --------
   New text: (Section 4.4.1)
   ---------

   If the SGP determines that the received RK was already registered,
   and the Local-Routing-Key-Identifier in the RK is the same as the
   Local-Routing-Key-Identifier that was used to register the RK, then
   the SGP treats the REG REQ as a duplicate and returns a Registration
   Response message to the ASP, containing a Registration Result
   "Successful" and the Routing Context associated with the RK.

   If the SGP determines that the received RK was already registered,
   and the Local-Routing-Key-Identifier in the RK does not match the
   Local-Routing-Key-Identifier that was used to register the RK, the
   the SGP treats the REG REQ as an overlapping RK request and returns
   a Registration Response message to the ASP, containing a Registration
   Result "Error - Cannot Support Unique Routing". This error applies
   whether the Routing Key is Active or Inactive.

This would only push the SGP into maintain, along with the fact that the
ASP is registered against the RK, the Local-Routing-Key-Id that was used
by the ASP to register.  4 bytes.  Not too much.

Then an AS can send as many REG REQ as it likes, as long as it uses the
same LRK-Id, it will get the same response.

Sound good?

--brian

On Tue, 15 Jul 2003, Zannin, Francois wrote:

> Brian,
>
> This is true that SCTP is a very reliable transport layer but it may
> happen that a M3UA on top of SCTP not "bug free" can lost a received
> message (maybe due to local congestion?). In that case, if the ASP has
> lost the reg_resp, than it cannot deregister the lost RC (because it is
> not known), the guilty ASP is then unable to recover the missing RC
> unless maybe to send an ASP-DOWN/ASP-UP in order to allows again the AS
> registration. Then, all already registered AS on the same associations
> are lost!
>
> If the Routing KEY is the same as the one already registered, isn't it
> possible for the SG to repeat the same RC?
>
> If the AS (RC) was already ACTIVE on this association, yes it is an
> error, then the AS mUST return to the INACTIVE state.
>
> Francois.
>

--
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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

---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003


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



From exim@www1.ietf.org  Tue Jul 15 15:52:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA01869
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 15:52:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cVqE-00077v-F1
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 15:52:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6FJq2eQ027384
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 15:52:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cVqC-000777-KH; Tue, 15 Jul 2003 15:52:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cVpm-00076g-Po
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 15:51:35 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA01848
	for <SIGTRAN@ietf.org>; Tue, 15 Jul 2003 15:51:30 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cVpl-0004vd-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 15:51:33 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[yN+WPrlt9E7KxU5063PPHa2EJ3KwvgBg])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cVpU-0004vV-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 15:51:16 -0400
Received: from ns.pigworks.openss7.net (IDENT:XehDZDT5MoVsTUCrGtKK3fdYsFZAu4Uo@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6FJonP09695;
	Tue, 15 Jul 2003 13:50:49 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6FJonI14336;
	Tue, 15 Jul 2003 13:50:49 -0600
Date: Tue, 15 Jul 2003 13:50:49 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Umesh Vats <umesh@ccpu.com>
Cc: SIGTRAN e-mail list <SIGTRAN@ietf.org>, Vivek Toky <vivektoky@ccpu.com>
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Message-ID: <20030715135049.B14123@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Umesh Vats <umesh@ccpu.com>,
	SIGTRAN e-mail list <SIGTRAN@ietf.org>,
	Vivek Toky <vivektoky@ccpu.com>
References: <20030714231306.C31916@openss7.org> <ACEMIJONPBBPDLOKCABHEEEACBAA.umesh@ccpu.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <ACEMIJONPBBPDLOKCABHEEEACBAA.umesh@ccpu.com>; from umesh@ccpu.com on Tue, Jul 15, 2003 at 12:28:39PM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6FJonP09695
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Umesh,

What should ASP do if it loses data.  What should ASP do if it loses
ASP xxxx Ack.  What should ASP do if it loses mandatory Notify.  What
should ASP do....

The SGP should not lose messagez, or if it does, send ASP Inactive Ack
or ASP Down Ack to reset the association.

If it loses messages without detecting them, there is nothing to do but
recreate SCTP in M3UA (a rather absurd proposition).

--brian


On Tue, 15 Jul 2003, Umesh Vats wrote:

> REG-RSP from SGP to ASP ,may be lost at SG itself before it reaches SCT=
P
> due to local problems at SG. In this case, what should ASP do ?
>=20
> I don't understand why sending of RC from SG poses any problem.
> SG already knows about the RK and RC. It just has to return the RC.
>=20
> Umesh
>=20
>=20
> -----Original Message-----
> From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf O=
f
> Brian F. G. Bidulock
> Sent: Monday, July 14, 2003 10:13 PM
> To: Umesh Vats
> Cc: SIGTRAN e-mail list; Vivek Toky
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>=20
>=20
> Umesh,
>=20
> On Mon, 14 Jul 2003, Umesh Vats wrote:
>=20
> > Brian,
> >
> > I am sure you have sufficient background of highly-available/distribu=
ted
> > systems as your implementation is 100% fault tolerant with
> > nil (0%) message loss.
> >
> > In the given scenario, what if host1 receives REG-RSP from host3 but
> before
> > updating
> > the RC received in message to common data, host1 goes down.
>=20
> An host1 does not acknowledge the message to host3, so when host2 comes
> up, host3 sends the REG RSP again to host2.
>=20
> >
> > In this case, host2 doesn't know the RC and RK relation.
>=20
> The RC is in the REG RSP that host3 resends to host2.  The RK is in com=
mon
> data along with the Local-Routing-Key-Identifier that is also contained=
 in
> the REG RSP.  host2 will have no problem correlating the REG RSP with t=
he
> REG REQ, as would host1 (on the LRK-Id).
>=20
> > Performing ASP-DOWN procedure will make ASP DOWN for all the Applicat=
ion
> > Servers for which traffic exchange might be happening.
>=20
> So, don't.  Have host3 send the REG RSP message not acknowledged by hos=
t1
> to host2 when host2 takes over.
>=20
> >
> > Infact, your implementation can use this idea to further decrease mes=
sage
> > loss from 0% to -1% percentage or less :)
>=20
> Your implementation does not need to lose REG RSP unless it has been
> designed to do so.  Don't do that.  Use techniques such as above so tha=
t
> you do not lose messages.  Do not expect M3UA to repeat messages to han=
dle
> the idiosynchrasies of your own internal architecture.
>=20
> The reason for sending "Cannot support unique routing" was because we s=
pent
> some great pains removing RK depedency from the ASP.  The ASP can be
> designed
> completely RK neutral and treat the RK as opaque.  Only the AS need kno=
w the
> internals of the RK.  One side effect of this is that mutliple AS at th=
e ASP
> may attempt to activate for the same RK.  The ASP cannot detect this an
> remain
> RK neutral.  So, when a second AS attemtps to register the same RK (and
> obtain
> its own unique RC), it is the SGP that refuses the registration with "C=
annot
> support unique routing", instead of requiring the ASP to maintain RKs a=
nd
> police AS/RC/RK uniqueness.  The RFC places the obligation at the SGP (=
M3UA,
> NIF or MTP) to enforce the 1:1:1 relationship of AS:RC:RK, because it i=
s the
> one that requires this relationship so that there is "unique routing" i=
n the
> SGP routing table.
>=20
> This is rather important to interoperability, because there is no mecha=
nism
> in the M3UA ASP specification to enforce this relationship, only the
> mechanism
> at the SGP.  Altering the behviour of the SGP would break this
> interoperability mechanism.
>=20
> There are probably three or four other mechanism within your architectu=
re
> that will solve the problem for you without changing M3UA and breaking =
this
> more important interoperability behavior.  So I cannot agree with the
> change.
>=20
> --brian
>=20
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>=20
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>=20
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Tue Jul 15 15:52:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA01871
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 15:52:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cVqE-00077w-F4
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 15:52:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6FJq2jb027383
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 15:52:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cVqD-00077F-20; Tue, 15 Jul 2003 15:52:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cVpm-00076h-S0
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 15:51:35 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA01849
	for <SIGTRAN@ietf.org>; Tue, 15 Jul 2003 15:51:30 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cVpl-0004vc-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 15:51:33 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[ICNKMgmtnTblqWy5il2BUQWwzb4PJ1Eo])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cVpU-0004v1-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 15:51:16 -0400
Received: from ns.pigworks.openss7.net (IDENT:SBpw4rubYqO9U7sZf/unLgWWG4Y82hcj@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6FJm7P09664;
	Tue, 15 Jul 2003 13:48:07 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6FJm6q14301;
	Tue, 15 Jul 2003 13:48:06 -0600
Date: Tue, 15 Jul 2003 13:48:06 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Umesh Vats <umesh@ccpu.com>
Cc: "Zannin, Francois" <francois.zannin@hp.com>,
        VIvek Toky <vivektoky@ccpu.com>,
        SIGTRAN e-mail list <SIGTRAN@ietf.org>
Subject: Re: [Sigtran] M3UA IG V04: Receiving REG for a RK already registered
Message-ID: <20030715134806.A14123@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Umesh Vats <umesh@ccpu.com>,
	"Zannin, Francois" <francois.zannin@hp.com>,
	VIvek Toky <vivektoky@ccpu.com>,
	SIGTRAN e-mail list <SIGTRAN@ietf.org>
References: <20030715043651.C1007@openss7.org> <ACEMIJONPBBPDLOKCABHKEEACBAA.umesh@ccpu.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <ACEMIJONPBBPDLOKCABHKEEACBAA.umesh@ccpu.com>; from umesh@ccpu.com on Tue, Jul 15, 2003 at 12:30:48PM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6FJm7P09664
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Umesh,

No, because by only comparing RK's the SG cannot distinguish between
a separate AS registering for the same RK from the same ASP.  I thought
I explained this several mails ago.

--brian

On Tue, 15 Jul 2003, Umesh Vats wrote:

> Whay does SG need to store the Local RK-Id ? Local RK-Id
> is just like transaction id for correlating REG-REQ and REG-RSP for
> one exchange of routing key registration procedure.
>=20
> To find out whether the RK request from an ASP is duplicate or not,
> SG can just compare the RK's already registered by the corresponding AS=
P.
>=20
> Umesh
>=20
> -----Original Message-----
> From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf O=
f
> Brian F. G. Bidulock
> Sent: Tuesday, July 15, 2003 3:37 AM
> To: Zannin, Francois
> Cc: Umesh Vats; VIvek Toky; SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA IG V04: Receiving REG for a RK already
> registered
>=20
>=20
> Francois,
>=20
> Well, not quite.  Even the buggy M3UA should be able to retain the
> Local-Routing-Key-Identifier from the original REG REQ.
> All that the ASP need do is reissue If the M3UA cannot maintain the
> state of the original REG REQ (i.e, retain the
> Local-Routing-Key-Identifier used) then it cannot correlate the REG RSP
> anyway.  the REG RSP with the same Local-Routing-Key-Identifier.  The
> SGP uses this to determine that the REG REQ is a repeat attempt an
> replies with a REG RSP.
>=20
> REG RSP("Cannot support unique routing") is intended as a reply to a RE=
G
> REQ from a different AS at the ASP.  This could be identified by the
> presence of a different Local-Routing-Key-Identifier than the original.
>=20
> I know that this not mentioned in the IG 04, but we discussed it.
>=20
> Example: an ASP sends a REG REQ and waits.  It gives up and sends
> another REG REQ.  It then receives two REG RSP.  We discussed that if a=
n
> ASP were to send two REG REQ for the same registration that they should
> contain the same LRK-Id.
>=20
> In the case where a REG REQ contains the same LRK-Id for the same RK,
> the response to the second could be the same as the response to the
> first. (Successful).
>=20
> Now, the objectionable lines are in IG 04:
>=20
>    --------
>    New text: (Section 4.4.1)
>    ---------
>=20
>    If the SGP determines that the received RK was already registered,
>    the SGP returns a Registration Response message to the ASP,
>    containing a Registration Result "Error - Cannot Support Unique
>    Routing". This error applies whether the Routing Key is Active or
>    Inactive.
>=20
> But, this (strangely) does not change the pertinent passages in the RFC
> which say:
>=20
>    If the SGP determines that a unique Routing Key cannot be created,
>    the SGP returns a Registration Response message to the ASP, with a
>    Registration Status of "Error - "Cannot Support Unique Routing"  An
>    incoming signalling message received at an SGP should not match
>    against more than one Routing Key.
>=20
> This error was really intended to preclude "overlapping" routing keys
> which are verbodden in M3UA.  The error was never intended for attempts
> at re-registration by the same AS.
>=20
> How does the following change sound?
>=20
>    --------
>    New text: (Section 4.4.1)
>    ---------
>=20
>    If the SGP determines that the received RK was already registered,
>    and the Local-Routing-Key-Identifier in the RK is the same as the
>    Local-Routing-Key-Identifier that was used to register the RK, then
>    the SGP treats the REG REQ as a duplicate and returns a Registration
>    Response message to the ASP, containing a Registration Result
>    "Successful" and the Routing Context associated with the RK.
>=20
>    If the SGP determines that the received RK was already registered,
>    and the Local-Routing-Key-Identifier in the RK does not match the
>    Local-Routing-Key-Identifier that was used to register the RK, the
>    the SGP treats the REG REQ as an overlapping RK request and returns
>    a Registration Response message to the ASP, containing a Registratio=
n
>    Result "Error - Cannot Support Unique Routing". This error applies
>    whether the Routing Key is Active or Inactive.
>=20
> This would only push the SGP into maintain, along with the fact that th=
e
> ASP is registered against the RK, the Local-Routing-Key-Id that was use=
d
> by the ASP to register.  4 bytes.  Not too much.
>=20
> Then an AS can send as many REG REQ as it likes, as long as it uses the
> same LRK-Id, it will get the same response.
>=20
> Sound good?
>=20
> --brian
>=20
> On Tue, 15 Jul 2003, Zannin, Francois wrote:
>=20
> > Brian,
> >
> > This is true that SCTP is a very reliable transport layer but it may
> > happen that a M3UA on top of SCTP not "bug free" can lost a received
> > message (maybe due to local congestion?). In that case, if the ASP ha=
s
> > lost the reg_resp, than it cannot deregister the lost RC (because it =
is
> > not known), the guilty ASP is then unable to recover the missing RC
> > unless maybe to send an ASP-DOWN/ASP-UP in order to allows again the =
AS
> > registration. Then, all already registered AS on the same association=
s
> > are lost!
> >
> > If the Routing KEY is the same as the one already registered, isn't i=
t
> > possible for the SG to repeat the same RC?
> >
> > If the AS (RC) was already ACTIVE on this association, yes it is an
> > error, then the AS mUST return to the INACTIVE state.
> >
> > Francois.
> >
>=20
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>=20
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>=20
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Tue Jul 15 16:02:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA02378
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 16:02:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cVzv-0007o6-6x
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 16:02:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6FK23kB030009
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 16:02:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cVzu-0007nb-9q; Tue, 15 Jul 2003 16:02:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cVyz-0007n8-8h
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 16:01:05 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA02324
	for <SIGTRAN@ietf.org>; Tue, 15 Jul 2003 16:01:00 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cVyx-00052J-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 16:01:03 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19cVym-0004zb-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 16:00:52 -0400
Received: from [172.16.1.40] (helo=ccpu00251)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19cVwj-0006KJ-0u; Tue, 15 Jul 2003 12:58:45 -0700
Reply-To: <umesh@ccpu.com>
From: "Umesh Vats" <umesh@ccpu.com>
To: <bidulock@openss7.org>
Cc: "SIGTRAN e-mail list" <SIGTRAN@ietf.org>
Subject: RE: [Sigtran] M3UA IG V04: Receiving REG for a RK already registered
Date: Tue, 15 Jul 2003 13:04:21 -0700
Message-ID: <ACEMIJONPBBPDLOKCABHKEEBCBAA.umesh@ccpu.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0)
In-Reply-To: <20030715134806.A14123@openss7.org>
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id QAA02327
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Why will there be two Application Servers with same RK at an
ASP ? There is 1:1 releationship between RK and AS. If an
ASP is sending same RK, this in an Error case no matter
what Local RK-Id is there in the REG-REQ.

SGP can respond with unsuccessful REG-RSP with Rk-Id and RC.

Umesh



-----Original Message-----
From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
Sent: Tuesday, July 15, 2003 12:48 PM
To: Umesh Vats
Cc: Zannin, Francois; VIvek Toky; SIGTRAN e-mail list
Subject: Re: [Sigtran] M3UA IG V04: Receiving REG for a RK already
registered


Umesh,

No, because by only comparing RK's the SG cannot distinguish between
a separate AS registering for the same RK from the same ASP.  I thought
I explained this several mails ago.

--brian

On Tue, 15 Jul 2003, Umesh Vats wrote:

> Whay does SG need to store the Local RK-Id ? Local RK-Id
> is just like transaction id for correlating REG-REQ and REG-RSP for
> one exchange of routing key registration procedure.
>
> To find out whether the RK request from an ASP is duplicate or not,
> SG can just compare the RK's already registered by the corresponding AS=
P.
>
> Umesh
>
> -----Original Message-----
> From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf O=
f
> Brian F. G. Bidulock
> Sent: Tuesday, July 15, 2003 3:37 AM
> To: Zannin, Francois
> Cc: Umesh Vats; VIvek Toky; SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA IG V04: Receiving REG for a RK already
> registered
>
>
> Francois,
>
> Well, not quite.  Even the buggy M3UA should be able to retain the
> Local-Routing-Key-Identifier from the original REG REQ.
> All that the ASP need do is reissue If the M3UA cannot maintain the
> state of the original REG REQ (i.e, retain the
> Local-Routing-Key-Identifier used) then it cannot correlate the REG RSP
> anyway.  the REG RSP with the same Local-Routing-Key-Identifier.  The
> SGP uses this to determine that the REG REQ is a repeat attempt an
> replies with a REG RSP.
>
> REG RSP("Cannot support unique routing") is intended as a reply to a RE=
G
> REQ from a different AS at the ASP.  This could be identified by the
> presence of a different Local-Routing-Key-Identifier than the original.
>
> I know that this not mentioned in the IG 04, but we discussed it.
>
> Example: an ASP sends a REG REQ and waits.  It gives up and sends
> another REG REQ.  It then receives two REG RSP.  We discussed that if a=
n
> ASP were to send two REG REQ for the same registration that they should
> contain the same LRK-Id.
>
> In the case where a REG REQ contains the same LRK-Id for the same RK,
> the response to the second could be the same as the response to the
> first. (Successful).
>
> Now, the objectionable lines are in IG 04:
>
>    --------
>    New text: (Section 4.4.1)
>    ---------
>
>    If the SGP determines that the received RK was already registered,
>    the SGP returns a Registration Response message to the ASP,
>    containing a Registration Result "Error - Cannot Support Unique
>    Routing". This error applies whether the Routing Key is Active or
>    Inactive.
>
> But, this (strangely) does not change the pertinent passages in the RFC
> which say:
>
>    If the SGP determines that a unique Routing Key cannot be created,
>    the SGP returns a Registration Response message to the ASP, with a
>    Registration Status of "Error - "Cannot Support Unique Routing"  An
>    incoming signalling message received at an SGP should not match
>    against more than one Routing Key.
>
> This error was really intended to preclude "overlapping" routing keys
> which are verbodden in M3UA.  The error was never intended for attempts
> at re-registration by the same AS.
>
> How does the following change sound?
>
>    --------
>    New text: (Section 4.4.1)
>    ---------
>
>    If the SGP determines that the received RK was already registered,
>    and the Local-Routing-Key-Identifier in the RK is the same as the
>    Local-Routing-Key-Identifier that was used to register the RK, then
>    the SGP treats the REG REQ as a duplicate and returns a Registration
>    Response message to the ASP, containing a Registration Result
>    "Successful" and the Routing Context associated with the RK.
>
>    If the SGP determines that the received RK was already registered,
>    and the Local-Routing-Key-Identifier in the RK does not match the
>    Local-Routing-Key-Identifier that was used to register the RK, the
>    the SGP treats the REG REQ as an overlapping RK request and returns
>    a Registration Response message to the ASP, containing a Registratio=
n
>    Result "Error - Cannot Support Unique Routing". This error applies
>    whether the Routing Key is Active or Inactive.
>
> This would only push the SGP into maintain, along with the fact that th=
e
> ASP is registered against the RK, the Local-Routing-Key-Id that was use=
d
> by the ASP to register.  4 bytes.  Not too much.
>
> Then an AS can send as many REG REQ as it likes, as long as it uses the
> same LRK-Id, it will get the same response.
>
> Sound good?
>
> --brian
>
> On Tue, 15 Jul 2003, Zannin, Francois wrote:
>
> > Brian,
> >
> > This is true that SCTP is a very reliable transport layer but it may
> > happen that a M3UA on top of SCTP not "bug free" can lost a received
> > message (maybe due to local congestion?). In that case, if the ASP ha=
s
> > lost the reg_resp, than it cannot deregister the lost RC (because it =
is
> > not known), the guilty ASP is then unable to recover the missing RC
> > unless maybe to send an ASP-DOWN/ASP-UP in order to allows again the =
AS
> > registration. Then, all already registered AS on the same association=
s
> > are lost!
> >
> > If the Routing KEY is the same as the one already registered, isn't i=
t
> > possible for the SG to repeat the same RC?
> >
> > If the AS (RC) was already ACTIVE on this association, yes it is an
> > error, then the AS mUST return to the INACTIVE state.
> >
> > Francois.
> >
>
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

--
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003


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



From exim@www1.ietf.org  Tue Jul 15 16:05:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA02508
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 16:05:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cW2n-00085A-NQ
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 16:05:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6FK51q7031063
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 16:05:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cW2n-00084v-K6; Tue, 15 Jul 2003 16:05:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cW2e-00083t-DO
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 16:04:52 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA02486
	for <SIGTRAN@ietf.org>; Tue, 15 Jul 2003 16:04:47 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cW2Y-000545-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 16:04:46 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19cW2N-000540-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 16:04:36 -0400
Received: from [172.16.1.40] (helo=ccpu00251)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19cW2G-0006i0-4F; Tue, 15 Jul 2003 13:04:28 -0700
Reply-To: <umesh@ccpu.com>
From: "Umesh Vats" <umesh@ccpu.com>
To: <bidulock@openss7.org>
Cc: "SIGTRAN e-mail list" <SIGTRAN@ietf.org>
Subject: RE: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Date: Tue, 15 Jul 2003 13:10:04 -0700
Message-ID: <ACEMIJONPBBPDLOKCABHAEECCBAA.umesh@ccpu.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0)
In-Reply-To: <20030715135049.B14123@openss7.org>
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id QAA02487
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

If ASP looses ASPSM/ASPTM/DRKM/SSNM messages(which are consumed by M3UA,
it should be able to recover.

For SSNM, subsequent SSNM messages or DAUD procedure take care of message
loss.
For if ASP-ACTIVE_ACK response is lost, ASP can resend ASP-ACTIVE.

But for loss of REG-RSP, the only recovery now possible is ASP-DOWN
procedure which is far worse that deregistering just one RK.

Umesh

-----Original Message-----
From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
Sent: Tuesday, July 15, 2003 12:51 PM
To: Umesh Vats
Cc: SIGTRAN e-mail list; Vivek Toky
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP


Umesh,

What should ASP do if it loses data.  What should ASP do if it loses
ASP xxxx Ack.  What should ASP do if it loses mandatory Notify.  What
should ASP do....

The SGP should not lose messagez, or if it does, send ASP Inactive Ack
or ASP Down Ack to reset the association.

If it loses messages without detecting them, there is nothing to do but
recreate SCTP in M3UA (a rather absurd proposition).

--brian


On Tue, 15 Jul 2003, Umesh Vats wrote:

> REG-RSP from SGP to ASP ,may be lost at SG itself before it reaches SCT=
P
> due to local problems at SG. In this case, what should ASP do ?
>
> I don't understand why sending of RC from SG poses any problem.
> SG already knows about the RK and RC. It just has to return the RC.
>
> Umesh
>
>
> -----Original Message-----
> From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf O=
f
> Brian F. G. Bidulock
> Sent: Monday, July 14, 2003 10:13 PM
> To: Umesh Vats
> Cc: SIGTRAN e-mail list; Vivek Toky
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>
>
> Umesh,
>
> On Mon, 14 Jul 2003, Umesh Vats wrote:
>
> > Brian,
> >
> > I am sure you have sufficient background of highly-available/distribu=
ted
> > systems as your implementation is 100% fault tolerant with
> > nil (0%) message loss.
> >
> > In the given scenario, what if host1 receives REG-RSP from host3 but
> before
> > updating
> > the RC received in message to common data, host1 goes down.
>
> An host1 does not acknowledge the message to host3, so when host2 comes
> up, host3 sends the REG RSP again to host2.
>
> >
> > In this case, host2 doesn't know the RC and RK relation.
>
> The RC is in the REG RSP that host3 resends to host2.  The RK is in com=
mon
> data along with the Local-Routing-Key-Identifier that is also contained=
 in
> the REG RSP.  host2 will have no problem correlating the REG RSP with t=
he
> REG REQ, as would host1 (on the LRK-Id).
>
> > Performing ASP-DOWN procedure will make ASP DOWN for all the Applicat=
ion
> > Servers for which traffic exchange might be happening.
>
> So, don't.  Have host3 send the REG RSP message not acknowledged by hos=
t1
> to host2 when host2 takes over.
>
> >
> > Infact, your implementation can use this idea to further decrease
message
> > loss from 0% to -1% percentage or less :)
>
> Your implementation does not need to lose REG RSP unless it has been
> designed to do so.  Don't do that.  Use techniques such as above so tha=
t
> you do not lose messages.  Do not expect M3UA to repeat messages to han=
dle
> the idiosynchrasies of your own internal architecture.
>
> The reason for sending "Cannot support unique routing" was because we
spent
> some great pains removing RK depedency from the ASP.  The ASP can be
> designed
> completely RK neutral and treat the RK as opaque.  Only the AS need kno=
w
the
> internals of the RK.  One side effect of this is that mutliple AS at th=
e
ASP
> may attempt to activate for the same RK.  The ASP cannot detect this an
> remain
> RK neutral.  So, when a second AS attemtps to register the same RK (and
> obtain
> its own unique RC), it is the SGP that refuses the registration with
"Cannot
> support unique routing", instead of requiring the ASP to maintain RKs a=
nd
> police AS/RC/RK uniqueness.  The RFC places the obligation at the SGP
(M3UA,
> NIF or MTP) to enforce the 1:1:1 relationship of AS:RC:RK, because it i=
s
the
> one that requires this relationship so that there is "unique routing" i=
n
the
> SGP routing table.
>
> This is rather important to interoperability, because there is no
mechanism
> in the M3UA ASP specification to enforce this relationship, only the
> mechanism
> at the SGP.  Altering the behviour of the SGP would break this
> interoperability mechanism.
>
> There are probably three or four other mechanism within your architectu=
re
> that will solve the problem for you without changing M3UA and breaking
this
> more important interoperability behavior.  So I cannot agree with the
> change.
>
> --brian
>
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

--
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003


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



From exim@www1.ietf.org  Tue Jul 15 17:17:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA05306
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 17:17:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cXAW-0002xw-3q
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 17:17:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6FLH4bd011394
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 17:17:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cXAT-0002xS-By; Tue, 15 Jul 2003 17:17:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cXA5-0002xF-Lq
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 17:16:38 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA05295
	for <SIGTRAN@ietf.org>; Tue, 15 Jul 2003 17:16:32 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cXA3-0005mM-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 17:16:35 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[B9RNT04sFVY6fHuoouBf6ftrPJ1CYr9v])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cX9n-0005lz-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 17:16:19 -0400
Received: from ns.pigworks.openss7.net (IDENT:YSBbunq7jVwY2dau6WBTYJOYV7qwAKtS@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6FLFVP11150;
	Tue, 15 Jul 2003 15:15:31 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6FLFU715839;
	Tue, 15 Jul 2003 15:15:30 -0600
Date: Tue, 15 Jul 2003 15:15:30 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Umesh Vats <umesh@ccpu.com>
Cc: SIGTRAN e-mail list <SIGTRAN@ietf.org>
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Message-ID: <20030715151530.C14123@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Umesh Vats <umesh@ccpu.com>,
	SIGTRAN e-mail list <SIGTRAN@ietf.org>
References: <20030715135049.B14123@openss7.org> <ACEMIJONPBBPDLOKCABHAEECCBAA.umesh@ccpu.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <ACEMIJONPBBPDLOKCABHAEECCBAA.umesh@ccpu.com>; from umesh@ccpu.com on Tue, Jul 15, 2003 at 01:10:04PM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6FLFVP11150
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Umesh,

An SG designed to meet SS7 requirements will not have any problems
losing M3UA messages.

--brian

On Tue, 15 Jul 2003, Umesh Vats wrote:

> If ASP looses ASPSM/ASPTM/DRKM/SSNM messages(which are consumed by M3UA=
,
> it should be able to recover.
>=20
> For SSNM, subsequent SSNM messages or DAUD procedure take care of messa=
ge
> loss.
> For if ASP-ACTIVE_ACK response is lost, ASP can resend ASP-ACTIVE.
>=20
> But for loss of REG-RSP, the only recovery now possible is ASP-DOWN
> procedure which is far worse that deregistering just one RK.
>=20
> Umesh
>=20
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Tuesday, July 15, 2003 12:51 PM
> To: Umesh Vats
> Cc: SIGTRAN e-mail list; Vivek Toky
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>=20
>=20
> Umesh,
>=20
> What should ASP do if it loses data.  What should ASP do if it loses
> ASP xxxx Ack.  What should ASP do if it loses mandatory Notify.  What
> should ASP do....
>=20
> The SGP should not lose messagez, or if it does, send ASP Inactive Ack
> or ASP Down Ack to reset the association.
>=20
> If it loses messages without detecting them, there is nothing to do but
> recreate SCTP in M3UA (a rather absurd proposition).
>=20
> --brian
>=20
>=20
> On Tue, 15 Jul 2003, Umesh Vats wrote:
>=20
> > REG-RSP from SGP to ASP ,may be lost at SG itself before it reaches S=
CTP
> > due to local problems at SG. In this case, what should ASP do ?
> >
> > I don't understand why sending of RC from SG poses any problem.
> > SG already knows about the RK and RC. It just has to return the RC.
> >
> > Umesh
> >
> >
> > -----Original Message-----
> > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf=
 Of
> > Brian F. G. Bidulock
> > Sent: Monday, July 14, 2003 10:13 PM
> > To: Umesh Vats
> > Cc: SIGTRAN e-mail list; Vivek Toky
> > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> >
> >
> > Umesh,
> >
> > On Mon, 14 Jul 2003, Umesh Vats wrote:
> >
> > > Brian,
> > >
> > > I am sure you have sufficient background of highly-available/distri=
buted
> > > systems as your implementation is 100% fault tolerant with
> > > nil (0%) message loss.
> > >
> > > In the given scenario, what if host1 receives REG-RSP from host3 bu=
t
> > before
> > > updating
> > > the RC received in message to common data, host1 goes down.
> >
> > An host1 does not acknowledge the message to host3, so when host2 com=
es
> > up, host3 sends the REG RSP again to host2.
> >
> > >
> > > In this case, host2 doesn't know the RC and RK relation.
> >
> > The RC is in the REG RSP that host3 resends to host2.  The RK is in c=
ommon
> > data along with the Local-Routing-Key-Identifier that is also contain=
ed in
> > the REG RSP.  host2 will have no problem correlating the REG RSP with=
 the
> > REG REQ, as would host1 (on the LRK-Id).
> >
> > > Performing ASP-DOWN procedure will make ASP DOWN for all the Applic=
ation
> > > Servers for which traffic exchange might be happening.
> >
> > So, don't.  Have host3 send the REG RSP message not acknowledged by h=
ost1
> > to host2 when host2 takes over.
> >
> > >
> > > Infact, your implementation can use this idea to further decrease
> message
> > > loss from 0% to -1% percentage or less :)
> >
> > Your implementation does not need to lose REG RSP unless it has been
> > designed to do so.  Don't do that.  Use techniques such as above so t=
hat
> > you do not lose messages.  Do not expect M3UA to repeat messages to h=
andle
> > the idiosynchrasies of your own internal architecture.
> >
> > The reason for sending "Cannot support unique routing" was because we
> spent
> > some great pains removing RK depedency from the ASP.  The ASP can be
> > designed
> > completely RK neutral and treat the RK as opaque.  Only the AS need k=
now
> the
> > internals of the RK.  One side effect of this is that mutliple AS at =
the
> ASP
> > may attempt to activate for the same RK.  The ASP cannot detect this =
an
> > remain
> > RK neutral.  So, when a second AS attemtps to register the same RK (a=
nd
> > obtain
> > its own unique RC), it is the SGP that refuses the registration with
> "Cannot
> > support unique routing", instead of requiring the ASP to maintain RKs=
 and
> > police AS/RC/RK uniqueness.  The RFC places the obligation at the SGP
> (M3UA,
> > NIF or MTP) to enforce the 1:1:1 relationship of AS:RC:RK, because it=
 is
> the
> > one that requires this relationship so that there is "unique routing"=
 in
> the
> > SGP routing table.
> >
> > This is rather important to interoperability, because there is no
> mechanism
> > in the M3UA ASP specification to enforce this relationship, only the
> > mechanism
> > at the SGP.  Altering the behviour of the SGP would break this
> > interoperability mechanism.
> >
> > There are probably three or four other mechanism within your architec=
ture
> > that will solve the problem for you without changing M3UA and breakin=
g
> this
> > more important interoperability behavior.  So I cannot agree with the
> > change.
> >
> > --brian
> >
> > --
> > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =
=A6
> > bidulock@openss7.org    =A6 world; the unreasonable one persists in  =
=A6
> > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =
=A6
> >                         =A6 Therefore  all  progress  depends on the =
=A6
> >                         =A6 unreasonable man. -- George Bernard Shaw =
=A6
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> > ---
> > Incoming mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > ---
> > Outgoing mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>=20
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>=20
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>=20
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Tue Jul 15 17:18:26 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA05359
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 17:18:26 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cXBR-000304-2c
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 17:18:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6FLI176011528
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 17:18:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cXBQ-0002zp-V4; Tue, 15 Jul 2003 17:18:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cXAT-0002xL-Bq
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 17:17:01 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA05298
	for <SIGTRAN@ietf.org>; Tue, 15 Jul 2003 17:16:54 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cXAP-0005mZ-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 17:16:57 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[Aa7bZPkwa2u7+uS/pN0CZWD7ZtjviNpt])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cXA9-0005mT-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 17:16:41 -0400
Received: from ns.pigworks.openss7.net (IDENT:CoWVEmKzjnjIEWuxouUokpKxUsgl1dBO@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6FLGUP11165;
	Tue, 15 Jul 2003 15:16:30 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6FLGUD15860;
	Tue, 15 Jul 2003 15:16:30 -0600
Date: Tue, 15 Jul 2003 15:16:30 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Umesh Vats <umesh@ccpu.com>
Cc: SIGTRAN e-mail list <SIGTRAN@ietf.org>
Subject: Re: [Sigtran] M3UA IG V04: Receiving REG for a RK already registered
Message-ID: <20030715151630.D14123@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Umesh Vats <umesh@ccpu.com>,
	SIGTRAN e-mail list <SIGTRAN@ietf.org>
References: <20030715134806.A14123@openss7.org> <ACEMIJONPBBPDLOKCABHKEEBCBAA.umesh@ccpu.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <ACEMIJONPBBPDLOKCABHKEEBCBAA.umesh@ccpu.com>; from umesh@ccpu.com on Tue, Jul 15, 2003 at 01:04:21PM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6FLGUP11165
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Umesh,

Read back in this thread.  Do a visual grep for "RK neutral".

--brian

On Tue, 15 Jul 2003, Umesh Vats wrote:

> Why will there be two Application Servers with same RK at an
> ASP ? There is 1:1 releationship between RK and AS. If an
> ASP is sending same RK, this in an Error case no matter
> what Local RK-Id is there in the REG-REQ.
>=20
> SGP can respond with unsuccessful REG-RSP with Rk-Id and RC.
>=20
> Umesh
>=20
>=20
>=20
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Tuesday, July 15, 2003 12:48 PM
> To: Umesh Vats
> Cc: Zannin, Francois; VIvek Toky; SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA IG V04: Receiving REG for a RK already
> registered
>=20
>=20
> Umesh,
>=20
> No, because by only comparing RK's the SG cannot distinguish between
> a separate AS registering for the same RK from the same ASP.  I thought
> I explained this several mails ago.
>=20
> --brian
>=20
> On Tue, 15 Jul 2003, Umesh Vats wrote:
>=20
> > Whay does SG need to store the Local RK-Id ? Local RK-Id
> > is just like transaction id for correlating REG-REQ and REG-RSP for
> > one exchange of routing key registration procedure.
> >
> > To find out whether the RK request from an ASP is duplicate or not,
> > SG can just compare the RK's already registered by the corresponding =
ASP.
> >
> > Umesh
> >
> > -----Original Message-----
> > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf=
 Of
> > Brian F. G. Bidulock
> > Sent: Tuesday, July 15, 2003 3:37 AM
> > To: Zannin, Francois
> > Cc: Umesh Vats; VIvek Toky; SIGTRAN e-mail list
> > Subject: Re: [Sigtran] M3UA IG V04: Receiving REG for a RK already
> > registered
> >
> >
> > Francois,
> >
> > Well, not quite.  Even the buggy M3UA should be able to retain the
> > Local-Routing-Key-Identifier from the original REG REQ.
> > All that the ASP need do is reissue If the M3UA cannot maintain the
> > state of the original REG REQ (i.e, retain the
> > Local-Routing-Key-Identifier used) then it cannot correlate the REG R=
SP
> > anyway.  the REG RSP with the same Local-Routing-Key-Identifier.  The
> > SGP uses this to determine that the REG REQ is a repeat attempt an
> > replies with a REG RSP.
> >
> > REG RSP("Cannot support unique routing") is intended as a reply to a =
REG
> > REQ from a different AS at the ASP.  This could be identified by the
> > presence of a different Local-Routing-Key-Identifier than the origina=
l.
> >
> > I know that this not mentioned in the IG 04, but we discussed it.
> >
> > Example: an ASP sends a REG REQ and waits.  It gives up and sends
> > another REG REQ.  It then receives two REG RSP.  We discussed that if=
 an
> > ASP were to send two REG REQ for the same registration that they shou=
ld
> > contain the same LRK-Id.
> >
> > In the case where a REG REQ contains the same LRK-Id for the same RK,
> > the response to the second could be the same as the response to the
> > first. (Successful).
> >
> > Now, the objectionable lines are in IG 04:
> >
> >    --------
> >    New text: (Section 4.4.1)
> >    ---------
> >
> >    If the SGP determines that the received RK was already registered,
> >    the SGP returns a Registration Response message to the ASP,
> >    containing a Registration Result "Error - Cannot Support Unique
> >    Routing". This error applies whether the Routing Key is Active or
> >    Inactive.
> >
> > But, this (strangely) does not change the pertinent passages in the R=
FC
> > which say:
> >
> >    If the SGP determines that a unique Routing Key cannot be created,
> >    the SGP returns a Registration Response message to the ASP, with a
> >    Registration Status of "Error - "Cannot Support Unique Routing"  A=
n
> >    incoming signalling message received at an SGP should not match
> >    against more than one Routing Key.
> >
> > This error was really intended to preclude "overlapping" routing keys
> > which are verbodden in M3UA.  The error was never intended for attemp=
ts
> > at re-registration by the same AS.
> >
> > How does the following change sound?
> >
> >    --------
> >    New text: (Section 4.4.1)
> >    ---------
> >
> >    If the SGP determines that the received RK was already registered,
> >    and the Local-Routing-Key-Identifier in the RK is the same as the
> >    Local-Routing-Key-Identifier that was used to register the RK, the=
n
> >    the SGP treats the REG REQ as a duplicate and returns a Registrati=
on
> >    Response message to the ASP, containing a Registration Result
> >    "Successful" and the Routing Context associated with the RK.
> >
> >    If the SGP determines that the received RK was already registered,
> >    and the Local-Routing-Key-Identifier in the RK does not match the
> >    Local-Routing-Key-Identifier that was used to register the RK, the
> >    the SGP treats the REG REQ as an overlapping RK request and return=
s
> >    a Registration Response message to the ASP, containing a Registrat=
ion
> >    Result "Error - Cannot Support Unique Routing". This error applies
> >    whether the Routing Key is Active or Inactive.
> >
> > This would only push the SGP into maintain, along with the fact that =
the
> > ASP is registered against the RK, the Local-Routing-Key-Id that was u=
sed
> > by the ASP to register.  4 bytes.  Not too much.
> >
> > Then an AS can send as many REG REQ as it likes, as long as it uses t=
he
> > same LRK-Id, it will get the same response.
> >
> > Sound good?
> >
> > --brian
> >
> > On Tue, 15 Jul 2003, Zannin, Francois wrote:
> >
> > > Brian,
> > >
> > > This is true that SCTP is a very reliable transport layer but it ma=
y
> > > happen that a M3UA on top of SCTP not "bug free" can lost a receive=
d
> > > message (maybe due to local congestion?). In that case, if the ASP =
has
> > > lost the reg_resp, than it cannot deregister the lost RC (because i=
t is
> > > not known), the guilty ASP is then unable to recover the missing RC
> > > unless maybe to send an ASP-DOWN/ASP-UP in order to allows again th=
e AS
> > > registration. Then, all already registered AS on the same associati=
ons
> > > are lost!
> > >
> > > If the Routing KEY is the same as the one already registered, isn't=
 it
> > > possible for the SG to repeat the same RC?
> > >
> > > If the AS (RC) was already ACTIVE on this association, yes it is an
> > > error, then the AS mUST return to the INACTIVE state.
> > >
> > > Francois.
> > >
> >
> > --
> > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =
=A6
> > bidulock@openss7.org    =A6 world; the unreasonable one persists in  =
=A6
> > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =
=A6
> >                         =A6 Therefore  all  progress  depends on the =
=A6
> >                         =A6 unreasonable man. -- George Bernard Shaw =
=A6
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> > ---
> > Incoming mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > ---
> > Outgoing mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>=20
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>=20
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>=20
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Tue Jul 15 17:48:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA06283
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 17:48:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cXeU-0004I3-Ar
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 17:48:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6FLm26R016490
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 17:48:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cXeT-0004Hg-4G; Tue, 15 Jul 2003 17:48:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cXdW-0004C6-Vy
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 17:47:03 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA06181
	for <SIGTRAN@ietf.org>; Tue, 15 Jul 2003 17:46:57 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cXdU-00062c-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 17:47:00 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19cXdJ-00062V-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 17:46:49 -0400
Received: from [172.16.1.203] (helo=CCPU00186)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19cXcm-0005hp-78; Tue, 15 Jul 2003 14:46:16 -0700
From: "VIvek Toky" <vivektoky@ccpu.com>
To: <bidulock@openss7.org>, "Umesh Vats" <umesh@ccpu.com>
Cc: "SIGTRAN e-mail list" <SIGTRAN@ietf.org>
Subject: RE: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Date: Tue, 15 Jul 2003 14:50:57 -0700
Message-ID: <HKEBKFBNPIDALDOKLGKDOEHMCDAA.vivektoky@ccpu.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.6604 (9.0.2911.0)
In-Reply-To: <20030715151530.C14123@openss7.org>
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id RAA06182
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Brian,

  You said
   "The reason for sending "Cannot support unique routing" was because we
spent
some great pains removing RK depedency from the ASP.  The ASP can be
designed
completely RK neutral and treat the RK as opaque.  Only the AS need know =
the
internals of the RK."

Why do you want ASPs to be totally unawar of whats inside the RK, If SGPs
have to do the routing based on RKs then ASPs also have to route the
incoming messages to the destined AS (locally). If different ASs try to
register same routing key then it can be checked at ASP itself without ev=
en
sending the REG Request message out.

Can you explain more about the interoperability scenarios you mentioned..

-Vivek.

-----Original Message-----
From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf Of
Brian F. G. Bidulock
Sent: Tuesday, July 15, 2003 2:16 PM
To: Umesh Vats
Cc: SIGTRAN e-mail list
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP


Umesh,

An SG designed to meet SS7 requirements will not have any problems
losing M3UA messages.

--brian

On Tue, 15 Jul 2003, Umesh Vats wrote:

> If ASP looses ASPSM/ASPTM/DRKM/SSNM messages(which are consumed by M3UA=
,
> it should be able to recover.
>
> For SSNM, subsequent SSNM messages or DAUD procedure take care of messa=
ge
> loss.
> For if ASP-ACTIVE_ACK response is lost, ASP can resend ASP-ACTIVE.
>
> But for loss of REG-RSP, the only recovery now possible is ASP-DOWN
> procedure which is far worse that deregistering just one RK.
>
> Umesh
>
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Tuesday, July 15, 2003 12:51 PM
> To: Umesh Vats
> Cc: SIGTRAN e-mail list; Vivek Toky
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>
>
> Umesh,
>
> What should ASP do if it loses data.  What should ASP do if it loses
> ASP xxxx Ack.  What should ASP do if it loses mandatory Notify.  What
> should ASP do....
>
> The SGP should not lose messagez, or if it does, send ASP Inactive Ack
> or ASP Down Ack to reset the association.
>
> If it loses messages without detecting them, there is nothing to do but
> recreate SCTP in M3UA (a rather absurd proposition).
>
> --brian
>
>
> On Tue, 15 Jul 2003, Umesh Vats wrote:
>
> > REG-RSP from SGP to ASP ,may be lost at SG itself before it reaches S=
CTP
> > due to local problems at SG. In this case, what should ASP do ?
> >
> > I don't understand why sending of RC from SG poses any problem.
> > SG already knows about the RK and RC. It just has to return the RC.
> >
> > Umesh
> >
> >
> > -----Original Message-----
> > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf=
 Of
> > Brian F. G. Bidulock
> > Sent: Monday, July 14, 2003 10:13 PM
> > To: Umesh Vats
> > Cc: SIGTRAN e-mail list; Vivek Toky
> > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> >
> >
> > Umesh,
> >
> > On Mon, 14 Jul 2003, Umesh Vats wrote:
> >
> > > Brian,
> > >
> > > I am sure you have sufficient background of
highly-available/distributed
> > > systems as your implementation is 100% fault tolerant with
> > > nil (0%) message loss.
> > >
> > > In the given scenario, what if host1 receives REG-RSP from host3 bu=
t
> > before
> > > updating
> > > the RC received in message to common data, host1 goes down.
> >
> > An host1 does not acknowledge the message to host3, so when host2 com=
es
> > up, host3 sends the REG RSP again to host2.
> >
> > >
> > > In this case, host2 doesn't know the RC and RK relation.
> >
> > The RC is in the REG RSP that host3 resends to host2.  The RK is in
common
> > data along with the Local-Routing-Key-Identifier that is also contain=
ed
in
> > the REG RSP.  host2 will have no problem correlating the REG RSP with
the
> > REG REQ, as would host1 (on the LRK-Id).
> >
> > > Performing ASP-DOWN procedure will make ASP DOWN for all the
Application
> > > Servers for which traffic exchange might be happening.
> >
> > So, don't.  Have host3 send the REG RSP message not acknowledged by
host1
> > to host2 when host2 takes over.
> >
> > >
> > > Infact, your implementation can use this idea to further decrease
> message
> > > loss from 0% to -1% percentage or less :)
> >
> > Your implementation does not need to lose REG RSP unless it has been
> > designed to do so.  Don't do that.  Use techniques such as above so t=
hat
> > you do not lose messages.  Do not expect M3UA to repeat messages to
handle
> > the idiosynchrasies of your own internal architecture.
> >
> > The reason for sending "Cannot support unique routing" was because we
> spent
> > some great pains removing RK depedency from the ASP.  The ASP can be
> > designed
> > completely RK neutral and treat the RK as opaque.  Only the AS need k=
now
> the
> > internals of the RK.  One side effect of this is that mutliple AS at =
the
> ASP
> > may attempt to activate for the same RK.  The ASP cannot detect this =
an
> > remain
> > RK neutral.  So, when a second AS attemtps to register the same RK (a=
nd
> > obtain
> > its own unique RC), it is the SGP that refuses the registration with
> "Cannot
> > support unique routing", instead of requiring the ASP to maintain RKs
and
> > police AS/RC/RK uniqueness.  The RFC places the obligation at the SGP
> (M3UA,
> > NIF or MTP) to enforce the 1:1:1 relationship of AS:RC:RK, because it=
 is
> the
> > one that requires this relationship so that there is "unique routing"=
 in
> the
> > SGP routing table.
> >
> > This is rather important to interoperability, because there is no
> mechanism
> > in the M3UA ASP specification to enforce this relationship, only the
> > mechanism
> > at the SGP.  Altering the behviour of the SGP would break this
> > interoperability mechanism.
> >
> > There are probably three or four other mechanism within your
architecture
> > that will solve the problem for you without changing M3UA and breakin=
g
> this
> > more important interoperability behavior.  So I cannot agree with the
> > change.
> >
> > --brian
> >
> > --
> > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =
=A6
> > bidulock@openss7.org    =A6 world; the unreasonable one persists in  =
=A6
> > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =
=A6
> >                         =A6 Therefore  all  progress  depends on the =
=A6
> >                         =A6 unreasonable man. -- George Bernard Shaw =
=A6
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> > ---
> > Incoming mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > ---
> > Outgoing mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

--
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

_______________________________________________
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 exim@www1.ietf.org  Tue Jul 15 18:21:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA08592
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 18:21:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cYAS-0005oa-0P
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 18:21:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6FML3Bi022346
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 18:21:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cYAQ-0005n3-8t; Tue, 15 Jul 2003 18:21:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cYA8-0005mK-9U
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 18:20:44 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA08548
	for <SIGTRAN@ietf.org>; Tue, 15 Jul 2003 18:20:37 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cYA4-0006OL-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 18:20:40 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19cY9u-0006O6-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 18:20:30 -0400
Received: from [172.16.1.40] (helo=ccpu00251)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19cY9a-0008DO-0G; Tue, 15 Jul 2003 15:20:10 -0700
Reply-To: <umesh@ccpu.com>
From: "Umesh Vats" <umesh@ccpu.com>
To: <bidulock@openss7.org>
Cc: "SIGTRAN e-mail list" <SIGTRAN@ietf.org>
Subject: RE: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Date: Tue, 15 Jul 2003 15:25:46 -0700
Message-ID: <ACEMIJONPBBPDLOKCABHEEEDCBAA.umesh@ccpu.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0)
In-Reply-To: <20030715151530.C14123@openss7.org>
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id SAA08549
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Brian,

REG-RSP message with unsuccesful registration status and Local RK-Id
but no RC serves far less purpose than REG-RSP with Local-RK and RC in
all practical SGP/ASP impplementations.

I don't see any reason why a better approach for handling errors shouldn'=
t
be used if its possible.

Umesh


-----Original Message-----
From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
Sent: Tuesday, July 15, 2003 2:16 PM
To: Umesh Vats
Cc: SIGTRAN e-mail list
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP


Umesh,

An SG designed to meet SS7 requirements will not have any problems
losing M3UA messages.

--brian

On Tue, 15 Jul 2003, Umesh Vats wrote:

> If ASP looses ASPSM/ASPTM/DRKM/SSNM messages(which are consumed by M3UA=
,
> it should be able to recover.
>
> For SSNM, subsequent SSNM messages or DAUD procedure take care of messa=
ge
> loss.
> For if ASP-ACTIVE_ACK response is lost, ASP can resend ASP-ACTIVE.
>
> But for loss of REG-RSP, the only recovery now possible is ASP-DOWN
> procedure which is far worse that deregistering just one RK.
>
> Umesh
>
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Tuesday, July 15, 2003 12:51 PM
> To: Umesh Vats
> Cc: SIGTRAN e-mail list; Vivek Toky
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>
>
> Umesh,
>
> What should ASP do if it loses data.  What should ASP do if it loses
> ASP xxxx Ack.  What should ASP do if it loses mandatory Notify.  What
> should ASP do....
>
> The SGP should not lose messagez, or if it does, send ASP Inactive Ack
> or ASP Down Ack to reset the association.
>
> If it loses messages without detecting them, there is nothing to do but
> recreate SCTP in M3UA (a rather absurd proposition).
>
> --brian
>
>
> On Tue, 15 Jul 2003, Umesh Vats wrote:
>
> > REG-RSP from SGP to ASP ,may be lost at SG itself before it reaches S=
CTP
> > due to local problems at SG. In this case, what should ASP do ?
> >
> > I don't understand why sending of RC from SG poses any problem.
> > SG already knows about the RK and RC. It just has to return the RC.
> >
> > Umesh
> >
> >
> > -----Original Message-----
> > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf=
 Of
> > Brian F. G. Bidulock
> > Sent: Monday, July 14, 2003 10:13 PM
> > To: Umesh Vats
> > Cc: SIGTRAN e-mail list; Vivek Toky
> > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> >
> >
> > Umesh,
> >
> > On Mon, 14 Jul 2003, Umesh Vats wrote:
> >
> > > Brian,
> > >
> > > I am sure you have sufficient background of
highly-available/distributed
> > > systems as your implementation is 100% fault tolerant with
> > > nil (0%) message loss.
> > >
> > > In the given scenario, what if host1 receives REG-RSP from host3 bu=
t
> > before
> > > updating
> > > the RC received in message to common data, host1 goes down.
> >
> > An host1 does not acknowledge the message to host3, so when host2 com=
es
> > up, host3 sends the REG RSP again to host2.
> >
> > >
> > > In this case, host2 doesn't know the RC and RK relation.
> >
> > The RC is in the REG RSP that host3 resends to host2.  The RK is in
common
> > data along with the Local-Routing-Key-Identifier that is also contain=
ed
in
> > the REG RSP.  host2 will have no problem correlating the REG RSP with
the
> > REG REQ, as would host1 (on the LRK-Id).
> >
> > > Performing ASP-DOWN procedure will make ASP DOWN for all the
Application
> > > Servers for which traffic exchange might be happening.
> >
> > So, don't.  Have host3 send the REG RSP message not acknowledged by
host1
> > to host2 when host2 takes over.
> >
> > >
> > > Infact, your implementation can use this idea to further decrease
> message
> > > loss from 0% to -1% percentage or less :)
> >
> > Your implementation does not need to lose REG RSP unless it has been
> > designed to do so.  Don't do that.  Use techniques such as above so t=
hat
> > you do not lose messages.  Do not expect M3UA to repeat messages to
handle
> > the idiosynchrasies of your own internal architecture.
> >
> > The reason for sending "Cannot support unique routing" was because we
> spent
> > some great pains removing RK depedency from the ASP.  The ASP can be
> > designed
> > completely RK neutral and treat the RK as opaque.  Only the AS need k=
now
> the
> > internals of the RK.  One side effect of this is that mutliple AS at =
the
> ASP
> > may attempt to activate for the same RK.  The ASP cannot detect this =
an
> > remain
> > RK neutral.  So, when a second AS attemtps to register the same RK (a=
nd
> > obtain
> > its own unique RC), it is the SGP that refuses the registration with
> "Cannot
> > support unique routing", instead of requiring the ASP to maintain RKs
and
> > police AS/RC/RK uniqueness.  The RFC places the obligation at the SGP
> (M3UA,
> > NIF or MTP) to enforce the 1:1:1 relationship of AS:RC:RK, because it=
 is
> the
> > one that requires this relationship so that there is "unique routing"=
 in
> the
> > SGP routing table.
> >
> > This is rather important to interoperability, because there is no
> mechanism
> > in the M3UA ASP specification to enforce this relationship, only the
> > mechanism
> > at the SGP.  Altering the behviour of the SGP would break this
> > interoperability mechanism.
> >
> > There are probably three or four other mechanism within your
architecture
> > that will solve the problem for you without changing M3UA and breakin=
g
> this
> > more important interoperability behavior.  So I cannot agree with the
> > change.
> >
> > --brian
> >
> > --
> > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =
=A6
> > bidulock@openss7.org    =A6 world; the unreasonable one persists in  =
=A6
> > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =
=A6
> >                         =A6 Therefore  all  progress  depends on the =
=A6
> >                         =A6 unreasonable man. -- George Bernard Shaw =
=A6
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> >
> > ---
> > Incoming mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > ---
> > Outgoing mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

--
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003


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



From exim@www1.ietf.org  Tue Jul 15 18:23:25 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA08683
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 18:23:25 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cYCL-0005yn-Ak
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 18:23:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6FMN11i022970
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 18:23:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cYCL-0005yO-6D; Tue, 15 Jul 2003 18:23:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cYBS-0005wn-S3
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 18:22:06 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA08615
	for <SIGTRAN@ietf.org>; Tue, 15 Jul 2003 18:22:00 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cYBM-0006P7-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 18:22:00 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[h9pL5+weFFjWoGHCJCwlmr/La1zIfJQ6])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cYB6-0006Os-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 18:21:44 -0400
Received: from ns.pigworks.openss7.net (IDENT:7CkOaAreVBIhFgB4JvH8IlbD+FjDIHrn@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6FMLOP12143;
	Tue, 15 Jul 2003 16:21:24 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6FMLOW17080;
	Tue, 15 Jul 2003 16:21:24 -0600
Date: Tue, 15 Jul 2003 16:21:23 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: VIvek Toky <vivektoky@ccpu.com>
Cc: Umesh Vats <umesh@ccpu.com>, SIGTRAN e-mail list <SIGTRAN@ietf.org>
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Message-ID: <20030715162123.A16479@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: VIvek Toky <vivektoky@ccpu.com>,
	Umesh Vats <umesh@ccpu.com>, SIGTRAN e-mail list <SIGTRAN@ietf.org>
References: <20030715151530.C14123@openss7.org> <HKEBKFBNPIDALDOKLGKDOEHMCDAA.vivektoky@ccpu.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <HKEBKFBNPIDALDOKLGKDOEHMCDAA.vivektoky@ccpu.com>; from vivektoky@ccpu.com on Tue, Jul 15, 2003 at 02:50:57PM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6FMLOP12143
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

VIvek,

All ASPs need to route messages locally to AS is the RC.  They need
not be aware of the RK.

--brian


On Tue, 15 Jul 2003, VIvek Toky wrote:

> Brian,
>=20
>   You said
>    "The reason for sending "Cannot support unique routing" was because =
we
> spent
> some great pains removing RK depedency from the ASP.  The ASP can be
> designed
> completely RK neutral and treat the RK as opaque.  Only the AS need kno=
w the
> internals of the RK."
>=20
> Why do you want ASPs to be totally unawar of whats inside the RK, If SG=
Ps
> have to do the routing based on RKs then ASPs also have to route the
> incoming messages to the destined AS (locally). If different ASs try to
> register same routing key then it can be checked at ASP itself without =
even
> sending the REG Request message out.
>=20
> Can you explain more about the interoperability scenarios you mentioned=
..
>=20
> -Vivek.
>=20
> -----Original Message-----
> From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf O=
f
> Brian F. G. Bidulock
> Sent: Tuesday, July 15, 2003 2:16 PM
> To: Umesh Vats
> Cc: SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>=20
>=20
> Umesh,
>=20
> An SG designed to meet SS7 requirements will not have any problems
> losing M3UA messages.
>=20
> --brian
>=20
> On Tue, 15 Jul 2003, Umesh Vats wrote:
>=20
> > If ASP looses ASPSM/ASPTM/DRKM/SSNM messages(which are consumed by M3=
UA,
> > it should be able to recover.
> >
> > For SSNM, subsequent SSNM messages or DAUD procedure take care of mes=
sage
> > loss.
> > For if ASP-ACTIVE_ACK response is lost, ASP can resend ASP-ACTIVE.
> >
> > But for loss of REG-RSP, the only recovery now possible is ASP-DOWN
> > procedure which is far worse that deregistering just one RK.
> >
> > Umesh
> >
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: Tuesday, July 15, 2003 12:51 PM
> > To: Umesh Vats
> > Cc: SIGTRAN e-mail list; Vivek Toky
> > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> >
> >
> > Umesh,
> >
> > What should ASP do if it loses data.  What should ASP do if it loses
> > ASP xxxx Ack.  What should ASP do if it loses mandatory Notify.  What
> > should ASP do....
> >
> > The SGP should not lose messagez, or if it does, send ASP Inactive Ac=
k
> > or ASP Down Ack to reset the association.
> >
> > If it loses messages without detecting them, there is nothing to do b=
ut
> > recreate SCTP in M3UA (a rather absurd proposition).
> >
> > --brian
> >
> >
> > On Tue, 15 Jul 2003, Umesh Vats wrote:
> >
> > > REG-RSP from SGP to ASP ,may be lost at SG itself before it reaches=
 SCTP
> > > due to local problems at SG. In this case, what should ASP do ?
> > >
> > > I don't understand why sending of RC from SG poses any problem.
> > > SG already knows about the RK and RC. It just has to return the RC.
> > >
> > > Umesh
> > >
> > >
> > > -----Original Message-----
> > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Beha=
lf Of
> > > Brian F. G. Bidulock
> > > Sent: Monday, July 14, 2003 10:13 PM
> > > To: Umesh Vats
> > > Cc: SIGTRAN e-mail list; Vivek Toky
> > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > >
> > >
> > > Umesh,
> > >
> > > On Mon, 14 Jul 2003, Umesh Vats wrote:
> > >
> > > > Brian,
> > > >
> > > > I am sure you have sufficient background of
> highly-available/distributed
> > > > systems as your implementation is 100% fault tolerant with
> > > > nil (0%) message loss.
> > > >
> > > > In the given scenario, what if host1 receives REG-RSP from host3 =
but
> > > before
> > > > updating
> > > > the RC received in message to common data, host1 goes down.
> > >
> > > An host1 does not acknowledge the message to host3, so when host2 c=
omes
> > > up, host3 sends the REG RSP again to host2.
> > >
> > > >
> > > > In this case, host2 doesn't know the RC and RK relation.
> > >
> > > The RC is in the REG RSP that host3 resends to host2.  The RK is in
> common
> > > data along with the Local-Routing-Key-Identifier that is also conta=
ined
> in
> > > the REG RSP.  host2 will have no problem correlating the REG RSP wi=
th
> the
> > > REG REQ, as would host1 (on the LRK-Id).
> > >
> > > > Performing ASP-DOWN procedure will make ASP DOWN for all the
> Application
> > > > Servers for which traffic exchange might be happening.
> > >
> > > So, don't.  Have host3 send the REG RSP message not acknowledged by
> host1
> > > to host2 when host2 takes over.
> > >
> > > >
> > > > Infact, your implementation can use this idea to further decrease
> > message
> > > > loss from 0% to -1% percentage or less :)
> > >
> > > Your implementation does not need to lose REG RSP unless it has bee=
n
> > > designed to do so.  Don't do that.  Use techniques such as above so=
 that
> > > you do not lose messages.  Do not expect M3UA to repeat messages to
> handle
> > > the idiosynchrasies of your own internal architecture.
> > >
> > > The reason for sending "Cannot support unique routing" was because =
we
> > spent
> > > some great pains removing RK depedency from the ASP.  The ASP can b=
e
> > > designed
> > > completely RK neutral and treat the RK as opaque.  Only the AS need=
 know
> > the
> > > internals of the RK.  One side effect of this is that mutliple AS a=
t the
> > ASP
> > > may attempt to activate for the same RK.  The ASP cannot detect thi=
s an
> > > remain
> > > RK neutral.  So, when a second AS attemtps to register the same RK =
(and
> > > obtain
> > > its own unique RC), it is the SGP that refuses the registration wit=
h
> > "Cannot
> > > support unique routing", instead of requiring the ASP to maintain R=
Ks
> and
> > > police AS/RC/RK uniqueness.  The RFC places the obligation at the S=
GP
> > (M3UA,
> > > NIF or MTP) to enforce the 1:1:1 relationship of AS:RC:RK, because =
it is
> > the
> > > one that requires this relationship so that there is "unique routin=
g" in
> > the
> > > SGP routing table.
> > >
> > > This is rather important to interoperability, because there is no
> > mechanism
> > > in the M3UA ASP specification to enforce this relationship, only th=
e
> > > mechanism
> > > at the SGP.  Altering the behviour of the SGP would break this
> > > interoperability mechanism.
> > >
> > > There are probably three or four other mechanism within your
> architecture
> > > that will solve the problem for you without changing M3UA and break=
ing
> > this
> > > more important interoperability behavior.  So I cannot agree with t=
he
> > > change.
> > >
> > > --brian
> > >
> > > --
> > > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to th=
e =A6
> > > bidulock@openss7.org    =A6 world; the unreasonable one persists in=
  =A6
> > > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself=
. =A6
> > >                         =A6 Therefore  all  progress  depends on th=
e =A6
> > >                         =A6 unreasonable man. -- George Bernard Sha=
w =A6
> > >
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >
> > > ---
> > > Incoming mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > > ---
> > > Outgoing mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > --
> > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =
=A6
> > bidulock@openss7.org    =A6 world; the unreasonable one persists in  =
=A6
> > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =
=A6
> >                         =A6 Therefore  all  progress  depends on the =
=A6
> >                         =A6 unreasonable man. -- George Bernard Shaw =
=A6
> >
> > ---
> > Incoming mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > ---
> > Outgoing mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>=20
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Tue Jul 15 18:26:26 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA08841
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 18:26:26 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cYFG-0006BJ-2X
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 18:26:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6FMQ2sq023743
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 18:26:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cYFF-0006Ai-Ee; Tue, 15 Jul 2003 18:26:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cYFC-0006AL-46
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 18:25:58 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA08791
	for <SIGTRAN@ietf.org>; Tue, 15 Jul 2003 18:25:51 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cYF9-0006SI-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 18:25:55 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[WPCHfauQbJG8ZPnYmCrCE0majALOcJS+])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cYEt-0006SF-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 18:25:39 -0400
Received: from ns.pigworks.openss7.net (IDENT:wL4si6mTFCdGwwK44vWETCN+uO3khp+V@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6FMPTP12202;
	Tue, 15 Jul 2003 16:25:29 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6FMPSM17140;
	Tue, 15 Jul 2003 16:25:28 -0600
Date: Tue, 15 Jul 2003 16:25:28 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Umesh Vats <umesh@ccpu.com>
Cc: SIGTRAN e-mail list <SIGTRAN@ietf.org>
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Message-ID: <20030715162528.B16479@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Umesh Vats <umesh@ccpu.com>,
	SIGTRAN e-mail list <SIGTRAN@ietf.org>
References: <20030715151530.C14123@openss7.org> <ACEMIJONPBBPDLOKCABHEEEDCBAA.umesh@ccpu.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <ACEMIJONPBBPDLOKCABHEEEDCBAA.umesh@ccpu.com>; from umesh@ccpu.com on Tue, Jul 15, 2003 at 03:25:46PM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6FMPTP12202
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Umesh,

If your SG loses the odd REG RSP have it send two (or, why not,
three) of each.  The ASP can disregard the duplicates.

Oh, you missed DATA and required Notify below.  Also, beat/beat-ack.
And what about unsolicited ASP Inactive Ack and ASP Down Ack.

Tell me, why do you use SCTP?  All that reliabilty just so that your
design will throw away messages: seems like a waste.  You should use
UDP.

--brian

On Tue, 15 Jul 2003, Umesh Vats wrote:

> Brian,
>=20
> REG-RSP message with unsuccesful registration status and Local RK-Id
> but no RC serves far less purpose than REG-RSP with Local-RK and RC in
> all practical SGP/ASP impplementations.
>=20
> I don't see any reason why a better approach for handling errors should=
n't
> be used if its possible.
>=20
> Umesh
>=20
>=20
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Tuesday, July 15, 2003 2:16 PM
> To: Umesh Vats
> Cc: SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>=20
>=20
> Umesh,
>=20
> An SG designed to meet SS7 requirements will not have any problems
> losing M3UA messages.
>=20
> --brian
>=20
> On Tue, 15 Jul 2003, Umesh Vats wrote:
>=20
> > If ASP looses ASPSM/ASPTM/DRKM/SSNM messages(which are consumed by M3=
UA,
> > it should be able to recover.
> >
> > For SSNM, subsequent SSNM messages or DAUD procedure take care of mes=
sage
> > loss.
> > For if ASP-ACTIVE_ACK response is lost, ASP can resend ASP-ACTIVE.
> >
> > But for loss of REG-RSP, the only recovery now possible is ASP-DOWN
> > procedure which is far worse that deregistering just one RK.
> >
> > Umesh
> >
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: Tuesday, July 15, 2003 12:51 PM
> > To: Umesh Vats
> > Cc: SIGTRAN e-mail list; Vivek Toky
> > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> >
> >
> > Umesh,
> >
> > What should ASP do if it loses data.  What should ASP do if it loses
> > ASP xxxx Ack.  What should ASP do if it loses mandatory Notify.  What
> > should ASP do....
> >
> > The SGP should not lose messagez, or if it does, send ASP Inactive Ac=
k
> > or ASP Down Ack to reset the association.
> >
> > If it loses messages without detecting them, there is nothing to do b=
ut
> > recreate SCTP in M3UA (a rather absurd proposition).
> >
> > --brian
> >
> >
> > On Tue, 15 Jul 2003, Umesh Vats wrote:
> >
> > > REG-RSP from SGP to ASP ,may be lost at SG itself before it reaches=
 SCTP
> > > due to local problems at SG. In this case, what should ASP do ?
> > >
> > > I don't understand why sending of RC from SG poses any problem.
> > > SG already knows about the RK and RC. It just has to return the RC.
> > >
> > > Umesh
> > >
> > >
> > > -----Original Message-----
> > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Beha=
lf Of
> > > Brian F. G. Bidulock
> > > Sent: Monday, July 14, 2003 10:13 PM
> > > To: Umesh Vats
> > > Cc: SIGTRAN e-mail list; Vivek Toky
> > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > >
> > >
> > > Umesh,
> > >
> > > On Mon, 14 Jul 2003, Umesh Vats wrote:
> > >
> > > > Brian,
> > > >
> > > > I am sure you have sufficient background of
> highly-available/distributed
> > > > systems as your implementation is 100% fault tolerant with
> > > > nil (0%) message loss.
> > > >
> > > > In the given scenario, what if host1 receives REG-RSP from host3 =
but
> > > before
> > > > updating
> > > > the RC received in message to common data, host1 goes down.
> > >
> > > An host1 does not acknowledge the message to host3, so when host2 c=
omes
> > > up, host3 sends the REG RSP again to host2.
> > >
> > > >
> > > > In this case, host2 doesn't know the RC and RK relation.
> > >
> > > The RC is in the REG RSP that host3 resends to host2.  The RK is in
> common
> > > data along with the Local-Routing-Key-Identifier that is also conta=
ined
> in
> > > the REG RSP.  host2 will have no problem correlating the REG RSP wi=
th
> the
> > > REG REQ, as would host1 (on the LRK-Id).
> > >
> > > > Performing ASP-DOWN procedure will make ASP DOWN for all the
> Application
> > > > Servers for which traffic exchange might be happening.
> > >
> > > So, don't.  Have host3 send the REG RSP message not acknowledged by
> host1
> > > to host2 when host2 takes over.
> > >
> > > >
> > > > Infact, your implementation can use this idea to further decrease
> > message
> > > > loss from 0% to -1% percentage or less :)
> > >
> > > Your implementation does not need to lose REG RSP unless it has bee=
n
> > > designed to do so.  Don't do that.  Use techniques such as above so=
 that
> > > you do not lose messages.  Do not expect M3UA to repeat messages to
> handle
> > > the idiosynchrasies of your own internal architecture.
> > >
> > > The reason for sending "Cannot support unique routing" was because =
we
> > spent
> > > some great pains removing RK depedency from the ASP.  The ASP can b=
e
> > > designed
> > > completely RK neutral and treat the RK as opaque.  Only the AS need=
 know
> > the
> > > internals of the RK.  One side effect of this is that mutliple AS a=
t the
> > ASP
> > > may attempt to activate for the same RK.  The ASP cannot detect thi=
s an
> > > remain
> > > RK neutral.  So, when a second AS attemtps to register the same RK =
(and
> > > obtain
> > > its own unique RC), it is the SGP that refuses the registration wit=
h
> > "Cannot
> > > support unique routing", instead of requiring the ASP to maintain R=
Ks
> and
> > > police AS/RC/RK uniqueness.  The RFC places the obligation at the S=
GP
> > (M3UA,
> > > NIF or MTP) to enforce the 1:1:1 relationship of AS:RC:RK, because =
it is
> > the
> > > one that requires this relationship so that there is "unique routin=
g" in
> > the
> > > SGP routing table.
> > >
> > > This is rather important to interoperability, because there is no
> > mechanism
> > > in the M3UA ASP specification to enforce this relationship, only th=
e
> > > mechanism
> > > at the SGP.  Altering the behviour of the SGP would break this
> > > interoperability mechanism.
> > >
> > > There are probably three or four other mechanism within your
> architecture
> > > that will solve the problem for you without changing M3UA and break=
ing
> > this
> > > more important interoperability behavior.  So I cannot agree with t=
he
> > > change.
> > >
> > > --brian
> > >
> > > --
> > > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to th=
e =A6
> > > bidulock@openss7.org    =A6 world; the unreasonable one persists in=
  =A6
> > > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself=
. =A6
> > >                         =A6 Therefore  all  progress  depends on th=
e =A6
> > >                         =A6 unreasonable man. -- George Bernard Sha=
w =A6
> > >
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >
> > > ---
> > > Incoming mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > > ---
> > > Outgoing mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > --
> > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =
=A6
> > bidulock@openss7.org    =A6 world; the unreasonable one persists in  =
=A6
> > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =
=A6
> >                         =A6 Therefore  all  progress  depends on the =
=A6
> >                         =A6 unreasonable man. -- George Bernard Shaw =
=A6
> >
> > ---
> > Incoming mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > ---
> > Outgoing mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>=20
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>=20
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>=20
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Tue Jul 15 20:59:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA12830
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 20:59:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cadL-00048V-Ez
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 20:59:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6G0x3D3015899
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 20:59:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cadJ-00048C-5x; Tue, 15 Jul 2003 20:59:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cad1-000480-27
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 20:58:43 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA12826
	for <SIGTRAN@ietf.org>; Tue, 15 Jul 2003 20:58:38 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cacy-0007lL-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 20:58:40 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19cacn-0007l3-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 20:58:29 -0400
Received: from [172.16.1.40] (helo=ccpu00251)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19cacB-0002G4-2J; Tue, 15 Jul 2003 17:57:51 -0700
Reply-To: <umesh@ccpu.com>
From: "Umesh Vats" <umesh@ccpu.com>
To: <bidulock@openss7.org>
Cc: "SIGTRAN e-mail list" <SIGTRAN@ietf.org>
Subject: RE: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Date: Tue, 15 Jul 2003 18:03:28 -0700
Message-ID: <ACEMIJONPBBPDLOKCABHCEEFCBAA.umesh@ccpu.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0)
In-Reply-To: <20030715162528.B16479@openss7.org>
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id UAA12827
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Brian,

Many SGs(except yours) can lose the  REG-RSP in some or the other
node-level fault  scenarios.

If your implementation doesn't store RKs at ASPs, then why to penalise
other implementations who want to use RC for correlation with routing key.

RC is available in SG. So SG can send it in REG-RSP for the benefit of
ASP which don't work in the way your's do.

Thanks
Umesh


-----Original Message-----
From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
Sent: Tuesday, July 15, 2003 3:25 PM
To: Umesh Vats
Cc: SIGTRAN e-mail list
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP


Umesh,

If your SG loses the odd REG RSP have it send two (or, why not,
three) of each.  The ASP can disregard the duplicates.

Oh, you missed DATA and required Notify below.  Also, beat/beat-ack.
And what about unsolicited ASP Inactive Ack and ASP Down Ack.

Tell me, why do you use SCTP?  All that reliabilty just so that your
design will throw away messages: seems like a waste.  You should use
UDP.

--brian

On Tue, 15 Jul 2003, Umesh Vats wrote:

> Brian,
>
> REG-RSP message with unsuccesful registration status and Local RK-Id
> but no RC serves far less purpose than REG-RSP with Local-RK and RC in
> all practical SGP/ASP impplementations.
>
> I don't see any reason why a better approach for handling errors should=
n't
> be used if its possible.
>
> Umesh
>
>
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Tuesday, July 15, 2003 2:16 PM
> To: Umesh Vats
> Cc: SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>
>
> Umesh,
>
> An SG designed to meet SS7 requirements will not have any problems
> losing M3UA messages.
>
> --brian
>
> On Tue, 15 Jul 2003, Umesh Vats wrote:
>
> > If ASP looses ASPSM/ASPTM/DRKM/SSNM messages(which are consumed by M3=
UA,
> > it should be able to recover.
> >
> > For SSNM, subsequent SSNM messages or DAUD procedure take care of
message
> > loss.
> > For if ASP-ACTIVE_ACK response is lost, ASP can resend ASP-ACTIVE.
> >
> > But for loss of REG-RSP, the only recovery now possible is ASP-DOWN
> > procedure which is far worse that deregistering just one RK.
> >
> > Umesh
> >
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: Tuesday, July 15, 2003 12:51 PM
> > To: Umesh Vats
> > Cc: SIGTRAN e-mail list; Vivek Toky
> > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> >
> >
> > Umesh,
> >
> > What should ASP do if it loses data.  What should ASP do if it loses
> > ASP xxxx Ack.  What should ASP do if it loses mandatory Notify.  What
> > should ASP do....
> >
> > The SGP should not lose messagez, or if it does, send ASP Inactive Ac=
k
> > or ASP Down Ack to reset the association.
> >
> > If it loses messages without detecting them, there is nothing to do b=
ut
> > recreate SCTP in M3UA (a rather absurd proposition).
> >
> > --brian
> >
> >
> > On Tue, 15 Jul 2003, Umesh Vats wrote:
> >
> > > REG-RSP from SGP to ASP ,may be lost at SG itself before it reaches
SCTP
> > > due to local problems at SG. In this case, what should ASP do ?
> > >
> > > I don't understand why sending of RC from SG poses any problem.
> > > SG already knows about the RK and RC. It just has to return the RC.
> > >
> > > Umesh
> > >
> > >
> > > -----Original Message-----
> > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Beha=
lf
Of
> > > Brian F. G. Bidulock
> > > Sent: Monday, July 14, 2003 10:13 PM
> > > To: Umesh Vats
> > > Cc: SIGTRAN e-mail list; Vivek Toky
> > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > >
> > >
> > > Umesh,
> > >
> > > On Mon, 14 Jul 2003, Umesh Vats wrote:
> > >
> > > > Brian,
> > > >
> > > > I am sure you have sufficient background of
> highly-available/distributed
> > > > systems as your implementation is 100% fault tolerant with
> > > > nil (0%) message loss.
> > > >
> > > > In the given scenario, what if host1 receives REG-RSP from host3 =
but
> > > before
> > > > updating
> > > > the RC received in message to common data, host1 goes down.
> > >
> > > An host1 does not acknowledge the message to host3, so when host2
comes
> > > up, host3 sends the REG RSP again to host2.
> > >
> > > >
> > > > In this case, host2 doesn't know the RC and RK relation.
> > >
> > > The RC is in the REG RSP that host3 resends to host2.  The RK is in
> common
> > > data along with the Local-Routing-Key-Identifier that is also
contained
> in
> > > the REG RSP.  host2 will have no problem correlating the REG RSP wi=
th
> the
> > > REG REQ, as would host1 (on the LRK-Id).
> > >
> > > > Performing ASP-DOWN procedure will make ASP DOWN for all the
> Application
> > > > Servers for which traffic exchange might be happening.
> > >
> > > So, don't.  Have host3 send the REG RSP message not acknowledged by
> host1
> > > to host2 when host2 takes over.
> > >
> > > >
> > > > Infact, your implementation can use this idea to further decrease
> > message
> > > > loss from 0% to -1% percentage or less :)
> > >
> > > Your implementation does not need to lose REG RSP unless it has bee=
n
> > > designed to do so.  Don't do that.  Use techniques such as above so
that
> > > you do not lose messages.  Do not expect M3UA to repeat messages to
> handle
> > > the idiosynchrasies of your own internal architecture.
> > >
> > > The reason for sending "Cannot support unique routing" was because =
we
> > spent
> > > some great pains removing RK depedency from the ASP.  The ASP can b=
e
> > > designed
> > > completely RK neutral and treat the RK as opaque.  Only the AS need
know
> > the
> > > internals of the RK.  One side effect of this is that mutliple AS a=
t
the
> > ASP
> > > may attempt to activate for the same RK.  The ASP cannot detect thi=
s
an
> > > remain
> > > RK neutral.  So, when a second AS attemtps to register the same RK
(and
> > > obtain
> > > its own unique RC), it is the SGP that refuses the registration wit=
h
> > "Cannot
> > > support unique routing", instead of requiring the ASP to maintain R=
Ks
> and
> > > police AS/RC/RK uniqueness.  The RFC places the obligation at the S=
GP
> > (M3UA,
> > > NIF or MTP) to enforce the 1:1:1 relationship of AS:RC:RK, because =
it
is
> > the
> > > one that requires this relationship so that there is "unique routin=
g"
in
> > the
> > > SGP routing table.
> > >
> > > This is rather important to interoperability, because there is no
> > mechanism
> > > in the M3UA ASP specification to enforce this relationship, only th=
e
> > > mechanism
> > > at the SGP.  Altering the behviour of the SGP would break this
> > > interoperability mechanism.
> > >
> > > There are probably three or four other mechanism within your
> architecture
> > > that will solve the problem for you without changing M3UA and break=
ing
> > this
> > > more important interoperability behavior.  So I cannot agree with t=
he
> > > change.
> > >
> > > --brian
> > >
> > > --
> > > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to th=
e =A6
> > > bidulock@openss7.org    =A6 world; the unreasonable one persists in=
  =A6
> > > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself=
. =A6
> > >                         =A6 Therefore  all  progress  depends on th=
e =A6
> > >                         =A6 unreasonable man. -- George Bernard Sha=
w =A6
> > >
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >
> > > ---
> > > Incoming mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > > ---
> > > Outgoing mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > --
> > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =
=A6
> > bidulock@openss7.org    =A6 world; the unreasonable one persists in  =
=A6
> > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =
=A6
> >                         =A6 Therefore  all  progress  depends on the =
=A6
> >                         =A6 unreasonable man. -- George Bernard Shaw =
=A6
> >
> > ---
> > Incoming mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > ---
> > Outgoing mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

--
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003


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



From exim@www1.ietf.org  Tue Jul 15 21:18:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA13165
	for <sigtran-archive@odin.ietf.org>; Tue, 15 Jul 2003 21:18:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cavj-00055P-32
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 21:18:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6G1I3Vc019538
	for sigtran-archive@odin.ietf.org; Tue, 15 Jul 2003 21:18:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cavh-00054t-Ok; Tue, 15 Jul 2003 21:18:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cav9-00054V-JN
	for sigtran@optimus.ietf.org; Tue, 15 Jul 2003 21:17:27 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA13150
	for <SIGTRAN@ietf.org>; Tue, 15 Jul 2003 21:17:22 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cav7-00006y-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 21:17:25 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19cauw-00006m-00
	for SIGTRAN@ietf.org; Tue, 15 Jul 2003 21:17:14 -0400
Received: from [172.16.1.203] (helo=CCPU00186)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19caud-0003BZ-5G; Tue, 15 Jul 2003 18:16:55 -0700
From: "VIvek Toky" <vivektoky@ccpu.com>
To: <bidulock@openss7.org>
Cc: "Umesh Vats" <umesh@ccpu.com>, "SIGTRAN e-mail list" <SIGTRAN@ietf.org>
Subject: RE: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Date: Tue, 15 Jul 2003 18:21:36 -0700
Message-ID: <HKEBKFBNPIDALDOKLGKDIEHOCDAA.vivektoky@ccpu.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.6604 (9.0.2911.0)
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Importance: Normal
In-Reply-To: <20030715162123.A16479@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id VAA13151
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Hi Brian,
  Can you give me pointer to a section in RFC or IGv4 where its mentioned
"ASP must use only RC to route messages locally to AS".. If your ASP
implementation does not maintain routing keys, why do you want SGPs to do
the work for you.

 Even if an implementaion routes messages locally based on RC, it can sti=
ll
validate RK while sending the REG Request message out, I don't see a prob=
lem
with that. If you don't want to modify your datastructures, that doesn't
mean M3UA protocol should change for you.

-Vivek

-----Original Message-----
From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf Of
Brian F. G. Bidulock
Sent: Tuesday, July 15, 2003 3:21 PM
To: VIvek Toky
Cc: Umesh Vats; SIGTRAN e-mail list
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP


VIvek,

All ASPs need to route messages locally to AS is the RC.  They need
not be aware of the RK.

--brian


On Tue, 15 Jul 2003, VIvek Toky wrote:

> Brian,
>
>   You said
>    "The reason for sending "Cannot support unique routing" was because =
we
> spent
> some great pains removing RK depedency from the ASP.  The ASP can be
> designed
> completely RK neutral and treat the RK as opaque.  Only the AS need kno=
w
the
> internals of the RK."
>
> Why do you want ASPs to be totally unawar of whats inside the RK, If SG=
Ps
> have to do the routing based on RKs then ASPs also have to route the
> incoming messages to the destined AS (locally). If different ASs try to
> register same routing key then it can be checked at ASP itself without
even
> sending the REG Request message out.
>
> Can you explain more about the interoperability scenarios you mentioned=
..
>
> -Vivek.
>
> -----Original Message-----
> From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf O=
f
> Brian F. G. Bidulock
> Sent: Tuesday, July 15, 2003 2:16 PM
> To: Umesh Vats
> Cc: SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>
>
> Umesh,
>
> An SG designed to meet SS7 requirements will not have any problems
> losing M3UA messages.
>
> --brian
>
> On Tue, 15 Jul 2003, Umesh Vats wrote:
>
> > If ASP looses ASPSM/ASPTM/DRKM/SSNM messages(which are consumed by M3=
UA,
> > it should be able to recover.
> >
> > For SSNM, subsequent SSNM messages or DAUD procedure take care of
message
> > loss.
> > For if ASP-ACTIVE_ACK response is lost, ASP can resend ASP-ACTIVE.
> >
> > But for loss of REG-RSP, the only recovery now possible is ASP-DOWN
> > procedure which is far worse that deregistering just one RK.
> >
> > Umesh
> >
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: Tuesday, July 15, 2003 12:51 PM
> > To: Umesh Vats
> > Cc: SIGTRAN e-mail list; Vivek Toky
> > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> >
> >
> > Umesh,
> >
> > What should ASP do if it loses data.  What should ASP do if it loses
> > ASP xxxx Ack.  What should ASP do if it loses mandatory Notify.  What
> > should ASP do....
> >
> > The SGP should not lose messagez, or if it does, send ASP Inactive Ac=
k
> > or ASP Down Ack to reset the association.
> >
> > If it loses messages without detecting them, there is nothing to do b=
ut
> > recreate SCTP in M3UA (a rather absurd proposition).
> >
> > --brian
> >
> >
> > On Tue, 15 Jul 2003, Umesh Vats wrote:
> >
> > > REG-RSP from SGP to ASP ,may be lost at SG itself before it reaches
SCTP
> > > due to local problems at SG. In this case, what should ASP do ?
> > >
> > > I don't understand why sending of RC from SG poses any problem.
> > > SG already knows about the RK and RC. It just has to return the RC.
> > >
> > > Umesh
> > >
> > >
> > > -----Original Message-----
> > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Beha=
lf
Of
> > > Brian F. G. Bidulock
> > > Sent: Monday, July 14, 2003 10:13 PM
> > > To: Umesh Vats
> > > Cc: SIGTRAN e-mail list; Vivek Toky
> > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > >
> > >
> > > Umesh,
> > >
> > > On Mon, 14 Jul 2003, Umesh Vats wrote:
> > >
> > > > Brian,
> > > >
> > > > I am sure you have sufficient background of
> highly-available/distributed
> > > > systems as your implementation is 100% fault tolerant with
> > > > nil (0%) message loss.
> > > >
> > > > In the given scenario, what if host1 receives REG-RSP from host3 =
but
> > > before
> > > > updating
> > > > the RC received in message to common data, host1 goes down.
> > >
> > > An host1 does not acknowledge the message to host3, so when host2
comes
> > > up, host3 sends the REG RSP again to host2.
> > >
> > > >
> > > > In this case, host2 doesn't know the RC and RK relation.
> > >
> > > The RC is in the REG RSP that host3 resends to host2.  The RK is in
> common
> > > data along with the Local-Routing-Key-Identifier that is also
contained
> in
> > > the REG RSP.  host2 will have no problem correlating the REG RSP wi=
th
> the
> > > REG REQ, as would host1 (on the LRK-Id).
> > >
> > > > Performing ASP-DOWN procedure will make ASP DOWN for all the
> Application
> > > > Servers for which traffic exchange might be happening.
> > >
> > > So, don't.  Have host3 send the REG RSP message not acknowledged by
> host1
> > > to host2 when host2 takes over.
> > >
> > > >
> > > > Infact, your implementation can use this idea to further decrease
> > message
> > > > loss from 0% to -1% percentage or less :)
> > >
> > > Your implementation does not need to lose REG RSP unless it has bee=
n
> > > designed to do so.  Don't do that.  Use techniques such as above so
that
> > > you do not lose messages.  Do not expect M3UA to repeat messages to
> handle
> > > the idiosynchrasies of your own internal architecture.
> > >
> > > The reason for sending "Cannot support unique routing" was because =
we
> > spent
> > > some great pains removing RK depedency from the ASP.  The ASP can b=
e
> > > designed
> > > completely RK neutral and treat the RK as opaque.  Only the AS need
know
> > the
> > > internals of the RK.  One side effect of this is that mutliple AS a=
t
the
> > ASP
> > > may attempt to activate for the same RK.  The ASP cannot detect thi=
s
an
> > > remain
> > > RK neutral.  So, when a second AS attemtps to register the same RK
(and
> > > obtain
> > > its own unique RC), it is the SGP that refuses the registration wit=
h
> > "Cannot
> > > support unique routing", instead of requiring the ASP to maintain R=
Ks
> and
> > > police AS/RC/RK uniqueness.  The RFC places the obligation at the S=
GP
> > (M3UA,
> > > NIF or MTP) to enforce the 1:1:1 relationship of AS:RC:RK, because =
it
is
> > the
> > > one that requires this relationship so that there is "unique routin=
g"
in
> > the
> > > SGP routing table.
> > >
> > > This is rather important to interoperability, because there is no
> > mechanism
> > > in the M3UA ASP specification to enforce this relationship, only th=
e
> > > mechanism
> > > at the SGP.  Altering the behviour of the SGP would break this
> > > interoperability mechanism.
> > >
> > > There are probably three or four other mechanism within your
> architecture
> > > that will solve the problem for you without changing M3UA and break=
ing
> > this
> > > more important interoperability behavior.  So I cannot agree with t=
he
> > > change.
> > >
> > > --brian
> > >
> > > --
> > > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to th=
e =A6
> > > bidulock@openss7.org    =A6 world; the unreasonable one persists in=
  =A6
> > > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself=
. =A6
> > >                         =A6 Therefore  all  progress  depends on th=
e =A6
> > >                         =A6 unreasonable man. -- George Bernard Sha=
w =A6
> > >
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > >
> > > ---
> > > Incoming mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > > ---
> > > Outgoing mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > --
> > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =
=A6
> > bidulock@openss7.org    =A6 world; the unreasonable one persists in  =
=A6
> > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =
=A6
> >                         =A6 Therefore  all  progress  depends on the =
=A6
> >                         =A6 unreasonable man. -- George Bernard Shaw =
=A6
> >
> > ---
> > Incoming mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > ---
> > Outgoing mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

_______________________________________________
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 exim@www1.ietf.org  Wed Jul 16 01:27:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA18398
	for <sigtran-archive@odin.ietf.org>; Wed, 16 Jul 2003 01:27:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ceoh-0006H5-Fh
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 01:27:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6G5R3qw024108
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 01:27:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ceof-0006Fj-Gs; Wed, 16 Jul 2003 01:27:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ceo5-0006El-P3
	for sigtran@optimus.ietf.org; Wed, 16 Jul 2003 01:26:26 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA18304
	for <SIGTRAN@ietf.org>; Wed, 16 Jul 2003 01:26:21 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ceo2-0002lu-00
	for SIGTRAN@ietf.org; Wed, 16 Jul 2003 01:26:22 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[WRQiknWxZqmlma2DwxkN/zCzGMavShl9])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cenm-0002lV-00
	for SIGTRAN@ietf.org; Wed, 16 Jul 2003 01:26:06 -0400
Received: from ns.pigworks.openss7.net (IDENT:NG4duNqKPnDL7OE5jL6wivuMciwKa1O+@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6G5PeP18687;
	Tue, 15 Jul 2003 23:25:40 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6G5PeY21978;
	Tue, 15 Jul 2003 23:25:40 -0600
Date: Tue, 15 Jul 2003 23:25:39 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: VIvek Toky <vivektoky@ccpu.com>
Cc: Umesh Vats <umesh@ccpu.com>, SIGTRAN e-mail list <SIGTRAN@ietf.org>
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Message-ID: <20030715232539.A17550@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: VIvek Toky <vivektoky@ccpu.com>,
	Umesh Vats <umesh@ccpu.com>, SIGTRAN e-mail list <SIGTRAN@ietf.org>
References: <20030715162123.A16479@openss7.org> <HKEBKFBNPIDALDOKLGKDIEHOCDAA.vivektoky@ccpu.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <HKEBKFBNPIDALDOKLGKDIEHOCDAA.vivektoky@ccpu.com>; from vivektoky@ccpu.com on Tue, Jul 15, 2003 at 06:21:36PM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6G5PeP18687
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

VIvek,

Validate it at the ASP all you want.  In the RFC it is the SG's responsib=
ility
to guarantee unique routing, not the ASP's.

--brian

On Tue, 15 Jul 2003, VIvek Toky wrote:

> Hi Brian,
>   Can you give me pointer to a section in RFC or IGv4 where its mention=
ed
> "ASP must use only RC to route messages locally to AS".. If your ASP
> implementation does not maintain routing keys, why do you want SGPs to =
do
> the work for you.
>=20
>  Even if an implementaion routes messages locally based on RC, it can s=
till
> validate RK while sending the REG Request message out, I don't see a pr=
oblem
> with that. If you don't want to modify your datastructures, that doesn'=
t
> mean M3UA protocol should change for you.
>=20
> -Vivek
>=20
> -----Original Message-----
> From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf O=
f
> Brian F. G. Bidulock
> Sent: Tuesday, July 15, 2003 3:21 PM
> To: VIvek Toky
> Cc: Umesh Vats; SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>=20
>=20
> VIvek,
>=20
> All ASPs need to route messages locally to AS is the RC.  They need
> not be aware of the RK.
>=20
> --brian
>=20
>=20
> On Tue, 15 Jul 2003, VIvek Toky wrote:
>=20
> > Brian,
> >
> >   You said
> >    "The reason for sending "Cannot support unique routing" was becaus=
e we
> > spent
> > some great pains removing RK depedency from the ASP.  The ASP can be
> > designed
> > completely RK neutral and treat the RK as opaque.  Only the AS need k=
now
> the
> > internals of the RK."
> >
> > Why do you want ASPs to be totally unawar of whats inside the RK, If =
SGPs
> > have to do the routing based on RKs then ASPs also have to route the
> > incoming messages to the destined AS (locally). If different ASs try =
to
> > register same routing key then it can be checked at ASP itself withou=
t
> even
> > sending the REG Request message out.
> >
> > Can you explain more about the interoperability scenarios you mention=
ed..
> >
> > -Vivek.
> >
> > -----Original Message-----
> > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf=
 Of
> > Brian F. G. Bidulock
> > Sent: Tuesday, July 15, 2003 2:16 PM
> > To: Umesh Vats
> > Cc: SIGTRAN e-mail list
> > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> >
> >
> > Umesh,
> >
> > An SG designed to meet SS7 requirements will not have any problems
> > losing M3UA messages.
> >
> > --brian
> >
> > On Tue, 15 Jul 2003, Umesh Vats wrote:
> >
> > > If ASP looses ASPSM/ASPTM/DRKM/SSNM messages(which are consumed by =
M3UA,
> > > it should be able to recover.
> > >
> > > For SSNM, subsequent SSNM messages or DAUD procedure take care of
> message
> > > loss.
> > > For if ASP-ACTIVE_ACK response is lost, ASP can resend ASP-ACTIVE.
> > >
> > > But for loss of REG-RSP, the only recovery now possible is ASP-DOWN
> > > procedure which is far worse that deregistering just one RK.
> > >
> > > Umesh
> > >
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: Tuesday, July 15, 2003 12:51 PM
> > > To: Umesh Vats
> > > Cc: SIGTRAN e-mail list; Vivek Toky
> > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > >
> > >
> > > Umesh,
> > >
> > > What should ASP do if it loses data.  What should ASP do if it lose=
s
> > > ASP xxxx Ack.  What should ASP do if it loses mandatory Notify.  Wh=
at
> > > should ASP do....
> > >
> > > The SGP should not lose messagez, or if it does, send ASP Inactive =
Ack
> > > or ASP Down Ack to reset the association.
> > >
> > > If it loses messages without detecting them, there is nothing to do=
 but
> > > recreate SCTP in M3UA (a rather absurd proposition).
> > >
> > > --brian
> > >
> > >
> > > On Tue, 15 Jul 2003, Umesh Vats wrote:
> > >
> > > > REG-RSP from SGP to ASP ,may be lost at SG itself before it reach=
es
> SCTP
> > > > due to local problems at SG. In this case, what should ASP do ?
> > > >
> > > > I don't understand why sending of RC from SG poses any problem.
> > > > SG already knows about the RK and RC. It just has to return the R=
C.
> > > >
> > > > Umesh
> > > >
> > > >
> > > > -----Original Message-----
> > > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Be=
half
> Of
> > > > Brian F. G. Bidulock
> > > > Sent: Monday, July 14, 2003 10:13 PM
> > > > To: Umesh Vats
> > > > Cc: SIGTRAN e-mail list; Vivek Toky
> > > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > > >
> > > >
> > > > Umesh,
> > > >
> > > > On Mon, 14 Jul 2003, Umesh Vats wrote:
> > > >
> > > > > Brian,
> > > > >
> > > > > I am sure you have sufficient background of
> > highly-available/distributed
> > > > > systems as your implementation is 100% fault tolerant with
> > > > > nil (0%) message loss.
> > > > >
> > > > > In the given scenario, what if host1 receives REG-RSP from host=
3 but
> > > > before
> > > > > updating
> > > > > the RC received in message to common data, host1 goes down.
> > > >
> > > > An host1 does not acknowledge the message to host3, so when host2
> comes
> > > > up, host3 sends the REG RSP again to host2.
> > > >
> > > > >
> > > > > In this case, host2 doesn't know the RC and RK relation.
> > > >
> > > > The RC is in the REG RSP that host3 resends to host2.  The RK is =
in
> > common
> > > > data along with the Local-Routing-Key-Identifier that is also
> contained
> > in
> > > > the REG RSP.  host2 will have no problem correlating the REG RSP =
with
> > the
> > > > REG REQ, as would host1 (on the LRK-Id).
> > > >
> > > > > Performing ASP-DOWN procedure will make ASP DOWN for all the
> > Application
> > > > > Servers for which traffic exchange might be happening.
> > > >
> > > > So, don't.  Have host3 send the REG RSP message not acknowledged =
by
> > host1
> > > > to host2 when host2 takes over.
> > > >
> > > > >
> > > > > Infact, your implementation can use this idea to further decrea=
se
> > > message
> > > > > loss from 0% to -1% percentage or less :)
> > > >
> > > > Your implementation does not need to lose REG RSP unless it has b=
een
> > > > designed to do so.  Don't do that.  Use techniques such as above =
so
> that
> > > > you do not lose messages.  Do not expect M3UA to repeat messages =
to
> > handle
> > > > the idiosynchrasies of your own internal architecture.
> > > >
> > > > The reason for sending "Cannot support unique routing" was becaus=
e we
> > > spent
> > > > some great pains removing RK depedency from the ASP.  The ASP can=
 be
> > > > designed
> > > > completely RK neutral and treat the RK as opaque.  Only the AS ne=
ed
> know
> > > the
> > > > internals of the RK.  One side effect of this is that mutliple AS=
 at
> the
> > > ASP
> > > > may attempt to activate for the same RK.  The ASP cannot detect t=
his
> an
> > > > remain
> > > > RK neutral.  So, when a second AS attemtps to register the same R=
K
> (and
> > > > obtain
> > > > its own unique RC), it is the SGP that refuses the registration w=
ith
> > > "Cannot
> > > > support unique routing", instead of requiring the ASP to maintain=
 RKs
> > and
> > > > police AS/RC/RK uniqueness.  The RFC places the obligation at the=
 SGP
> > > (M3UA,
> > > > NIF or MTP) to enforce the 1:1:1 relationship of AS:RC:RK, becaus=
e it
> is
> > > the
> > > > one that requires this relationship so that there is "unique rout=
ing"
> in
> > > the
> > > > SGP routing table.
> > > >
> > > > This is rather important to interoperability, because there is no
> > > mechanism
> > > > in the M3UA ASP specification to enforce this relationship, only =
the
> > > > mechanism
> > > > at the SGP.  Altering the behviour of the SGP would break this
> > > > interoperability mechanism.
> > > >
> > > > There are probably three or four other mechanism within your
> > architecture
> > > > that will solve the problem for you without changing M3UA and bre=
aking
> > > this
> > > > more important interoperability behavior.  So I cannot agree with=
 the
> > > > change.
> > > >
> > > > --brian
> > > >
> > > > --
> > > > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to =
the =A6
> > > > bidulock@openss7.org    =A6 world; the unreasonable one persists =
in  =A6
> > > > http://www.openss7.org/ =A6 trying  to adapt the  world  to himse=
lf. =A6
> > > >                         =A6 Therefore  all  progress  depends on =
the =A6
> > > >                         =A6 unreasonable man. -- George Bernard S=
haw =A6
> > > >
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > >
> > > > ---
> > > > Incoming mail is certified Virus Free.
> > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > > >
> > > > ---
> > > > Outgoing mail is certified Virus Free.
> > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > > --
> > > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to th=
e =A6
> > > bidulock@openss7.org    =A6 world; the unreasonable one persists in=
  =A6
> > > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself=
. =A6
> > >                         =A6 Therefore  all  progress  depends on th=
e =A6
> > >                         =A6 unreasonable man. -- George Bernard Sha=
w =A6
> > >
> > > ---
> > > Incoming mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > > ---
> > > Outgoing mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > --
> > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =
=A6
> > bidulock@openss7.org    =A6 world; the unreasonable one persists in  =
=A6
> > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =
=A6
> >                         =A6 Therefore  all  progress  depends on the =
=A6
> >                         =A6 unreasonable man. -- George Bernard Shaw =
=A6
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
>=20
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Wed Jul 16 01:28:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA18447
	for <sigtran-archive@odin.ietf.org>; Wed, 16 Jul 2003 01:28:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cepd-0006M4-Na
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 01:28:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6G5S1Zn024423
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 01:28:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cepd-0006Lp-Jv; Wed, 16 Jul 2003 01:28:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ceon-0006I9-1y
	for sigtran@optimus.ietf.org; Wed, 16 Jul 2003 01:27:09 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA18377
	for <SIGTRAN@ietf.org>; Wed, 16 Jul 2003 01:27:05 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ceoj-0002n3-00
	for SIGTRAN@ietf.org; Wed, 16 Jul 2003 01:27:05 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[7PWSQSkSm+6QkGbfIcIfNqgqQe+ab8p8])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ceoT-0002mm-00
	for SIGTRAN@ietf.org; Wed, 16 Jul 2003 01:26:49 -0400
Received: from ns.pigworks.openss7.net (IDENT:A0yiLGtkAf0L6uGC3e/sS8OpTrnp7MNv@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6G5QeP18699;
	Tue, 15 Jul 2003 23:26:40 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6G5QeK21985;
	Tue, 15 Jul 2003 23:26:40 -0600
Date: Tue, 15 Jul 2003 23:26:39 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Umesh Vats <umesh@ccpu.com>
Cc: SIGTRAN e-mail list <SIGTRAN@ietf.org>
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Message-ID: <20030715232639.B17550@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Umesh Vats <umesh@ccpu.com>,
	SIGTRAN e-mail list <SIGTRAN@ietf.org>
References: <20030715162528.B16479@openss7.org> <ACEMIJONPBBPDLOKCABHCEEFCBAA.umesh@ccpu.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <ACEMIJONPBBPDLOKCABHCEEFCBAA.umesh@ccpu.com>; from umesh@ccpu.com on Tue, Jul 15, 2003 at 06:03:28PM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6G5QeP18699
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Umesh,

On Tue, 15 Jul 2003, Umesh Vats wrote:

> Brian,
>=20
> Many SGs(except yours) can lose the  REG-RSP in some or the other
> node-level fault  scenarios.

Who's?  Yours?  I didn't know that you spoke for both me and everybody
else.

--brian

>=20
> If your implementation doesn't store RKs at ASPs, then why to penalise
> other implementations who want to use RC for correlation with routing k=
ey.
>=20
> RC is available in SG. So SG can send it in REG-RSP for the benefit of
> ASP which don't work in the way your's do.
>=20
> Thanks
> Umesh
>=20
>=20
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Tuesday, July 15, 2003 3:25 PM
> To: Umesh Vats
> Cc: SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>=20
>=20
> Umesh,
>=20
> If your SG loses the odd REG RSP have it send two (or, why not,
> three) of each.  The ASP can disregard the duplicates.
>=20
> Oh, you missed DATA and required Notify below.  Also, beat/beat-ack.
> And what about unsolicited ASP Inactive Ack and ASP Down Ack.
>=20
> Tell me, why do you use SCTP?  All that reliabilty just so that your
> design will throw away messages: seems like a waste.  You should use
> UDP.
>=20
> --brian
>=20
> On Tue, 15 Jul 2003, Umesh Vats wrote:
>=20
> > Brian,
> >
> > REG-RSP message with unsuccesful registration status and Local RK-Id
> > but no RC serves far less purpose than REG-RSP with Local-RK and RC i=
n
> > all practical SGP/ASP impplementations.
> >
> > I don't see any reason why a better approach for handling errors shou=
ldn't
> > be used if its possible.
> >
> > Umesh
> >
> >
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: Tuesday, July 15, 2003 2:16 PM
> > To: Umesh Vats
> > Cc: SIGTRAN e-mail list
> > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> >
> >
> > Umesh,
> >
> > An SG designed to meet SS7 requirements will not have any problems
> > losing M3UA messages.
> >
> > --brian
> >
> > On Tue, 15 Jul 2003, Umesh Vats wrote:
> >
> > > If ASP looses ASPSM/ASPTM/DRKM/SSNM messages(which are consumed by =
M3UA,
> > > it should be able to recover.
> > >
> > > For SSNM, subsequent SSNM messages or DAUD procedure take care of
> message
> > > loss.
> > > For if ASP-ACTIVE_ACK response is lost, ASP can resend ASP-ACTIVE.
> > >
> > > But for loss of REG-RSP, the only recovery now possible is ASP-DOWN
> > > procedure which is far worse that deregistering just one RK.
> > >
> > > Umesh
> > >
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: Tuesday, July 15, 2003 12:51 PM
> > > To: Umesh Vats
> > > Cc: SIGTRAN e-mail list; Vivek Toky
> > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > >
> > >
> > > Umesh,
> > >
> > > What should ASP do if it loses data.  What should ASP do if it lose=
s
> > > ASP xxxx Ack.  What should ASP do if it loses mandatory Notify.  Wh=
at
> > > should ASP do....
> > >
> > > The SGP should not lose messagez, or if it does, send ASP Inactive =
Ack
> > > or ASP Down Ack to reset the association.
> > >
> > > If it loses messages without detecting them, there is nothing to do=
 but
> > > recreate SCTP in M3UA (a rather absurd proposition).
> > >
> > > --brian
> > >
> > >
> > > On Tue, 15 Jul 2003, Umesh Vats wrote:
> > >
> > > > REG-RSP from SGP to ASP ,may be lost at SG itself before it reach=
es
> SCTP
> > > > due to local problems at SG. In this case, what should ASP do ?
> > > >
> > > > I don't understand why sending of RC from SG poses any problem.
> > > > SG already knows about the RK and RC. It just has to return the R=
C.
> > > >
> > > > Umesh
> > > >
> > > >
> > > > -----Original Message-----
> > > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Be=
half
> Of
> > > > Brian F. G. Bidulock
> > > > Sent: Monday, July 14, 2003 10:13 PM
> > > > To: Umesh Vats
> > > > Cc: SIGTRAN e-mail list; Vivek Toky
> > > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > > >
> > > >
> > > > Umesh,
> > > >
> > > > On Mon, 14 Jul 2003, Umesh Vats wrote:
> > > >
> > > > > Brian,
> > > > >
> > > > > I am sure you have sufficient background of
> > highly-available/distributed
> > > > > systems as your implementation is 100% fault tolerant with
> > > > > nil (0%) message loss.
> > > > >
> > > > > In the given scenario, what if host1 receives REG-RSP from host=
3 but
> > > > before
> > > > > updating
> > > > > the RC received in message to common data, host1 goes down.
> > > >
> > > > An host1 does not acknowledge the message to host3, so when host2
> comes
> > > > up, host3 sends the REG RSP again to host2.
> > > >
> > > > >
> > > > > In this case, host2 doesn't know the RC and RK relation.
> > > >
> > > > The RC is in the REG RSP that host3 resends to host2.  The RK is =
in
> > common
> > > > data along with the Local-Routing-Key-Identifier that is also
> contained
> > in
> > > > the REG RSP.  host2 will have no problem correlating the REG RSP =
with
> > the
> > > > REG REQ, as would host1 (on the LRK-Id).
> > > >
> > > > > Performing ASP-DOWN procedure will make ASP DOWN for all the
> > Application
> > > > > Servers for which traffic exchange might be happening.
> > > >
> > > > So, don't.  Have host3 send the REG RSP message not acknowledged =
by
> > host1
> > > > to host2 when host2 takes over.
> > > >
> > > > >
> > > > > Infact, your implementation can use this idea to further decrea=
se
> > > message
> > > > > loss from 0% to -1% percentage or less :)
> > > >
> > > > Your implementation does not need to lose REG RSP unless it has b=
een
> > > > designed to do so.  Don't do that.  Use techniques such as above =
so
> that
> > > > you do not lose messages.  Do not expect M3UA to repeat messages =
to
> > handle
> > > > the idiosynchrasies of your own internal architecture.
> > > >
> > > > The reason for sending "Cannot support unique routing" was becaus=
e we
> > > spent
> > > > some great pains removing RK depedency from the ASP.  The ASP can=
 be
> > > > designed
> > > > completely RK neutral and treat the RK as opaque.  Only the AS ne=
ed
> know
> > > the
> > > > internals of the RK.  One side effect of this is that mutliple AS=
 at
> the
> > > ASP
> > > > may attempt to activate for the same RK.  The ASP cannot detect t=
his
> an
> > > > remain
> > > > RK neutral.  So, when a second AS attemtps to register the same R=
K
> (and
> > > > obtain
> > > > its own unique RC), it is the SGP that refuses the registration w=
ith
> > > "Cannot
> > > > support unique routing", instead of requiring the ASP to maintain=
 RKs
> > and
> > > > police AS/RC/RK uniqueness.  The RFC places the obligation at the=
 SGP
> > > (M3UA,
> > > > NIF or MTP) to enforce the 1:1:1 relationship of AS:RC:RK, becaus=
e it
> is
> > > the
> > > > one that requires this relationship so that there is "unique rout=
ing"
> in
> > > the
> > > > SGP routing table.
> > > >
> > > > This is rather important to interoperability, because there is no
> > > mechanism
> > > > in the M3UA ASP specification to enforce this relationship, only =
the
> > > > mechanism
> > > > at the SGP.  Altering the behviour of the SGP would break this
> > > > interoperability mechanism.
> > > >
> > > > There are probably three or four other mechanism within your
> > architecture
> > > > that will solve the problem for you without changing M3UA and bre=
aking
> > > this
> > > > more important interoperability behavior.  So I cannot agree with=
 the
> > > > change.
> > > >
> > > > --brian
> > > >
> > > > --
> > > > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to =
the =A6
> > > > bidulock@openss7.org    =A6 world; the unreasonable one persists =
in  =A6
> > > > http://www.openss7.org/ =A6 trying  to adapt the  world  to himse=
lf. =A6
> > > >                         =A6 Therefore  all  progress  depends on =
the =A6
> > > >                         =A6 unreasonable man. -- George Bernard S=
haw =A6
> > > >
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > >
> > > > ---
> > > > Incoming mail is certified Virus Free.
> > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > > >
> > > > ---
> > > > Outgoing mail is certified Virus Free.
> > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > > --
> > > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to th=
e =A6
> > > bidulock@openss7.org    =A6 world; the unreasonable one persists in=
  =A6
> > > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself=
. =A6
> > >                         =A6 Therefore  all  progress  depends on th=
e =A6
> > >                         =A6 unreasonable man. -- George Bernard Sha=
w =A6
> > >
> > > ---
> > > Incoming mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > > ---
> > > Outgoing mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > --
> > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =
=A6
> > bidulock@openss7.org    =A6 world; the unreasonable one persists in  =
=A6
> > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =
=A6
> >                         =A6 Therefore  all  progress  depends on the =
=A6
> >                         =A6 unreasonable man. -- George Bernard Shaw =
=A6
> >
> > ---
> > Incoming mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > ---
> > Outgoing mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>=20
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>=20
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>=20
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Wed Jul 16 01:38:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA18780
	for <sigtran-archive@odin.ietf.org>; Wed, 16 Jul 2003 01:38:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cezK-00071Z-M3
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 01:38:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6G5c29Q026980
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 01:38:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cezJ-00070v-Jq; Wed, 16 Jul 2003 01:38:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ceyn-0006vr-EK
	for sigtran@optimus.ietf.org; Wed, 16 Jul 2003 01:37:29 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA18755
	for <sigtran@ietf.org>; Wed, 16 Jul 2003 01:37:25 -0400 (EDT)
From: agurmukhani@hss.hns.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ceyk-0002vH-00
	for sigtran@ietf.org; Wed, 16 Jul 2003 01:37:26 -0400
Received: from hindon.hss.co.in ([202.54.26.202])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cey6-0002ts-00
	for sigtran@ietf.org; Wed, 16 Jul 2003 01:37:14 -0400
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h6G5Vgs27002;
	Wed, 16 Jul 2003 11:01:42 +0530 (IST)
Received: from dlfmail1.hss.hns.com (localhost [127.0.0.1])
	by ultra.hss.co.in (8.10.0/8.10.0) with ESMTP id h6G5ZOQ19485;
	Wed, 16 Jul 2003 11:05:25 +0530 (IST)
To: javier.pastor-balbas@ece.ericsson.se
Cc: sigtran@ietf.org
Date: Wed, 16 Jul 2003 10:58:40 +0530
Message-ID: <OF951580A8.4346E664-ON65256D64.0040C8D3@hss.hns.com>
X-MIMETrack: Serialize by Router on Dlfmail1/HSS(Release 6.0|September 26, 2002) at 07/16/2003
 10:57:36 AM
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Subject: [Sigtran] M3UA MIB Draft 05
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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 ...



In the M3UA MIB draft 05 , in the SGP/IPSP Assciation table (
m3uaSgpIpspAssocTable), the attributes included contain
m3uaSgpIpspAssocReqOutStreams. but the same is replaced by Minout and
Maxin streams  later in the draft.
IMO, this needs to be aligned at two  places.

Also   Requested Outstreams was present in draft04  as a part of the
m3uaSpEpTable  , is there any specific reason to exclude the same.



Thanks and Regards
Anjali
www.hssworld.com
Hughes Software Systems


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



From exim@www1.ietf.org  Wed Jul 16 03:13:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA17966
	for <sigtran-archive@odin.ietf.org>; Wed, 16 Jul 2003 03:13:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cgTG-0005eB-Mg
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 03:13:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6G7D2S0021707
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 03:13:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cgTE-0005cW-4P; Wed, 16 Jul 2003 03:13:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cgSt-0005bR-Su
	for sigtran@optimus.ietf.org; Wed, 16 Jul 2003 03:12:39 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA17941
	for <sigtran@ietf.org>; Wed, 16 Jul 2003 03:12:36 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cgSr-0003rq-00
	for sigtran@ietf.org; Wed, 16 Jul 2003 03:12:37 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[gbc7Gw33pbkE90n4vYqPgwvwlAsMUyWu])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cgSa-0003rb-00
	for sigtran@ietf.org; Wed, 16 Jul 2003 03:12:21 -0400
Received: from ns.pigworks.openss7.net (IDENT:4BkBWpj9twd6ksXaJwAJxlCrh6vFp8M1@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6G7BeP20477;
	Wed, 16 Jul 2003 01:11:40 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6G7Bep23523;
	Wed, 16 Jul 2003 01:11:40 -0600
Date: Wed, 16 Jul 2003 01:11:40 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: kakaul@hss.hns.com
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
Message-ID: <20030716011140.A23441@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: kakaul@hss.hns.com, sigtran@ietf.org
References: <OF62FF9733.E9CC03AC-ON65256D64.004838B3@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: <OF62FF9733.E9CC03AC-ON65256D64.004838B3@hss.hns.com>; from kakaul@hss.hns.com on Tue, Jul 15, 2003 at 06:59:37PM +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>

kakaul,

An SS7 SEP at the position of the ASP, where the SGs are STPs would
not maintain congestion state on a per-route (per-SG) basis.  So
what makes you think that the ASP needs to?

Also, SCON() in this configuration corresponds to TFC (not MTP-STATUS)
and therefore cannot signal abatement.  (TFC does not signal abatement,
the lack of a TFC does.)  In the same ways, the ASP can use the lack
of a SCON to indicate abatement.

So, a) SCON(1) and SCON(3) in your example are congestion signals
(neither is an abatement signal); b) there is no need to maintain congestion
status on a per-SG basis at the ASP ('cause the SS7 SEP doesn't).

This might lead back to the discustion why ASPs don't send RCT, but the
simple answer is that each SG as STP can send RCT on all the ASP's behalf
as it would for an "Alias" point code (ANSI T1.111/2000).

Because this is getting into a narrower and narrower corner of the
specification, I think we are safe to strike the mention of the timer
from the IG.

--brian

kakaul@hss.hns.com wrote:                                                           (Tue, 15 Jul 2003 18:59:37)
> 
> 
> 
> 
> Brian
>    some comments inline
> 
> 
> 
> 
> "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/15/2003
> 12:16:38 PM
> 
> Please respond to bidulock@openss7.org
> 
> Sent by:    sigtran-admin@ietf.org
> 
> 
> To:    Kamesh Kaul/HSS@HSS
> cc:    sigtran@ietf.org
> 
> Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> 
> 
> kakaul,
> 
> If there is no interoperablity issue then it is implementation dependent.
> A description of a procedure for the timer could be added as an
> IMPLEMENTATION
> NOTE making a suggestion to the implementor of one way to address local
> issues
> at the ASP, but it cannot be required.
> 
> But looking at your description, this mechanism would be a duplication of
> the
> mechanism used by SCCP, ISUP, BICC, which use a dual-timer approach to
> detecting congestion abatement.  Why would it be necessary for the ASP to
> do
> this as well?
> >> It is also done in MTP3 for nation options with priority do you mean to
> say that RCT mechanism is not required in MTP3.
> 
> Also, you should not that SCON received from one SG is no different than
> SCON
> received from another.  The SCON represents a statistical sampling of
> congestion in the SS7 network between the ASP and the destination, and is
> not
> referenced to an SG.  Therefore, the ASP should never perform this
> procedure
> on a per-SG basis.  The SCON received from either SG will generate an
> MTP-STATUS primitive to the AS, and the AS will run suitable timers at the
> MTP-User (where the responsibility lies to avoid sending messages of a
> priority less that the congestion state).
> >> In national option with priority MTP3 user is given MTP-STATUS
> indication only if point code (routeset )congestion status
> changes.Similarly at ASP MTP-STATUS indication shall be given to the user
> part only if derived status of affected point code changes as shown in
> figure below :
> 
> 
>                               SCON (1)+--+----++
>                              <------- |        |
>                           +-----------+  SG1   +-----+
>                           |           |        |     |
>                           |           +--------+     |      SS7 NODE
>            +--+----++     |                          |    +----------+
>            |        |     |                          |    |   PC     |
>            |  ASP   +-----+                          +----+  X-Y-Z   |
>            |        |     |                          |    |          |
>            +--------+     |                          |    +----------+
>                           |          +--+----++      |
>              AS:A-B-C     |          |        |      |
>                           +----------+  SG2   +------+
>                             <------- |        |
>                              SCON(3) +--------+
> 
> MTP-User will be given only MTP-STATUS indication with congestion status 3
> if SCON(3) reaches before SCON(1).So for this you have to maintain
> congestion status of affected point code through each SG and ultimately
> given the Derived congestion status to MTP-User.
> 
> Understand that SCON is not an M3UA congestion indication: it is an SS7
> network congestion indication that is directed at the MTP-User.  The ASP
> need
> only translate SCON messages to MTP-STATUS primitives to the AS to which
> they
> correspond.  A AS that conforms to MTP-User standards specifications with
> deal
> with the MTP-STATUS primitives appropriately.
> >>It is only true in international and national networks without priority
> option.
> 
> It is also important to understand that the ASP should not change the
> routing
> of messages on the basis of a received SCON message.  Where applicable and
> for
> applicable SS7 standards specifications, the SG will send DRST under very
> specific and standardized congestion situations.  The ASP is already
> expected
> to perform some functions of MTP routing in response to DRST.
> >> Here i definitely agree with you that rerouting of messages should not
> be done at ASP to avoid missequencing.
> 
> So, now I don't understand why you expect the ASP to run this timer at all.
> >> Some implementations may want to provision congestion of affected point
> code through supervisory congestion timer, they should be allowed to do so.
> 
> Please explain.
> 
> --brian
> 
> kakaul@hss.hns.com wrote:
> (Tue, 15 Jul 2003 10:46:03)
> >
> >
> >
> >
> > Brian,
> >    There is no interoperability issue involved in it and also i dont see
> > any need of peer knowing the existence of this timer.It is local to ASP
> and
> > is used by ASP to supervise the congestion status of the affected point
> > code reachable through a given SG.ASP after receiving an SCON message
> with
> > congestion status greater than 0 can start this timer and if no SCON
> > message is received from peer till the expiry of congestion timer ASP can
> > decrement affected point code  congestion status by 1 and start the timer
> > provisioning of congestion again if congestion status of affected point
> > code has not reached 0 or no congestion.
> >
> > Regards
> > Kamesh Kaul
> > www.hssworld.com
> > Hughes Software Systems
> >
> >
> >
> >
> > "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/15/2003
> > 10:28:40 AM
> >
> > Please respond to bidulock@openss7.org
> >
> > Sent by:    sigtran-admin@ietf.org
> >
> >
> > To:    Kamesh Kaul/HSS@HSS
> > cc:    "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
> >        sigtran@ietf.org
> >
> > Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> >
> >
> > kakaul,
> >
> > Then add text describing how to use the timer in an interoperable way.
> > Also, please describe why the peer needs to know about the existence
> > of this timer to remain interoperable.  Otherwise, there would not be
> > any need to describe it.
> >
> > --brian
> >
> >
> > kakaul@hss.hns.com wrote:
> > (Tue, 15 Jul 2003 09:35:27)
> > >
> > >
> > >
> > >
> > > Hi Javier,
> > >                     This timer may be used by some implementations to
> > > supervise the congestion status of point code if SCON messages send by
> > SGP
> > > are lost in transit to ASP.It was discussed earlier and was agreed upon
> > by
> > > all to keep this timer.I disagree with removing the mention of the
> timer.
> > >
> > > Regards
> > > Kamesh Kaul
> > > www.hssworld.com
> > > Hughes Software Systems
> > >
> > >
> > >
> > >
> > > "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>@ietf.org on
> > > 07/14/2003 04:05:58 PM
> > >
> > > Sent by:    sigtran-admin@ietf.org
> > >
> > >
> > > To:    "'Brian F. G. Bidulock '" <bidulock@openss7.org>
> > > cc:    "''sigtran@ietf.org' '" <sigtran@ietf.org>
> > >
> > > Subject:    RE: [Sigtran] M3UA IG v04 Section 3.18
> > >
> > >
> > > Brian,
> > >
> > > We'll remove the mention of the timer, if nobody disagrees...
> > >
> > > Cheers // Javier.
> > >
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock
> > > To: J.Javier Pastor (EE/EEM)
> > > Cc: 'sigtran@ietf.org'
> > > Sent: 11/07/2003 11:07
> > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > >
> > > J.Javier,
> > >
> > > J.Javier Pastor (EE/EEM) wrote:
> > > (Fri, 11 Jul 2003 09:56:59)
> > > > Hi Brian,
> > > >
> > > > Pls, see below...
> > > >
> > > > > -----Original Message-----
> > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > Sent: jueves, 10 de julio de 2003 15:34
> > > > > To: J.Javier Pastor (EE/EEM)
> > > > > Cc: 'sigtran@ietf.org'
> > > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > > > >
> > > > >
> > > > > Javier,
> > > > >
> > > > > 3.18.3 Solution Description
> > > > >
> > > > >    Clarify that the ASP needs to indicate when the congestion
> > > > > level has
> > > > >    changed (including abatement).  Further, the ASP peer can
> > > maintain
> > > > >    a timer, if desired, in case the ASP fails to indicate
> congestion
> > > > >    abatement.
> > > > >
> > > > > Where are the procedures for this timer described?  This is the
> only
> > > place
> > > > > that the timer is mentioned.  Please describe the timer or remove
> > > the
> > > > > sentence saying that it has been described...
> > > > >
> > > >
> > > > Do you think is worth include the timer explanation into the new
> text?
> > > > Otherwise, we may remove that sentence.
> > >
> > > Please remove mention of the timer.
> > >
> > > --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
> >
> 
> --
> 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 exim@www1.ietf.org  Wed Jul 16 12:33:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA07890
	for <sigtran-archive@odin.ietf.org>; Wed, 16 Jul 2003 12:33:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cpDD-0005pA-F3
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 12:33:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6GGX37f022388
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 12:33:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cpDB-0005or-1F; Wed, 16 Jul 2003 12:33:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cpCu-0005oN-Rb
	for sigtran@optimus.ietf.org; Wed, 16 Jul 2003 12:32:44 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA07868
	for <sigtran@ietf.org>; Wed, 16 Jul 2003 12:32:39 -0400 (EDT)
From: kakaul@hss.hns.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cpCt-00010Q-00
	for sigtran@ietf.org; Wed, 16 Jul 2003 12:32:43 -0400
Received: from hindon.hss.co.in ([202.54.26.202])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cpCh-000109-00
	for sigtran@ietf.org; Wed, 16 Jul 2003 12:32:32 -0400
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h6GGSss05009;
	Wed, 16 Jul 2003 21:58:54 +0530 (IST)
Received: from dlfmail1.hss.hns.com (localhost [127.0.0.1])
	by ultra.hss.co.in (8.10.0/8.10.0) with ESMTP id h6GGWZp23562;
	Wed, 16 Jul 2003 22:02:36 +0530 (IST)
Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
To: bidulock@openss7.org
Cc: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.5  September 22, 2000
Message-ID: <OFAE52A198.69A2F46E-ON65256D65.00591206@hss.hns.com>
Date: Wed, 16 Jul 2003 21:57:11 +0530
X-MIMETrack: Serialize by Router on Dlfmail1/HSS(Release 6.0|September 26, 2002) at 07/16/2003
 09:54:45 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,
   Comments inline




"Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/16/2003
12:41:40 PM

Please respond to bidulock@openss7.org

Sent by:    sigtran-admin@ietf.org


To:    Kamesh Kaul/HSS@HSS
cc:    sigtran@ietf.org

Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18


kakaul,

An SS7 SEP at the position of the ASP, where the SGs are STPs would
not maintain congestion state on a per-route (per-SG) basis.  So
what makes you think that the ASP needs to?
[Kamesh]It would because route-set congestion status will change depending
on the derived congestion status of individual routes I am attaching
section 11.2.4 from Q.704


11.2.4 Signalling route set congestion (National option with congestion
       priorities)
In the case when the congestion status of a signalling route set changes as
a result of either the receipt of a transfer controlled message relating to
a particular destination (see 13.7) or an indication of local signalling
link congestion, or due to the signalling route-set-congestion-test
procedure (see 13.9) an indication is given from the Message Transfer Part
to the local level 4 informing it about the current congestion status of
the signalling route set. Each user then takes appropriate actions in order
to stop generation of signalling messages destined for the affected
signalling point with congestion priorities lower than the specified
congestion status. Messages received from the local level 4 with congestion
priorities lower than the current signalling route set congestion status
are discarded by the Message Transfer Part.
Also, SCON() in this configuration corresponds to TFC (not MTP-STATUS)
and therefore cannot signal abatement.  (TFC does not signal abatement,
the lack of a TFC does.)  In the same ways, the ASP can use the lack
of a SCON to indicate abatement.
[Kamesh]Yes they do and derived congestion status leads to MTP-STATUS

So, a) SCON(1) and SCON(3) in your example are congestion signals
(neither is an abatement signal); b) there is no need to maintain
congestion
status on a per-SG basis at the ASP ('cause the SS7 SEP doesn't).
[Kamesh]Yes there is need to maintain congestion status on the per-SG basis
at ASP because SG1 doing RCT procedure on the behalf of ASP may return
SCON(0) when it receives audit for the affected point code and if affected
point code is reachable through SG1 and SG2 as shown in previous figure SG2
returns SCON(2) then congestion status sent in MTP-STATUS is 2 because
affected point code is still congested.

This might lead back to the discustion why ASPs don't send RCT, but the
simple answer is that each SG as STP can send RCT on all the ASP's behalf
as it would for an "Alias" point code (ANSI T1.111/2000).

Because this is getting into a narrower and narrower corner of the
specification, I think we are safe to strike the mention of the timer
from the IG.
[Kamesh]I dont agree with striking timer
--brian

kakaul@hss.hns.com wrote:
(Tue, 15 Jul 2003 18:59:37)
>
>
>
>
> Brian
>    some comments inline
>
>
>
>
> "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/15/2003
> 12:16:38 PM
>
> Please respond to bidulock@openss7.org
>
> Sent by:    sigtran-admin@ietf.org
>
>
> To:    Kamesh Kaul/HSS@HSS
> cc:    sigtran@ietf.org
>
> Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
>
>
> kakaul,
>
> If there is no interoperablity issue then it is implementation dependent.
> A description of a procedure for the timer could be added as an
> IMPLEMENTATION
> NOTE making a suggestion to the implementor of one way to address local
> issues
> at the ASP, but it cannot be required.
>
> But looking at your description, this mechanism would be a duplication of
> the
> mechanism used by SCCP, ISUP, BICC, which use a dual-timer approach to
> detecting congestion abatement.  Why would it be necessary for the ASP to
> do
> this as well?
> >> It is also done in MTP3 for nation options with priority do you mean
to
> say that RCT mechanism is not required in MTP3.
>
> Also, you should not that SCON received from one SG is no different than
> SCON
> received from another.  The SCON represents a statistical sampling of
> congestion in the SS7 network between the ASP and the destination, and is
> not
> referenced to an SG.  Therefore, the ASP should never perform this
> procedure
> on a per-SG basis.  The SCON received from either SG will generate an
> MTP-STATUS primitive to the AS, and the AS will run suitable timers at
the
> MTP-User (where the responsibility lies to avoid sending messages of a
> priority less that the congestion state).
> >> In national option with priority MTP3 user is given MTP-STATUS
> indication only if point code (routeset )congestion status
> changes.Similarly at ASP MTP-STATUS indication shall be given to the user
> part only if derived status of affected point code changes as shown in
> figure below :
>
>
>                               SCON (1)+--+----++
>                              <------- |        |
>                           +-----------+  SG1   +-----+
>                           |           |        |     |
>                           |           +--------+     |      SS7 NODE
>            +--+----++     |                          |    +----------+
>            |        |     |                          |    |   PC     |
>            |  ASP   +-----+                          +----+  X-Y-Z   |
>            |        |     |                          |    |          |
>            +--------+     |                          |    +----------+
>                           |          +--+----++      |
>              AS:A-B-C     |          |        |      |
>                           +----------+  SG2   +------+
>                             <------- |        |
>                              SCON(3) +--------+
>
> MTP-User will be given only MTP-STATUS indication with congestion status
3
> if SCON(3) reaches before SCON(1).So for this you have to maintain
> congestion status of affected point code through each SG and ultimately
> given the Derived congestion status to MTP-User.
>
> Understand that SCON is not an M3UA congestion indication: it is an SS7
> network congestion indication that is directed at the MTP-User.  The ASP
> need
> only translate SCON messages to MTP-STATUS primitives to the AS to which
> they
> correspond.  A AS that conforms to MTP-User standards specifications with
> deal
> with the MTP-STATUS primitives appropriately.
> >>It is only true in international and national networks without priority
> option.
>
> It is also important to understand that the ASP should not change the
> routing
> of messages on the basis of a received SCON message.  Where applicable
and
> for
> applicable SS7 standards specifications, the SG will send DRST under very
> specific and standardized congestion situations.  The ASP is already
> expected
> to perform some functions of MTP routing in response to DRST.
> >> Here i definitely agree with you that rerouting of messages should not
> be done at ASP to avoid missequencing.
>
> So, now I don't understand why you expect the ASP to run this timer at
all.
> >> Some implementations may want to provision congestion of affected
point
> code through supervisory congestion timer, they should be allowed to do
so.
>
> Please explain.
>
> --brian
>
> kakaul@hss.hns.com wrote:
> (Tue, 15 Jul 2003 10:46:03)
> >
> >
> >
> >
> > Brian,
> >    There is no interoperability issue involved in it and also i dont
see
> > any need of peer knowing the existence of this timer.It is local to ASP
> and
> > is used by ASP to supervise the congestion status of the affected point
> > code reachable through a given SG.ASP after receiving an SCON message
> with
> > congestion status greater than 0 can start this timer and if no SCON
> > message is received from peer till the expiry of congestion timer ASP
can
> > decrement affected point code  congestion status by 1 and start the
timer
> > provisioning of congestion again if congestion status of affected point
> > code has not reached 0 or no congestion.
> >
> > Regards
> > Kamesh Kaul
> > www.hssworld.com
> > Hughes Software Systems
> >
> >
> >
> >
> > "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/15/2003
> > 10:28:40 AM
> >
> > Please respond to bidulock@openss7.org
> >
> > Sent by:    sigtran-admin@ietf.org
> >
> >
> > To:    Kamesh Kaul/HSS@HSS
> > cc:    "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
> >        sigtran@ietf.org
> >
> > Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> >
> >
> > kakaul,
> >
> > Then add text describing how to use the timer in an interoperable way.
> > Also, please describe why the peer needs to know about the existence
> > of this timer to remain interoperable.  Otherwise, there would not be
> > any need to describe it.
> >
> > --brian
> >
> >
> > kakaul@hss.hns.com wrote:
> > (Tue, 15 Jul 2003 09:35:27)
> > >
> > >
> > >
> > >
> > > Hi Javier,
> > >                     This timer may be used by some implementations to
> > > supervise the congestion status of point code if SCON messages send
by
> > SGP
> > > are lost in transit to ASP.It was discussed earlier and was agreed
upon
> > by
> > > all to keep this timer.I disagree with removing the mention of the
> timer.
> > >
> > > Regards
> > > Kamesh Kaul
> > > www.hssworld.com
> > > Hughes Software Systems
> > >
> > >
> > >
> > >
> > > "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>@ietf.org on
> > > 07/14/2003 04:05:58 PM
> > >
> > > Sent by:    sigtran-admin@ietf.org
> > >
> > >
> > > To:    "'Brian F. G. Bidulock '" <bidulock@openss7.org>
> > > cc:    "''sigtran@ietf.org' '" <sigtran@ietf.org>
> > >
> > > Subject:    RE: [Sigtran] M3UA IG v04 Section 3.18
> > >
> > >
> > > Brian,
> > >
> > > We'll remove the mention of the timer, if nobody disagrees...
> > >
> > > Cheers // Javier.
> > >
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock
> > > To: J.Javier Pastor (EE/EEM)
> > > Cc: 'sigtran@ietf.org'
> > > Sent: 11/07/2003 11:07
> > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > >
> > > J.Javier,
> > >
> > > J.Javier Pastor (EE/EEM) wrote:
> > > (Fri, 11 Jul 2003 09:56:59)
> > > > Hi Brian,
> > > >
> > > > Pls, see below...
> > > >
> > > > > -----Original Message-----
> > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > Sent: jueves, 10 de julio de 2003 15:34
> > > > > To: J.Javier Pastor (EE/EEM)
> > > > > Cc: 'sigtran@ietf.org'
> > > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > > > >
> > > > >
> > > > > Javier,
> > > > >
> > > > > 3.18.3 Solution Description
> > > > >
> > > > >    Clarify that the ASP needs to indicate when the congestion
> > > > > level has
> > > > >    changed (including abatement).  Further, the ASP peer can
> > > maintain
> > > > >    a timer, if desired, in case the ASP fails to indicate
> congestion
> > > > >    abatement.
> > > > >
> > > > > Where are the procedures for this timer described?  This is the
> only
> > > place
> > > > > that the timer is mentioned.  Please describe the timer or remove
> > > the
> > > > > sentence saying that it has been described...
> > > > >
> > > >
> > > > Do you think is worth include the timer explanation into the new
> text?
> > > > Otherwise, we may remove that sentence.
> > >
> > > Please remove mention of the timer.
> > >
> > > --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
> >
>
> --
> 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 exim@www1.ietf.org  Wed Jul 16 17:54:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA19095
	for <sigtran-archive@odin.ietf.org>; Wed, 16 Jul 2003 17:54:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cuDt-0008JD-Fr
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 17:54:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6GLs5o0031938
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 17:54:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cuDp-0008Iq-IK; Wed, 16 Jul 2003 17:54:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cuD2-0008Id-8p
	for sigtran@optimus.ietf.org; Wed, 16 Jul 2003 17:53:12 -0400
Received: from pt.com (ottgw.pt.com [206.191.2.130])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA19053
	for <sigtran@ietf.org>; Wed, 16 Jul 2003 17:53:05 -0400 (EDT)
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id RAA15244;
	Wed, 16 Jul 2003 17:47:26 -0400
Message-ID: <3F15C7F2.6070704@pt.com>
Date: Wed, 16 Jul 2003 17:47:30 -0400
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: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - RPO
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030712170744.A11077@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

Hi Brian,

If the side in RPO is sending data acknowledgments then it must also be 
delivering these messages to MTP3.  Is there some guarantee that MTP3 
will accept messages from a link that it has marked as blocked?  Or am I 
misunderstanding the intent of your proposal?

Andrew

Brian F. G. Bidulock wrote:

>M2PA-09 says:
>
> While there is a remote processor outage condition:
>
>   (a) Any User Data messages received from the peer after the Link
>       Status Processor Outage MUST NOT be acknowledged and MUST be
>       buffered.
>
>   (b) M2PA SHOULD continue to acknowledge User Data messages received
>       and accepted by MTP3 before the remote processor outage.
>
>In Q.703 the side experiencing PO cannot send MSUs or FISUs without indicating a
>recovery from processor outage.  M2PA can send User Data without indicating
>processor recovery.  There is no reason for the receiver of User Data not to
>process it even after receiving Link Status Processor Outage.
>
>Otherwise, if (a) above is followed, MSUs in the transmission and retransmission
>buffers at the time that local processor outage occured will be trapped.  If
>User Data is acknowledged by the peer, they will be processed by the receiver.
>
>Change M2PA to say:
>
> While there is a remote processour outage condition:
>
>   (a) M2PA SHOULD continue to acknowledge User Data messages received
>       and accepted by MTP3 regardless of the remote processor outage.
>
>This allows the TB/RTB on the side experiencing processor outage to drain.
>
>
>M2PA-09 says:
>
>  Note that because the Link Status and User Data messages are sent on
>  separate streams, it is possible for a User Data message sent after
>  the Link Status Processor Recovered message to arrive before
>  it. Therefore, when flushing buffers, M2PA must take care not to flush
>  messages sent after the remote processor recovered. M2PA SHOULD NOT
>  flush User Data messages with FSN greater than the FSN in the Link
>  Status Processor Recovered message. 
>
>MTP3 does not flush the RB after RPO or before RPR (just the RTB), unless it
>uses the remote processor outage to local processor outage latching option, and
>then it is only because of the LPO condition not the RPO condition.
>
>With the change above, M2PA receives and acknowledges User Data before, during
>and after remote processor outage, and comparison to any sequence numbers in the
>Link Status Processor Recovered is unnecessary.
>
>M2PA-09 says:
>
>  When the local processor outage condition ends, M2PA SHALL send a Link
>  Status Processor Recovered message to its peer. This message is used
>  to signal in the end of the processor outage condition, instead of an
>  MSU or FISU as in MTP2. The BSN value in this message is used by the
>  peer to resynchronize its sequence numbers, if this is required by
>  MTP2.
>
>After a flush buffers operation, the side that experienced PO uses the first
>received FISU/MSU to resync and discards the FISU/MSU.
>
>ANSI T1.111.3/2000 Section 8 says:
>
>  ... If level 3 requests level 2 [at the side experiencing processor outage] to
>  clear its buffers of message signal units, level 2 begins to continuously
>  transmit fill-in signal units and discards all message signal units that it
>  has received from remote level 2 but not yet delivered to local level 3 and
>  all message signal units that it has received from local level 3 but not yet
>  transmitted on the link.  The first fill-in signal unit transmitted
>  acknowledges all message signal units that have been accepted from remote
>  level 2 including any mesage signal units for which acknowledgements were
>  previously withheld because of level 2 congestion (see section 9).  Level 2
>  then begins accepting message signal units from remote Level 2. When the next
>  fill-in or message signal unit is received that passes the acceptance
>  procedure of section 4 and sections 5 or 6,
>
>   (1) all message signal units that are available for retransmission are
>       discarded [clear RTB].
>
>   (2) the forward sequence number and forward indicator bit of the
>       transmitted fill-in signal units are set equal to the backward sequence
>       number and backward indicator bit of the received signal unit,
>
>   (3) normal transmission of message signal units and fill-in signal units is
>       resumed, and
>
>   (4) level 3 is notified that the buffer clearing has been completed and
>       that normal operations has been resumed.
>
>So, you see, the side experiencing the processor outage, after clearing buffers,
>resynchronizes its FSN with the BSN of the first received FISU or MSU.  The side
>experiencing processor outage could use the first User Data message received
>after processor recovery to resyncronize.  However, if the peer has nothing to
>send, processor recovery will be delayed.
>
>Note, however, that resynchronization of sequence numbers for the transmit
>direction is necessary because MTP2 traps messages in the TB/RTB of the side
>experiencing processor outage because it must continuously transmit SIPO during
>processor outage.  Permitting the draining of the TB/RTB with the change above
>removes the need to flush the TB/RTB on the side experiencing processor outage,
>as well as removing the need to resynchronize sequence numbers.
>
>In the other direction, for messages queued to be sent to the side experiencing
>processour outage, flushing and resynchronization is still necessary (actions
>(1) to (4) per ANSI above).
>
>However, Q.703 permits FISUs/MSUs received by the side experiencing remote
>processor outage to discard the first and several subsequent FISU/MSUs when
>recovering.  There is even a test case in Q.781 (8.9) that shows an MSU being
>discarded (and negative acknowledged) in this fashion.  Because M2PA cannot
>discard and expect retransmission of a message, and because sychronization is
>not required when the TB/RTB of the side experiencing processor outage is
>permitted to drain with the changes above, M2PA cannot permit messages to be
>discarded in this fashion.
>
>M2PA-09 says:
>
>  If M2PA receives a Flush command from MTP3, M2PA SHALL discard the
>  incoming messages that were queued and unacknowledged during the
>  processor outage condition. Furthermore, M2PA SHALL discard messages
>  in the transmit and retransmit queues as required by MTP2. 
>
>To handle flushing and resynchronization properly in both directions, change the
>above to read as follows:
>
>Change M2PA to say:
>
>  If M2PA receives a Flush command from MTP3, M2PA SHALL discard the
>  incoming messages that were queued and unacknowledged during the
>  processor outage condition.  Furthermore, the M2PA experiencing
>  processor outage SHALL NOT discard message from the transmit and
>  retransmit queues and SHALL NOT perform procedures to resynchronize
>  sequence numbers, but SHALL continue normal transmission of User Data.
>  The M2PA experiencing remote processor outage, SHALL discard messages in
>  the transmit and retransmit queues as required by MTP2, and MAY perform
>  resynchronization of sequence numbers using the BSN from the received
>  Link Status Processor Recovered message.  However, during
>  resychronization, M2PA SHALL NOT discard any received User Data
>  messages.
>
>As Q.2140 handles processor outage by putting the link out of service, also add:
>
>  When M2PA experiences a local processor outage, it MAY put the link out
>  of service by sending a Link Status Out of Service message if allowed by
>  the applicable MTP2 standard (e.g, [Q.2140]).
>
>I think that wraps up processor outage problems.
>
>--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 exim@www1.ietf.org  Wed Jul 16 18:03:25 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA19588
	for <sigtran-archive@odin.ietf.org>; Wed, 16 Jul 2003 18:03:25 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cuMX-0000RQ-86
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 18:03:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6GM31rF001675
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 18:03:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cuMX-0000Qu-3v; Wed, 16 Jul 2003 18:03:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cuMB-0000QV-DN
	for sigtran@optimus.ietf.org; Wed, 16 Jul 2003 18:02:39 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA19540
	for <sigtran@ietf.org>; Wed, 16 Jul 2003 18:02:33 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cuM8-00040P-00
	for sigtran@ietf.org; Wed, 16 Jul 2003 18:02:36 -0400
Received: from ottgw.pt.com ([206.191.2.130] helo=pt.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19cuLs-00040I-00
	for sigtran@ietf.org; Wed, 16 Jul 2003 18:02:20 -0400
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id SAA30826;
	Wed, 16 Jul 2003 18:01:36 -0400
Message-ID: <3F15CB44.1010204@pt.com>
Date: Wed, 16 Jul 2003 18:01:40 -0400
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 last call comment - ANSI primitives for 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>
Content-Transfer-Encoding: 7bit

Hi Tom,

The ANSI primitives for exiting processor outage seem to be Resume, 
Clear Buffers and Clear RTB.  The corresponding ITU primitives 
(continue, flush, ... ) are described explicitly in M2PA draft 9. 
 Should the ANSI primitives also be described explicitly?

Andrew



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



From exim@www1.ietf.org  Wed Jul 16 18:13:26 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA20752
	for <sigtran-archive@odin.ietf.org>; Wed, 16 Jul 2003 18:13:26 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cuWE-0001a7-80
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 18:13:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6GMD2cU006073
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 18:13:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cuWD-0001Zi-D2; Wed, 16 Jul 2003 18:13:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cuVN-0001ZD-8r
	for sigtran@optimus.ietf.org; Wed, 16 Jul 2003 18:12:09 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA20608
	for <sigtran@ietf.org>; Wed, 16 Jul 2003 18:12:03 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cuVK-00044U-00
	for sigtran@ietf.org; Wed, 16 Jul 2003 18:12:06 -0400
Received: from ottgw.pt.com ([206.191.2.130] helo=pt.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19cuV4-000441-00
	for sigtran@ietf.org; Wed, 16 Jul 2003 18:11:50 -0400
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id SAA08774;
	Wed, 16 Jul 2003 18:10:59 -0400
Message-ID: <3F15CD78.6020209@pt.com>
Date: Wed, 16 Jul 2003 18:11:04 -0400
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: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - T6
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030712002921.A21136@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

Hi Brian,

Comments inline.



Brian F. G. Bidulock wrote:

>M2PA-09 says:
>
>  
>
>>M2PA SHALL continue transmitting and acknowledging messages while it
>>is in receive congestion.
>>    
>>
>
>Q.703 says:
>
>| The receiving level 2 also withholds acknowledgement of the message signal
>| unit, which triggered off the congestion detection, and of message signal
>| units received during the congestion situation; that is fill-in signal units
>| or message signal units are sent as usual, but with the backward sequence
>| number and backwards indicator bit assigned the values which are contained
>| in the last transmitted signal unit before the congestion is recognized.
>
>It is rather important that the busy end not acknowledge messages while it is
>in receive congestion.  This is because if the receive congestion results in
>link failure, acknowledged messages caught at the congested receiver will be
>unavailable for retransmission on an alternate link at the transmitter.
>
Are you assuming here that the Busy end is not delivering messages to MTP3?

>
>Also, delivering these messages to MTP3 late will aggravate missequencing of
>these delayed messages with subsequent messages transmitted on the alternate
>link after changeover.
>
>Change M2PA to say:
>
>< M2PA SHALL continue transmitting message while it is in receive congestion
>< but MUST NOT acknowledge the message that triggered sending of the Link
>< Status Busy message, nor any messages received before the sending of Link
>< Status Busy Ended.
>
>M2PA-09 says:
>
>  
>
>>When the peer M2PA receives the first Link Status Busy message, it
>>SHALL start the Remote Congestion timer T6. Additional Link Status
>>Busy messages received while T6 is running do not cause T6 to be
>>reset.
>>    
>>
>
>Q.703 says:
>
>| At the remote end of the signalling link, every reception of a link status
>| singal unit containing indication "B" causes the excessive delay of
>| acknolwedgement timer T7 to be restarted, if the timer is already running.
>| In addition first reception of the link status signal unit containing a
>| status indication "B" starts a longer supervision timer T7 (see 12.3) only
>| if there are message signal units in the retransmission buffer.  Should
>| timer T6 expire, link failure indication is generated.
>
>Change M2PA to say:
>
>< When the peer M2PA receives the first Link Status Busy message, it SHALL
>< start the Remote Congestion timer T6 if there are messages in the
>< retransmission buffer awaiting acknowledgement (i.e, T7 is running).
>< In addition, M2PA will stop the T7 timer if it is running.  Additional Link
>< Status Busy messages received while T6 is running do not cause T6 to be
>< reset and do not cause T7 to be started.  Also, while T6 is running, T7
>< SHALL NOT be started.
>
>< When the peer M2PA receives the Link Status Busy Ended message and T6
>< has not expired, it SHALL stop T6 if T6 is running and start T7 if there are
>< messages awaiting acknowledgement in the retransmission buffer.
>
>Now, I know that we said that Link Status messages are not used to acknowledge
>DATA messages, but it makes sense to permit the Link Status Busy Ended message
>to acknowledge outstanding User Data messages, so that upon its receipt, T7
>might not need to be started if all messages in the retransmission buffer are
>acknolwedged by the Link Status Busy Ended message.
>
>The above also kind of changes the idea of T7 metering the oldest message in
>the retransmission buffer.  It would now measure the oldest message in the
>retransmission buffer since the receipt of the last Link Status Busy Ended.
>
>M2PA-09 says:
>
>  
>
>>The peer M2PA SHOULD continue transmitting and acknowledging messages
>>while its T6 timer is running, i.e., while the other end is Busy.
>>    
>>
>
>But T6 is not necessarily running during the busy condition.
>
I don't understand this comment.  T6 is started when LSB is received and 
stopped when LSBE is received.  Doesn't that mean that T6 is always 
running if the peer is busy?

Thanks for any info.

Regards,
Andrew

>
>Change M2PA to say:
>
>< The peer M2PA SHOULD continue transmitting and acknowledging messages
>< while the other end is Busy: i.e, between receiving Link Status Busy
>< and Link Status Busy Ended.
>
>--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 exim@www1.ietf.org  Wed Jul 16 19:49:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA22846
	for <sigtran-archive@odin.ietf.org>; Wed, 16 Jul 2003 19:49:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cw18-0006iA-TF
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 19:49:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6GNn2sM025792
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 19:49:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cw17-0006hl-Jn; Wed, 16 Jul 2003 19:49:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cw15-0006ha-2I
	for sigtran@optimus.ietf.org; Wed, 16 Jul 2003 19:48:59 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA22843
	for <SIGTRAN@ietf.org>; Wed, 16 Jul 2003 19:48:55 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cw13-0004ke-00
	for SIGTRAN@ietf.org; Wed, 16 Jul 2003 19:48:57 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19cw0r-0004kN-00
	for SIGTRAN@ietf.org; Wed, 16 Jul 2003 19:48:46 -0400
Received: from [172.16.1.203] (helo=CCPU00186)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19cw08-0007YA-Ks; Wed, 16 Jul 2003 16:48:00 -0700
From: "VIvek Toky" <vivektoky@ccpu.com>
To: <bidulock@openss7.org>
Cc: "Umesh Vats" <umesh@ccpu.com>, "SIGTRAN e-mail list" <SIGTRAN@ietf.org>
Subject: RE: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Date: Wed, 16 Jul 2003 16:52:37 -0700
Message-ID: <HKEBKFBNPIDALDOKLGKDIEICCDAA.vivektoky@ccpu.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.6604 (9.0.2911.0)
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
In-Reply-To: <20030715232539.A17550@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id TAA22844
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Brian,
  If the same ASP is serving multiple AS it can also cross check the
uniqueness of the routing Keys among those AS. That way we don't even nee=
d
to send the REG Request messages to a peer node.
  Even if multiple ASs (in same ASP) manage to send REG request messages =
for
same routing keys, SGP can anyway return Response message to the ASP,
containing a Registration Result "Error - Cannot Support Unique Routing" =
but
can also send RC related to the routing key already registered. This woul=
d
help implementations to recover from data inconsistencies (if present),
without stopping all the traffic for that ASP.

-Vivek


-----Original Message-----
From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf Of
Brian F. G. Bidulock
Sent: Tuesday, July 15, 2003 10:26 PM
To: VIvek Toky
Cc: Umesh Vats; SIGTRAN e-mail list
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP


VIvek,

Validate it at the ASP all you want.  In the RFC it is the SG's
responsibility
to guarantee unique routing, not the ASP's.

--brian

On Tue, 15 Jul 2003, VIvek Toky wrote:

> Hi Brian,
>   Can you give me pointer to a section in RFC or IGv4 where its mention=
ed
> "ASP must use only RC to route messages locally to AS".. If your ASP
> implementation does not maintain routing keys, why do you want SGPs to =
do
> the work for you.
>
>  Even if an implementaion routes messages locally based on RC, it can
still
> validate RK while sending the REG Request message out, I don't see a
problem
> with that. If you don't want to modify your datastructures, that doesn'=
t
> mean M3UA protocol should change for you.
>
> -Vivek
>
> -----Original Message-----
> From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf O=
f
> Brian F. G. Bidulock
> Sent: Tuesday, July 15, 2003 3:21 PM
> To: VIvek Toky
> Cc: Umesh Vats; SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>
>
> VIvek,
>
> All ASPs need to route messages locally to AS is the RC.  They need
> not be aware of the RK.
>
> --brian
>
>
> On Tue, 15 Jul 2003, VIvek Toky wrote:
>
> > Brian,
> >
> >   You said
> >    "The reason for sending "Cannot support unique routing" was becaus=
e
we
> > spent
> > some great pains removing RK depedency from the ASP.  The ASP can be
> > designed
> > completely RK neutral and treat the RK as opaque.  Only the AS need k=
now
> the
> > internals of the RK."
> >
> > Why do you want ASPs to be totally unawar of whats inside the RK, If
SGPs
> > have to do the routing based on RKs then ASPs also have to route the
> > incoming messages to the destined AS (locally). If different ASs try =
to
> > register same routing key then it can be checked at ASP itself withou=
t
> even
> > sending the REG Request message out.
> >
> > Can you explain more about the interoperability scenarios you
mentioned..
> >
> > -Vivek.
> >
> > -----Original Message-----
> > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf=
 Of
> > Brian F. G. Bidulock
> > Sent: Tuesday, July 15, 2003 2:16 PM
> > To: Umesh Vats
> > Cc: SIGTRAN e-mail list
> > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> >
> >
> > Umesh,
> >
> > An SG designed to meet SS7 requirements will not have any problems
> > losing M3UA messages.
> >
> > --brian
> >
> > On Tue, 15 Jul 2003, Umesh Vats wrote:
> >
> > > If ASP looses ASPSM/ASPTM/DRKM/SSNM messages(which are consumed by
M3UA,
> > > it should be able to recover.
> > >
> > > For SSNM, subsequent SSNM messages or DAUD procedure take care of
> message
> > > loss.
> > > For if ASP-ACTIVE_ACK response is lost, ASP can resend ASP-ACTIVE.
> > >
> > > But for loss of REG-RSP, the only recovery now possible is ASP-DOWN
> > > procedure which is far worse that deregistering just one RK.
> > >
> > > Umesh
> > >
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: Tuesday, July 15, 2003 12:51 PM
> > > To: Umesh Vats
> > > Cc: SIGTRAN e-mail list; Vivek Toky
> > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > >
> > >
> > > Umesh,
> > >
> > > What should ASP do if it loses data.  What should ASP do if it lose=
s
> > > ASP xxxx Ack.  What should ASP do if it loses mandatory Notify.  Wh=
at
> > > should ASP do....
> > >
> > > The SGP should not lose messagez, or if it does, send ASP Inactive =
Ack
> > > or ASP Down Ack to reset the association.
> > >
> > > If it loses messages without detecting them, there is nothing to do
but
> > > recreate SCTP in M3UA (a rather absurd proposition).
> > >
> > > --brian
> > >
> > >
> > > On Tue, 15 Jul 2003, Umesh Vats wrote:
> > >
> > > > REG-RSP from SGP to ASP ,may be lost at SG itself before it reach=
es
> SCTP
> > > > due to local problems at SG. In this case, what should ASP do ?
> > > >
> > > > I don't understand why sending of RC from SG poses any problem.
> > > > SG already knows about the RK and RC. It just has to return the R=
C.
> > > >
> > > > Umesh
> > > >
> > > >
> > > > -----Original Message-----
> > > > From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On
Behalf
> Of
> > > > Brian F. G. Bidulock
> > > > Sent: Monday, July 14, 2003 10:13 PM
> > > > To: Umesh Vats
> > > > Cc: SIGTRAN e-mail list; Vivek Toky
> > > > Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
> > > >
> > > >
> > > > Umesh,
> > > >
> > > > On Mon, 14 Jul 2003, Umesh Vats wrote:
> > > >
> > > > > Brian,
> > > > >
> > > > > I am sure you have sufficient background of
> > highly-available/distributed
> > > > > systems as your implementation is 100% fault tolerant with
> > > > > nil (0%) message loss.
> > > > >
> > > > > In the given scenario, what if host1 receives REG-RSP from host=
3
but
> > > > before
> > > > > updating
> > > > > the RC received in message to common data, host1 goes down.
> > > >
> > > > An host1 does not acknowledge the message to host3, so when host2
> comes
> > > > up, host3 sends the REG RSP again to host2.
> > > >
> > > > >
> > > > > In this case, host2 doesn't know the RC and RK relation.
> > > >
> > > > The RC is in the REG RSP that host3 resends to host2.  The RK is =
in
> > common
> > > > data along with the Local-Routing-Key-Identifier that is also
> contained
> > in
> > > > the REG RSP.  host2 will have no problem correlating the REG RSP
with
> > the
> > > > REG REQ, as would host1 (on the LRK-Id).
> > > >
> > > > > Performing ASP-DOWN procedure will make ASP DOWN for all the
> > Application
> > > > > Servers for which traffic exchange might be happening.
> > > >
> > > > So, don't.  Have host3 send the REG RSP message not acknowledged =
by
> > host1
> > > > to host2 when host2 takes over.
> > > >
> > > > >
> > > > > Infact, your implementation can use this idea to further decrea=
se
> > > message
> > > > > loss from 0% to -1% percentage or less :)
> > > >
> > > > Your implementation does not need to lose REG RSP unless it has b=
een
> > > > designed to do so.  Don't do that.  Use techniques such as above =
so
> that
> > > > you do not lose messages.  Do not expect M3UA to repeat messages =
to
> > handle
> > > > the idiosynchrasies of your own internal architecture.
> > > >
> > > > The reason for sending "Cannot support unique routing" was becaus=
e
we
> > > spent
> > > > some great pains removing RK depedency from the ASP.  The ASP can=
 be
> > > > designed
> > > > completely RK neutral and treat the RK as opaque.  Only the AS ne=
ed
> know
> > > the
> > > > internals of the RK.  One side effect of this is that mutliple AS=
 at
> the
> > > ASP
> > > > may attempt to activate for the same RK.  The ASP cannot detect t=
his
> an
> > > > remain
> > > > RK neutral.  So, when a second AS attemtps to register the same R=
K
> (and
> > > > obtain
> > > > its own unique RC), it is the SGP that refuses the registration w=
ith
> > > "Cannot
> > > > support unique routing", instead of requiring the ASP to maintain
RKs
> > and
> > > > police AS/RC/RK uniqueness.  The RFC places the obligation at the
SGP
> > > (M3UA,
> > > > NIF or MTP) to enforce the 1:1:1 relationship of AS:RC:RK, becaus=
e
it
> is
> > > the
> > > > one that requires this relationship so that there is "unique
routing"
> in
> > > the
> > > > SGP routing table.
> > > >
> > > > This is rather important to interoperability, because there is no
> > > mechanism
> > > > in the M3UA ASP specification to enforce this relationship, only =
the
> > > > mechanism
> > > > at the SGP.  Altering the behviour of the SGP would break this
> > > > interoperability mechanism.
> > > >
> > > > There are probably three or four other mechanism within your
> > architecture
> > > > that will solve the problem for you without changing M3UA and
breaking
> > > this
> > > > more important interoperability behavior.  So I cannot agree with
the
> > > > change.
> > > >
> > > > --brian
> > > >
> > > > --
> > > > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to =
the =A6
> > > > bidulock@openss7.org    =A6 world; the unreasonable one persists =
in  =A6
> > > > http://www.openss7.org/ =A6 trying  to adapt the  world  to himse=
lf. =A6
> > > >                         =A6 Therefore  all  progress  depends on =
the =A6
> > > >                         =A6 unreasonable man. -- George Bernard S=
haw =A6
> > > >
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > >
> > > > ---
> > > > Incoming mail is certified Virus Free.
> > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > > >
> > > > ---
> > > > Outgoing mail is certified Virus Free.
> > > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > > --
> > > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to th=
e =A6
> > > bidulock@openss7.org    =A6 world; the unreasonable one persists in=
  =A6
> > > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself=
. =A6
> > >                         =A6 Therefore  all  progress  depends on th=
e =A6
> > >                         =A6 unreasonable man. -- George Bernard Sha=
w =A6
> > >
> > > ---
> > > Incoming mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> > >
> > > ---
> > > Outgoing mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
> >
> > --
> > Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =
=A6
> > bidulock@openss7.org    =A6 world; the unreasonable one persists in  =
=A6
> > http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =
=A6
> >                         =A6 Therefore  all  progress  depends on the =
=A6
> >                         =A6 unreasonable man. -- George Bernard Shaw =
=A6
> >
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
>
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

_______________________________________________
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 exim@www1.ietf.org  Wed Jul 16 20:53:10 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA24164
	for <sigtran-archive@odin.ietf.org>; Wed, 16 Jul 2003 20:53:10 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cx0m-0001Y7-4u
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 20:52:44 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6H0qi0u005948
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 20:52:44 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cx05-0001Xb-1Y; Wed, 16 Jul 2003 20:52:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cwzg-0001XM-5p
	for sigtran@optimus.ietf.org; Wed, 16 Jul 2003 20:51:36 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA24128
	for <SIGTRAN@ietf.org>; Wed, 16 Jul 2003 20:51:31 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cwzd-0005Fm-00
	for SIGTRAN@ietf.org; Wed, 16 Jul 2003 20:51:33 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[k+fJ3j/J4VCW6sQs8fxKzStl4esyOYdZ])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cwzN-0005Fj-00
	for SIGTRAN@ietf.org; Wed, 16 Jul 2003 20:51:17 -0400
Received: from ns.pigworks.openss7.net (IDENT:ZOHrihDUdE764heIlZfK1a3XFy1aIshk@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6H0oqP06520;
	Wed, 16 Jul 2003 18:50:52 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6H0opB07616;
	Wed, 16 Jul 2003 18:50:51 -0600
Date: Wed, 16 Jul 2003 18:50:51 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: VIvek Toky <vivektoky@ccpu.com>
Cc: Umesh Vats <umesh@ccpu.com>, SIGTRAN e-mail list <SIGTRAN@ietf.org>
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Message-ID: <20030716185051.D6725@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: VIvek Toky <vivektoky@ccpu.com>,
	Umesh Vats <umesh@ccpu.com>, SIGTRAN e-mail list <SIGTRAN@ietf.org>
References: <20030715232539.A17550@openss7.org> <HKEBKFBNPIDALDOKLGKDIEICCDAA.vivektoky@ccpu.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <HKEBKFBNPIDALDOKLGKDIEICCDAA.vivektoky@ccpu.com>; from vivektoky@ccpu.com on Wed, Jul 16, 2003 at 04:52:37PM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6H0oqP06520
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

VIvek,

Except the purpose of "Error - Cannot Support Unique Routing" was for
overlapping routing keys.  In the case where the requested RK overlaps wi=
th a
number of RKs, there are multiple existing RC values.  Which one would it
return?  Also, because there are overlapping RKs does not mean that the
requesting ASP is registered for any of them.  In that case any of the
overlapping RCs included would have either no meaning or could be poorly
interpreted by the ASP.  That is if one returns an RC value to which the =
ASP
is not registered and the ASP send DEREG REQ, it will generate another er=
ror.
Which is bad.  No, if the RC is actually registered to the ASP and usable=
 by
the ASP, the return should be Success.  And this is what I proposed.  Tha=
t if
the LRK-Id is the same as the LRK-Id that the ASP is registered against t=
he RK
and the RK matches, then the return would be Success.  In the overlapping
case, or where the RK matches and the LRK-Id does not match, the result w=
ould
be "Error - Cannot Support Unique Routing".

Using the LRK-Id serves both needs.

Another possibility would be to return a different error code for a match=
ing
already registered RK.  Such as "Error - Already Registered" in which cas=
e the
RC could be returned insted of 0 as you suggest.  This would give the ASP
enough information to either start using the registered RC, or use the RC=
 in
a DEREG REQ.  That text would look something like:

     12             Error - Routing Key Already Registered

  If the SGP determines that the received RK is already registered, the
  SGP returns a Registration Response message to the ASP, containing a
  Regsistration Result "Error - Routing Key Already Registered" and
  includes the Routing Context value of the existing Routing Key in the
  Registration Response.

We would also have to change the passage that says that RC is always
zero for outher than "Successful" to other than "Successful" or "Error -
Routing Key Already Registered".

This would also permit an ASP to audit existing registrations by sending
a REG REQ for an existing RK and comparing the returned RC.

--brian

On Wed, 16 Jul 2003, VIvek Toky wrote:

> Brian,
>   If the same ASP is serving multiple AS it can also cross check the
> uniqueness of the routing Keys among those AS. That way we don't even n=
eed
> to send the REG Request messages to a peer node.
>   Even if multiple ASs (in same ASP) manage to send REG request message=
s for
> same routing keys, SGP can anyway return Response message to the ASP,
> containing a Registration Result "Error - Cannot Support Unique Routing=
" but
> can also send RC related to the routing key already registered. This wo=
uld
> help implementations to recover from data inconsistencies (if present),
> without stopping all the traffic for that ASP.
>=20
> -Vivek
>=20
>=20

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Wed Jul 16 20:59:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA24305
	for <sigtran-archive@odin.ietf.org>; Wed, 16 Jul 2003 20:59:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cx6r-0001sM-Ts
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 20:59:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6H0x1YQ007205
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 20:59:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cx6r-0001ry-8c; Wed, 16 Jul 2003 20:59:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cx6J-0001re-Pt
	for sigtran@optimus.ietf.org; Wed, 16 Jul 2003 20:58:27 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA24286
	for <sigtran@ietf.org>; Wed, 16 Jul 2003 20:58:23 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cx6H-0005IU-00
	for sigtran@ietf.org; Wed, 16 Jul 2003 20:58:25 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[5SpqwPm6jsX82OUEOu0r7C4N9Ef4syqi])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cx61-0005IR-00
	for sigtran@ietf.org; Wed, 16 Jul 2003 20:58:09 -0400
Received: from ns.pigworks.openss7.net (IDENT:5LK0HqHuCl/lH4BnpcB367Bofrenlxs6@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6H0vwP06619;
	Wed, 16 Jul 2003 18:57:58 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6H0vwo07725;
	Wed, 16 Jul 2003 18:57:58 -0600
Date: Wed, 16 Jul 2003 18:57:58 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Andrew Booth <abooth@pt.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - T6
Message-ID: <20030716185758.E6725@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Andrew Booth <abooth@pt.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030712002921.A21136@openss7.org> <3F15CD78.6020209@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: <3F15CD78.6020209@pt.com>; from abooth@pt.com on Wed, Jul 16, 2003 at 06:11:04PM -0400
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, 16 Jul 2003 18:11:04)
> Hi Brian,
> 
> Comments inline.
> 
> 
> 
> Brian F. G. Bidulock wrote:
> 
> >M2PA-09 says:
> >
> >  
> >
> >>M2PA SHALL continue transmitting and acknowledging messages while it
> >>is in receive congestion.
> >>    
> >>
> >
> >Q.703 says:
> >
> >| The receiving level 2 also withholds acknowledgement of the message signal
> >| unit, which triggered off the congestion detection, and of message signal
> >| units received during the congestion situation; that is fill-in signal units
> >| or message signal units are sent as usual, but with the backward sequence
> >| number and backwards indicator bit assigned the values which are contained
> >| in the last transmitted signal unit before the congestion is recognized.
> >
> >It is rather important that the busy end not acknowledge messages while it is
> >in receive congestion.  This is because if the receive congestion results in
> >link failure, acknowledged messages caught at the congested receiver will be
> >unavailable for retransmission on an alternate link at the transmitter.
> >
> Are you assuming here that the Busy end is not delivering messages to MTP3?

Not really assuming.  Q.703 only sends SIB when it is in Congestion Discard.
Meaning that it is receive congested to the point that it is discarding
messages.  So I suppose the discarded messages cannot be delivered to MTP3.
Messages in queue (causing the congestion) are still delivered.

Also, Q.703 does not stop sending SIB until there is no receive congestion.
So the receive side passes through Congestion Accept to No Congestion before
it will stop SIBing.  This is the necessary hysteresis to avoid congestion
thrashing.

-X-snip-X-
> >>    
> >>
> >
> >But T6 is not necessarily running during the busy condition.
> >
> I don't understand this comment.  T6 is started when LSB is received and 
> stopped when LSBE is received.  Doesn't that mean that T6 is always 
> running if the peer is busy?

Q.703 on "first reception of status indication "B" starts a longer supervision
timer T6 (see 12.3) only if there are message signal units in the
retransmission buffer".  That is, T6 is only running if the remote is in
receive congestion and the local side has unacknowledged messages.  There is
the case where the remote is congested and there are no unacknowledged
messages. (The last acknowledge message triggered the receive congestion
discard situation.)  In this latter case, Q.703, T1.111.3 do not start T6.

So it is not true that T6 is always running when the peer is in receive
congestion.  In some cases the peer has send BUSY and T6 is not running.

--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 exim@www1.ietf.org  Wed Jul 16 21:11:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA24570
	for <sigtran-archive@odin.ietf.org>; Wed, 16 Jul 2003 21:11:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cxIV-0002Zz-9W
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 21:11:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6H1B3C0009909
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 21:11:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cxIU-0002Za-5q; Wed, 16 Jul 2003 21:11:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cxHo-0002ZJ-4j
	for sigtran@optimus.ietf.org; Wed, 16 Jul 2003 21:10:20 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA24559
	for <SIGTRAN@ietf.org>; Wed, 16 Jul 2003 21:10:15 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cxHl-0005Nt-00
	for SIGTRAN@ietf.org; Wed, 16 Jul 2003 21:10:17 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19cxHa-0005No-00
	for SIGTRAN@ietf.org; Wed, 16 Jul 2003 21:10:07 -0400
Received: from [172.16.1.40] (helo=ccpu00251)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19cxH3-00042P-JY; Wed, 16 Jul 2003 18:09:33 -0700
Reply-To: <umesh@ccpu.com>
From: "Umesh Vats" <umesh@ccpu.com>
To: <bidulock@openss7.org>, "VIvek Toky" <vivektoky@ccpu.com>
Cc: "SIGTRAN e-mail list" <SIGTRAN@ietf.org>
Subject: RE: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Date: Wed, 16 Jul 2003 18:15:14 -0700
Message-ID: <ACEMIJONPBBPDLOKCABHKEEOCBAA.umesh@ccpu.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0)
In-Reply-To: <20030716185051.D6725@openss7.org>
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id VAA24560
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Brian,

"Another possibility would be to return a different error code for a
matching
already registered RK.  Such as "Error - Already Registered" in which cas=
e
the
RC could be returned insted of 0 as you suggest."

This looks like a good idea. The new error code in REG-RSP along with RC =
can
be
interpreted more clearly and will be more useful for the ASPs.

Now "Error - Cannot Support Unique Routing" can be returned only for
overlapping
routing keys and not for the same RKs.

Umesh

-----Original Message-----
From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
Sent: Wednesday, July 16, 2003 5:51 PM
To: VIvek Toky
Cc: Umesh Vats; SIGTRAN e-mail list
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP


VIvek,

Except the purpose of "Error - Cannot Support Unique Routing" was for
overlapping routing keys.  In the case where the requested RK overlaps wi=
th
a
number of RKs, there are multiple existing RC values.  Which one would it
return?  Also, because there are overlapping RKs does not mean that the
requesting ASP is registered for any of them.  In that case any of the
overlapping RCs included would have either no meaning or could be poorly
interpreted by the ASP.  That is if one returns an RC value to which the =
ASP
is not registered and the ASP send DEREG REQ, it will generate another
error.
Which is bad.  No, if the RC is actually registered to the ASP and usable=
 by
the ASP, the return should be Success.  And this is what I proposed.  Tha=
t
if
the LRK-Id is the same as the LRK-Id that the ASP is registered against t=
he
RK
and the RK matches, then the return would be Success.  In the overlapping
case, or where the RK matches and the LRK-Id does not match, the result
would
be "Error - Cannot Support Unique Routing".

Using the LRK-Id serves both needs.

Another possibility would be to return a different error code for a match=
ing
already registered RK.  Such as "Error - Already Registered" in which cas=
e
the
RC could be returned insted of 0 as you suggest.  This would give the ASP
enough information to either start using the registered RC, or use the RC=
 in
a DEREG REQ.  That text would look something like:

     12             Error - Routing Key Already Registered

  If the SGP determines that the received RK is already registered, the
  SGP returns a Registration Response message to the ASP, containing a
  Regsistration Result "Error - Routing Key Already Registered" and
  includes the Routing Context value of the existing Routing Key in the
  Registration Response.

We would also have to change the passage that says that RC is always
zero for outher than "Successful" to other than "Successful" or "Error -
Routing Key Already Registered".

This would also permit an ASP to audit existing registrations by sending
a REG REQ for an existing RK and comparing the returned RC.

--brian

On Wed, 16 Jul 2003, VIvek Toky wrote:

> Brian,
>   If the same ASP is serving multiple AS it can also cross check the
> uniqueness of the routing Keys among those AS. That way we don't even n=
eed
> to send the REG Request messages to a peer node.
>   Even if multiple ASs (in same ASP) manage to send REG request message=
s
for
> same routing keys, SGP can anyway return Response message to the ASP,
> containing a Registration Result "Error - Cannot Support Unique Routing=
"
but
> can also send RC related to the routing key already registered. This wo=
uld
> help implementations to recover from data inconsistencies (if present),
> without stopping all the traffic for that ASP.
>
> -Vivek
>
>

--
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003


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



From exim@www1.ietf.org  Wed Jul 16 21:13:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA24614
	for <sigtran-archive@odin.ietf.org>; Wed, 16 Jul 2003 21:13:26 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cxKP-0002mC-6Z
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 21:13:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6H1D12t010665
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 21:13:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cxKO-0002ll-Vn; Wed, 16 Jul 2003 21:13:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cxJc-0002lG-Ro
	for sigtran@optimus.ietf.org; Wed, 16 Jul 2003 21:12:12 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA24593
	for <sigtran@ietf.org>; Wed, 16 Jul 2003 21:12:08 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cxJa-0005OK-00
	for sigtran@ietf.org; Wed, 16 Jul 2003 21:12:10 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[JDLTzcY0Hh9wIepDpk7UXj0nrIo0pyKV])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cxJK-0005OE-00
	for sigtran@ietf.org; Wed, 16 Jul 2003 21:11:54 -0400
Received: from ns.pigworks.openss7.net (IDENT:2VqOsdyp4mrGx61xnmnO9J31Olt6/V7T@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6H1BiP06954;
	Wed, 16 Jul 2003 19:11:44 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6H1BhE08100;
	Wed, 16 Jul 2003 19:11:43 -0600
Date: Wed, 16 Jul 2003 19:11:43 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Andrew Booth <abooth@pt.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - RPO
Message-ID: <20030716191143.F6725@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Andrew Booth <abooth@pt.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030712170744.A11077@openss7.org> <3F15C7F2.6070704@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: <3F15C7F2.6070704@pt.com>; from abooth@pt.com on Wed, Jul 16, 2003 at 05:47:30PM -0400
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,

Yes, we discussed this.  The proposal was to permit both situations.

If you have a MTP3 that will not accept MSUs after RPO you have another
problem: messages send by the remote before sending LSPO, but which arrived
at the peer after the LSPO because the LPSO was received (and supposedly
expedited) on stream 0.

To accomodate such implementations of MTP3 (completely valid implementations
from the SS7 standards perspective), it would be a better idea to send LSPO
on stream 1.  Then LSPR also should be sent on stream 1, otherwise there is
the possibiliy that User Data sent after sending LSPR will arrive before the
LSPR.

The side experiencing PO should accept the acknowledgement of any message
sent before or during PO.  This permits the TB/RTB to drain and has no
interaction with L3.

The statement below really says that the side experienceing RPO SHOULD
continue to acknowledge menssages received and accepted by MTP3.  If MTP3
will not accept messages during the RPO condition, then of course they 
cannot be acknowledged.  However, if MTP3 is accepting messages during
RPO (as some equally valid implementations do), then they SHOULD be
acknowledged.

This leaves both options open.  The current M2PA-09 text does not leave the
option open to continue delivering to MTP3 during RPO if MTP3 is still ready
to accept messages.

This also allows the M2PA working with an MTP3 that accepts MSUs during RPO
to not look in the sequence numbers of LSPO and LSPR sent on stream 0 to
determine which received messages preceded or followed the RPO condition and
which were between LSPO and LSPR (becaues they are all delivered to MTP3).

Make sense?

--brian


Andrew Booth wrote:                                                                 (Wed, 16 Jul 2003 17:47:30)
> Hi Brian,
> 
> If the side in RPO is sending data acknowledgments then it must also be 
> delivering these messages to MTP3.  Is there some guarantee that MTP3 
> will accept messages from a link that it has marked as blocked?  Or am I 
> misunderstanding the intent of your proposal?
> 
> Andrew
> 
> Brian F. G. Bidulock wrote:
> 
> >M2PA-09 says:
> >
> > While there is a remote processor outage condition:
> >
> >   (a) Any User Data messages received from the peer after the Link
> >       Status Processor Outage MUST NOT be acknowledged and MUST be
> >       buffered.
> >
> >   (b) M2PA SHOULD continue to acknowledge User Data messages received
> >       and accepted by MTP3 before the remote processor outage.
> >
> >In Q.703 the side experiencing PO cannot send MSUs or FISUs without indicating a
> >recovery from processor outage.  M2PA can send User Data without indicating
> >processor recovery.  There is no reason for the receiver of User Data not to
> >process it even after receiving Link Status Processor Outage.
> >
> >Otherwise, if (a) above is followed, MSUs in the transmission and retransmission
> >buffers at the time that local processor outage occured will be trapped.  If
> >User Data is acknowledged by the peer, they will be processed by the receiver.
> >
> >Change M2PA to say:
> >
> > While there is a remote processour outage condition:
> >
> >   (a) M2PA SHOULD continue to acknowledge User Data messages received
> >       and accepted by MTP3 regardless of the remote processor outage.
> >
> >This allows the TB/RTB on the side experiencing processor outage to drain.
> >
> >
> >M2PA-09 says:
> >
> >  Note that because the Link Status and User Data messages are sent on
> >  separate streams, it is possible for a User Data message sent after
> >  the Link Status Processor Recovered message to arrive before
> >  it. Therefore, when flushing buffers, M2PA must take care not to flush
> >  messages sent after the remote processor recovered. M2PA SHOULD NOT
> >  flush User Data messages with FSN greater than the FSN in the Link
> >  Status Processor Recovered message. 
> >
> >MTP3 does not flush the RB after RPO or before RPR (just the RTB), unless it
> >uses the remote processor outage to local processor outage latching option, and
> >then it is only because of the LPO condition not the RPO condition.
> >
> >With the change above, M2PA receives and acknowledges User Data before, during
> >and after remote processor outage, and comparison to any sequence numbers in the
> >Link Status Processor Recovered is unnecessary.
> >
> >M2PA-09 says:
> >
> >  When the local processor outage condition ends, M2PA SHALL send a Link
> >  Status Processor Recovered message to its peer. This message is used
> >  to signal in the end of the processor outage condition, instead of an
> >  MSU or FISU as in MTP2. The BSN value in this message is used by the
> >  peer to resynchronize its sequence numbers, if this is required by
> >  MTP2.
> >
> >After a flush buffers operation, the side that experienced PO uses the first
> >received FISU/MSU to resync and discards the FISU/MSU.
> >
> >ANSI T1.111.3/2000 Section 8 says:
> >
> >  ... If level 3 requests level 2 [at the side experiencing processor outage] to
> >  clear its buffers of message signal units, level 2 begins to continuously
> >  transmit fill-in signal units and discards all message signal units that it
> >  has received from remote level 2 but not yet delivered to local level 3 and
> >  all message signal units that it has received from local level 3 but not yet
> >  transmitted on the link.  The first fill-in signal unit transmitted
> >  acknowledges all message signal units that have been accepted from remote
> >  level 2 including any mesage signal units for which acknowledgements were
> >  previously withheld because of level 2 congestion (see section 9).  Level 2
> >  then begins accepting message signal units from remote Level 2. When the next
> >  fill-in or message signal unit is received that passes the acceptance
> >  procedure of section 4 and sections 5 or 6,
> >
> >   (1) all message signal units that are available for retransmission are
> >       discarded [clear RTB].
> >
> >   (2) the forward sequence number and forward indicator bit of the
> >       transmitted fill-in signal units are set equal to the backward sequence
> >       number and backward indicator bit of the received signal unit,
> >
> >   (3) normal transmission of message signal units and fill-in signal units is
> >       resumed, and
> >
> >   (4) level 3 is notified that the buffer clearing has been completed and
> >       that normal operations has been resumed.
> >
> >So, you see, the side experiencing the processor outage, after clearing buffers,
> >resynchronizes its FSN with the BSN of the first received FISU or MSU.  The side
> >experiencing processor outage could use the first User Data message received
> >after processor recovery to resyncronize.  However, if the peer has nothing to
> >send, processor recovery will be delayed.
> >
> >Note, however, that resynchronization of sequence numbers for the transmit
> >direction is necessary because MTP2 traps messages in the TB/RTB of the side
> >experiencing processor outage because it must continuously transmit SIPO during
> >processor outage.  Permitting the draining of the TB/RTB with the change above
> >removes the need to flush the TB/RTB on the side experiencing processor outage,
> >as well as removing the need to resynchronize sequence numbers.
> >
> >In the other direction, for messages queued to be sent to the side experiencing
> >processour outage, flushing and resynchronization is still necessary (actions
> >(1) to (4) per ANSI above).
> >
> >However, Q.703 permits FISUs/MSUs received by the side experiencing remote
> >processor outage to discard the first and several subsequent FISU/MSUs when
> >recovering.  There is even a test case in Q.781 (8.9) that shows an MSU being
> >discarded (and negative acknowledged) in this fashion.  Because M2PA cannot
> >discard and expect retransmission of a message, and because sychronization is
> >not required when the TB/RTB of the side experiencing processor outage is
> >permitted to drain with the changes above, M2PA cannot permit messages to be
> >discarded in this fashion.
> >
> >M2PA-09 says:
> >
> >  If M2PA receives a Flush command from MTP3, M2PA SHALL discard the
> >  incoming messages that were queued and unacknowledged during the
> >  processor outage condition. Furthermore, M2PA SHALL discard messages
> >  in the transmit and retransmit queues as required by MTP2. 
> >
> >To handle flushing and resynchronization properly in both directions, change the
> >above to read as follows:
> >
> >Change M2PA to say:
> >
> >  If M2PA receives a Flush command from MTP3, M2PA SHALL discard the
> >  incoming messages that were queued and unacknowledged during the
> >  processor outage condition.  Furthermore, the M2PA experiencing
> >  processor outage SHALL NOT discard message from the transmit and
> >  retransmit queues and SHALL NOT perform procedures to resynchronize
> >  sequence numbers, but SHALL continue normal transmission of User Data.
> >  The M2PA experiencing remote processor outage, SHALL discard messages in
> >  the transmit and retransmit queues as required by MTP2, and MAY perform
> >  resynchronization of sequence numbers using the BSN from the received
> >  Link Status Processor Recovered message.  However, during
> >  resychronization, M2PA SHALL NOT discard any received User Data
> >  messages.
> >
> >As Q.2140 handles processor outage by putting the link out of service, also add:
> >
> >  When M2PA experiences a local processor outage, it MAY put the link out
> >  of service by sending a Link Status Out of Service message if allowed by
> >  the applicable MTP2 standard (e.g, [Q.2140]).
> >
> >I think that wraps up processor outage problems.
> >
> >--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.
> 

-- 
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 exim@www1.ietf.org  Wed Jul 16 21:14:26 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA24669
	for <sigtran-archive@odin.ietf.org>; Wed, 16 Jul 2003 21:14:26 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cxLM-0002r7-SI
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 21:14:00 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6H1E0o2010965
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 21:14:00 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cxLM-0002qk-Nf; Wed, 16 Jul 2003 21:14:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cxKY-0002nl-J3
	for sigtran@optimus.ietf.org; Wed, 16 Jul 2003 21:13:10 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA24612
	for <SIGTRAN@ietf.org>; Wed, 16 Jul 2003 21:13:05 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cxKW-0005Ok-00
	for SIGTRAN@ietf.org; Wed, 16 Jul 2003 21:13:08 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[WR0sPyg3AG2bYh5kLEQfgAM+93LECiwB])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cxKF-0005Ob-00
	for SIGTRAN@ietf.org; Wed, 16 Jul 2003 21:12:52 -0400
Received: from ns.pigworks.openss7.net (IDENT:X/YyyCRETNU6qutxiIy2kPyfjbo6KbP6@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6H1CfP06965;
	Wed, 16 Jul 2003 19:12:41 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6H1Cf508107;
	Wed, 16 Jul 2003 19:12:41 -0600
Date: Wed, 16 Jul 2003 19:12:41 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Umesh Vats <umesh@ccpu.com>
Cc: VIvek Toky <vivektoky@ccpu.com>, SIGTRAN e-mail list <SIGTRAN@ietf.org>
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Message-ID: <20030716191241.G6725@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Umesh Vats <umesh@ccpu.com>,
	VIvek Toky <vivektoky@ccpu.com>,
	SIGTRAN e-mail list <SIGTRAN@ietf.org>
References: <20030716185051.D6725@openss7.org> <ACEMIJONPBBPDLOKCABHKEEOCBAA.umesh@ccpu.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <ACEMIJONPBBPDLOKCABHKEEOCBAA.umesh@ccpu.com>; from umesh@ccpu.com on Wed, Jul 16, 2003 at 06:15:14PM -0700
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6H1CfP06965
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Umesh,

Good. I think we are agreed then.

Other comments?

--brian

On Wed, 16 Jul 2003, Umesh Vats wrote:

> Brian,
>=20
> "Another possibility would be to return a different error code for a
> matching
> already registered RK.  Such as "Error - Already Registered" in which c=
ase
> the
> RC could be returned insted of 0 as you suggest."
>=20
> This looks like a good idea. The new error code in REG-RSP along with R=
C can
> be
> interpreted more clearly and will be more useful for the ASPs.
>=20
> Now "Error - Cannot Support Unique Routing" can be returned only for
> overlapping
> routing keys and not for the same RKs.
>=20
> Umesh
>=20
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Wednesday, July 16, 2003 5:51 PM
> To: VIvek Toky
> Cc: Umesh Vats; SIGTRAN e-mail list
> Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP
>=20
>=20
> VIvek,
>=20
> Except the purpose of "Error - Cannot Support Unique Routing" was for
> overlapping routing keys.  In the case where the requested RK overlaps =
with
> a
> number of RKs, there are multiple existing RC values.  Which one would =
it
> return?  Also, because there are overlapping RKs does not mean that the
> requesting ASP is registered for any of them.  In that case any of the
> overlapping RCs included would have either no meaning or could be poorl=
y
> interpreted by the ASP.  That is if one returns an RC value to which th=
e ASP
> is not registered and the ASP send DEREG REQ, it will generate another
> error.
> Which is bad.  No, if the RC is actually registered to the ASP and usab=
le by
> the ASP, the return should be Success.  And this is what I proposed.  T=
hat
> if
> the LRK-Id is the same as the LRK-Id that the ASP is registered against=
 the
> RK
> and the RK matches, then the return would be Success.  In the overlappi=
ng
> case, or where the RK matches and the LRK-Id does not match, the result
> would
> be "Error - Cannot Support Unique Routing".
>=20
> Using the LRK-Id serves both needs.
>=20
> Another possibility would be to return a different error code for a mat=
ching
> already registered RK.  Such as "Error - Already Registered" in which c=
ase
> the
> RC could be returned insted of 0 as you suggest.  This would give the A=
SP
> enough information to either start using the registered RC, or use the =
RC in
> a DEREG REQ.  That text would look something like:
>=20
>      12             Error - Routing Key Already Registered
>=20
>   If the SGP determines that the received RK is already registered, the
>   SGP returns a Registration Response message to the ASP, containing a
>   Regsistration Result "Error - Routing Key Already Registered" and
>   includes the Routing Context value of the existing Routing Key in the
>   Registration Response.
>=20
> We would also have to change the passage that says that RC is always
> zero for outher than "Successful" to other than "Successful" or "Error =
-
> Routing Key Already Registered".
>=20
> This would also permit an ASP to audit existing registrations by sendin=
g
> a REG REQ for an existing RK and comparing the returned RC.
>=20
> --brian
>=20
> On Wed, 16 Jul 2003, VIvek Toky wrote:
>=20
> > Brian,
> >   If the same ASP is serving multiple AS it can also cross check the
> > uniqueness of the routing Keys among those AS. That way we don't even=
 need
> > to send the REG Request messages to a peer node.
> >   Even if multiple ASs (in same ASP) manage to send REG request messa=
ges
> for
> > same routing keys, SGP can anyway return Response message to the ASP,
> > containing a Registration Result "Error - Cannot Support Unique Routi=
ng"
> but
> > can also send RC related to the routing key already registered. This =
would
> > help implementations to recover from data inconsistencies (if present=
),
> > without stopping all the traffic for that ASP.
> >
> > -Vivek
> >
> >
>=20
> --
> Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
> bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
> http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
>                         =A6 Therefore  all  progress  depends on the =A6
>                         =A6 unreasonable man. -- George Bernard Shaw =A6
>=20
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003
>=20
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.500 / Virus Database: 298 - Release Date: 7/10/2003

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Wed Jul 16 21:14:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA24684
	for <sigtran-archive@odin.ietf.org>; Wed, 16 Jul 2003 21:14:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cxLN-0002uP-JY
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 21:14:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6H1E1I1011164
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 21:14:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cxLN-0002t9-C5; Wed, 16 Jul 2003 21:14:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cxLK-0002qZ-DX
	for sigtran@optimus.ietf.org; Wed, 16 Jul 2003 21:13:58 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA24649
	for <SIGTRAN@ietf.org>; Wed, 16 Jul 2003 21:13:53 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cxLH-0005PH-00
	for SIGTRAN@ietf.org; Wed, 16 Jul 2003 21:13:55 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19cxL7-0005PB-00
	for SIGTRAN@ietf.org; Wed, 16 Jul 2003 21:13:45 -0400
Received: from [172.16.1.203] (helo=CCPU00186)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19cxKy-0004E1-S0; Wed, 16 Jul 2003 18:13:36 -0700
From: "VIvek Toky" <vivektoky@ccpu.com>
To: <bidulock@openss7.org>
Cc: "Umesh Vats" <umesh@ccpu.com>, "SIGTRAN e-mail list" <SIGTRAN@ietf.org>
Subject: RE: [Sigtran] M3UA : RC in unsuccesful REG-RSP
Date: Wed, 16 Jul 2003 18:18:20 -0700
Message-ID: <HKEBKFBNPIDALDOKLGKDIEIDCDAA.vivektoky@ccpu.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.6604 (9.0.2911.0)
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
In-Reply-To: <20030716185051.D6725@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id VAA24650
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Brian,
 I think new error code(Error - Routing Key Already Registered) is fine f=
or
me. Also this error code better explains the actual error.

thanks,
-Vivek.

-----Original Message-----
From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf Of
Brian F. G. Bidulock
Sent: Wednesday, July 16, 2003 5:51 PM
To: VIvek Toky
Cc: Umesh Vats; SIGTRAN e-mail list
Subject: Re: [Sigtran] M3UA : RC in unsuccesful REG-RSP


VIvek,

Except the purpose of "Error - Cannot Support Unique Routing" was for
overlapping routing keys.  In the case where the requested RK overlaps wi=
th
a
number of RKs, there are multiple existing RC values.  Which one would it
return?  Also, because there are overlapping RKs does not mean that the
requesting ASP is registered for any of them.  In that case any of the
overlapping RCs included would have either no meaning or could be poorly
interpreted by the ASP.  That is if one returns an RC value to which the =
ASP
is not registered and the ASP send DEREG REQ, it will generate another
error.
Which is bad.  No, if the RC is actually registered to the ASP and usable=
 by
the ASP, the return should be Success.  And this is what I proposed.  Tha=
t
if
the LRK-Id is the same as the LRK-Id that the ASP is registered against t=
he
RK
and the RK matches, then the return would be Success.  In the overlapping
case, or where the RK matches and the LRK-Id does not match, the result
would
be "Error - Cannot Support Unique Routing".

Using the LRK-Id serves both needs.

Another possibility would be to return a different error code for a match=
ing
already registered RK.  Such as "Error - Already Registered" in which cas=
e
the
RC could be returned insted of 0 as you suggest.  This would give the ASP
enough information to either start using the registered RC, or use the RC=
 in
a DEREG REQ.  That text would look something like:

     12             Error - Routing Key Already Registered

  If the SGP determines that the received RK is already registered, the
  SGP returns a Registration Response message to the ASP, containing a
  Regsistration Result "Error - Routing Key Already Registered" and
  includes the Routing Context value of the existing Routing Key in the
  Registration Response.

We would also have to change the passage that says that RC is always
zero for outher than "Successful" to other than "Successful" or "Error -
Routing Key Already Registered".

This would also permit an ASP to audit existing registrations by sending
a REG REQ for an existing RK and comparing the returned RC.

--brian

On Wed, 16 Jul 2003, VIvek Toky wrote:

> Brian,
>   If the same ASP is serving multiple AS it can also cross check the
> uniqueness of the routing Keys among those AS. That way we don't even n=
eed
> to send the REG Request messages to a peer node.
>   Even if multiple ASs (in same ASP) manage to send REG request message=
s
for
> same routing keys, SGP can anyway return Response message to the ASP,
> containing a Registration Result "Error - Cannot Support Unique Routing=
"
but
> can also send RC related to the routing key already registered. This wo=
uld
> help implementations to recover from data inconsistencies (if present),
> without stopping all the traffic for that ASP.
>
> -Vivek
>
>

--
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

_______________________________________________
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 exim@www1.ietf.org  Wed Jul 16 21:17:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA24794
	for <sigtran-archive@odin.ietf.org>; Wed, 16 Jul 2003 21:17:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cxOH-00039R-Ou
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 21:17:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6H1H1XN012099
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 21:17:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cxOH-000392-D8; Wed, 16 Jul 2003 21:17:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cxNl-00038o-Li
	for sigtran@optimus.ietf.org; Wed, 16 Jul 2003 21:16:29 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA24756
	for <sigtran@ietf.org>; Wed, 16 Jul 2003 21:16:25 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cxNj-0005T8-00
	for sigtran@ietf.org; Wed, 16 Jul 2003 21:16:27 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[1zjML6VrkuXiNpvoPIsvMGOs8C5OvKeO])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cxNT-0005T5-00
	for sigtran@ietf.org; Wed, 16 Jul 2003 21:16:11 -0400
Received: from ns.pigworks.openss7.net (IDENT:+RPz4aYFiJp5HHb0eZeFEseDqDGqb+z4@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6H1G1P07019;
	Wed, 16 Jul 2003 19:16:01 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6H1G0408186;
	Wed, 16 Jul 2003 19:16:00 -0600
Date: Wed, 16 Jul 2003 19:16:00 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: kakaul@hss.hns.com
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
Message-ID: <20030716191600.H6725@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: kakaul@hss.hns.com, sigtran@ietf.org
References: <OFAE52A198.69A2F46E-ON65256D65.00591206@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: <OFAE52A198.69A2F46E-ON65256D65.00591206@hss.hns.com>; from kakaul@hss.hns.com on Wed, Jul 16, 2003 at 09:57:11PM +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>

kakaul,

You are confusing congestion at an STP due to the conestion of signalling
links on a route with routeset congestion.  Although the congestion of a
signalling link can trigger sending TFC, the route upon which the TFC is
received has nothing to do with congestion.

Put an SEP where the ASP is and tell me where in the SS7 specs it says
that a TFC received from an STP is correlated to the route upon which it
was received.  A: it doesn't.

The passages you quote correspond to transmit congestion on the SCTP
associations between the ASP and the SG.  (SCON is not involved there.)

--brian

kakaul@hss.hns.com wrote:                                                           (Wed, 16 Jul 2003 21:57:11)
> 
> 
> 
> 
> Brian,
>    Comments inline
> 
> 
> 
> 
> "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/16/2003
> 12:41:40 PM
> 
> Please respond to bidulock@openss7.org
> 
> Sent by:    sigtran-admin@ietf.org
> 
> 
> To:    Kamesh Kaul/HSS@HSS
> cc:    sigtran@ietf.org
> 
> Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> 
> 
> kakaul,
> 
> An SS7 SEP at the position of the ASP, where the SGs are STPs would
> not maintain congestion state on a per-route (per-SG) basis.  So
> what makes you think that the ASP needs to?
> [Kamesh]It would because route-set congestion status will change depending
> on the derived congestion status of individual routes I am attaching
> section 11.2.4 from Q.704
> 
> 
> 11.2.4 Signalling route set congestion (National option with congestion
>        priorities)
> In the case when the congestion status of a signalling route set changes as
> a result of either the receipt of a transfer controlled message relating to
> a particular destination (see 13.7) or an indication of local signalling
> link congestion, or due to the signalling route-set-congestion-test
> procedure (see 13.9) an indication is given from the Message Transfer Part
> to the local level 4 informing it about the current congestion status of
> the signalling route set. Each user then takes appropriate actions in order
> to stop generation of signalling messages destined for the affected
> signalling point with congestion priorities lower than the specified
> congestion status. Messages received from the local level 4 with congestion
> priorities lower than the current signalling route set congestion status
> are discarded by the Message Transfer Part.
> Also, SCON() in this configuration corresponds to TFC (not MTP-STATUS)
> and therefore cannot signal abatement.  (TFC does not signal abatement,
> the lack of a TFC does.)  In the same ways, the ASP can use the lack
> of a SCON to indicate abatement.
> [Kamesh]Yes they do and derived congestion status leads to MTP-STATUS
> 
> So, a) SCON(1) and SCON(3) in your example are congestion signals
> (neither is an abatement signal); b) there is no need to maintain
> congestion
> status on a per-SG basis at the ASP ('cause the SS7 SEP doesn't).
> [Kamesh]Yes there is need to maintain congestion status on the per-SG basis
> at ASP because SG1 doing RCT procedure on the behalf of ASP may return
> SCON(0) when it receives audit for the affected point code and if affected
> point code is reachable through SG1 and SG2 as shown in previous figure SG2
> returns SCON(2) then congestion status sent in MTP-STATUS is 2 because
> affected point code is still congested.
> 
> This might lead back to the discustion why ASPs don't send RCT, but the
> simple answer is that each SG as STP can send RCT on all the ASP's behalf
> as it would for an "Alias" point code (ANSI T1.111/2000).
> 
> Because this is getting into a narrower and narrower corner of the
> specification, I think we are safe to strike the mention of the timer
> from the IG.
> [Kamesh]I dont agree with striking timer
> --brian
> 
> kakaul@hss.hns.com wrote:
> (Tue, 15 Jul 2003 18:59:37)
> >
> >
> >
> >
> > Brian
> >    some comments inline
> >
> >
> >
> >
> > "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/15/2003
> > 12:16:38 PM
> >
> > Please respond to bidulock@openss7.org
> >
> > Sent by:    sigtran-admin@ietf.org
> >
> >
> > To:    Kamesh Kaul/HSS@HSS
> > cc:    sigtran@ietf.org
> >
> > Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> >
> >
> > kakaul,
> >
> > If there is no interoperablity issue then it is implementation dependent.
> > A description of a procedure for the timer could be added as an
> > IMPLEMENTATION
> > NOTE making a suggestion to the implementor of one way to address local
> > issues
> > at the ASP, but it cannot be required.
> >
> > But looking at your description, this mechanism would be a duplication of
> > the
> > mechanism used by SCCP, ISUP, BICC, which use a dual-timer approach to
> > detecting congestion abatement.  Why would it be necessary for the ASP to
> > do
> > this as well?
> > >> It is also done in MTP3 for nation options with priority do you mean
> to
> > say that RCT mechanism is not required in MTP3.
> >
> > Also, you should not that SCON received from one SG is no different than
> > SCON
> > received from another.  The SCON represents a statistical sampling of
> > congestion in the SS7 network between the ASP and the destination, and is
> > not
> > referenced to an SG.  Therefore, the ASP should never perform this
> > procedure
> > on a per-SG basis.  The SCON received from either SG will generate an
> > MTP-STATUS primitive to the AS, and the AS will run suitable timers at
> the
> > MTP-User (where the responsibility lies to avoid sending messages of a
> > priority less that the congestion state).
> > >> In national option with priority MTP3 user is given MTP-STATUS
> > indication only if point code (routeset )congestion status
> > changes.Similarly at ASP MTP-STATUS indication shall be given to the user
> > part only if derived status of affected point code changes as shown in
> > figure below :
> >
> >
> >                               SCON (1)+--+----++
> >                              <------- |        |
> >                           +-----------+  SG1   +-----+
> >                           |           |        |     |
> >                           |           +--------+     |      SS7 NODE
> >            +--+----++     |                          |    +----------+
> >            |        |     |                          |    |   PC     |
> >            |  ASP   +-----+                          +----+  X-Y-Z   |
> >            |        |     |                          |    |          |
> >            +--------+     |                          |    +----------+
> >                           |          +--+----++      |
> >              AS:A-B-C     |          |        |      |
> >                           +----------+  SG2   +------+
> >                             <------- |        |
> >                              SCON(3) +--------+
> >
> > MTP-User will be given only MTP-STATUS indication with congestion status
> 3
> > if SCON(3) reaches before SCON(1).So for this you have to maintain
> > congestion status of affected point code through each SG and ultimately
> > given the Derived congestion status to MTP-User.
> >
> > Understand that SCON is not an M3UA congestion indication: it is an SS7
> > network congestion indication that is directed at the MTP-User.  The ASP
> > need
> > only translate SCON messages to MTP-STATUS primitives to the AS to which
> > they
> > correspond.  A AS that conforms to MTP-User standards specifications with
> > deal
> > with the MTP-STATUS primitives appropriately.
> > >>It is only true in international and national networks without priority
> > option.
> >
> > It is also important to understand that the ASP should not change the
> > routing
> > of messages on the basis of a received SCON message.  Where applicable
> and
> > for
> > applicable SS7 standards specifications, the SG will send DRST under very
> > specific and standardized congestion situations.  The ASP is already
> > expected
> > to perform some functions of MTP routing in response to DRST.
> > >> Here i definitely agree with you that rerouting of messages should not
> > be done at ASP to avoid missequencing.
> >
> > So, now I don't understand why you expect the ASP to run this timer at
> all.
> > >> Some implementations may want to provision congestion of affected
> point
> > code through supervisory congestion timer, they should be allowed to do
> so.
> >
> > Please explain.
> >
> > --brian
> >
> > kakaul@hss.hns.com wrote:
> > (Tue, 15 Jul 2003 10:46:03)
> > >
> > >
> > >
> > >
> > > Brian,
> > >    There is no interoperability issue involved in it and also i dont
> see
> > > any need of peer knowing the existence of this timer.It is local to ASP
> > and
> > > is used by ASP to supervise the congestion status of the affected point
> > > code reachable through a given SG.ASP after receiving an SCON message
> > with
> > > congestion status greater than 0 can start this timer and if no SCON
> > > message is received from peer till the expiry of congestion timer ASP
> can
> > > decrement affected point code  congestion status by 1 and start the
> timer
> > > provisioning of congestion again if congestion status of affected point
> > > code has not reached 0 or no congestion.
> > >
> > > Regards
> > > Kamesh Kaul
> > > www.hssworld.com
> > > Hughes Software Systems
> > >
> > >
> > >
> > >
> > > "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/15/2003
> > > 10:28:40 AM
> > >
> > > Please respond to bidulock@openss7.org
> > >
> > > Sent by:    sigtran-admin@ietf.org
> > >
> > >
> > > To:    Kamesh Kaul/HSS@HSS
> > > cc:    "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
> > >        sigtran@ietf.org
> > >
> > > Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> > >
> > >
> > > kakaul,
> > >
> > > Then add text describing how to use the timer in an interoperable way.
> > > Also, please describe why the peer needs to know about the existence
> > > of this timer to remain interoperable.  Otherwise, there would not be
> > > any need to describe it.
> > >
> > > --brian
> > >
> > >
> > > kakaul@hss.hns.com wrote:
> > > (Tue, 15 Jul 2003 09:35:27)
> > > >
> > > >
> > > >
> > > >
> > > > Hi Javier,
> > > >                     This timer may be used by some implementations to
> > > > supervise the congestion status of point code if SCON messages send
> by
> > > SGP
> > > > are lost in transit to ASP.It was discussed earlier and was agreed
> upon
> > > by
> > > > all to keep this timer.I disagree with removing the mention of the
> > timer.
> > > >
> > > > Regards
> > > > Kamesh Kaul
> > > > www.hssworld.com
> > > > Hughes Software Systems
> > > >
> > > >
> > > >
> > > >
> > > > "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>@ietf.org on
> > > > 07/14/2003 04:05:58 PM
> > > >
> > > > Sent by:    sigtran-admin@ietf.org
> > > >
> > > >
> > > > To:    "'Brian F. G. Bidulock '" <bidulock@openss7.org>
> > > > cc:    "''sigtran@ietf.org' '" <sigtran@ietf.org>
> > > >
> > > > Subject:    RE: [Sigtran] M3UA IG v04 Section 3.18
> > > >
> > > >
> > > > Brian,
> > > >
> > > > We'll remove the mention of the timer, if nobody disagrees...
> > > >
> > > > Cheers // Javier.
> > > >
> > > > -----Original Message-----
> > > > From: Brian F. G. Bidulock
> > > > To: J.Javier Pastor (EE/EEM)
> > > > Cc: 'sigtran@ietf.org'
> > > > Sent: 11/07/2003 11:07
> > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > > >
> > > > J.Javier,
> > > >
> > > > J.Javier Pastor (EE/EEM) wrote:
> > > > (Fri, 11 Jul 2003 09:56:59)
> > > > > Hi Brian,
> > > > >
> > > > > Pls, see below...
> > > > >
> > > > > > -----Original Message-----
> > > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > > Sent: jueves, 10 de julio de 2003 15:34
> > > > > > To: J.Javier Pastor (EE/EEM)
> > > > > > Cc: 'sigtran@ietf.org'
> > > > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > > > > >
> > > > > >
> > > > > > Javier,
> > > > > >
> > > > > > 3.18.3 Solution Description
> > > > > >
> > > > > >    Clarify that the ASP needs to indicate when the congestion
> > > > > > level has
> > > > > >    changed (including abatement).  Further, the ASP peer can
> > > > maintain
> > > > > >    a timer, if desired, in case the ASP fails to indicate
> > congestion
> > > > > >    abatement.
> > > > > >
> > > > > > Where are the procedures for this timer described?  This is the
> > only
> > > > place
> > > > > > that the timer is mentioned.  Please describe the timer or remove
> > > > the
> > > > > > sentence saying that it has been described...
> > > > > >
> > > > >
> > > > > Do you think is worth include the timer explanation into the new
> > text?
> > > > > Otherwise, we may remove that sentence.
> > > >
> > > > Please remove mention of the timer.
> > > >
> > > > --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
> > >
> >
> > --
> > 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 exim@www1.ietf.org  Wed Jul 16 21:19:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA24850
	for <sigtran-archive@odin.ietf.org>; Wed, 16 Jul 2003 21:19:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cxQD-0003F0-C2
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 21:19:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6H1J1f0012436
	for sigtran-archive@odin.ietf.org; Wed, 16 Jul 2003 21:19:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cxQD-0003EU-8c; Wed, 16 Jul 2003 21:19:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19cxPh-0003E7-Iq
	for sigtran@optimus.ietf.org; Wed, 16 Jul 2003 21:18:29 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA24824
	for <sigtran@ietf.org>; Wed, 16 Jul 2003 21:18:24 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cxPe-0005UT-00
	for sigtran@ietf.org; Wed, 16 Jul 2003 21:18:27 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[oHtQeaRLak6n0GiCPTlcGW064CFsWeh9])
	by ietf-mx with esmtp (Exim 4.12)
	id 19cxPO-0005UI-00
	for sigtran@ietf.org; Wed, 16 Jul 2003 21:18:10 -0400
Received: from ns.pigworks.openss7.net (IDENT:1Wu4JRzC/tpor4QtjtacmjeeF8jLlGk2@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6H1HmP07043;
	Wed, 16 Jul 2003 19:17:48 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6H1Hmh08269;
	Wed, 16 Jul 2003 19:17:48 -0600
Date: Wed, 16 Jul 2003 19:17:48 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Andrew Booth <abooth@pt.com>
Cc: sigtran@ietf.org, tom.george@alcatel.com
Subject: Re: [Sigtran] M2PA last call comment - ANSI primitives for LPO
Message-ID: <20030716191748.I6725@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Andrew Booth <abooth@pt.com>, sigtran@ietf.org,
	tom.george@alcatel.com
References: <3F15CB44.1010204@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: <3F15CB44.1010204@pt.com>; from abooth@pt.com on Wed, Jul 16, 2003 at 06:01:40PM -0400
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,

Either that or say "such as" or "e.g." in front of the ITU primitives.
Or, say something more generic instead of Continue or Resume.

--brian

Andrew Booth wrote:                                                                 (Wed, 16 Jul 2003 18:01:40)
> Hi Tom,
> 
> The ANSI primitives for exiting processor outage seem to be Resume, 
> Clear Buffers and Clear RTB.  The corresponding ITU primitives 
> (continue, flush, ... ) are described explicitly in M2PA draft 9. 
>  Should the ANSI primitives also be described explicitly?
> 
> Andrew
> 
> 
> 
> _______________________________________________
> 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 exim@www1.ietf.org  Thu Jul 17 17:23:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA23157
	for <sigtran-archive@odin.ietf.org>; Thu, 17 Jul 2003 17:23:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dGDR-0000VW-Iq
	for sigtran-archive@odin.ietf.org; Thu, 17 Jul 2003 17:23:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6HLN5EQ001946
	for sigtran-archive@odin.ietf.org; Thu, 17 Jul 2003 17:23:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dGDN-0000V9-BD; Thu, 17 Jul 2003 17:23:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dGCQ-0000Tx-Oq
	for sigtran@optimus.ietf.org; Thu, 17 Jul 2003 17:22:02 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA23118
	for <SIGTRAN@IETF.org>; Thu, 17 Jul 2003 17:21:57 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dGCO-0001Vl-00
	for SIGTRAN@IETF.org; Thu, 17 Jul 2003 17:22:00 -0400
Received: from auds953.usa.alcatel.com ([143.209.238.6])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dGCD-0001VK-00
	for SIGTRAN@IETF.org; Thu, 17 Jul 2003 17:21:49 -0400
Received: from alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6HLKqLX016299;
	Thu, 17 Jul 2003 16:20:53 -0500 (CDT)
Message-ID: <3F171334.6070308@alcatel.com>
Date: Thu, 17 Jul 2003 16:20:52 -0500
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
User-Agent: Mozilla/5.0 (X11; U; SunOS sun4u; en-US; rv:1.0.1) Gecko/20020920 Netscape/7.0
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: SIGTRAN e-mail list <SIGTRAN@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - T1 Editorial
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030711224130.A20278@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

Brian,

I don't have a problem with your suggested paragraph, except that is a 
bit wordy. I put in the qualification "in the case where MTP2 would send 
a FISU after proving is complete" to cover the cases you mention. Send 
LS Ready when MTP2 would send the FISU. Elsewhere in the document it 
says to send LS PO where MTP2 sends SIPO, and User Data where MTP2 sends 
an MSU.

But the judge of readability is the reader. Let me know if you still 
think I should change it.

tomG.

Brian F. G. Bidulock wrote:
> I have been updating the M2PA-TEST spec and have noticed a few things in
> M2PA-09...
> 
> M2PA-09 Says: (top of Page 27):
> 
>   The Link Status Ready message replaces the FISU of MTP2 that is sent
>   at the end of the proving period. The Link Status Ready message is
>   used to verify that both ends have completed proving. When M2PA starts
>   timer T1, it SHALL send a Link Status Ready message to its peer in the
>   case where MTP2 would send a FISU after proving is complete. M2PA MAY
>   send additional Link Status Ready messages while timer T1 is running.
> 
> This is all that is mention on T1.
> 
> In MTP2 T1 is started when the state machine enters the "Aligned Ready"
> or "Aligned Not Ready" state.  Only from the "Aligned Ready" state does
> MTP2 send FISU.  From "Aligned Not Ready" state, it sends SIPO.  Also,
> if data is pending transmission, MTP2 can enter the "Aligned Ready" state
> by sending an MSU rather than a FISU.
> 
> The problem is that M2PA-09 says "When M2PA starts timer T1, it SHALL send
> a Link Status message...", when in fact if M2PA follows Q.703 could start
> T1 and send a DATA message, or could start T1 and send a PROCESSOR-OUTAGE
> message.
> 
> Perhaps we could say:
> 
>   The Link Status Ready message replaces the FISU, the Data message replaces
>   the MSU, and the Link Status Processor Outage message replaces the SIPO,
>   that would be sent at the end of the proving period.  The Link Status Ready
>   Data or Link Status Processor Outage message is used to verify that both
>   ends have completed proving.  Whem M2PA starts timer T1, it SHALL send a
>   Link Status Ready, Data or Link Status Processor Outage message to is peer
>   in the case where MTP2 would send a FISU, MSU or SIPO ater proving is
>   complete.  M2PA MAY send additional Link Status Ready, Data or Link Status
>   Processor Outage messages while timer T1 is running.
> 
> Ok?
> 
> --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.  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 exim@www1.ietf.org  Thu Jul 17 18:18:26 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA25423
	for <sigtran-archive@odin.ietf.org>; Thu, 17 Jul 2003 18:18:26 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dH4c-0002WK-7V
	for sigtran-archive@odin.ietf.org; Thu, 17 Jul 2003 18:18:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6HMI2w8009655
	for sigtran-archive@odin.ietf.org; Thu, 17 Jul 2003 18:18:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dH4b-0002VK-Hj; Thu, 17 Jul 2003 18:18:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dH4O-0002Uv-8P
	for sigtran@optimus.ietf.org; Thu, 17 Jul 2003 18:17:48 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA25400
	for <sigtran@ietf.org>; Thu, 17 Jul 2003 18:17:37 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dH4G-0001ps-00
	for sigtran@ietf.org; Thu, 17 Jul 2003 18:17:40 -0400
Received: from auds953.usa.alcatel.com ([143.209.238.6])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dH45-0001pa-00
	for sigtran@ietf.org; Thu, 17 Jul 2003 18:17:29 -0400
Received: from alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6HMGsLX022697;
	Thu, 17 Jul 2003 17:16:54 -0500 (CDT)
Message-ID: <3F172056.9080909@alcatel.com>
Date: Thu, 17 Jul 2003 17:16:54 -0500
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
User-Agent: Mozilla/5.0 (X11; U; SunOS sun4u; en-US; rv:1.0.1) Gecko/20020920 Netscape/7.0
X-Accept-Language: en-us, en
MIME-Version: 1.0
CC: IETF sigtran <sigtran@ietf.org>
Subject: Re: FW: [Sigtran] M2PA Last Call Comment
References: <D73559B38D51D4119CB800508B9527B505AC1965@mchh266e.mchh.siemens.de> <20030714034015.A13132@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

Brian,

In the non-IP realm, is it possible for a Q.703 to talk to a Q.2140? 
What would be done in the scenario you described?

tomG.

Brian F. G. Bidulock wrote:
> Hanns Juergen,
> 
> Consider that one side follows Q.703 and requires proving while the
> other side follows Q.2140 and foregoes proving.  The following will
> occur:
   [...]


-- 
* 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.  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 exim@www1.ietf.org  Thu Jul 17 19:34:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA26990
	for <sigtran-archive@odin.ietf.org>; Thu, 17 Jul 2003 19:34:33 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dIGC-00052j-Vf
	for sigtran-archive@odin.ietf.org; Thu, 17 Jul 2003 19:34:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6HNY4kj019385
	for sigtran-archive@odin.ietf.org; Thu, 17 Jul 2003 19:34:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dIG8-00052M-Np; Thu, 17 Jul 2003 19:34:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dIFl-000526-Ix
	for sigtran@optimus.ietf.org; Thu, 17 Jul 2003 19:33:38 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA26944
	for <sigtran@ietf.org>; Thu, 17 Jul 2003 19:33:34 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dIFk-0002Lg-00
	for sigtran@ietf.org; Thu, 17 Jul 2003 19:33:36 -0400
Received: from web41804.mail.yahoo.com ([66.218.93.138])
	by ietf-mx with smtp (Exim 4.12)
	id 19dIFZ-0002LR-00
	for sigtran@ietf.org; Thu, 17 Jul 2003 19:33:25 -0400
Message-ID: <20030717233229.55677.qmail@web41804.mail.yahoo.com>
Received: from [65.198.40.51] by web41804.mail.yahoo.com via HTTP; Thu, 17 Jul 2003 18:32:29 CDT
Date: Thu, 17 Jul 2003 18:32:29 -0500 (CDT)
From: =?iso-8859-1?q?Pablo=20Salinas?= <pablopesb@yahoo.com>
To: bidulock@openss7.org
Cc: sigtran@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="0-1379191101-1058484749=:54945"
Subject: [Sigtran] SDL and SDT parameteres for M2UA stack
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

--0-1379191101-1058484749=:54945
Content-Type: text/plain; charset=iso-8859-1
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id TAA26945
Content-Transfer-Encoding: quoted-printable

Brian,
        I couldn't find more info about SDL's and SDT's. So, may I ask yo=
u a few more questions:
1. Is this a SDL and SDT typical scenario?
SDLI's represent CIC's and SDTI's represent SSP's
2. From which SS7's protocol layer do SDLI and SDTI come from? I mean, wi=
ch layer is responsible for selecting this values?
3. Suppose I want to initiate a SGP and a MGC. The code for initializatio=
n in the SGP and the MGC should have to call this=20
functions:
a) initiate_protocol_stack -> Initiates the M2UA protocol stack.
x) add_sg -> Receives the SG identifier (for local managment only). This =
function should only be called in the MGC.
b) add_sgp -> Receives the list of IP adresses:Port number attached to th=
is SGP, the port number it should listen in order to=20
support dynamic registration (default:2904), the max and min number of st=
reams per sctp association and the SGP identifier=20
(for local managment only).
c) add_asp -> Receives the ASP identifier (for local managment only), the=
 max and min number of streams per sctp association,=20
the (local and remote)IP addresses:Port number (at the ASP and at the SGP=
) used for the sctp association (Example:=20
65.198.40.50:3005, 65.198.40.51:3005).
d) add_as -> Receives the AS identifier (for local managment only), the i=
nterface identifiers the AS responds to and the ASP=20
identifiers of the ASP's associated to this AS.

At what function call, should the SDLI and SDTI parameters be specified?
I believe they should be specified in the add_as function call.
If that is the case, how can the SG and the MGC know which Interface Iden=
tifier is associated with every AS, in order
send the MAUP messages with the rigth Interface Identifier parameter?
I am thinking about a scenario where SG and MGC support dynamic registrat=
ion; but at the begin there are pre-registered
ASP's and AS's.

               thanks again,
                            Pablo Salinas




---------------------------------
Do You Yahoo!?
Todo lo que quieres saber de Estados Unidos, Am=E9rica Latina y el resto =
del Mundo.
Vis=EDta Yahoo! Noticias.

--0-1379191101-1058484749=:54945
Content-Type: text/html; charset=iso-8859-1
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id TAA26945
Content-Transfer-Encoding: quoted-printable

<DIV>Brian,<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; I couldn't find=
 more info about SDL's and SDT's. So, may I ask you a few more questions:=
</DIV>
<DIV>1. Is this a SDL and SDT typical scenario?<BR>SDLI's represent CIC's=
 and SDTI's represent SSP's</DIV>
<DIV>2. From which SS7's protocol layer do SDLI and SDTI come from? I mea=
n, wich layer is responsible for selecting this values?</DIV>
<DIV>3. Suppose I want to initiate a SGP and a MGC. The code for initiali=
zation in the SGP and the MGC should have to call this </DIV>
<DIV>functions:</DIV>
<DIV>a) initiate_protocol_stack -&gt; Initiates the M2UA protocol stack.<=
/DIV>
<DIV>x) add_sg -&gt; Receives the SG identifier (for local managment only=
). This function should only be called in the MGC.</DIV>
<DIV>b) add_sgp -&gt; Receives the list of IP adresses:Port number attach=
ed to this SGP, the port number it should listen in order to </DIV>
<DIV>support dynamic registration (default:2904), the max and min number =
of streams per sctp association and the SGP identifier </DIV>
<DIV>(for local managment only).</DIV>
<DIV>c) add_asp -&gt; Receives the ASP identifier (for local managment on=
ly), the max and min number of streams per sctp association, </DIV>
<DIV>the (local and remote)IP addresses:Port number (at the ASP and at th=
e SGP) used for the sctp association (Example: </DIV>
<DIV>65.198.40.50:3005, 65.198.40.51:3005).</DIV>
<DIV>d) add_as -&gt; Receives the AS identifier (for local managment only=
), the interface identifiers the AS responds to and the ASP </DIV>
<DIV>identifiers of the ASP's associated to this AS.</DIV>
<DIV><BR>At what function call, should the SDLI and SDTI parameters be sp=
ecified?<BR>I believe they should be specified in the add_as function cal=
l.</DIV>
<DIV>If that is the case, how can the SG and the MGC know which Interface=
 Identifier is associated with every AS, in order<BR>send the MAUP messag=
es with the rigth Interface Identifier parameter?</DIV>
<DIV>I am thinking about a scenario where SG and MGC support dynamic regi=
stration; but at the begin there are pre-registered<BR>ASP's and AS's.</D=
IV>
<DIV><BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp; thanks again,<BR>&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; Pablo Salinas<BR><=
/DIV><p><br><hr size=3D1 noshade><b>Do You Yahoo!?</b><br>
<a href=3Dhttp://espanol.yahoo.com/mail_tagline/*http://espanol.news.yaho=
o.com><img src=3D"http://us.i1.yimg.com/us.yimg.com/i/not/dailynews.gif" =
width=3D30 height=3D26 border=3D0 align=3Dleft></a>
Todo lo que quieres saber de Estados Unidos, Am=E9rica Latina y el resto =
del Mundo.<br>
Vis=EDta <a href=3Dhttp://espanol.yahoo.com/mail_tagline/*http://espanol.=
news.yahoo.com>Yahoo! Noticias</a>.<br>
--0-1379191101-1058484749=:54945--

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



From exim@www1.ietf.org  Thu Jul 17 19:55:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA27571
	for <sigtran-archive@odin.ietf.org>; Thu, 17 Jul 2003 19:55:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dIaU-00068B-2E
	for sigtran-archive@odin.ietf.org; Thu, 17 Jul 2003 19:55:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6HNt2wi023563
	for sigtran-archive@odin.ietf.org; Thu, 17 Jul 2003 19:55:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dIaT-00067l-9F; Thu, 17 Jul 2003 19:55:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dIZl-00063b-LD
	for sigtran@optimus.ietf.org; Thu, 17 Jul 2003 19:54:17 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA27559
	for <SIGTRAN@ietf.org>; Thu, 17 Jul 2003 19:54:14 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dIZj-0002W0-00
	for SIGTRAN@ietf.org; Thu, 17 Jul 2003 19:54:15 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[lTpvCoEmiVG/y6aRCvwP7qeMSWkMtJjm])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dIZT-0002VE-00
	for SIGTRAN@ietf.org; Thu, 17 Jul 2003 19:54:00 -0400
Received: from ns.pigworks.openss7.net (IDENT:XkuN1eGNz1ZqdQ1gj57Pg7otFxfl0wG0@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6HNqNP30163;
	Thu, 17 Jul 2003 17:52:23 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6HNqNj29268;
	Thu, 17 Jul 2003 17:52:23 -0600
Date: Thu, 17 Jul 2003 17:52:23 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tom George <Tom.George@alcatel.com>
Cc: SIGTRAN e-mail list <SIGTRAN@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - T1 Editorial
Message-ID: <20030717175223.A29221@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tom George <Tom.George@alcatel.com>,
	SIGTRAN e-mail list <SIGTRAN@ietf.org>
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030711224130.A20278@openss7.org> <3F171334.6070308@alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3F171334.6070308@alcatel.com>; from Tom.George@alcatel.com on Thu, Jul 17, 2003 at 04:20:52PM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6HNqNP30163
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Tom,

Maybe change "would send" to "sends"???

Also, it may only send additional LSR where MTP2 sends additional FISU?

I think the intent is unchanged, maybe we can find clearer wording for th=
e
final.

--brian

On Thu, 17 Jul 2003, Tom George wrote:

> Brian,
>=20
> I don't have a problem with your suggested paragraph, except that is a=20
> bit wordy. I put in the qualification "in the case where MTP2 would sen=
d=20
> a FISU after proving is complete" to cover the cases you mention. Send=20
> LS Ready when MTP2 would send the FISU. Elsewhere in the document it=20
> says to send LS PO where MTP2 sends SIPO, and User Data where MTP2 send=
s=20
> an MSU.
>=20
> But the judge of readability is the reader. Let me know if you still=20
> think I should change it.
>=20
> tomG.
>=20
> Brian F. G. Bidulock wrote:
> > I have been updating the M2PA-TEST spec and have noticed a few things=
 in
> > M2PA-09...
> >=20
> > M2PA-09 Says: (top of Page 27):
> >=20
> >   The Link Status Ready message replaces the FISU of MTP2 that is sen=
t
> >   at the end of the proving period. The Link Status Ready message is
> >   used to verify that both ends have completed proving. When M2PA sta=
rts
> >   timer T1, it SHALL send a Link Status Ready message to its peer in =
the
> >   case where MTP2 would send a FISU after proving is complete. M2PA M=
AY
> >   send additional Link Status Ready messages while timer T1 is runnin=
g.
> >=20
> > This is all that is mention on T1.
> >=20
> > In MTP2 T1 is started when the state machine enters the "Aligned Read=
y"
> > or "Aligned Not Ready" state.  Only from the "Aligned Ready" state do=
es
> > MTP2 send FISU.  From "Aligned Not Ready" state, it sends SIPO.  Also=
,
> > if data is pending transmission, MTP2 can enter the "Aligned Ready" s=
tate
> > by sending an MSU rather than a FISU.
> >=20
> > The problem is that M2PA-09 says "When M2PA starts timer T1, it SHALL=
 send
> > a Link Status message...", when in fact if M2PA follows Q.703 could s=
tart
> > T1 and send a DATA message, or could start T1 and send a PROCESSOR-OU=
TAGE
> > message.
> >=20
> > Perhaps we could say:
> >=20
> >   The Link Status Ready message replaces the FISU, the Data message r=
eplaces
> >   the MSU, and the Link Status Processor Outage message replaces the =
SIPO,
> >   that would be sent at the end of the proving period.  The Link Stat=
us Ready
> >   Data or Link Status Processor Outage message is used to verify that=
 both
> >   ends have completed proving.  Whem M2PA starts timer T1, it SHALL s=
end a
> >   Link Status Ready, Data or Link Status Processor Outage message to =
is peer
> >   in the case where MTP2 would send a FISU, MSU or SIPO ater proving =
is
> >   complete.  M2PA MAY send additional Link Status Ready, Data or Link=
 Status
> >   Processor Outage messages while timer T1 is running.
> >=20
> > Ok?
> >=20
> > --brian
> >=20
>=20
>=20
> --=20
> * 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.  I do not purport to    *
> *   advertise or represent its position, operations, or activities.   *
>=20
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Thu Jul 17 19:56:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA27597
	for <sigtran-archive@odin.ietf.org>; Thu, 17 Jul 2003 19:56:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dIbR-0006DF-J0
	for sigtran-archive@odin.ietf.org; Thu, 17 Jul 2003 19:56:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6HNu1Q6023862
	for sigtran-archive@odin.ietf.org; Thu, 17 Jul 2003 19:56:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dIbR-0006Cm-Fl; Thu, 17 Jul 2003 19:56:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dIaq-00068c-VZ
	for sigtran@optimus.ietf.org; Thu, 17 Jul 2003 19:55:25 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA27569
	for <sigtran@ietf.org>; Thu, 17 Jul 2003 19:55:21 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dIap-0002WP-00
	for sigtran@ietf.org; Thu, 17 Jul 2003 19:55:23 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[qWg3g75TKjZDQsdsL40f8KV+V3T6A0h7])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dIaZ-0002WL-00
	for sigtran@ietf.org; Thu, 17 Jul 2003 19:55:07 -0400
Received: from ns.pigworks.openss7.net (IDENT:R4FSXAnXEXhb+gPDEtTgb15NdDeKkYf8@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6HNsuP30185;
	Thu, 17 Jul 2003 17:54:56 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6HNsux29286;
	Thu, 17 Jul 2003 17:54:56 -0600
Date: Thu, 17 Jul 2003 17:54:56 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tom George <Tom.George@alcatel.com>
Cc: IETF sigtran <sigtran@ietf.org>
Subject: Re: FW: [Sigtran] M2PA Last Call Comment
Message-ID: <20030717175456.B29221@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tom George <Tom.George@alcatel.com>,
	IETF sigtran <sigtran@ietf.org>
References: <D73559B38D51D4119CB800508B9527B505AC1965@mchh266e.mchh.siemens.de> <20030714034015.A13132@openss7.org> <3F172056.9080909@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: <3F172056.9080909@alcatel.com>; from Tom.George@alcatel.com on Thu, Jul 17, 2003 at 05:16:54PM -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>

Tom,

Q.703 does not give the option to talk to a Q.2140 link.  But it
appears that M2PA allows a Q.703-state-machine-based M2PA link to
talk to a Q.2140-state-machine-based M2PA link.  That is were the
problem ensues.

--brian

Tom George wrote:                                                                      (Thu, 17 Jul 2003 17:16:54)
> Brian,
> 
> In the non-IP realm, is it possible for a Q.703 to talk to a Q.2140? 
> What would be done in the scenario you described?
> 
> tomG.
> 
> Brian F. G. Bidulock wrote:
> > Hanns Juergen,
> > 
> > Consider that one side follows Q.703 and requires proving while the
> > other side follows Q.2140 and foregoes proving.  The following will
> > occur:
>    [...]
> 
> 
> -- 
> * 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.  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 exim@www1.ietf.org  Thu Jul 17 20:08:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA27846
	for <sigtran-archive@odin.ietf.org>; Thu, 17 Jul 2003 20:08:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dIn4-0006mt-Fr
	for sigtran-archive@odin.ietf.org; Thu, 17 Jul 2003 20:08:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6I082cB026085
	for sigtran-archive@odin.ietf.org; Thu, 17 Jul 2003 20:08:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dIn4-0006mU-0h; Thu, 17 Jul 2003 20:08:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dIma-0006i7-SH
	for sigtran@optimus.ietf.org; Thu, 17 Jul 2003 20:07:32 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA27840
	for <sigtran@ietf.org>; Thu, 17 Jul 2003 20:07:29 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dImY-0002cK-00
	for sigtran@ietf.org; Thu, 17 Jul 2003 20:07:30 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[Q44rEhLYB70fNTFTDTEA/+ufKuu0eBUo])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dImJ-0002cE-00
	for sigtran@ietf.org; Thu, 17 Jul 2003 20:07:15 -0400
Received: from ns.pigworks.openss7.net (IDENT:gKjN5QuzoXLcdjLpSd/Ribtjlll2zUTP@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6I075P30527;
	Thu, 17 Jul 2003 18:07:05 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6I074J29656;
	Thu, 17 Jul 2003 18:07:04 -0600
Date: Thu, 17 Jul 2003 18:07:04 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Pablo Salinas <pablopesb@yahoo.com>
Cc: sigtran@ietf.org
Message-ID: <20030717180704.C29221@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Pablo Salinas <pablopesb@yahoo.com>, sigtran@ietf.org
References: <20030717233229.55677.qmail@web41804.mail.yahoo.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20030717233229.55677.qmail@web41804.mail.yahoo.com>; from pablopesb@yahoo.com on Thu, Jul 17, 2003 at 06:32:29PM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6I075P30527
Content-Transfer-Encoding: quoted-printable
Subject: [Sigtran] Re: SDL and SDT parameteres for M2UA stack
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Pablo,

See Q.704 Clause 12 and particularly section 12.3, 12.4, 12.5 and 12.6.

Also see Q.751-1, section 7.1.7 (managed object class definition for a
Signalling Data Link Termination Point) and 7.1.14 (managed object class
definitionf for Signalling Terminal).

The signTermId attribute of the signTerm managed object class is the SDTI.

The signDataLinkTpId attribute of the signDataLink managed object class i=
s
the SDLI.

For a SG that does not support automatic allocation of signalling termina=
ls or
singalling data links (see 12.3/Q.704, 12.4/Q.704) the signDataLinkTp and
signTerm are bound at configuration.  Registration is then not necessary.
Each IID identifies a static signTerm/signDataLinkTp pair.  Registration
supports automatic allocation of signalling terminals or signalling data =
links
or both, by permitting, via registration, a signalling terminal (signTerm=
) to
be associated with a signalling data link (signDataLinkTp) and assigned a=
n
IID. =20

--brian

Pablo Salinas wrote:                                                     =
            (Thu, 17 Jul 2003 18:32:29)
>=20
>    Brian,
>            I couldn't find more info about SDL's and SDT's. So, may I a=
sk
>    you a few more questions:
>=20
>    1. Is this a SDL and SDT typical scenario?
>    SDLI's represent CIC's and SDTI's represent SSP's
>=20
>    2. From which SS7's protocol layer do SDLI and SDTI come from? I mea=
n,
>    wich layer is responsible for selecting this values?
>=20
>    3.  Suppose  I  want  to  initiate  a  SGP  and  a  MGC.  The code f=
or
>    initialization in the SGP and the MGC should have to call this
>=20
>    functions:
>=20
>    a) initiate_protocol_stack -> Initiates the M2UA protocol stack.
>=20
>    x)  add_sg  ->  Receives the SG identifier (for local managment only=
).
>    This function should only be called in the MGC.
>=20
>    b) add_sgp -> Receives the list of IP adresses:Port number attached =
to
>    this SGP, the port number it should listen in order to
>=20
>    support dynamic registration (default:2904), the max and min number =
of
>    streams per sctp association and the SGP identifier
>=20
>    (for local managment only).
>=20
>    c)  add_asp -> Receives the ASP identifier (for local managment only=
),
>    the max and min number of streams per sctp association,
>=20
>    the  (local and remote)IP addresses:Port number (at the ASP and at t=
he
>    SGP) used for the sctp association (Example:
>=20
>    65.198.40.50:3005, 65.198.40.51:3005).
>=20
>    d)  add_as  ->  Receives the AS identifier (for local managment only=
),
>    the interface identifiers the AS responds to and the ASP
>=20
>    identifiers of the ASP's associated to this AS.
>=20
>    At  what  function  call,  should  the  SDLI  and  SDTI  parameters =
be
>    specified?
>    I believe they should be specified in the add_as function call.
>=20
>    If  that  is the case, how can the SG and the MGC know which Interfa=
ce
>    Identifier is associated with every AS, in order
>    send the MAUP messages with the rigth Interface Identifier parameter=
?
>=20
>    I  am  thinking  about  a  scenario  where  SG and MGC support dynam=
ic
>    registration; but at the begin there are pre-registered
>    ASP's and AS's.
>=20
>                   thanks again,
>                                Pablo Salinas
>      _________________________________________________________________
>=20
>    Do You Yahoo!?
>    [1][dailynews.gif]  Todo  lo  que  quieres  saber  de  Estados Unido=
s,
>    Am=E9rica Latina y el resto del Mundo.
>    Vis=EDta [2]Yahoo! Noticias.
>=20
> References
>=20
>    1. http://espanol.yahoo.com/mail_tagline/*http://espanol.news.yahoo.=
com
>    2. http://espanol.yahoo.com/mail_tagline/*http://espanol.news.yahoo.=
com

--=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



From exim@www1.ietf.org  Fri Jul 18 01:01:32 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA04416
	for <sigtran-archive@odin.ietf.org>; Fri, 18 Jul 2003 01:01:32 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dNMe-0007iN-7w
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 01:01:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6I514B8029650
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 01:01:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dNMa-0007hz-6j; Fri, 18 Jul 2003 01:01:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dNLd-0007hC-OG
	for sigtran@optimus.ietf.org; Fri, 18 Jul 2003 01:00:02 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA04360
	for <sigtran@ietf.org>; Fri, 18 Jul 2003 00:59:56 -0400 (EDT)
From: kakaul@hss.hns.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dNLa-0004m1-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 00:59:58 -0400
Received: from [202.54.26.202] (helo=hindon.hss.co.in)
	by ietf-mx with esmtp (Exim 4.12)
	id 19dNLO-0004lg-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 00:59:47 -0400
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h6I4mWs03963;
	Fri, 18 Jul 2003 10:18:52 +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 ESMTP id h6I4pGx08428;
	Fri, 18 Jul 2003 10:21:57 +0530 (IST)
Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
To: bidulock@openss7.org
Cc: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.5  September 22, 2000
Message-ID: <OF45BDDDE4.AB370171-ON65256D67.00151D03@hss.hns.com>
Date: Fri, 18 Jul 2003 10:15:37 +0530
X-MIMETrack: Serialize by Router on Sandesh/HSS(Release 6.0|September 26, 2002) at 18/07/2003
 10:15:31 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 ,
   I agree with you that in SS7 SEP congestion status is only maintained
per routeset basis and not per route basis, beacause due to RCT procedure
there is no need of maintaining congestion status  on per route basis.Now
putting ASP as SEP you have no RCT procedure defined, but as you say SG's
connected with ASP perform  RCT on the behalf of ASP.Assuming affected
point code x-y-z is reachable through two SG's SG1 and SG2.Now ASP receives
SCON(1) from SG1 and update the point code (X-Y-Z) congestion status  to 1
and MTP3 users are informed with MTP-STATUS with congestion status 1.Now
ASP has to start DAUD procedure for affected point code x-y-z per SG and
ASP will continue to send DAUD messages with affected point code X-Y-Z to
SG1 untill ASP is sure that point code(x-y-z) congestion status  is 0
(no-congestion) through SG1.Mean- while SG1 might be doing RCT procedure on
the behalf of ASP for affected point code X-Y-Z.There are actually two
options for ASP to stop sending DAUD messages to SG1  with affected point
code x-y-z. Either ASP maintains congestion status per SG route or mark the
SG1 route for affected point code x-y-z congested by using some flag so
that ASP can know when to stop DAUD procedure for affected point code per
SG.Roueset (x-y-z) congestion status is just there to indicate MTP-STATUS
primitives.ASP will AUDIT affected point code per SG till congestion
abates, for that ASP has to know when to start and stop sending DAUD
messages for affected congested point code to a given SG which in a way
implies ASP has to maintain either congestion status or flag (indicationg
occurance and abatement of congestion).Now for abatement of affected point
code congestion status there are two approaches either SCON(0) received
from SG for affected point code or timer supervision because once you have
indicated congestion status greater than 0 to MTP3 user you have to
indicate congestion abatement with congestion status 0 to MTP3 user this
statement is only true if the network appearance associated with affected
point code is of national option with congestion priority.I have taken two
SG's in my above example to explain that affected point code  X-Y-Z
congestion status can get updated to 3 on receiving SCON(3) from SG2.Now
you have to send DAUD messages to both SG1 and SG2 till abatement of
congestion occures. Please comment

Regards
Kamesh Kaul
www.hssworld.com
Hughes Software Systems





"Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/17/2003
06:46:00 AM

Please respond to bidulock@openss7.org

Sent by:    sigtran-admin@ietf.org


To:    Kamesh Kaul/HSS@HSS
cc:    sigtran@ietf.org

Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18


kakaul,

You are confusing congestion at an STP due to the conestion of signalling
links on a route with routeset congestion.  Although the congestion of a
signalling link can trigger sending TFC, the route upon which the TFC is
received has nothing to do with congestion.

Put an SEP where the ASP is and tell me where in the SS7 specs it says
that a TFC received from an STP is correlated to the route upon which it
was received.  A: it doesn't.

The passages you quote correspond to transmit congestion on the SCTP
associations between the ASP and the SG.  (SCON is not involved there.)

--brian

kakaul@hss.hns.com wrote:
(Wed, 16 Jul 2003 21:57:11)
>
>
>
>
> Brian,
>    Comments inline
>
>
>
>
> "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/16/2003
> 12:41:40 PM
>
> Please respond to bidulock@openss7.org
>
> Sent by:    sigtran-admin@ietf.org
>
>
> To:    Kamesh Kaul/HSS@HSS
> cc:    sigtran@ietf.org
>
> Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
>
>
> kakaul,
>
> An SS7 SEP at the position of the ASP, where the SGs are STPs would
> not maintain congestion state on a per-route (per-SG) basis.  So
> what makes you think that the ASP needs to?
> [Kamesh]It would because route-set congestion status will change
depending
> on the derived congestion status of individual routes I am attaching
> section 11.2.4 from Q.704
>
>
> 11.2.4 Signalling route set congestion (National option with congestion
>        priorities)
> In the case when the congestion status of a signalling route set changes
as
> a result of either the receipt of a transfer controlled message relating
to
> a particular destination (see 13.7) or an indication of local signalling
> link congestion, or due to the signalling route-set-congestion-test
> procedure (see 13.9) an indication is given from the Message Transfer
Part
> to the local level 4 informing it about the current congestion status of
> the signalling route set. Each user then takes appropriate actions in
order
> to stop generation of signalling messages destined for the affected
> signalling point with congestion priorities lower than the specified
> congestion status. Messages received from the local level 4 with
congestion
> priorities lower than the current signalling route set congestion status
> are discarded by the Message Transfer Part.
> Also, SCON() in this configuration corresponds to TFC (not MTP-STATUS)
> and therefore cannot signal abatement.  (TFC does not signal abatement,
> the lack of a TFC does.)  In the same ways, the ASP can use the lack
> of a SCON to indicate abatement.
> [Kamesh]Yes they do and derived congestion status leads to MTP-STATUS
>
> So, a) SCON(1) and SCON(3) in your example are congestion signals
> (neither is an abatement signal); b) there is no need to maintain
> congestion
> status on a per-SG basis at the ASP ('cause the SS7 SEP doesn't).
> [Kamesh]Yes there is need to maintain congestion status on the per-SG
basis
> at ASP because SG1 doing RCT procedure on the behalf of ASP may return
> SCON(0) when it receives audit for the affected point code and if
affected
> point code is reachable through SG1 and SG2 as shown in previous figure
SG2
> returns SCON(2) then congestion status sent in MTP-STATUS is 2 because
> affected point code is still congested.
>
> This might lead back to the discustion why ASPs don't send RCT, but the
> simple answer is that each SG as STP can send RCT on all the ASP's behalf
> as it would for an "Alias" point code (ANSI T1.111/2000).
>
> Because this is getting into a narrower and narrower corner of the
> specification, I think we are safe to strike the mention of the timer
> from the IG.
> [Kamesh]I dont agree with striking timer
> --brian
>
> kakaul@hss.hns.com wrote:
> (Tue, 15 Jul 2003 18:59:37)
> >
> >
> >
> >
> > Brian
> >    some comments inline
> >
> >
> >
> >
> > "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/15/2003
> > 12:16:38 PM
> >
> > Please respond to bidulock@openss7.org
> >
> > Sent by:    sigtran-admin@ietf.org
> >
> >
> > To:    Kamesh Kaul/HSS@HSS
> > cc:    sigtran@ietf.org
> >
> > Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> >
> >
> > kakaul,
> >
> > If there is no interoperablity issue then it is implementation
dependent.
> > A description of a procedure for the timer could be added as an
> > IMPLEMENTATION
> > NOTE making a suggestion to the implementor of one way to address local
> > issues
> > at the ASP, but it cannot be required.
> >
> > But looking at your description, this mechanism would be a duplication
of
> > the
> > mechanism used by SCCP, ISUP, BICC, which use a dual-timer approach to
> > detecting congestion abatement.  Why would it be necessary for the ASP
to
> > do
> > this as well?
> > >> It is also done in MTP3 for nation options with priority do you mean
> to
> > say that RCT mechanism is not required in MTP3.
> >
> > Also, you should not that SCON received from one SG is no different
than
> > SCON
> > received from another.  The SCON represents a statistical sampling of
> > congestion in the SS7 network between the ASP and the destination, and
is
> > not
> > referenced to an SG.  Therefore, the ASP should never perform this
> > procedure
> > on a per-SG basis.  The SCON received from either SG will generate an
> > MTP-STATUS primitive to the AS, and the AS will run suitable timers at
> the
> > MTP-User (where the responsibility lies to avoid sending messages of a
> > priority less that the congestion state).
> > >> In national option with priority MTP3 user is given MTP-STATUS
> > indication only if point code (routeset )congestion status
> > changes.Similarly at ASP MTP-STATUS indication shall be given to the
user
> > part only if derived status of affected point code changes as shown in
> > figure below :
> >
> >
> >                               SCON (1)+--+----++
> >                              <------- |        |
> >                           +-----------+  SG1   +-----+
> >                           |           |        |     |
> >                           |           +--------+     |      SS7 NODE
> >            +--+----++     |                          |    +----------+
> >            |        |     |                          |    |   PC     |
> >            |  ASP   +-----+                          +----+  X-Y-Z   |
> >            |        |     |                          |    |          |
> >            +--------+     |                          |    +----------+
> >                           |          +--+----++      |
> >              AS:A-B-C     |          |        |      |
> >                           +----------+  SG2   +------+
> >                             <------- |        |
> >                              SCON(3) +--------+
> >
> > MTP-User will be given only MTP-STATUS indication with congestion
status
> 3
> > if SCON(3) reaches before SCON(1).So for this you have to maintain
> > congestion status of affected point code through each SG and ultimately
> > given the Derived congestion status to MTP-User.
> >
> > Understand that SCON is not an M3UA congestion indication: it is an SS7
> > network congestion indication that is directed at the MTP-User.  The
ASP
> > need
> > only translate SCON messages to MTP-STATUS primitives to the AS to
which
> > they
> > correspond.  A AS that conforms to MTP-User standards specifications
with
> > deal
> > with the MTP-STATUS primitives appropriately.
> > >>It is only true in international and national networks without
priority
> > option.
> >
> > It is also important to understand that the ASP should not change the
> > routing
> > of messages on the basis of a received SCON message.  Where applicable
> and
> > for
> > applicable SS7 standards specifications, the SG will send DRST under
very
> > specific and standardized congestion situations.  The ASP is already
> > expected
> > to perform some functions of MTP routing in response to DRST.
> > >> Here i definitely agree with you that rerouting of messages should
not
> > be done at ASP to avoid missequencing.
> >
> > So, now I don't understand why you expect the ASP to run this timer at
> all.
> > >> Some implementations may want to provision congestion of affected
> point
> > code through supervisory congestion timer, they should be allowed to do
> so.
> >
> > Please explain.
> >
> > --brian
> >
> > kakaul@hss.hns.com wrote:
> > (Tue, 15 Jul 2003 10:46:03)
> > >
> > >
> > >
> > >
> > > Brian,
> > >    There is no interoperability issue involved in it and also i dont
> see
> > > any need of peer knowing the existence of this timer.It is local to
ASP
> > and
> > > is used by ASP to supervise the congestion status of the affected
point
> > > code reachable through a given SG.ASP after receiving an SCON message
> > with
> > > congestion status greater than 0 can start this timer and if no SCON
> > > message is received from peer till the expiry of congestion timer ASP
> can
> > > decrement affected point code  congestion status by 1 and start the
> timer
> > > provisioning of congestion again if congestion status of affected
point
> > > code has not reached 0 or no congestion.
> > >
> > > Regards
> > > Kamesh Kaul
> > > www.hssworld.com
> > > Hughes Software Systems
> > >
> > >
> > >
> > >
> > > "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/15/2003
> > > 10:28:40 AM
> > >
> > > Please respond to bidulock@openss7.org
> > >
> > > Sent by:    sigtran-admin@ietf.org
> > >
> > >
> > > To:    Kamesh Kaul/HSS@HSS
> > > cc:    "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
> > >        sigtran@ietf.org
> > >
> > > Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> > >
> > >
> > > kakaul,
> > >
> > > Then add text describing how to use the timer in an interoperable
way.
> > > Also, please describe why the peer needs to know about the existence
> > > of this timer to remain interoperable.  Otherwise, there would not be
> > > any need to describe it.
> > >
> > > --brian
> > >
> > >
> > > kakaul@hss.hns.com wrote:
> > > (Tue, 15 Jul 2003 09:35:27)
> > > >
> > > >
> > > >
> > > >
> > > > Hi Javier,
> > > >                     This timer may be used by some implementations
to
> > > > supervise the congestion status of point code if SCON messages send
> by
> > > SGP
> > > > are lost in transit to ASP.It was discussed earlier and was agreed
> upon
> > > by
> > > > all to keep this timer.I disagree with removing the mention of the
> > timer.
> > > >
> > > > Regards
> > > > Kamesh Kaul
> > > > www.hssworld.com
> > > > Hughes Software Systems
> > > >
> > > >
> > > >
> > > >
> > > > "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>@ietf.org
on
> > > > 07/14/2003 04:05:58 PM
> > > >
> > > > Sent by:    sigtran-admin@ietf.org
> > > >
> > > >
> > > > To:    "'Brian F. G. Bidulock '" <bidulock@openss7.org>
> > > > cc:    "''sigtran@ietf.org' '" <sigtran@ietf.org>
> > > >
> > > > Subject:    RE: [Sigtran] M3UA IG v04 Section 3.18
> > > >
> > > >
> > > > Brian,
> > > >
> > > > We'll remove the mention of the timer, if nobody disagrees...
> > > >
> > > > Cheers // Javier.
> > > >
> > > > -----Original Message-----
> > > > From: Brian F. G. Bidulock
> > > > To: J.Javier Pastor (EE/EEM)
> > > > Cc: 'sigtran@ietf.org'
> > > > Sent: 11/07/2003 11:07
> > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > > >
> > > > J.Javier,
> > > >
> > > > J.Javier Pastor (EE/EEM) wrote:
> > > > (Fri, 11 Jul 2003 09:56:59)
> > > > > Hi Brian,
> > > > >
> > > > > Pls, see below...
> > > > >
> > > > > > -----Original Message-----
> > > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > > Sent: jueves, 10 de julio de 2003 15:34
> > > > > > To: J.Javier Pastor (EE/EEM)
> > > > > > Cc: 'sigtran@ietf.org'
> > > > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > > > > >
> > > > > >
> > > > > > Javier,
> > > > > >
> > > > > > 3.18.3 Solution Description
> > > > > >
> > > > > >    Clarify that the ASP needs to indicate when the congestion
> > > > > > level has
> > > > > >    changed (including abatement).  Further, the ASP peer can
> > > > maintain
> > > > > >    a timer, if desired, in case the ASP fails to indicate
> > congestion
> > > > > >    abatement.
> > > > > >
> > > > > > Where are the procedures for this timer described?  This is the
> > only
> > > > place
> > > > > > that the timer is mentioned.  Please describe the timer or
remove
> > > > the
> > > > > > sentence saying that it has been described...
> > > > > >
> > > > >
> > > > > Do you think is worth include the timer explanation into the new
> > text?
> > > > > Otherwise, we may remove that sentence.
> > > >
> > > > Please remove mention of the timer.
> > > >
> > > > --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
> > >
> >
> > --
> > 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 exim@www1.ietf.org  Fri Jul 18 03:08:32 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA20287
	for <sigtran-archive@odin.ietf.org>; Fri, 18 Jul 2003 03:08:32 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dPLZ-0005xQ-Nr
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 03:08:06 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6I785ws022896
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 03:08:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dPLV-0005wE-O2; Fri, 18 Jul 2003 03:08:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dPLA-0005sk-5F
	for sigtran@optimus.ietf.org; Fri, 18 Jul 2003 03:07:40 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA20236
	for <sigtran@ietf.org>; Fri, 18 Jul 2003 03:07:36 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dPL6-0005lv-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 03:07:36 -0400
Received: from david.siemens.de ([192.35.17.14])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dPKv-0005lg-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 03:07:25 -0400
Received: from mail2.siemens.de (mail2.siemens.de [139.25.208.11])
	by david.siemens.de (8.11.7/8.11.7) with ESMTP id h6I776I11864;
	Fri, 18 Jul 2003 09:07:06 +0200 (MEST)
Received: from moody.mchh.siemens.de (moody.mchh.siemens.de [139.21.205.85])
	by mail2.siemens.de (8.11.7/8.11.7) with ESMTP id h6I776r01128;
	Fri, 18 Jul 2003 09:07:06 +0200 (MEST)
Received: from mchh253e.mchh.siemens.de (mchh253e.mchh.siemens.de [139.21.200.63])
	by moody.mchh.siemens.de (8.9.3/8.9.1) with ESMTP id JAA29987;
	Fri, 18 Jul 2003 09:07:05 +0200 (MET DST)
Received: by mchh253e.mchh.siemens.de with Internet Mail Service (5.5.2653.19)
	id <3945KWPY>; Fri, 18 Jul 2003 09:06:30 +0200
Message-ID: <D73559B38D51D4119CB800508B9527B505AC1990@mchh266e.mchh.siemens.de>
From: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>,
        Tom George
	 <Tom.George@alcatel.com>
Cc: IETF sigtran <sigtran@ietf.org>
Subject: RE: FW: [Sigtran] M2PA Last Call Comment
Date: Fri, 18 Jul 2003 09:06:19 +0200
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>

Folks,

First of all we need a clarification of terms "Q.703-state-machine-based M2PA link" or "Q.2140-..." 
I'm not aware of any occurrences of these terms.

before spending more bytes on discussions, let's see what ITU-T or ANSI are doing in similiar situations:

 Question: does STP A care about the fact that STPB has links according to Q.2140? 
 Answer: NO, NOT AT ALL, STP A operator might not even know that STP B has Q.2140 links.  

        +-----------+                 +-----------+
   Q.703|           |                 |           |   Q.2140
  ------+           |                 |           +--------
        |           |       Q.703     |           |
        | STP A     +-----------------+   STP B   |
        |           |                 |           |
   Q.703|           |                 |           |   Q.2140
  ------+           |                 |           +--------
        |           |                 |           |
        +-----------+                 +-----------+

Let's go to an IP based link between STP A and STP B

Question: should STP A care about the fact that STP B has links according to Q.2140?
Answer: I see no necessity to deviate from ANSI and ITU-T.


        +-----------+                 +-----------+
   Q.703|           |                 |           |   Q.2140
  ------+           |                 |           +--------
        |           |       M2PA      |           |
        | STP A     +-----------------+   STP B   |
        |           |                 |           |
   Q.703|           |                 |           |   Q.2140
  ------+           |                 |           +--------
        |           |                 |           |
        +-----------+                 +-----------+

If STP A wants to prove, he should do that and STP B has to support STP A in that procedure and vice versa.
But I see no necessity why both sides should perform similiar. 

In all other cases STP A would have to be informed if e.g. STP-B migrates from pure Q.703 to pure Q.2140 and that's not acceptable.

greetings 

HannsJuergen



> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: Friday, July 18, 2003 1:55 AM
> To: Tom George
> Cc: IETF sigtran
> Subject: Re: FW: [Sigtran] M2PA Last Call Comment
> 
> 
> Tom,
> 
> Q.703 does not give the option to talk to a Q.2140 link.  But it
> appears that M2PA allows a Q.703-state-machine-based M2PA link to
> talk to a Q.2140-state-machine-based M2PA link.  That is were the
> problem ensues.
> 
> --brian
> 
> Tom George wrote:                                             
>                          (Thu, 17 Jul 2003 17:16:54)
> > Brian,
> > 
> > In the non-IP realm, is it possible for a Q.703 to talk to 
> a Q.2140? 
> > What would be done in the scenario you described?
> > 
> > tomG.
> > 
> > Brian F. G. Bidulock wrote:
> > > Hanns Juergen,
> > > 
> > > Consider that one side follows Q.703 and requires proving 
> while the
> > > other side follows Q.2140 and foregoes proving.  The 
> following will
> > > occur:
> >    [...]
> > 
> > 
> > -- 
> > * 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.  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 exim@www1.ietf.org  Fri Jul 18 03:42:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA21013
	for <sigtran-archive@odin.ietf.org>; Fri, 18 Jul 2003 03:42:32 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dPsS-0007VB-LJ
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 03:42:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6I7g4N6028838
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 03:42:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dPsP-0007Um-Gj; Fri, 18 Jul 2003 03:42:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dPsE-0007UT-D8
	for sigtran@optimus.ietf.org; Fri, 18 Jul 2003 03:41:50 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA21006
	for <sigtran@ietf.org>; Fri, 18 Jul 2003 03:41:48 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dPsC-000656-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 03:41:48 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[PS+RS3caUzcqoklWSspsJJZ3mIrmFslY])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dPrv-00064J-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 03:41:31 -0400
Received: from ns.pigworks.openss7.net (IDENT:X50z1NdOZtMUPcGGgRJNmLKK5W0g+gcU@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6I7eMP05462;
	Fri, 18 Jul 2003 01:40:22 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6I7eM303757;
	Fri, 18 Jul 2003 01:40:22 -0600
Date: Fri, 18 Jul 2003 01:40:22 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: kakaul@hss.hns.com
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
Message-ID: <20030718014022.B3547@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: kakaul@hss.hns.com, sigtran@ietf.org
References: <OF45BDDDE4.AB370171-ON65256D67.00151D03@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: <OF45BDDDE4.AB370171-ON65256D67.00151D03@hss.hns.com>; from kakaul@hss.hns.com on Fri, Jul 18, 2003 at 10:15:37AM +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>

kakaul,

I don't think the ASP needs to do all that.  Instead of DAUD, it can simply
lower the congestion status of the routeset after a time interval and indicate
abatement to the National MTP-User.  The next message sent by the MTP-User
will illicit SCON if the abatement was premature (without sending any message
on the SS7 network) and the timer can be reset.  Only one timer per routeset
is necessary.

In the multiple SG scenario SCON is comparable to TFC, not MTP-STATUS.
SCON(0) is not necessary in the multiple SG scenario, the ASP can abate
after a time interval and rely on responsive SCON from the SG.  If no
SCON is delivered, then abatement to no congestion sticks.

Also, understand that the Nation MTP (e.g. ANSI) does not indicate abatement
in all situations: only when the congestion is remote.  When the congestion
is a result of congestion of a signalling link, abatement is not indicated.
(e.g. see ANSI T1.111.4/2000 SDLs).

--brian


kakaul@hss.hns.com wrote:                                                           (Fri, 18 Jul 2003 10:15:37)
> 
> 
> 
> 
> Brian ,
>    I agree with you that in SS7 SEP congestion status is only maintained
> per routeset basis and not per route basis, beacause due to RCT procedure
> there is no need of maintaining congestion status  on per route basis.Now
> putting ASP as SEP you have no RCT procedure defined, but as you say SG's
> connected with ASP perform  RCT on the behalf of ASP.Assuming affected
> point code x-y-z is reachable through two SG's SG1 and SG2.Now ASP receives
> SCON(1) from SG1 and update the point code (X-Y-Z) congestion status  to 1
> and MTP3 users are informed with MTP-STATUS with congestion status 1.Now
> ASP has to start DAUD procedure for affected point code x-y-z per SG and
> ASP will continue to send DAUD messages with affected point code X-Y-Z to
> SG1 untill ASP is sure that point code(x-y-z) congestion status  is 0
> (no-congestion) through SG1.Mean- while SG1 might be doing RCT procedure on
> the behalf of ASP for affected point code X-Y-Z.There are actually two
> options for ASP to stop sending DAUD messages to SG1  with affected point
> code x-y-z. Either ASP maintains congestion status per SG route or mark the
> SG1 route for affected point code x-y-z congested by using some flag so
> that ASP can know when to stop DAUD procedure for affected point code per
> SG.Roueset (x-y-z) congestion status is just there to indicate MTP-STATUS
> primitives.ASP will AUDIT affected point code per SG till congestion
> abates, for that ASP has to know when to start and stop sending DAUD
> messages for affected congested point code to a given SG which in a way
> implies ASP has to maintain either congestion status or flag (indicationg
> occurance and abatement of congestion).Now for abatement of affected point
> code congestion status there are two approaches either SCON(0) received
> from SG for affected point code or timer supervision because once you have
> indicated congestion status greater than 0 to MTP3 user you have to
> indicate congestion abatement with congestion status 0 to MTP3 user this
> statement is only true if the network appearance associated with affected
> point code is of national option with congestion priority.I have taken two
> SG's in my above example to explain that affected point code  X-Y-Z
> congestion status can get updated to 3 on receiving SCON(3) from SG2.Now
> you have to send DAUD messages to both SG1 and SG2 till abatement of
> congestion occures. Please comment
> 
> Regards
> Kamesh Kaul
> www.hssworld.com
> Hughes Software Systems
> 
> 
> 
> 
> 
> "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/17/2003
> 06:46:00 AM
> 
> Please respond to bidulock@openss7.org
> 
> Sent by:    sigtran-admin@ietf.org
> 
> 
> To:    Kamesh Kaul/HSS@HSS
> cc:    sigtran@ietf.org
> 
> Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> 
> 
> kakaul,
> 
> You are confusing congestion at an STP due to the conestion of signalling
> links on a route with routeset congestion.  Although the congestion of a
> signalling link can trigger sending TFC, the route upon which the TFC is
> received has nothing to do with congestion.
> 
> Put an SEP where the ASP is and tell me where in the SS7 specs it says
> that a TFC received from an STP is correlated to the route upon which it
> was received.  A: it doesn't.
> 
> The passages you quote correspond to transmit congestion on the SCTP
> associations between the ASP and the SG.  (SCON is not involved there.)
> 
> --brian
> 
> kakaul@hss.hns.com wrote:
> (Wed, 16 Jul 2003 21:57:11)
> >
> >
> >
> >
> > Brian,
> >    Comments inline
> >
> >
> >
> >
> > "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/16/2003
> > 12:41:40 PM
> >
> > Please respond to bidulock@openss7.org
> >
> > Sent by:    sigtran-admin@ietf.org
> >
> >
> > To:    Kamesh Kaul/HSS@HSS
> > cc:    sigtran@ietf.org
> >
> > Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> >
> >
> > kakaul,
> >
> > An SS7 SEP at the position of the ASP, where the SGs are STPs would
> > not maintain congestion state on a per-route (per-SG) basis.  So
> > what makes you think that the ASP needs to?
> > [Kamesh]It would because route-set congestion status will change
> depending
> > on the derived congestion status of individual routes I am attaching
> > section 11.2.4 from Q.704
> >
> >
> > 11.2.4 Signalling route set congestion (National option with congestion
> >        priorities)
> > In the case when the congestion status of a signalling route set changes
> as
> > a result of either the receipt of a transfer controlled message relating
> to
> > a particular destination (see 13.7) or an indication of local signalling
> > link congestion, or due to the signalling route-set-congestion-test
> > procedure (see 13.9) an indication is given from the Message Transfer
> Part
> > to the local level 4 informing it about the current congestion status of
> > the signalling route set. Each user then takes appropriate actions in
> order
> > to stop generation of signalling messages destined for the affected
> > signalling point with congestion priorities lower than the specified
> > congestion status. Messages received from the local level 4 with
> congestion
> > priorities lower than the current signalling route set congestion status
> > are discarded by the Message Transfer Part.
> > Also, SCON() in this configuration corresponds to TFC (not MTP-STATUS)
> > and therefore cannot signal abatement.  (TFC does not signal abatement,
> > the lack of a TFC does.)  In the same ways, the ASP can use the lack
> > of a SCON to indicate abatement.
> > [Kamesh]Yes they do and derived congestion status leads to MTP-STATUS
> >
> > So, a) SCON(1) and SCON(3) in your example are congestion signals
> > (neither is an abatement signal); b) there is no need to maintain
> > congestion
> > status on a per-SG basis at the ASP ('cause the SS7 SEP doesn't).
> > [Kamesh]Yes there is need to maintain congestion status on the per-SG
> basis
> > at ASP because SG1 doing RCT procedure on the behalf of ASP may return
> > SCON(0) when it receives audit for the affected point code and if
> affected
> > point code is reachable through SG1 and SG2 as shown in previous figure
> SG2
> > returns SCON(2) then congestion status sent in MTP-STATUS is 2 because
> > affected point code is still congested.
> >
> > This might lead back to the discustion why ASPs don't send RCT, but the
> > simple answer is that each SG as STP can send RCT on all the ASP's behalf
> > as it would for an "Alias" point code (ANSI T1.111/2000).
> >
> > Because this is getting into a narrower and narrower corner of the
> > specification, I think we are safe to strike the mention of the timer
> > from the IG.
> > [Kamesh]I dont agree with striking timer
> > --brian
> >
> > kakaul@hss.hns.com wrote:
> > (Tue, 15 Jul 2003 18:59:37)
> > >
> > >
> > >
> > >
> > > Brian
> > >    some comments inline
> > >
> > >
> > >
> > >
> > > "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/15/2003
> > > 12:16:38 PM
> > >
> > > Please respond to bidulock@openss7.org
> > >
> > > Sent by:    sigtran-admin@ietf.org
> > >
> > >
> > > To:    Kamesh Kaul/HSS@HSS
> > > cc:    sigtran@ietf.org
> > >
> > > Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> > >
> > >
> > > kakaul,
> > >
> > > If there is no interoperablity issue then it is implementation
> dependent.
> > > A description of a procedure for the timer could be added as an
> > > IMPLEMENTATION
> > > NOTE making a suggestion to the implementor of one way to address local
> > > issues
> > > at the ASP, but it cannot be required.
> > >
> > > But looking at your description, this mechanism would be a duplication
> of
> > > the
> > > mechanism used by SCCP, ISUP, BICC, which use a dual-timer approach to
> > > detecting congestion abatement.  Why would it be necessary for the ASP
> to
> > > do
> > > this as well?
> > > >> It is also done in MTP3 for nation options with priority do you mean
> > to
> > > say that RCT mechanism is not required in MTP3.
> > >
> > > Also, you should not that SCON received from one SG is no different
> than
> > > SCON
> > > received from another.  The SCON represents a statistical sampling of
> > > congestion in the SS7 network between the ASP and the destination, and
> is
> > > not
> > > referenced to an SG.  Therefore, the ASP should never perform this
> > > procedure
> > > on a per-SG basis.  The SCON received from either SG will generate an
> > > MTP-STATUS primitive to the AS, and the AS will run suitable timers at
> > the
> > > MTP-User (where the responsibility lies to avoid sending messages of a
> > > priority less that the congestion state).
> > > >> In national option with priority MTP3 user is given MTP-STATUS
> > > indication only if point code (routeset )congestion status
> > > changes.Similarly at ASP MTP-STATUS indication shall be given to the
> user
> > > part only if derived status of affected point code changes as shown in
> > > figure below :
> > >
> > >
> > >                               SCON (1)+--+----++
> > >                              <------- |        |
> > >                           +-----------+  SG1   +-----+
> > >                           |           |        |     |
> > >                           |           +--------+     |      SS7 NODE
> > >            +--+----++     |                          |    +----------+
> > >            |        |     |                          |    |   PC     |
> > >            |  ASP   +-----+                          +----+  X-Y-Z   |
> > >            |        |     |                          |    |          |
> > >            +--------+     |                          |    +----------+
> > >                           |          +--+----++      |
> > >              AS:A-B-C     |          |        |      |
> > >                           +----------+  SG2   +------+
> > >                             <------- |        |
> > >                              SCON(3) +--------+
> > >
> > > MTP-User will be given only MTP-STATUS indication with congestion
> status
> > 3
> > > if SCON(3) reaches before SCON(1).So for this you have to maintain
> > > congestion status of affected point code through each SG and ultimately
> > > given the Derived congestion status to MTP-User.
> > >
> > > Understand that SCON is not an M3UA congestion indication: it is an SS7
> > > network congestion indication that is directed at the MTP-User.  The
> ASP
> > > need
> > > only translate SCON messages to MTP-STATUS primitives to the AS to
> which
> > > they
> > > correspond.  A AS that conforms to MTP-User standards specifications
> with
> > > deal
> > > with the MTP-STATUS primitives appropriately.
> > > >>It is only true in international and national networks without
> priority
> > > option.
> > >
> > > It is also important to understand that the ASP should not change the
> > > routing
> > > of messages on the basis of a received SCON message.  Where applicable
> > and
> > > for
> > > applicable SS7 standards specifications, the SG will send DRST under
> very
> > > specific and standardized congestion situations.  The ASP is already
> > > expected
> > > to perform some functions of MTP routing in response to DRST.
> > > >> Here i definitely agree with you that rerouting of messages should
> not
> > > be done at ASP to avoid missequencing.
> > >
> > > So, now I don't understand why you expect the ASP to run this timer at
> > all.
> > > >> Some implementations may want to provision congestion of affected
> > point
> > > code through supervisory congestion timer, they should be allowed to do
> > so.
> > >
> > > Please explain.
> > >
> > > --brian
> > >
> > > kakaul@hss.hns.com wrote:
> > > (Tue, 15 Jul 2003 10:46:03)
> > > >
> > > >
> > > >
> > > >
> > > > Brian,
> > > >    There is no interoperability issue involved in it and also i dont
> > see
> > > > any need of peer knowing the existence of this timer.It is local to
> ASP
> > > and
> > > > is used by ASP to supervise the congestion status of the affected
> point
> > > > code reachable through a given SG.ASP after receiving an SCON message
> > > with
> > > > congestion status greater than 0 can start this timer and if no SCON
> > > > message is received from peer till the expiry of congestion timer ASP
> > can
> > > > decrement affected point code  congestion status by 1 and start the
> > timer
> > > > provisioning of congestion again if congestion status of affected
> point
> > > > code has not reached 0 or no congestion.
> > > >
> > > > Regards
> > > > Kamesh Kaul
> > > > www.hssworld.com
> > > > Hughes Software Systems
> > > >
> > > >
> > > >
> > > >
> > > > "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/15/2003
> > > > 10:28:40 AM
> > > >
> > > > Please respond to bidulock@openss7.org
> > > >
> > > > Sent by:    sigtran-admin@ietf.org
> > > >
> > > >
> > > > To:    Kamesh Kaul/HSS@HSS
> > > > cc:    "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
> > > >        sigtran@ietf.org
> > > >
> > > > Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> > > >
> > > >
> > > > kakaul,
> > > >
> > > > Then add text describing how to use the timer in an interoperable
> way.
> > > > Also, please describe why the peer needs to know about the existence
> > > > of this timer to remain interoperable.  Otherwise, there would not be
> > > > any need to describe it.
> > > >
> > > > --brian
> > > >
> > > >
> > > > kakaul@hss.hns.com wrote:
> > > > (Tue, 15 Jul 2003 09:35:27)
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > Hi Javier,
> > > > >                     This timer may be used by some implementations
> to
> > > > > supervise the congestion status of point code if SCON messages send
> > by
> > > > SGP
> > > > > are lost in transit to ASP.It was discussed earlier and was agreed
> > upon
> > > > by
> > > > > all to keep this timer.I disagree with removing the mention of the
> > > timer.
> > > > >
> > > > > Regards
> > > > > Kamesh Kaul
> > > > > www.hssworld.com
> > > > > Hughes Software Systems
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>@ietf.org
> on
> > > > > 07/14/2003 04:05:58 PM
> > > > >
> > > > > Sent by:    sigtran-admin@ietf.org
> > > > >
> > > > >
> > > > > To:    "'Brian F. G. Bidulock '" <bidulock@openss7.org>
> > > > > cc:    "''sigtran@ietf.org' '" <sigtran@ietf.org>
> > > > >
> > > > > Subject:    RE: [Sigtran] M3UA IG v04 Section 3.18
> > > > >
> > > > >
> > > > > Brian,
> > > > >
> > > > > We'll remove the mention of the timer, if nobody disagrees...
> > > > >
> > > > > Cheers // Javier.
> > > > >
> > > > > -----Original Message-----
> > > > > From: Brian F. G. Bidulock
> > > > > To: J.Javier Pastor (EE/EEM)
> > > > > Cc: 'sigtran@ietf.org'
> > > > > Sent: 11/07/2003 11:07
> > > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > > > >
> > > > > J.Javier,
> > > > >
> > > > > J.Javier Pastor (EE/EEM) wrote:
> > > > > (Fri, 11 Jul 2003 09:56:59)
> > > > > > Hi Brian,
> > > > > >
> > > > > > Pls, see below...
> > > > > >
> > > > > > > -----Original Message-----
> > > > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > > > Sent: jueves, 10 de julio de 2003 15:34
> > > > > > > To: J.Javier Pastor (EE/EEM)
> > > > > > > Cc: 'sigtran@ietf.org'
> > > > > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > > > > > >
> > > > > > >
> > > > > > > Javier,
> > > > > > >
> > > > > > > 3.18.3 Solution Description
> > > > > > >
> > > > > > >    Clarify that the ASP needs to indicate when the congestion
> > > > > > > level has
> > > > > > >    changed (including abatement).  Further, the ASP peer can
> > > > > maintain
> > > > > > >    a timer, if desired, in case the ASP fails to indicate
> > > congestion
> > > > > > >    abatement.
> > > > > > >
> > > > > > > Where are the procedures for this timer described?  This is the
> > > only
> > > > > place
> > > > > > > that the timer is mentioned.  Please describe the timer or
> remove
> > > > > the
> > > > > > > sentence saying that it has been described...
> > > > > > >
> > > > > >
> > > > > > Do you think is worth include the timer explanation into the new
> > > text?
> > > > > > Otherwise, we may remove that sentence.
> > > > >
> > > > > Please remove mention of the timer.
> > > > >
> > > > > --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
> > > >
> > >
> > > --
> > > 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 exim@www1.ietf.org  Fri Jul 18 11:18:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA03678
	for <sigtran-archive@odin.ietf.org>; Fri, 18 Jul 2003 11:18:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dWzj-0002Il-Su
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 11:18:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6IFI3Xg008841
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 11:18:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dWzg-0002IK-Ni; Fri, 18 Jul 2003 11:18:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dWzX-0002I7-Dn
	for sigtran@optimus.ietf.org; Fri, 18 Jul 2003 11:17:51 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA03667
	for <sigtran@ietf.org>; Fri, 18 Jul 2003 11:17:47 -0400 (EDT)
From: kakaul@hss.hns.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dWzV-0001dq-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 11:17:49 -0400
Received: from hindon.hss.co.in ([202.54.26.202])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dWzJ-0001bP-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 11:17:37 -0400
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h6IFAas07099;
	Fri, 18 Jul 2003 20:40:36 +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 ESMTP id h6IFEJJ11500;
	Fri, 18 Jul 2003 20:44:20 +0530 (IST)
Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
To: bidulock@openss7.org
Cc: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.5  September 22, 2000
Message-ID: <OFF6C0FC9C.F9FD3D92-ON65256D67.00472D25@hss.hns.com>
Date: Fri, 18 Jul 2003 20:38:57 +0530
X-MIMETrack: Serialize by Router on Sandesh/HSS(Release 6.0|September 26, 2002) at 18/07/2003
 08:37:42 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>





Brian,
   I agree with running congestion timer per affected route-set(point code)
basis as you have mentioned. But now the point is that in SS7, SEP runs the
RCT congestion supervisory timers after sending RCT message with current
congestion status of the affected route-set so that if the TFC for that
message is not received congestion abatement procedure will be supervised
by timers.So it means that in SS7, SEP RCT message checks the clearance of
path (for congestion) to affected point code and then monitors the
congestion abatement by timers.But in ASP there is no RCT message so we in
a way have to depend on DAUD message per SG to simulate RCT message and
monitor congestion by one timer per affected point code .But for that we
need to know to which SG DAUD message should be sent. SS7, SEP need not
bother about sending RCT per route because it sends RCT message as if it
has got dummy message from user part for affected point code with current
routeset congestion set in priority.For ASP to figure out to which SG DAUD
has to be send either congestion status or flag(signifying occurance or
abatement) has to be maintained per SG.Please comment.

Do you agree with not striking of timer mention in IG.

Regards
Kamesh Kaul
www.hssworld.com
Hughes Software Systems




"Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/18/2003
01:10:22 PM

Please respond to bidulock@openss7.org

Sent by:    sigtran-admin@ietf.org


To:    Kamesh Kaul/HSS@HSS
cc:    sigtran@ietf.org

Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18


kakaul,

I don't think the ASP needs to do all that.  Instead of DAUD, it can simply
lower the congestion status of the routeset after a time interval and
indicate
abatement to the National MTP-User.  The next message sent by the MTP-User
will illicit SCON if the abatement was premature (without sending any
message
on the SS7 network) and the timer can be reset.  Only one timer per
routeset
is necessary.

In the multiple SG scenario SCON is comparable to TFC, not MTP-STATUS.
SCON(0) is not necessary in the multiple SG scenario, the ASP can abate
after a time interval and rely on responsive SCON from the SG.  If no
SCON is delivered, then abatement to no congestion sticks.

Also, understand that the Nation MTP (e.g. ANSI) does not indicate
abatement
in all situations: only when the congestion is remote.  When the congestion
is a result of congestion of a signalling link, abatement is not indicated.
(e.g. see ANSI T1.111.4/2000 SDLs).

--brian


kakaul@hss.hns.com wrote:
(Fri, 18 Jul 2003 10:15:37)
>
>
>
>
> Brian ,
>    I agree with you that in SS7 SEP congestion status is only maintained
> per routeset basis and not per route basis, beacause due to RCT procedure
> there is no need of maintaining congestion status  on per route basis.Now
> putting ASP as SEP you have no RCT procedure defined, but as you say SG's
> connected with ASP perform  RCT on the behalf of ASP.Assuming affected
> point code x-y-z is reachable through two SG's SG1 and SG2.Now ASP
receives
> SCON(1) from SG1 and update the point code (X-Y-Z) congestion status  to
1
> and MTP3 users are informed with MTP-STATUS with congestion status 1.Now
> ASP has to start DAUD procedure for affected point code x-y-z per SG and
> ASP will continue to send DAUD messages with affected point code X-Y-Z to
> SG1 untill ASP is sure that point code(x-y-z) congestion status  is 0
> (no-congestion) through SG1.Mean- while SG1 might be doing RCT procedure
on
> the behalf of ASP for affected point code X-Y-Z.There are actually two
> options for ASP to stop sending DAUD messages to SG1  with affected point
> code x-y-z. Either ASP maintains congestion status per SG route or mark
the
> SG1 route for affected point code x-y-z congested by using some flag so
> that ASP can know when to stop DAUD procedure for affected point code per
> SG.Roueset (x-y-z) congestion status is just there to indicate MTP-STATUS
> primitives.ASP will AUDIT affected point code per SG till congestion
> abates, for that ASP has to know when to start and stop sending DAUD
> messages for affected congested point code to a given SG which in a way
> implies ASP has to maintain either congestion status or flag (indicationg
> occurance and abatement of congestion).Now for abatement of affected
point
> code congestion status there are two approaches either SCON(0) received
> from SG for affected point code or timer supervision because once you
have
> indicated congestion status greater than 0 to MTP3 user you have to
> indicate congestion abatement with congestion status 0 to MTP3 user this
> statement is only true if the network appearance associated with affected
> point code is of national option with congestion priority.I have taken
two
> SG's in my above example to explain that affected point code  X-Y-Z
> congestion status can get updated to 3 on receiving SCON(3) from SG2.Now
> you have to send DAUD messages to both SG1 and SG2 till abatement of
> congestion occures. Please comment
>
> Regards
> Kamesh Kaul
> www.hssworld.com
> Hughes Software Systems
>
>
>
>
>
> "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/17/2003
> 06:46:00 AM
>
> Please respond to bidulock@openss7.org
>
> Sent by:    sigtran-admin@ietf.org
>
>
> To:    Kamesh Kaul/HSS@HSS
> cc:    sigtran@ietf.org
>
> Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
>
>
> kakaul,
>
> You are confusing congestion at an STP due to the conestion of signalling
> links on a route with routeset congestion.  Although the congestion of a
> signalling link can trigger sending TFC, the route upon which the TFC is
> received has nothing to do with congestion.
>
> Put an SEP where the ASP is and tell me where in the SS7 specs it says
> that a TFC received from an STP is correlated to the route upon which it
> was received.  A: it doesn't.
>
> The passages you quote correspond to transmit congestion on the SCTP
> associations between the ASP and the SG.  (SCON is not involved there.)
>
> --brian
>
> kakaul@hss.hns.com wrote:
> (Wed, 16 Jul 2003 21:57:11)
> >
> >
> >
> >
> > Brian,
> >    Comments inline
> >
> >
> >
> >
> > "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/16/2003
> > 12:41:40 PM
> >
> > Please respond to bidulock@openss7.org
> >
> > Sent by:    sigtran-admin@ietf.org
> >
> >
> > To:    Kamesh Kaul/HSS@HSS
> > cc:    sigtran@ietf.org
> >
> > Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> >
> >
> > kakaul,
> >
> > An SS7 SEP at the position of the ASP, where the SGs are STPs would
> > not maintain congestion state on a per-route (per-SG) basis.  So
> > what makes you think that the ASP needs to?
> > [Kamesh]It would because route-set congestion status will change
> depending
> > on the derived congestion status of individual routes I am attaching
> > section 11.2.4 from Q.704
> >
> >
> > 11.2.4 Signalling route set congestion (National option with congestion
> >        priorities)
> > In the case when the congestion status of a signalling route set
changes
> as
> > a result of either the receipt of a transfer controlled message
relating
> to
> > a particular destination (see 13.7) or an indication of local
signalling
> > link congestion, or due to the signalling route-set-congestion-test
> > procedure (see 13.9) an indication is given from the Message Transfer
> Part
> > to the local level 4 informing it about the current congestion status
of
> > the signalling route set. Each user then takes appropriate actions in
> order
> > to stop generation of signalling messages destined for the affected
> > signalling point with congestion priorities lower than the specified
> > congestion status. Messages received from the local level 4 with
> congestion
> > priorities lower than the current signalling route set congestion
status
> > are discarded by the Message Transfer Part.
> > Also, SCON() in this configuration corresponds to TFC (not MTP-STATUS)
> > and therefore cannot signal abatement.  (TFC does not signal abatement,
> > the lack of a TFC does.)  In the same ways, the ASP can use the lack
> > of a SCON to indicate abatement.
> > [Kamesh]Yes they do and derived congestion status leads to MTP-STATUS
> >
> > So, a) SCON(1) and SCON(3) in your example are congestion signals
> > (neither is an abatement signal); b) there is no need to maintain
> > congestion
> > status on a per-SG basis at the ASP ('cause the SS7 SEP doesn't).
> > [Kamesh]Yes there is need to maintain congestion status on the per-SG
> basis
> > at ASP because SG1 doing RCT procedure on the behalf of ASP may return
> > SCON(0) when it receives audit for the affected point code and if
> affected
> > point code is reachable through SG1 and SG2 as shown in previous figure
> SG2
> > returns SCON(2) then congestion status sent in MTP-STATUS is 2 because
> > affected point code is still congested.
> >
> > This might lead back to the discustion why ASPs don't send RCT, but the
> > simple answer is that each SG as STP can send RCT on all the ASP's
behalf
> > as it would for an "Alias" point code (ANSI T1.111/2000).
> >
> > Because this is getting into a narrower and narrower corner of the
> > specification, I think we are safe to strike the mention of the timer
> > from the IG.
> > [Kamesh]I dont agree with striking timer
> > --brian
> >
> > kakaul@hss.hns.com wrote:
> > (Tue, 15 Jul 2003 18:59:37)
> > >
> > >
> > >
> > >
> > > Brian
> > >    some comments inline
> > >
> > >
> > >
> > >
> > > "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/15/2003
> > > 12:16:38 PM
> > >
> > > Please respond to bidulock@openss7.org
> > >
> > > Sent by:    sigtran-admin@ietf.org
> > >
> > >
> > > To:    Kamesh Kaul/HSS@HSS
> > > cc:    sigtran@ietf.org
> > >
> > > Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> > >
> > >
> > > kakaul,
> > >
> > > If there is no interoperablity issue then it is implementation
> dependent.
> > > A description of a procedure for the timer could be added as an
> > > IMPLEMENTATION
> > > NOTE making a suggestion to the implementor of one way to address
local
> > > issues
> > > at the ASP, but it cannot be required.
> > >
> > > But looking at your description, this mechanism would be a
duplication
> of
> > > the
> > > mechanism used by SCCP, ISUP, BICC, which use a dual-timer approach
to
> > > detecting congestion abatement.  Why would it be necessary for the
ASP
> to
> > > do
> > > this as well?
> > > >> It is also done in MTP3 for nation options with priority do you
mean
> > to
> > > say that RCT mechanism is not required in MTP3.
> > >
> > > Also, you should not that SCON received from one SG is no different
> than
> > > SCON
> > > received from another.  The SCON represents a statistical sampling of
> > > congestion in the SS7 network between the ASP and the destination,
and
> is
> > > not
> > > referenced to an SG.  Therefore, the ASP should never perform this
> > > procedure
> > > on a per-SG basis.  The SCON received from either SG will generate an
> > > MTP-STATUS primitive to the AS, and the AS will run suitable timers
at
> > the
> > > MTP-User (where the responsibility lies to avoid sending messages of
a
> > > priority less that the congestion state).
> > > >> In national option with priority MTP3 user is given MTP-STATUS
> > > indication only if point code (routeset )congestion status
> > > changes.Similarly at ASP MTP-STATUS indication shall be given to the
> user
> > > part only if derived status of affected point code changes as shown
in
> > > figure below :
> > >
> > >
> > >                               SCON (1)+--+----++
> > >                              <------- |        |
> > >                           +-----------+  SG1   +-----+
> > >                           |           |        |     |
> > >                           |           +--------+     |      SS7 NODE
> > >            +--+----++     |                          |
+----------+
> > >            |        |     |                          |    |   PC
|
> > >            |  ASP   +-----+                          +----+  X-Y-Z
|
> > >            |        |     |                          |    |
|
> > >            +--------+     |                          |
+----------+
> > >                           |          +--+----++      |
> > >              AS:A-B-C     |          |        |      |
> > >                           +----------+  SG2   +------+
> > >                             <------- |        |
> > >                              SCON(3) +--------+
> > >
> > > MTP-User will be given only MTP-STATUS indication with congestion
> status
> > 3
> > > if SCON(3) reaches before SCON(1).So for this you have to maintain
> > > congestion status of affected point code through each SG and
ultimately
> > > given the Derived congestion status to MTP-User.
> > >
> > > Understand that SCON is not an M3UA congestion indication: it is an
SS7
> > > network congestion indication that is directed at the MTP-User.  The
> ASP
> > > need
> > > only translate SCON messages to MTP-STATUS primitives to the AS to
> which
> > > they
> > > correspond.  A AS that conforms to MTP-User standards specifications
> with
> > > deal
> > > with the MTP-STATUS primitives appropriately.
> > > >>It is only true in international and national networks without
> priority
> > > option.
> > >
> > > It is also important to understand that the ASP should not change the
> > > routing
> > > of messages on the basis of a received SCON message.  Where
applicable
> > and
> > > for
> > > applicable SS7 standards specifications, the SG will send DRST under
> very
> > > specific and standardized congestion situations.  The ASP is already
> > > expected
> > > to perform some functions of MTP routing in response to DRST.
> > > >> Here i definitely agree with you that rerouting of messages should
> not
> > > be done at ASP to avoid missequencing.
> > >
> > > So, now I don't understand why you expect the ASP to run this timer
at
> > all.
> > > >> Some implementations may want to provision congestion of affected
> > point
> > > code through supervisory congestion timer, they should be allowed to
do
> > so.
> > >
> > > Please explain.
> > >
> > > --brian
> > >
> > > kakaul@hss.hns.com wrote:
> > > (Tue, 15 Jul 2003 10:46:03)
> > > >
> > > >
> > > >
> > > >
> > > > Brian,
> > > >    There is no interoperability issue involved in it and also i
dont
> > see
> > > > any need of peer knowing the existence of this timer.It is local to
> ASP
> > > and
> > > > is used by ASP to supervise the congestion status of the affected
> point
> > > > code reachable through a given SG.ASP after receiving an SCON
message
> > > with
> > > > congestion status greater than 0 can start this timer and if no
SCON
> > > > message is received from peer till the expiry of congestion timer
ASP
> > can
> > > > decrement affected point code  congestion status by 1 and start the
> > timer
> > > > provisioning of congestion again if congestion status of affected
> point
> > > > code has not reached 0 or no congestion.
> > > >
> > > > Regards
> > > > Kamesh Kaul
> > > > www.hssworld.com
> > > > Hughes Software Systems
> > > >
> > > >
> > > >
> > > >
> > > > "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on
07/15/2003
> > > > 10:28:40 AM
> > > >
> > > > Please respond to bidulock@openss7.org
> > > >
> > > > Sent by:    sigtran-admin@ietf.org
> > > >
> > > >
> > > > To:    Kamesh Kaul/HSS@HSS
> > > > cc:    "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
> > > >        sigtran@ietf.org
> > > >
> > > > Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> > > >
> > > >
> > > > kakaul,
> > > >
> > > > Then add text describing how to use the timer in an interoperable
> way.
> > > > Also, please describe why the peer needs to know about the
existence
> > > > of this timer to remain interoperable.  Otherwise, there would not
be
> > > > any need to describe it.
> > > >
> > > > --brian
> > > >
> > > >
> > > > kakaul@hss.hns.com wrote:
> > > > (Tue, 15 Jul 2003 09:35:27)
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > Hi Javier,
> > > > >                     This timer may be used by some
implementations
> to
> > > > > supervise the congestion status of point code if SCON messages
send
> > by
> > > > SGP
> > > > > are lost in transit to ASP.It was discussed earlier and was
agreed
> > upon
> > > > by
> > > > > all to keep this timer.I disagree with removing the mention of
the
> > > timer.
 > > > > >
> > > > > Regards
> > > > > Kamesh Kaul
> > > > > www.hssworld.com
> > > > > Hughes Software Systems
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > "J.Javier Pastor (EE/EEM)"
<j.javier.pastor@ericsson.com>@ietf.org
> on
> > > > > 07/14/2003 04:05:58 PM
> > > > >
> > > > > Sent by:    sigtran-admin@ietf.org
> > > > >
> > > > >
> > > > > To:    "'Brian F. G. Bidulock '" <bidulock@openss7.org>
> > > > > cc:    "''sigtran@ietf.org' '" <sigtran@ietf.org>
> > > > >
> > > > > Subject:    RE: [Sigtran] M3UA IG v04 Section 3.18
> > > > >
> > > > >
> > > > > Brian,
> > > > >
> > > > > We'll remove the mention of the timer, if nobody disagrees...
> > > > >
> > > > > Cheers // Javier.
> > > > >
> > > > > -----Original Message-----
> > > > > From: Brian F. G. Bidulock
> > > > > To: J.Javier Pastor (EE/EEM)
> > > > > Cc: 'sigtran@ietf.org'
> > > > > Sent: 11/07/2003 11:07
> > > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > > > >
> > > > > J.Javier,
> > > > >
> > > > > J.Javier Pastor (EE/EEM) wrote:
> > > > > (Fri, 11 Jul 2003 09:56:59)
> > > > > > Hi Brian,
> > > > > >
> > > > > > Pls, see below...
> > > > > >
> > > > > > > -----Original Message-----
> > > > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > > > Sent: jueves, 10 de julio de 2003 15:34
> > > > > > > To: J.Javier Pastor (EE/EEM)
> > > > > > > Cc: 'sigtran@ietf.org'
> > > > > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > > > > > >
> > > > > > >
> > > > > > > Javier,
> > > > > > >
> > > > > > > 3.18.3 Solution Description
> > > > > > >
> > > > > > >    Clarify that the ASP needs to indicate when the congestion
> > > > > > > level has
> > > > > > >    changed (including abatement).  Further, the ASP peer can
> > > > > maintain
> > > > > > >    a timer, if desired, in case the ASP fails to indicate
> > > congestion
> > > > > > >    abatement.
> > > > > > >
> > > > > > > Where are the procedures for this timer described?  This is
the
> > > only
> > > > > place
> > > > > > > that the timer is mentioned.  Please describe the timer or
> remove
> > > > > the
> > > > > > > sentence saying that it has been described...
> > > > > > >
> > > > > >
> > > > > > Do you think is worth include the timer explanation into the
new
> > > text?
> > > > > > Otherwise, we may remove that sentence.
> > > > >
> > > > > Please remove mention of the timer.
> > > > >
> > > > > --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
> > > >
> > >
> > > --
> > > 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



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



From exim@www1.ietf.org  Fri Jul 18 13:26:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA05989
	for <sigtran-archive@odin.ietf.org>; Fri, 18 Jul 2003 13:26:32 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dYzc-0006Fp-Ie
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 13:26:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6IHQ4ru024035
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 13:26:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dYza-0006FQ-1d; Fri, 18 Jul 2003 13:26:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dYzK-0006FB-UA
	for sigtran@optimus.ietf.org; Fri, 18 Jul 2003 13:25:47 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA05954
	for <sigtran@ietf.org>; Fri, 18 Jul 2003 13:25:43 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dYzH-0002BH-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 13:25:43 -0400
Received: from ottgw.pt.com ([206.191.2.130] helo=pt.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19dYz6-0002B2-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 13:25:32 -0400
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id NAA23977;
	Fri, 18 Jul 2003 13:24:30 -0400
Message-ID: <3F182D52.5040005@pt.com>
Date: Fri, 18 Jul 2003 13:24:34 -0400
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: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - T6
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030712002921.A21136@openss7.org> <3F15CD78.6020209@pt.com> <20030716185758.E6725@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

Hi Brian,

Thanks for the clarifications, they make sense.

I still have some concerns about your proposal though. As you pointed 
out, an MTP2 that is sending SIB is in congestion discard. Since the 
busy MTP2 is discarding messages then it should be able to alleviate 
receive congestion. Since M2PA is not able to discard messages though 
I'm not sure what the analogue of "congestion discard" should be. The 
options seem to be:
1) pass the messages to MTP3 and continue to ack them, hoping that 
receive congestion will abate (thus allowing the RB to drain)
2) queue the messages at M2PA without acknowledgment and hope that 
receive congestion abates (here the RB grows and so do the TB/RTB of the 
sender)
3) introduce retransmissions to M2PA so the M2PA layer can discard 
messages (for behaviour more like MTP2). This option is here only for 
completeness since I don't think anyone really wants to see 
retransmissions in M2PA.

I think either option (1) or (2) is acceptable in that if the link fails 
any undelivered messages can be retrieved from the sending side.
It seems to me that your are advocating option (2), but I don't see how 
this is going to allow receive congestion to clear, it seems like the RB 
at the Busy M2PA will just keep growing if it can't deliver messages to 
MTP3. This also means that if the link fails then the changeover has to 
recover more messages from the senders TB/RTB, introducing a larger 
burst to the alternate links. Also, if you ever have to drop a message 
due to lack of resources then you have no choice but to fail the link or 
silently lose messages... and if you're busy it is probably the most 
likely time to run out of resources.

Thoughts?
Andrew

Brian F. G. Bidulock wrote:

>Andrew,
>
>Andrew Booth wrote:                                                                 (Wed, 16 Jul 2003 18:11:04)
>  
>
>>Hi Brian,
>>
>>Comments inline.
>>
>>
>>
>>Brian F. G. Bidulock wrote:
>>
>>    
>>
>>>M2PA-09 says:
>>>
>>> 
>>>
>>>      
>>>
>>>>M2PA SHALL continue transmitting and acknowledging messages while it
>>>>is in receive congestion.
>>>>   
>>>>
>>>>        
>>>>
>>>Q.703 says:
>>>
>>>| The receiving level 2 also withholds acknowledgement of the message signal
>>>| unit, which triggered off the congestion detection, and of message signal
>>>| units received during the congestion situation; that is fill-in signal units
>>>| or message signal units are sent as usual, but with the backward sequence
>>>| number and backwards indicator bit assigned the values which are contained
>>>| in the last transmitted signal unit before the congestion is recognized.
>>>
>>>It is rather important that the busy end not acknowledge messages while it is
>>>in receive congestion.  This is because if the receive congestion results in
>>>link failure, acknowledged messages caught at the congested receiver will be
>>>unavailable for retransmission on an alternate link at the transmitter.
>>>
>>>      
>>>
>>Are you assuming here that the Busy end is not delivering messages to MTP3?
>>    
>>
>
>Not really assuming.  Q.703 only sends SIB when it is in Congestion Discard.
>Meaning that it is receive congested to the point that it is discarding
>messages.  So I suppose the discarded messages cannot be delivered to MTP3.
>Messages in queue (causing the congestion) are still delivered.
>
>Also, Q.703 does not stop sending SIB until there is no receive congestion.
>So the receive side passes through Congestion Accept to No Congestion before
>it will stop SIBing.  This is the necessary hysteresis to avoid congestion
>thrashing.
>
>-X-snip-X-
>  
>
>>>>   
>>>>
>>>>        
>>>>
>>>But T6 is not necessarily running during the busy condition.
>>>
>>>      
>>>
>>I don't understand this comment.  T6 is started when LSB is received and 
>>stopped when LSBE is received.  Doesn't that mean that T6 is always 
>>running if the peer is busy?
>>    
>>
>
>Q.703 on "first reception of status indication "B" starts a longer supervision
>timer T6 (see 12.3) only if there are message signal units in the
>retransmission buffer".  That is, T6 is only running if the remote is in
>receive congestion and the local side has unacknowledged messages.  There is
>the case where the remote is congested and there are no unacknowledged
>messages. (The last acknowledge message triggered the receive congestion
>discard situation.)  In this latter case, Q.703, T1.111.3 do not start T6.
>
>So it is not true that T6 is always running when the peer is in receive
>congestion.  In some cases the peer has send BUSY and T6 is not running.
>
>--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 exim@www1.ietf.org  Fri Jul 18 13:48:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA06252
	for <sigtran-archive@odin.ietf.org>; Fri, 18 Jul 2003 13:48:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dZKt-0006yN-1B
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 13:48:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6IHm39I026796
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 13:48:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dZKr-0006xw-Co; Fri, 18 Jul 2003 13:48:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dZKf-0006wb-Gh
	for sigtran@optimus.ietf.org; Fri, 18 Jul 2003 13:47:49 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA06244
	for <sigtran@ietf.org>; Fri, 18 Jul 2003 13:47:46 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dZKd-0002Fh-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 13:47:47 -0400
Received: from ottgw.pt.com ([206.191.2.130] helo=pt.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19dZKS-0002FY-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 13:47:36 -0400
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id NAA12951;
	Fri, 18 Jul 2003 13:46:49 -0400
Message-ID: <3F18328D.1080706@pt.com>
Date: Fri, 18 Jul 2003 13:46:53 -0400
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: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - RPO
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030712170744.A11077@openss7.org> <3F15C7F2.6070704@pt.com> <20030716191143.F6725@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

Hi Brian,

I agree that it makes sense to support both scenarios, where the MTP may 
accept user data after RPO or not.  Do both approaches work with the 
transparent interworking function you mentioned earlier?  It seems like 
in that case the interworking function could not properly process Data 
received after RPO without leaving RPO.

More comments inline.

Brian F. G. Bidulock wrote:

>Andrew,
>
>Yes, we discussed this.  The proposal was to permit both situations.
>
>If you have a MTP3 that will not accept MSUs after RPO you have another
>problem: messages send by the remote before sending LSPO, but which arrived
>at the peer after the LSPO because the LPSO was received (and supposedly
>expedited) on stream 0.
>
Agreed.  They have to queued also, and will potentially be lost by a flush.

>
>To accomodate such implementations of MTP3 (completely valid implementations
>from the SS7 standards perspective), it would be a better idea to send LSPO
>on stream 1.  Then LSPR also should be sent on stream 1, otherwise there is
>the possibiliy that User Data sent after sending LSPR will arrive before the
>LSPR.
>
I'm solidly on the fence here.  I agree that M2PA gets more complicated 
if LSPO/LSPR are not sent on the data stream, I have mixed feelings 
about whether the potential performance benefits outweigh the complexity.

>
>The side experiencing PO should accept the acknowledgement of any message
>sent before or during PO.  This permits the TB/RTB to drain and has no
>interaction with L3.
>
>The statement below really says that the side experienceing RPO SHOULD
>continue to acknowledge menssages received and accepted by MTP3.  If MTP3
>will not accept messages during the RPO condition, then of course they 
>cannot be acknowledged.  However, if MTP3 is accepting messages during
>RPO (as some equally valid implementations do), then they SHOULD be
>acknowledged.
>
Sounds good.

>
>This leaves both options open.  The current M2PA-09 text does not leave the
>option open to continue delivering to MTP3 during RPO if MTP3 is still ready
>to accept messages.
>
Sounds good to me.  It seems important though to keep the text that 
describes how the receive buffers are handled in PO for implementations 
that cannot deliver DATA after RPO.

>
>This also allows the M2PA working with an MTP3 that accepts MSUs during RPO
>to not look in the sequence numbers of LSPO and LSPR sent on stream 0 to
>determine which received messages preceded or followed the RPO condition and
>which were between LSPO and LSPR (becaues they are all delivered to MTP3).
>
>Make sense?
>
Yes, thanks.

Andrew

>
>--brian
>
>
>Andrew Booth wrote:                                                                 (Wed, 16 Jul 2003 17:47:30)
>  
>
>>Hi Brian,
>>
>>If the side in RPO is sending data acknowledgments then it must also be 
>>delivering these messages to MTP3.  Is there some guarantee that MTP3 
>>will accept messages from a link that it has marked as blocked?  Or am I 
>>misunderstanding the intent of your proposal?
>>
>>Andrew
>>
>>Brian F. G. Bidulock wrote:
>>
>>    
>>
>>>M2PA-09 says:
>>>
>>>While there is a remote processor outage condition:
>>>
>>>  (a) Any User Data messages received from the peer after the Link
>>>      Status Processor Outage MUST NOT be acknowledged and MUST be
>>>      buffered.
>>>
>>>  (b) M2PA SHOULD continue to acknowledge User Data messages received
>>>      and accepted by MTP3 before the remote processor outage.
>>>
>>>In Q.703 the side experiencing PO cannot send MSUs or FISUs without indicating a
>>>recovery from processor outage.  M2PA can send User Data without indicating
>>>processor recovery.  There is no reason for the receiver of User Data not to
>>>process it even after receiving Link Status Processor Outage.
>>>
>>>Otherwise, if (a) above is followed, MSUs in the transmission and retransmission
>>>buffers at the time that local processor outage occured will be trapped.  If
>>>User Data is acknowledged by the peer, they will be processed by the receiver.
>>>
>>>Change M2PA to say:
>>>
>>>While there is a remote processour outage condition:
>>>
>>>  (a) M2PA SHOULD continue to acknowledge User Data messages received
>>>      and accepted by MTP3 regardless of the remote processor outage.
>>>
>>>This allows the TB/RTB on the side experiencing processor outage to drain.
>>>
>>>
>>>M2PA-09 says:
>>>
>>> Note that because the Link Status and User Data messages are sent on
>>> separate streams, it is possible for a User Data message sent after
>>> the Link Status Processor Recovered message to arrive before
>>> it. Therefore, when flushing buffers, M2PA must take care not to flush
>>> messages sent after the remote processor recovered. M2PA SHOULD NOT
>>> flush User Data messages with FSN greater than the FSN in the Link
>>> Status Processor Recovered message. 
>>>
>>>MTP3 does not flush the RB after RPO or before RPR (just the RTB), unless it
>>>uses the remote processor outage to local processor outage latching option, and
>>>then it is only because of the LPO condition not the RPO condition.
>>>
>>>With the change above, M2PA receives and acknowledges User Data before, during
>>>and after remote processor outage, and comparison to any sequence numbers in the
>>>Link Status Processor Recovered is unnecessary.
>>>
>>>
>>>M2PA-09 says:
>>>
>>> When the local processor outage condition ends, M2PA SHALL send a Link
>>> Status Processor Recovered message to its peer. This message is used
>>> to signal in the end of the processor outage condition, instead of an
>>> MSU or FISU as in MTP2. The BSN value in this message is used by the
>>> peer to resynchronize its sequence numbers, if this is required by
>>> MTP2.
>>>
>>>After a flush buffers operation, the side that experienced PO uses the first
>>>received FISU/MSU to resync and discards the FISU/MSU.
>>>
>>>ANSI T1.111.3/2000 Section 8 says:
>>>
>>> ... If level 3 requests level 2 [at the side experiencing processor outage] to
>>> clear its buffers of message signal units, level 2 begins to continuously
>>> transmit fill-in signal units and discards all message signal units that it
>>> has received from remote level 2 but not yet delivered to local level 3 and
>>> all message signal units that it has received from local level 3 but not yet
>>> transmitted on the link.  The first fill-in signal unit transmitted
>>> acknowledges all message signal units that have been accepted from remote
>>> level 2 including any mesage signal units for which acknowledgements were
>>> previously withheld because of level 2 congestion (see section 9).  Level 2
>>> then begins accepting message signal units from remote Level 2. When the next
>>> fill-in or message signal unit is received that passes the acceptance
>>> procedure of section 4 and sections 5 or 6,
>>>
>>>  (1) all message signal units that are available for retransmission are
>>>      discarded [clear RTB].
>>>
>>>  (2) the forward sequence number and forward indicator bit of the
>>>      transmitted fill-in signal units are set equal to the backward sequence
>>>      number and backward indicator bit of the received signal unit,
>>>
>>>  (3) normal transmission of message signal units and fill-in signal units is
>>>      resumed, and
>>>
>>>  (4) level 3 is notified that the buffer clearing has been completed and
>>>      that normal operations has been resumed.
>>>
>>>So, you see, the side experiencing the processor outage, after clearing buffers,
>>>resynchronizes its FSN with the BSN of the first received FISU or MSU.  The side
>>>experiencing processor outage could use the first User Data message received
>>>after processor recovery to resyncronize.  However, if the peer has nothing to
>>>send, processor recovery will be delayed.
>>>
>>>Note, however, that resynchronization of sequence numbers for the transmit
>>>direction is necessary because MTP2 traps messages in the TB/RTB of the side
>>>experiencing processor outage because it must continuously transmit SIPO during
>>>processor outage.  Permitting the draining of the TB/RTB with the change above
>>>removes the need to flush the TB/RTB on the side experiencing processor outage,
>>>as well as removing the need to resynchronize sequence numbers.
>>>
>>>In the other direction, for messages queued to be sent to the side experiencing
>>>processour outage, flushing and resynchronization is still necessary (actions
>>>(1) to (4) per ANSI above).
>>>
>>>However, Q.703 permits FISUs/MSUs received by the side experiencing remote
>>>processor outage to discard the first and several subsequent FISU/MSUs when
>>>recovering.  There is even a test case in Q.781 (8.9) that shows an MSU being
>>>discarded (and negative acknowledged) in this fashion.  Because M2PA cannot
>>>discard and expect retransmission of a message, and because sychronization is
>>>not required when the TB/RTB of the side experiencing processor outage is
>>>permitted to drain with the changes above, M2PA cannot permit messages to be
>>>discarded in this fashion.
>>>
>>>M2PA-09 says:
>>>
>>> If M2PA receives a Flush command from MTP3, M2PA SHALL discard the
>>> incoming messages that were queued and unacknowledged during the
>>> processor outage condition. Furthermore, M2PA SHALL discard messages
>>> in the transmit and retransmit queues as required by MTP2. 
>>>
>>>To handle flushing and resynchronization properly in both directions, change the
>>>above to read as follows:
>>>
>>>Change M2PA to say:
>>>
>>> If M2PA receives a Flush command from MTP3, M2PA SHALL discard the
>>> incoming messages that were queued and unacknowledged during the
>>> processor outage condition.  Furthermore, the M2PA experiencing
>>> processor outage SHALL NOT discard message from the transmit and
>>> retransmit queues and SHALL NOT perform procedures to resynchronize
>>> sequence numbers, but SHALL continue normal transmission of User Data.
>>> The M2PA experiencing remote processor outage, SHALL discard messages in
>>> the transmit and retransmit queues as required by MTP2, and MAY perform
>>> resynchronization of sequence numbers using the BSN from the received
>>> Link Status Processor Recovered message.  However, during
>>> resychronization, M2PA SHALL NOT discard any received User Data
>>> messages.
>>>
>>>As Q.2140 handles processor outage by putting the link out of service, also add:
>>>
>>> When M2PA experiences a local processor outage, it MAY put the link out
>>> of service by sending a Link Status Out of Service message if allowed by
>>> the applicable MTP2 standard (e.g, [Q.2140]).
>>>
>>>I think that wraps up processor outage problems.
>>>
>>>--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.
>>
>>    
>>
>
>  
>

-- 
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 exim@www1.ietf.org  Fri Jul 18 13:56:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA06462
	for <sigtran-archive@odin.ietf.org>; Fri, 18 Jul 2003 13:56:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dZSd-0007Do-IE
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 13:56:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6IHu3Ua027755
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 13:56:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dZSc-0007DN-IQ; Fri, 18 Jul 2003 13:56:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dZSG-0007D6-K5
	for sigtran@optimus.ietf.org; Fri, 18 Jul 2003 13:55:40 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA06454
	for <sigtran@ietf.org>; Fri, 18 Jul 2003 13:55:37 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dZSE-0002IL-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 13:55:38 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[uHSNE+ZPv0Ge6ObYW5hnyT9hYZKJmv/R])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dZRy-0002Hx-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 13:55:22 -0400
Received: from ns.pigworks.openss7.net (IDENT:an66PoyZAePTu1jttOcYX3Grv49T66xr@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6IHsVP16505;
	Fri, 18 Jul 2003 11:54:31 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6IHsVh15268;
	Fri, 18 Jul 2003 11:54:31 -0600
Date: Fri, 18 Jul 2003 11:54:31 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Andrew Booth <abooth@pt.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - T6
Message-ID: <20030718115431.B14760@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Andrew Booth <abooth@pt.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030712002921.A21136@openss7.org> <3F15CD78.6020209@pt.com> <20030716185758.E6725@openss7.org> <3F182D52.5040005@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: <3F182D52.5040005@pt.com>; from abooth@pt.com on Fri, Jul 18, 2003 at 01:24:34PM -0400
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:                                                                 (Fri, 18 Jul 2003 13:24:34)
> Hi Brian,
> 
> Thanks for the clarifications, they make sense.
> 
> I still have some concerns about your proposal though. As you pointed 
> out, an MTP2 that is sending SIB is in congestion discard. Since the 
> busy MTP2 is discarding messages then it should be able to alleviate 
> receive congestion. Since M2PA is not able to discard messages though 
> I'm not sure what the analogue of "congestion discard" should be. The 
> options seem to be:
> 1) pass the messages to MTP3 and continue to ack them, hoping that 
> receive congestion will abate (thus allowing the RB to drain)

In some implementations Congestion Discard only occurs once it is not
possible to pass messages to MTP3 (e.g. queues to MTP3 from MTP2 are full)
leaving the only choice to discard.

> 2) queue the messages at M2PA without acknowledgment and hope that 
> receive congestion abates (here the RB grows and so do the TB/RTB of the 
> sender)

This is what I was proposing instead of (1).  M2PA's RB will not grow
after LSB is sent.  This is because the sender MUST stop sending User
Data once LSB is received.  The sender's TB/RTB will grow, but that is
what happens in Q.703 after SIB anyway.

> 3) introduce retransmissions to M2PA so the M2PA layer can discard 
> messages (for behaviour more like MTP2). This option is here only for 
> completeness since I don't think anyone really wants to see 
> retransmissions in M2PA.

Not necessary, (2) works fine.

> any undelivered messages can be retrieved from the sending side.
> It seems to me that your are advocating option (2), but I don't see how 
> this is going to allow receive congestion to clear, it seems like the RB 
> at the Busy M2PA will just keep growing if it can't deliver messages to 
> MTP3.

The RB should not grow after LSB.  The sender SHOULD NOT transmit messages
after receiving LSB.  Yes, Q.703 keeps sending event after SIB, but the
other end is dicarding so they are useless transmission at that point.  There
is no reason for the sender to keep sending after LSB and before LSBE.

> This also means that if the link fails then the changeover has to 
> recover more messages from the senders TB/RTB, introducing a larger 
> burst to the alternate links.

This is no different than Q.703.  Q.703 withhold acknowledgements and discards
messages during receive congestion.  It even discards them, so the NACK when
busy ends causes a burst on the same link even if changeover has not occured.
Q.703 receiver in congestion does not ack messages, so TB/RTB growth in M2PA
will be no worse than Q.703.

> Also, if you ever have to drop a message due to lack of resources then you
> have no choice but to fail the link or silently lose messages... and if
> you're busy it is probably the most likely time to run out of resources.

We'll just say that when the M2PA peer receives LS Busy, that it MUST NOT
send User Data until it receives LS Busy Ended.  That will curtail RB growth,
and follow Q.703 closely.

In addition the the changes I mentioned before, change:

  The peer M2PA SHOULD continue transmitting and acknowledging messages
  while its T6 timer is running, i.e. while the other end is busy.

 to:

  The peer M2PA SHOULD continue receiving and acknowledging messages while
  the other end is busy, but MUST NOT send User Data after receiving Link
  Status Busy and before receiving Link Status Busy Ended.

It think it was a type to say "transmitting" here anyway.

--brian


> 
> Thoughts?
> Andrew
> 
> Brian F. G. Bidulock wrote:
> 
> >Andrew,
> >
> >Andrew Booth wrote:                                                                 (Wed, 16 Jul 2003 18:11:04)
> >  
> >
> >>Hi Brian,
> >>
> >>Comments inline.
> >>
> >>
> >>
> >>Brian F. G. Bidulock wrote:
> >>
> >>    
> >>
> >>>M2PA-09 says:
> >>>
> >>> 
> >>>
> >>>      
> >>>
> >>>>M2PA SHALL continue transmitting and acknowledging messages while it
> >>>>is in receive congestion.
> >>>>   
> >>>>
> >>>>        
> >>>>
> >>>Q.703 says:
> >>>
> >>>| The receiving level 2 also withholds acknowledgement of the message signal
> >>>| unit, which triggered off the congestion detection, and of message signal
> >>>| units received during the congestion situation; that is fill-in signal units
> >>>| or message signal units are sent as usual, but with the backward sequence
> >>>| number and backwards indicator bit assigned the values which are contained
> >>>| in the last transmitted signal unit before the congestion is recognized.
> >>>
> >>>It is rather important that the busy end not acknowledge messages while it is
> >>>in receive congestion.  This is because if the receive congestion results in
> >>>link failure, acknowledged messages caught at the congested receiver will be
> >>>unavailable for retransmission on an alternate link at the transmitter.
> >>>
> >>>      
> >>>
> >>Are you assuming here that the Busy end is not delivering messages to MTP3?
> >>    
> >>
> >
> >Not really assuming.  Q.703 only sends SIB when it is in Congestion Discard.
> >Meaning that it is receive congested to the point that it is discarding
> >messages.  So I suppose the discarded messages cannot be delivered to MTP3.
> >Messages in queue (causing the congestion) are still delivered.
> >
> >Also, Q.703 does not stop sending SIB until there is no receive congestion.
> >So the receive side passes through Congestion Accept to No Congestion before
> >it will stop SIBing.  This is the necessary hysteresis to avoid congestion
> >thrashing.
> >
> >-X-snip-X-
> >  
> >
> >>>>   
> >>>>
> >>>>        
> >>>>
> >>>But T6 is not necessarily running during the busy condition.
> >>>
> >>>      
> >>>
> >>I don't understand this comment.  T6 is started when LSB is received and 
> >>stopped when LSBE is received.  Doesn't that mean that T6 is always 
> >>running if the peer is busy?
> >>    
> >>
> >
> >Q.703 on "first reception of status indication "B" starts a longer supervision
> >timer T6 (see 12.3) only if there are message signal units in the
> >retransmission buffer".  That is, T6 is only running if the remote is in
> >receive congestion and the local side has unacknowledged messages.  There is
> >the case where the remote is congested and there are no unacknowledged
> >messages. (The last acknowledge message triggered the receive congestion
> >discard situation.)  In this latter case, Q.703, T1.111.3 do not start T6.
> >
> >So it is not true that T6 is always running when the peer is in receive
> >congestion.  In some cases the peer has send BUSY and T6 is not running.
> >
> >--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.
> 

-- 
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 exim@www1.ietf.org  Fri Jul 18 14:02:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA06614
	for <sigtran-archive@odin.ietf.org>; Fri, 18 Jul 2003 14:02:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dZYP-0007Pb-Mm
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 14:02:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6II21Ka028482
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 14:02:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dZYP-0007P9-8a; Fri, 18 Jul 2003 14:02:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dZXn-0007Ny-3I
	for sigtran@optimus.ietf.org; Fri, 18 Jul 2003 14:01:23 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA06589
	for <sigtran@ietf.org>; Fri, 18 Jul 2003 14:01:20 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dZXk-0002KA-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 14:01:20 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[FvZ0fQ+GrKiNiagiKYQy66EVyAWbqI/P])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dZXU-0002K2-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 14:01:04 -0400
Received: from ns.pigworks.openss7.net (IDENT:DXBMcINml7ZnSaH3LvM8RjnwFu59B/xN@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6II0sP16849;
	Fri, 18 Jul 2003 12:00:54 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6II0sE15425;
	Fri, 18 Jul 2003 12:00:54 -0600
Date: Fri, 18 Jul 2003 12:00:53 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Andrew Booth <abooth@pt.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - RPO
Message-ID: <20030718120053.C14760@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Andrew Booth <abooth@pt.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030712170744.A11077@openss7.org> <3F15C7F2.6070704@pt.com> <20030716191143.F6725@openss7.org> <3F18328D.1080706@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: <3F18328D.1080706@pt.com>; from abooth@pt.com on Fri, Jul 18, 2003 at 01:46:53PM -0400
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:                                                                 (Fri, 18 Jul 2003 13:46:53)
> Hi Brian,
> 
> I agree that it makes sense to support both scenarios, where the MTP may 
> accept user data after RPO or not.  Do both approaches work with the 
> transparent interworking function you mentioned earlier?  It seems like 
> in that case the interworking function could not properly process Data 
> received after RPO without leaving RPO.

I think so.  The clearing buffers is the same.  The side in RPO is just
permitted to acknowledge messages where it was not before.

> 
> More comments inline.
> 
> Brian F. G. Bidulock wrote:
> 
> >Andrew,
> >
> >Yes, we discussed this.  The proposal was to permit both situations.
> >
> >If you have a MTP3 that will not accept MSUs after RPO you have another
> >problem: messages send by the remote before sending LSPO, but which arrived
> >at the peer after the LSPO because the LPSO was received (and supposedly
> >expedited) on stream 0.
> >
> Agreed.  They have to queued also, and will potentially be lost by a flush.
> 
> >
> >To accomodate such implementations of MTP3 (completely valid implementations
> >from the SS7 standards perspective), it would be a better idea to send LSPO
> >on stream 1.  Then LSPR also should be sent on stream 1, otherwise there is
> >the possibiliy that User Data sent after sending LSPR will arrive before the
> >LSPR.
> >
> I'm solidly on the fence here.  I agree that M2PA gets more complicated 
> if LSPO/LSPR are not sent on the data stream, I have mixed feelings 
> about whether the potential performance benefits outweigh the complexity.
> 
> >
> >The side experiencing PO should accept the acknowledgement of any message
> >sent before or during PO.  This permits the TB/RTB to drain and has no
> >interaction with L3.
> >
> >The statement below really says that the side experienceing RPO SHOULD
> >continue to acknowledge menssages received and accepted by MTP3.  If MTP3
> >will not accept messages during the RPO condition, then of course they 
> >cannot be acknowledged.  However, if MTP3 is accepting messages during
> >RPO (as some equally valid implementations do), then they SHOULD be
> >acknowledged.
> >
> Sounds good.
> 
> >
> >This leaves both options open.  The current M2PA-09 text does not leave the
> >option open to continue delivering to MTP3 during RPO if MTP3 is still ready
> >to accept messages.
> >
> Sounds good to me.  It seems important though to keep the text that 
> describes how the receive buffers are handled in PO for implementations 
> that cannot deliver DATA after RPO.

It is still there except for:
> >>>                                                 ... M2PA SHOULD NOT
> >>> flush User Data messages with FSN greater than the FSN in the Link
> >>> Status Processor Recovered message. 

The reason I removed this is that RPO latching to LPO option does exactly
that: flushes data messags with FSN greater than the FSN in the Link Status
Processor Recovered message.  What messages are flushed are really the
business of the flusher.

--brian

> 
> >
> >This also allows the M2PA working with an MTP3 that accepts MSUs during RPO
> >to not look in the sequence numbers of LSPO and LSPR sent on stream 0 to
> >determine which received messages preceded or followed the RPO condition and
> >which were between LSPO and LSPR (becaues they are all delivered to MTP3).
> >
> >Make sense?
> >
> Yes, thanks.
> 
> Andrew
> 
> >
> >--brian
> >
> >
> >Andrew Booth wrote:                                                                 (Wed, 16 Jul 2003 17:47:30)
> >  
> >
> >>Hi Brian,
> >>
> >>If the side in RPO is sending data acknowledgments then it must also be 
> >>delivering these messages to MTP3.  Is there some guarantee that MTP3 
> >>will accept messages from a link that it has marked as blocked?  Or am I 
> >>misunderstanding the intent of your proposal?
> >>
> >>Andrew
> >>
> >>Brian F. G. Bidulock wrote:
> >>
> >>    
> >>
> >>>M2PA-09 says:
> >>>
> >>>While there is a remote processor outage condition:
> >>>
> >>>  (a) Any User Data messages received from the peer after the Link
> >>>      Status Processor Outage MUST NOT be acknowledged and MUST be
> >>>      buffered.
> >>>
> >>>  (b) M2PA SHOULD continue to acknowledge User Data messages received
> >>>      and accepted by MTP3 before the remote processor outage.
> >>>
> >>>In Q.703 the side experiencing PO cannot send MSUs or FISUs without indicating a
> >>>recovery from processor outage.  M2PA can send User Data without indicating
> >>>processor recovery.  There is no reason for the receiver of User Data not to
> >>>process it even after receiving Link Status Processor Outage.
> >>>
> >>>Otherwise, if (a) above is followed, MSUs in the transmission and retransmission
> >>>buffers at the time that local processor outage occured will be trapped.  If
> >>>User Data is acknowledged by the peer, they will be processed by the receiver.
> >>>
> >>>Change M2PA to say:
> >>>
> >>>While there is a remote processour outage condition:
> >>>
> >>>  (a) M2PA SHOULD continue to acknowledge User Data messages received
> >>>      and accepted by MTP3 regardless of the remote processor outage.
> >>>
> >>>This allows the TB/RTB on the side experiencing processor outage to drain.
> >>>
> >>>
> >>>M2PA-09 says:
> >>>
> >>> Note that because the Link Status and User Data messages are sent on
> >>> separate streams, it is possible for a User Data message sent after
> >>> the Link Status Processor Recovered message to arrive before
> >>> it. Therefore, when flushing buffers, M2PA must take care not to flush
> >>> messages sent after the remote processor recovered. M2PA SHOULD NOT
> >>> flush User Data messages with FSN greater than the FSN in the Link
> >>> Status Processor Recovered message. 
> >>>
> >>>MTP3 does not flush the RB after RPO or before RPR (just the RTB), unless it
> >>>uses the remote processor outage to local processor outage latching option, and
> >>>then it is only because of the LPO condition not the RPO condition.
> >>>
> >>>With the change above, M2PA receives and acknowledges User Data before, during
> >>>and after remote processor outage, and comparison to any sequence numbers in the
> >>>Link Status Processor Recovered is unnecessary.
> >>>
> >>>
> >>>M2PA-09 says:
> >>>
> >>> When the local processor outage condition ends, M2PA SHALL send a Link
> >>> Status Processor Recovered message to its peer. This message is used
> >>> to signal in the end of the processor outage condition, instead of an
> >>> MSU or FISU as in MTP2. The BSN value in this message is used by the
> >>> peer to resynchronize its sequence numbers, if this is required by
> >>> MTP2.
> >>>
> >>>After a flush buffers operation, the side that experienced PO uses the first
> >>>received FISU/MSU to resync and discards the FISU/MSU.
> >>>
> >>>ANSI T1.111.3/2000 Section 8 says:
> >>>
> >>> ... If level 3 requests level 2 [at the side experiencing processor outage] to
> >>> clear its buffers of message signal units, level 2 begins to continuously
> >>> transmit fill-in signal units and discards all message signal units that it
> >>> has received from remote level 2 but not yet delivered to local level 3 and
> >>> all message signal units that it has received from local level 3 but not yet
> >>> transmitted on the link.  The first fill-in signal unit transmitted
> >>> acknowledges all message signal units that have been accepted from remote
> >>> level 2 including any mesage signal units for which acknowledgements were
> >>> previously withheld because of level 2 congestion (see section 9).  Level 2
> >>> then begins accepting message signal units from remote Level 2. When the next
> >>> fill-in or message signal unit is received that passes the acceptance
> >>> procedure of section 4 and sections 5 or 6,
> >>>
> >>>  (1) all message signal units that are available for retransmission are
> >>>      discarded [clear RTB].
> >>>
> >>>  (2) the forward sequence number and forward indicator bit of the
> >>>      transmitted fill-in signal units are set equal to the backward sequence
> >>>      number and backward indicator bit of the received signal unit,
> >>>
> >>>  (3) normal transmission of message signal units and fill-in signal units is
> >>>      resumed, and
> >>>
> >>>  (4) level 3 is notified that the buffer clearing has been completed and
> >>>      that normal operations has been resumed.
> >>>
> >>>So, you see, the side experiencing the processor outage, after clearing buffers,
> >>>resynchronizes its FSN with the BSN of the first received FISU or MSU.  The side
> >>>experiencing processor outage could use the first User Data message received
> >>>after processor recovery to resyncronize.  However, if the peer has nothing to
> >>>send, processor recovery will be delayed.
> >>>
> >>>Note, however, that resynchronization of sequence numbers for the transmit
> >>>direction is necessary because MTP2 traps messages in the TB/RTB of the side
> >>>experiencing processor outage because it must continuously transmit SIPO during
> >>>processor outage.  Permitting the draining of the TB/RTB with the change above
> >>>removes the need to flush the TB/RTB on the side experiencing processor outage,
> >>>as well as removing the need to resynchronize sequence numbers.
> >>>
> >>>In the other direction, for messages queued to be sent to the side experiencing
> >>>processour outage, flushing and resynchronization is still necessary (actions
> >>>(1) to (4) per ANSI above).
> >>>
> >>>However, Q.703 permits FISUs/MSUs received by the side experiencing remote
> >>>processor outage to discard the first and several subsequent FISU/MSUs when
> >>>recovering.  There is even a test case in Q.781 (8.9) that shows an MSU being
> >>>discarded (and negative acknowledged) in this fashion.  Because M2PA cannot
> >>>discard and expect retransmission of a message, and because sychronization is
> >>>not required when the TB/RTB of the side experiencing processor outage is
> >>>permitted to drain with the changes above, M2PA cannot permit messages to be
> >>>discarded in this fashion.
> >>>
> >>>M2PA-09 says:
> >>>
> >>> If M2PA receives a Flush command from MTP3, M2PA SHALL discard the
> >>> incoming messages that were queued and unacknowledged during the
> >>> processor outage condition. Furthermore, M2PA SHALL discard messages
> >>> in the transmit and retransmit queues as required by MTP2. 
> >>>
> >>>To handle flushing and resynchronization properly in both directions, change the
> >>>above to read as follows:
> >>>
> >>>Change M2PA to say:
> >>>
> >>> If M2PA receives a Flush command from MTP3, M2PA SHALL discard the
> >>> incoming messages that were queued and unacknowledged during the
> >>> processor outage condition.  Furthermore, the M2PA experiencing
> >>> processor outage SHALL NOT discard message from the transmit and
> >>> retransmit queues and SHALL NOT perform procedures to resynchronize
> >>> sequence numbers, but SHALL continue normal transmission of User Data.
> >>> The M2PA experiencing remote processor outage, SHALL discard messages in
> >>> the transmit and retransmit queues as required by MTP2, and MAY perform
> >>> resynchronization of sequence numbers using the BSN from the received
> >>> Link Status Processor Recovered message.  However, during
> >>> resychronization, M2PA SHALL NOT discard any received User Data
> >>> messages.
> >>>
> >>>As Q.2140 handles processor outage by putting the link out of service, also add:
> >>>
> >>> When M2PA experiences a local processor outage, it MAY put the link out
> >>> of service by sending a Link Status Out of Service message if allowed by
> >>> the applicable MTP2 standard (e.g, [Q.2140]).
> >>>
> >>>I think that wraps up processor outage problems.
> >>>
> >>>--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.
> >>
> >>    
> >>
> >
> >  
> >
> 
> -- 
> 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.
> 

-- 
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 exim@www1.ietf.org  Fri Jul 18 14:12:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA06994
	for <sigtran-archive@odin.ietf.org>; Fri, 18 Jul 2003 14:12:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dZi5-0007yt-Pz
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 14:12:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6IIC10Q030669
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 14:12:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dZi5-0007yQ-4A; Fri, 18 Jul 2003 14:12:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dZhW-0007yD-2Z
	for sigtran@optimus.ietf.org; Fri, 18 Jul 2003 14:11:26 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA06934
	for <sigtran@ietf.org>; Fri, 18 Jul 2003 14:11:23 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dZhT-0002On-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 14:11:23 -0400
Received: from ottgw.pt.com ([206.191.2.130] helo=pt.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19dZhI-0002Oe-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 14:11:12 -0400
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id OAA06777;
	Fri, 18 Jul 2003 14:10:26 -0400
Message-ID: <3F183816.8010700@pt.com>
Date: Fri, 18 Jul 2003 14:10:30 -0400
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: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - T6
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030712002921.A21136@openss7.org> <3F15CD78.6020209@pt.com> <20030716185758.E6725@openss7.org> <3F182D52.5040005@pt.com> <20030718115431.B14760@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

Hi Brian,

I handn't understood that you wanted the remote end to stop sending to 
the busy end.  Now that I understand that part your proposal makes a lot 
of sense.

Thanks for the clarification.

Andrew

Brian F. G. Bidulock wrote:

>Andrew,
>
>Andrew Booth wrote:                                                                 (Fri, 18 Jul 2003 13:24:34)
>  
>
>>Hi Brian,
>>
>>Thanks for the clarifications, they make sense.
>>
>>I still have some concerns about your proposal though. As you pointed 
>>out, an MTP2 that is sending SIB is in congestion discard. Since the 
>>busy MTP2 is discarding messages then it should be able to alleviate 
>>receive congestion. Since M2PA is not able to discard messages though 
>>I'm not sure what the analogue of "congestion discard" should be. The 
>>options seem to be:
>>1) pass the messages to MTP3 and continue to ack them, hoping that 
>>receive congestion will abate (thus allowing the RB to drain)
>>    
>>
>
>In some implementations Congestion Discard only occurs once it is not
>possible to pass messages to MTP3 (e.g. queues to MTP3 from MTP2 are full)
>leaving the only choice to discard.
>
>  
>
>>2) queue the messages at M2PA without acknowledgment and hope that 
>>receive congestion abates (here the RB grows and so do the TB/RTB of the 
>>sender)
>>    
>>
>
>This is what I was proposing instead of (1).  M2PA's RB will not grow
>after LSB is sent.  This is because the sender MUST stop sending User
>Data once LSB is received.  The sender's TB/RTB will grow, but that is
>what happens in Q.703 after SIB anyway.
>
>  
>
>>3) introduce retransmissions to M2PA so the M2PA layer can discard 
>>messages (for behaviour more like MTP2). This option is here only for 
>>completeness since I don't think anyone really wants to see 
>>retransmissions in M2PA.
>>    
>>
>
>Not necessary, (2) works fine.
>
>  
>
>>any undelivered messages can be retrieved from the sending side.
>>It seems to me that your are advocating option (2), but I don't see how 
>>this is going to allow receive congestion to clear, it seems like the RB 
>>at the Busy M2PA will just keep growing if it can't deliver messages to 
>>MTP3.
>>    
>>
>
>The RB should not grow after LSB.  The sender SHOULD NOT transmit messages
>after receiving LSB.  Yes, Q.703 keeps sending event after SIB, but the
>other end is dicarding so they are useless transmission at that point.  There
>is no reason for the sender to keep sending after LSB and before LSBE.
>
>  
>
>>This also means that if the link fails then the changeover has to 
>>recover more messages from the senders TB/RTB, introducing a larger 
>>burst to the alternate links.
>>    
>>
>
>This is no different than Q.703.  Q.703 withhold acknowledgements and discards
>messages during receive congestion.  It even discards them, so the NACK when
>busy ends causes a burst on the same link even if changeover has not occured.
>Q.703 receiver in congestion does not ack messages, so TB/RTB growth in M2PA
>will be no worse than Q.703.
>
>  
>
>>Also, if you ever have to drop a message due to lack of resources then you
>>have no choice but to fail the link or silently lose messages... and if
>>you're busy it is probably the most likely time to run out of resources.
>>    
>>
>
>We'll just say that when the M2PA peer receives LS Busy, that it MUST NOT
>send User Data until it receives LS Busy Ended.  That will curtail RB growth,
>and follow Q.703 closely.
>
>In addition the the changes I mentioned before, change:
>
>  The peer M2PA SHOULD continue transmitting and acknowledging messages
>  while its T6 timer is running, i.e. while the other end is busy.
>
> to:
>
>  The peer M2PA SHOULD continue receiving and acknowledging messages while
>  the other end is busy, but MUST NOT send User Data after receiving Link
>  Status Busy and before receiving Link Status Busy Ended.
>
>It think it was a type to say "transmitting" here anyway.
>
>--brian
>
>
>  
>
>>Thoughts?
>>Andrew
>>
>>Brian F. G. Bidulock wrote:
>>
>>    
>>
>>>Andrew,
>>>
>>>Andrew Booth wrote:                                                                 (Wed, 16 Jul 2003 18:11:04)
>>> 
>>>
>>>      
>>>
>>>>Hi Brian,
>>>>
>>>>Comments inline.
>>>>
>>>>
>>>>
>>>>Brian F. G. Bidulock wrote:
>>>>
>>>>   
>>>>
>>>>        
>>>>
>>>>>M2PA-09 says:
>>>>>
>>>>>
>>>>>
>>>>>     
>>>>>
>>>>>          
>>>>>
>>>>>>M2PA SHALL continue transmitting and acknowledging messages while it
>>>>>>is in receive congestion.
>>>>>>  
>>>>>>
>>>>>>       
>>>>>>
>>>>>>            
>>>>>>
>>>>>Q.703 says:
>>>>>
>>>>>| The receiving level 2 also withholds acknowledgement of the message signal
>>>>>| unit, which triggered off the congestion detection, and of message signal
>>>>>| units received during the congestion situation; that is fill-in signal units
>>>>>| or message signal units are sent as usual, but with the backward sequence
>>>>>| number and backwards indicator bit assigned the values which are contained
>>>>>| in the last transmitted signal unit before the congestion is recognized.
>>>>>
>>>>>It is rather important that the busy end not acknowledge messages while it is
>>>>>in receive congestion.  This is because if the receive congestion results in
>>>>>link failure, acknowledged messages caught at the congested receiver will be
>>>>>unavailable for retransmission on an alternate link at the transmitter.
>>>>>
>>>>>     
>>>>>
>>>>>          
>>>>>
>>>>Are you assuming here that the Busy end is not delivering messages to MTP3?
>>>>   
>>>>
>>>>        
>>>>
>>>Not really assuming.  Q.703 only sends SIB when it is in Congestion Discard.
>>>Meaning that it is receive congested to the point that it is discarding
>>>messages.  So I suppose the discarded messages cannot be delivered to MTP3.
>>>Messages in queue (causing the congestion) are still delivered.
>>>
>>>Also, Q.703 does not stop sending SIB until there is no receive congestion.
>>>So the receive side passes through Congestion Accept to No Congestion before
>>>it will stop SIBing.  This is the necessary hysteresis to avoid congestion
>>>thrashing.
>>>
>>>-X-snip-X-
>>> 
>>>
>>>      
>>>
>>>>>>  
>>>>>>
>>>>>>       
>>>>>>
>>>>>>            
>>>>>>
>>>>>But T6 is not necessarily running during the busy condition.
>>>>>
>>>>>     
>>>>>
>>>>>          
>>>>>
>>>>I don't understand this comment.  T6 is started when LSB is received and 
>>>>stopped when LSBE is received.  Doesn't that mean that T6 is always 
>>>>running if the peer is busy?
>>>>   
>>>>
>>>>        
>>>>
>>>Q.703 on "first reception of status indication "B" starts a longer supervision
>>>timer T6 (see 12.3) only if there are message signal units in the
>>>retransmission buffer".  That is, T6 is only running if the remote is in
>>>receive congestion and the local side has unacknowledged messages.  There is
>>>the case where the remote is congested and there are no unacknowledged
>>>messages. (The last acknowledge message triggered the receive congestion
>>>discard situation.)  In this latter case, Q.703, T1.111.3 do not start T6.
>>>
>>>So it is not true that T6 is always running when the peer is in receive
>>>congestion.  In some cases the peer has send BUSY and T6 is not running.
>>>
>>>--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.
>>
>>    
>>
>
>  
>

-- 
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 exim@www1.ietf.org  Fri Jul 18 15:26:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA10513
	for <sigtran-archive@odin.ietf.org>; Fri, 18 Jul 2003 15:26:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dari-0001v2-IK
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 15:26:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6IJQ2hj007370
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 15:26:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19darg-0001ud-K5; Fri, 18 Jul 2003 15:26:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dar5-0001uI-Js
	for sigtran@optimus.ietf.org; Fri, 18 Jul 2003 15:25:23 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA10462
	for <sigtran@ietf.org>; Fri, 18 Jul 2003 15:25:21 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dar1-0002u2-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 15:25:19 -0400
Received: from ottgw.pt.com ([206.191.2.130] helo=pt.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19daqq-0002tZ-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 15:25:08 -0400
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id PAA19814;
	Fri, 18 Jul 2003 15:24:05 -0400
Message-ID: <3F184959.2040103@pt.com>
Date: Fri, 18 Jul 2003 15:24:09 -0400
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: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - RPO
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030712170744.A11077@openss7.org> <3F15C7F2.6070704@pt.com> <20030716191143.F6725@openss7.org> <3F18328D.1080706@pt.com> <20030718120053.C14760@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

Hi Brian,

Some comments and questions below. I'll assume that M2PA continues to 
use separate stream 0 for LS PO and LS PR, and stream 1 for data.

Brian F. G. Bidulock wrote:

>Andrew,
>
>Andrew Booth wrote:                                                                 (Fri, 18 Jul 2003 13:46:53)
>  
>
>>Hi Brian,
>>
>>I agree that it makes sense to support both scenarios, where the MTP may 
>>accept user data after RPO or not.  Do both approaches work with the 
>>transparent interworking function you mentioned earlier?  It seems like 
>>in that case the interworking function could not properly process Data 
>>received after RPO without leaving RPO.
>>    
>>
>
>I think so.  The clearing buffers is the same.  The side in RPO is just
>permitted to acknowledge messages where it was not before.
>
Maybe I'm misunderstanding the role of the interworking function, but I 
assume it's for inline tunneling of MTP2 to M2PA with no MTP3 point code.

The case I'm wondering about is if the M2PA link receives LSPO followed 
by Data (suppose the messages were sent in the other order).
Once LSPO is passed to the interworking function I assume that it will 
send SIPO on the MTP2 link, after that if the IWF receives the Data from 
M2PA there's nothing it can do with it... if it sends to MTP2 the the 
MTP2 will see RPO as ended. On the other hand the message has been 
acknowledged by M2PA, if the link ends up dropping or if one side tries 
a sequenced changeover then the message is lost forever. This can be 
avoided if M2PA does not try to deliver the Data after RPO.

So I guess I'm wondering if the IWF is an example of an upper layer 
where it would be better not to pass data after RPO.

>
>  
>
>>More comments inline.
>>
>>Brian F. G. Bidulock wrote:
>>
>>    
>>
>>>Andrew,
>>>
>>>Yes, we discussed this.  The proposal was to permit both situations.
>>>
>>>If you have a MTP3 that will not accept MSUs after RPO you have another
>>>problem: messages send by the remote before sending LSPO, but which arrived
>>>at the peer after the LSPO because the LPSO was received (and supposedly
>>>expedited) on stream 0.
>>>
>>>      
>>>
>>Agreed.  They have to queued also, and will potentially be lost by a flush.
>>
>>    
>>
>>>To accomodate such implementations of MTP3 (completely valid implementations
>>>      
>>>
>>>from the SS7 standards perspective), it would be a better idea to send LSPO
>>    
>>
>>>on stream 1.  Then LSPR also should be sent on stream 1, otherwise there is
>>>the possibiliy that User Data sent after sending LSPR will arrive before the
>>>LSPR.
>>>
>>>      
>>>
>>I'm solidly on the fence here.  I agree that M2PA gets more complicated 
>>if LSPO/LSPR are not sent on the data stream, I have mixed feelings 
>>about whether the potential performance benefits outweigh the complexity.
>>
>>    
>>
>>>The side experiencing PO should accept the acknowledgement of any message
>>>sent before or during PO.  This permits the TB/RTB to drain and has no
>>>interaction with L3.
>>>
>>>The statement below really says that the side experienceing RPO SHOULD
>>>continue to acknowledge menssages received and accepted by MTP3.  If MTP3
>>>will not accept messages during the RPO condition, then of course they 
>>>cannot be acknowledged.  However, if MTP3 is accepting messages during
>>>RPO (as some equally valid implementations do), then they SHOULD be
>>>acknowledged.
>>>
>>>      
>>>
>>Sounds good.
>>
>>    
>>
>>>This leaves both options open.  The current M2PA-09 text does not leave the
>>>option open to continue delivering to MTP3 during RPO if MTP3 is still ready
>>>to accept messages.
>>>
>>>      
>>>
>>Sounds good to me.  It seems important though to keep the text that 
>>describes how the receive buffers are handled in PO for implementations 
>>that cannot deliver DATA after RPO.
>>    
>>
>
>It is still there except for:
>  
>
>>>>>                                                ... M2PA SHOULD NOT
>>>>>flush User Data messages with FSN greater than the FSN in the Link
>>>>>Status Processor Recovered message. 
>>>>>          
>>>>>
>
>The reason I removed this is that RPO latching to LPO option does exactly
>that: flushes data messags with FSN greater than the FSN in the Link Status
>Processor Recovered message.  What messages are flushed are really the
>business of the flusher.
>
For MTP's that can't send data after RPO, doesn't this case also come up 
when exiting from RPO? e.g. if one side receives LSPO then sometime 
after MTP3 T1 has expired it receives data and LSPR (again, that were 
sent in the wrong order). Then the DATA message is in the RB and should 
probably not be flushed when the MTP3 issues a flush buffers (or clear RTB).

In fact even if LSPR and data are sent on the same stream, there is the 
following scenario (I don't know how easy it is for this to happen, we 
may have to assume a retransmit of the LSPR so the LSPR and data arrive 
together after a changeback)...
Suppose that the end in RPO receives an LS PR which it forwards to MTP3, 
then receives data (which goes in the RB) and only then receives the 
flush/clear RTB from MTP3... In this case we probably want to flush the 
old messages in the RB but not the ones sent after LS PR...

Just wondering if the text on flushing buffers when exiting PO still has 
some use. Maybe an advisory comment would be enough...

Thoughts?
Andrew

>
>--brian
>
>  
>
>>>This also allows the M2PA working with an MTP3 that accepts MSUs during RPO
>>>to not look in the sequence numbers of LSPO and LSPR sent on stream 0 to
>>>determine which received messages preceded or followed the RPO condition and
>>>which were between LSPO and LSPR (becaues they are all delivered to MTP3).
>>>
>>>Make sense?
>>>
>>>      
>>>
>>Yes, thanks.
>>
>>Andrew
>>
>>    
>>
>>>--brian
>>>
>>>
>>>Andrew Booth wrote:                                                                 (Wed, 16 Jul 2003 17:47:30)
>>> 
>>>
>>>      
>>>
>>>>Hi Brian,
>>>>
>>>>If the side in RPO is sending data acknowledgments then it must also be 
>>>>delivering these messages to MTP3.  Is there some guarantee that MTP3 
>>>>will accept messages from a link that it has marked as blocked?  Or am I 
>>>>misunderstanding the intent of your proposal?
>>>>
>>>>Andrew
>>>>
>>>>Brian F. G. Bidulock wrote:
>>>>
>>>>   
>>>>
>>>>        
>>>>
>>>>>M2PA-09 says:
>>>>>
>>>>>While there is a remote processor outage condition:
>>>>>
>>>>> (a) Any User Data messages received from the peer after the Link
>>>>>     Status Processor Outage MUST NOT be acknowledged and MUST be
>>>>>     buffered.
>>>>>
>>>>> (b) M2PA SHOULD continue to acknowledge User Data messages received
>>>>>     and accepted by MTP3 before the remote processor outage.
>>>>>
>>>>>In Q.703 the side experiencing PO cannot send MSUs or FISUs without indicating a
>>>>>recovery from processor outage.  M2PA can send User Data without indicating
>>>>>processor recovery.  There is no reason for the receiver of User Data not to
>>>>>process it even after receiving Link Status Processor Outage.
>>>>>
>>>>>Otherwise, if (a) above is followed, MSUs in the transmission and retransmission
>>>>>buffers at the time that local processor outage occured will be trapped.  If
>>>>>User Data is acknowledged by the peer, they will be processed by the receiver.
>>>>>
>>>>>Change M2PA to say:
>>>>>
>>>>>While there is a remote processour outage condition:
>>>>>
>>>>> (a) M2PA SHOULD continue to acknowledge User Data messages received
>>>>>     and accepted by MTP3 regardless of the remote processor outage.
>>>>>
>>>>>This allows the TB/RTB on the side experiencing processor outage to drain.
>>>>>
>>>>>
>>>>>M2PA-09 says:
>>>>>
>>>>>Note that because the Link Status and User Data messages are sent on
>>>>>separate streams, it is possible for a User Data message sent after
>>>>>the Link Status Processor Recovered message to arrive before
>>>>>it. Therefore, when flushing buffers, M2PA must take care not to flush
>>>>>messages sent after the remote processor recovered. M2PA SHOULD NOT
>>>>>flush User Data messages with FSN greater than the FSN in the Link
>>>>>Status Processor Recovered message. 
>>>>>
>>>>>MTP3 does not flush the RB after RPO or before RPR (just the RTB), unless it
>>>>>uses the remote processor outage to local processor outage latching option, and
>>>>>then it is only because of the LPO condition not the RPO condition.
>>>>>
>>>>>With the change above, M2PA receives and acknowledges User Data before, during
>>>>>and after remote processor outage, and comparison to any sequence numbers in the
>>>>>Link Status Processor Recovered is unnecessary.
>>>>>
>>>>>
>>>>>M2PA-09 says:
>>>>>
>>>>>When the local processor outage condition ends, M2PA SHALL send a Link
>>>>>Status Processor Recovered message to its peer. This message is used
>>>>>to signal in the end of the processor outage condition, instead of an
>>>>>MSU or FISU as in MTP2. The BSN value in this message is used by the
>>>>>peer to resynchronize its sequence numbers, if this is required by
>>>>>MTP2.
>>>>>
>>>>>After a flush buffers operation, the side that experienced PO uses the first
>>>>>received FISU/MSU to resync and discards the FISU/MSU.
>>>>>
>>>>>ANSI T1.111.3/2000 Section 8 says:
>>>>>
>>>>>... If level 3 requests level 2 [at the side experiencing processor outage] to
>>>>>clear its buffers of message signal units, level 2 begins to continuously
>>>>>transmit fill-in signal units and discards all message signal units that it
>>>>>has received from remote level 2 but not yet delivered to local level 3 and
>>>>>all message signal units that it has received from local level 3 but not yet
>>>>>transmitted on the link.  The first fill-in signal unit transmitted
>>>>>acknowledges all message signal units that have been accepted from remote
>>>>>level 2 including any mesage signal units for which acknowledgements were
>>>>>previously withheld because of level 2 congestion (see section 9).  Level 2
>>>>>then begins accepting message signal units from remote Level 2. When the next
>>>>>fill-in or message signal unit is received that passes the acceptance
>>>>>procedure of section 4 and sections 5 or 6,
>>>>>
>>>>> (1) all message signal units that are available for retransmission are
>>>>>     discarded [clear RTB].
>>>>>
>>>>> (2) the forward sequence number and forward indicator bit of the
>>>>>     transmitted fill-in signal units are set equal to the backward sequence
>>>>>     number and backward indicator bit of the received signal unit,
>>>>>
>>>>> (3) normal transmission of message signal units and fill-in signal units is
>>>>>     resumed, and
>>>>>
>>>>> (4) level 3 is notified that the buffer clearing has been completed and
>>>>>     that normal operations has been resumed.
>>>>>
>>>>>So, you see, the side experiencing the processor outage, after clearing buffers,
>>>>>resynchronizes its FSN with the BSN of the first received FISU or MSU.  The side
>>>>>experiencing processor outage could use the first User Data message received
>>>>>after processor recovery to resyncronize.  However, if the peer has nothing to
>>>>>send, processor recovery will be delayed.
>>>>>
>>>>>Note, however, that resynchronization of sequence numbers for the transmit
>>>>>direction is necessary because MTP2 traps messages in the TB/RTB of the side
>>>>>experiencing processor outage because it must continuously transmit SIPO during
>>>>>processor outage.  Permitting the draining of the TB/RTB with the change above
>>>>>removes the need to flush the TB/RTB on the side experiencing processor outage,
>>>>>as well as removing the need to resynchronize sequence numbers.
>>>>>
>>>>>In the other direction, for messages queued to be sent to the side experiencing
>>>>>processour outage, flushing and resynchronization is still necessary (actions
>>>>>(1) to (4) per ANSI above).
>>>>>
>>>>>However, Q.703 permits FISUs/MSUs received by the side experiencing remote
>>>>>processor outage to discard the first and several subsequent FISU/MSUs when
>>>>>recovering.  There is even a test case in Q.781 (8.9) that shows an MSU being
>>>>>discarded (and negative acknowledged) in this fashion.  Because M2PA cannot
>>>>>discard and expect retransmission of a message, and because sychronization is
>>>>>not required when the TB/RTB of the side experiencing processor outage is
>>>>>permitted to drain with the changes above, M2PA cannot permit messages to be
>>>>>discarded in this fashion.
>>>>>
>>>>>M2PA-09 says:
>>>>>
>>>>>If M2PA receives a Flush command from MTP3, M2PA SHALL discard the
>>>>>incoming messages that were queued and unacknowledged during the
>>>>>processor outage condition. Furthermore, M2PA SHALL discard messages
>>>>>in the transmit and retransmit queues as required by MTP2. 
>>>>>
>>>>>To handle flushing and resynchronization properly in both directions, change the
>>>>>above to read as follows:
>>>>>
>>>>>Change M2PA to say:
>>>>>
>>>>>If M2PA receives a Flush command from MTP3, M2PA SHALL discard the
>>>>>incoming messages that were queued and unacknowledged during the
>>>>>processor outage condition.  Furthermore, the M2PA experiencing
>>>>>processor outage SHALL NOT discard message from the transmit and
>>>>>retransmit queues and SHALL NOT perform procedures to resynchronize
>>>>>sequence numbers, but SHALL continue normal transmission of User Data.
>>>>>The M2PA experiencing remote processor outage, SHALL discard messages in
>>>>>the transmit and retransmit queues as required by MTP2, and MAY perform
>>>>>resynchronization of sequence numbers using the BSN from the received
>>>>>Link Status Processor Recovered message.  However, during
>>>>>resychronization, M2PA SHALL NOT discard any received User Data
>>>>>messages.
>>>>>
>>>>>As Q.2140 handles processor outage by putting the link out of service, also add:
>>>>>
>>>>>When M2PA experiences a local processor outage, it MAY put the link out
>>>>>of service by sending a Link Status Out of Service message if allowed by
>>>>>the applicable MTP2 standard (e.g, [Q.2140]).
>>>>>
>>>>>I think that wraps up processor outage problems.
>>>>>
>>>>>--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.
>>>>
>>>>   
>>>>
>>>>        
>>>>
>>> 
>>>
>>>      
>>>
>>-- 
>>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.
>>
>>    
>>
>
>  
>

-- 
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 exim@www1.ietf.org  Fri Jul 18 15:43:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA11212
	for <sigtran-archive@odin.ietf.org>; Fri, 18 Jul 2003 15:43:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19db8A-0002cw-IX
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 15:43:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6IJh2AZ010091
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 15:43:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19db89-0002c4-Fc; Fri, 18 Jul 2003 15:43:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19db7n-0002Z7-KI
	for sigtran@optimus.ietf.org; Fri, 18 Jul 2003 15:42:39 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA11143
	for <sigtran@ietf.org>; Fri, 18 Jul 2003 15:42:37 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19db7m-00032O-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 15:42:38 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[gwzDufbCgb1SXDSoXm9OuvoSTHQmJOjv])
	by ietf-mx with esmtp (Exim 4.12)
	id 19db7W-00031e-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 15:42:22 -0400
Received: from ns.pigworks.openss7.net (IDENT:g1bIDrPk/WXXubsQrH2ZwUwr60AU2FQg@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6IJdvP18497;
	Fri, 18 Jul 2003 13:39:57 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6IJdvs16763;
	Fri, 18 Jul 2003 13:39:57 -0600
Date: Fri, 18 Jul 2003 13:39:57 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Andrew Booth <abooth@pt.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - RPO
Message-ID: <20030718133957.A16698@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Andrew Booth <abooth@pt.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030712170744.A11077@openss7.org> <3F15C7F2.6070704@pt.com> <20030716191143.F6725@openss7.org> <3F18328D.1080706@pt.com> <20030718120053.C14760@openss7.org> <3F184959.2040103@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: <3F184959.2040103@pt.com>; from abooth@pt.com on Fri, Jul 18, 2003 at 03:24:09PM -0400
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:                                                                 (Fri, 18 Jul 2003 15:24:09)
> Hi Brian,
> 
> Some comments and questions below. I'll assume that M2PA continues to 
> use separate stream 0 for LS PO and LS PR, and stream 1 for data.
> 
> Brian F. G. Bidulock wrote:
> 
> >Andrew,
> >
> >Andrew Booth wrote:                                                                 (Fri, 18 Jul 2003 13:46:53)
> >  
> >
> >>Hi Brian,
> >>
> >>I agree that it makes sense to support both scenarios, where the MTP may 
> >>accept user data after RPO or not.  Do both approaches work with the 
> >>transparent interworking function you mentioned earlier?  It seems like 
> >>in that case the interworking function could not properly process Data 
> >>received after RPO without leaving RPO.
> >>    
> >>
> >
> >I think so.  The clearing buffers is the same.  The side in RPO is just
> >permitted to acknowledge messages where it was not before.
> >
> Maybe I'm misunderstanding the role of the interworking function, but I 
> assume it's for inline tunneling of MTP2 to M2PA with no MTP3 point code.
> 
> The case I'm wondering about is if the M2PA link receives LSPO followed 
> by Data (suppose the messages were sent in the other order).
> Once LSPO is passed to the interworking function I assume that it will 
> send SIPO on the MTP2 link, after that if the IWF receives the Data from 
> M2PA there's nothing it can do with it... if it sends to MTP2 the the 
> MTP2 will see RPO as ended. On the other hand the message has been 
> acknowledged by M2PA, if the link ends up dropping or if one side tries 
> a sequenced changeover then the message is lost forever. This can be 
> avoided if M2PA does not try to deliver the Data after RPO.
> 
> So I guess I'm wondering if the IWF is an example of an upper layer 
> where it would be better not to pass data after RPO.

Oh, that IWF.  An IWF would probably be better to translate RPO into OOS.

[snip]
> >
> For MTP's that can't send data after RPO, doesn't this case also come up 
> when exiting from RPO? e.g. if one side receives LSPO then sometime 
> after MTP3 T1 has expired it receives data and LSPR (again, that were 
> sent in the wrong order). Then the DATA message is in the RB and should 
> probably not be flushed when the MTP3 issues a flush buffers (or clear RTB).
> 
> In fact even if LSPR and data are sent on the same stream, there is the 
> following scenario (I don't know how easy it is for this to happen, we 
> may have to assume a retransmit of the LSPR so the LSPR and data arrive 
> together after a changeback)...
> Suppose that the end in RPO receives an LS PR which it forwards to MTP3, 
> then receives data (which goes in the RB) and only then receives the 
> flush/clear RTB from MTP3... In this case we probably want to flush the 
> old messages in the RB but not the ones sent after LS PR...
> 
> Just wondering if the text on flushing buffers when exiting PO still has 
> some use. Maybe an advisory comment would be enough...

Sure.  Just as long as the sender will generate acknowledgements in RPO
I don't have a problem.  Also, T6 does not always run when in RPO.
Did you want to propose some alternate text to the text I proposed below?

--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 exim@www1.ietf.org  Fri Jul 18 17:36:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA14280
	for <sigtran-archive@odin.ietf.org>; Fri, 18 Jul 2003 17:36:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dctX-00062O-8T
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 17:36:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6ILa3SM023206
	for sigtran-archive@odin.ietf.org; Fri, 18 Jul 2003 17:36:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dctV-000621-P7; Fri, 18 Jul 2003 17:36:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19dcsl-00061b-M9
	for sigtran@optimus.ietf.org; Fri, 18 Jul 2003 17:35:15 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA14258
	for <sigtran@ietf.org>; Fri, 18 Jul 2003 17:35:11 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19dcsj-0003lx-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 17:35:13 -0400
Received: from ottgw.pt.com ([206.191.2.130] helo=pt.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19dcsY-0003la-00
	for sigtran@ietf.org; Fri, 18 Jul 2003 17:35:02 -0400
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id RAA22746;
	Fri, 18 Jul 2003 17:33:57 -0400
Message-ID: <3F1867CB.6060108@pt.com>
Date: Fri, 18 Jul 2003 17:34:03 -0400
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: "'sigtran@ietf.org'" <sigtran@ietf.org>, tom.george@alcatel.com
Subject: Re: [Sigtran] M2PA Last Call Comment - RPO
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030712170744.A11077@openss7.org> <3F15C7F2.6070704@pt.com> <20030716191143.F6725@openss7.org> <3F18328D.1080706@pt.com> <20030718120053.C14760@openss7.org> <3F184959.2040103@pt.com> <20030718133957.A16698@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

Hi Brian,


Brian F. G. Bidulock wrote:

>[snip]
>
>Sure.  Just as long as the sender will generate acknowledgements in RPO
>I don't have a problem.  Also, T6 does not always run when in RPO.
>Did you want to propose some alternate text to the text I proposed below?
>
>--brian
>
>  
>
What do you think of the following proposal, does it sum up both our points?
As a miscellaneous note I changed the LS PR to LS PO in the final 
paragraph; the motivation is that I think the FSN in the LSPO and LSPR 
should be the same (the side in LPO shouldn't be sending data) and 
because the Flush could arrive before the LS PR in which case the old 
text is clearly wrong. I'm not sure though if that change was the right 
one. Please let me know what you think.

Thanks,
Andrew

4.1.4 Processor Outage

The Link Status Processor Outage message replaces the SIPO message of
MTP2. Unlike MTP2, the message SHOULD NOT be transmitted
continuously. M2PA SHALL send a Link Status Processor Outage message
to its peer at the beginning of a processor outage condition where
MTP2 would send SIPO. M2PA MAY send additional Link Status Processor
Outage messages as long as that condition persists.

When the local processor outage condition ends, M2PA SHALL send a Link
Status Processor Recovered message to its peer. This message is used
to signal in the end of the processor outage condition, instead of an
MSU or FISU as in MTP2. The BSN value in this message is used by the
peer to resynchronize its sequence numbers, if this is required by
MTP2.

Some MTP3 implementations may be able to accept received User Data
while in Remote Processor Outage while others may not. While there
is a remote processor outage condition:

(a) M2PA SHOULD continue to acknowledge User Data Message received and
accepted by MTP3 regardless of the remote processor outage.

(b) Any User Data messages received from the peer after
the Link Status Processor Outage that may not be passed to MTP3
MUST NOT be acknowledged and MUST be buffered.

If M2PA receives a Flush command from MTP3, M2PA SHALL
discard the incoming messages that were queued and unacknowledged during the
processor outage condition. Furthermore, M2PA SHALL discard messages in the
transmit and retransmit queues as required by MTP2. Similar behaviour
is appropriate in response to corresponding primitives in other SS7 variants
such as the ANSI Clear RTB primitive.

Note that because the Link Status and User Data messages are sent on 
separate
streams, it is possible for a User Data message sent after the Link Status
Processor Recovered message to arrive before it. Therefore, when flushing
buffers, M2PA must take care not to flush messages sent after the remote
processor recovered. M2PA SHOULD NOT flush User Data messages with FSN 
greater
than the FSN in the last received Link Status Processor Outage message.
Note that similar conditions could occur if User Data is received after Link
Status Processor Recovered but before the Flush command from MTP3. Similar
behaviour is appropriate in response to comparable primitives in other 
SS7 variants,
such as the ANSI Clear RTB primitive.



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



From exim@www1.ietf.org  Mon Jul 21 07:10:52 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA24153
	for <sigtran-archive@odin.ietf.org>; Mon, 21 Jul 2003 07:10:52 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19eYYm-0000bW-QE
	for sigtran-archive@odin.ietf.org; Mon, 21 Jul 2003 07:10:28 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6LBASDP002303
	for sigtran-archive@odin.ietf.org; Mon, 21 Jul 2003 07:10:28 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19eYYL-0000aY-A4; Mon, 21 Jul 2003 07:10:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19eYXT-0000a0-4q
	for sigtran@optimus.ietf.org; Mon, 21 Jul 2003 07:09:07 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA24139
	for <sigtran@ietf.org>; Mon, 21 Jul 2003 07:09:00 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19eYXO-0001pL-00
	for sigtran@ietf.org; Mon, 21 Jul 2003 07:09:02 -0400
Received: from david.siemens.de ([192.35.17.14])
	by ietf-mx with esmtp (Exim 4.12)
	id 19eYXE-0001ou-00
	for sigtran@ietf.org; Mon, 21 Jul 2003 07:08:52 -0400
Received: from mail1.siemens.de (mail1.siemens.de [139.23.33.14])
	by david.siemens.de (8.11.7/8.11.7) with ESMTP id h6LB7v108694
	for <sigtran@ietf.org>; Mon, 21 Jul 2003 13:07:57 +0200 (MEST)
Received: from blues.mchh.siemens.de (blues.mchh.siemens.de [139.21.204.206])
	by mail1.siemens.de (8.11.7/8.11.7) with ESMTP id h6LB7uT08755
	for <sigtran@ietf.org>; Mon, 21 Jul 2003 13:07:56 +0200 (MEST)
Received: from mchh253e.mchh.siemens.de (mchh253e.mchh.siemens.de [139.21.200.63])
	by blues.mchh.siemens.de (8.9.3/8.9.1) with ESMTP id NAA14390
	for <sigtran@ietf.org>; Mon, 21 Jul 2003 13:06:45 +0200 (MET DST)
Received: by mchh253e.mchh.siemens.de with Internet Mail Service (5.5.2653.19)
	id <PFC11JM6>; Mon, 21 Jul 2003 13:07:19 +0200
Message-ID: <99522F0FD7EFD611A47D0002A58EDAB7C84A86@mchh2a8e.mchh.siemens.de>
From: Gaiser Franz <franz.gaiser@siemens.com>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject:  [Sigtran] M2PA Last Call Comment - LPO
Date: Mon, 21 Jul 2003 13:07:09 +0200
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 All,

implementing Processor Outage Handling of M2PA draft 09 the following questions came up:

The Processor Outage chapter 4.1.4 says

When the local processor outage conditions ends, M2PA shall send a Link
Status Processor Recovered message to its peer. This message is used
to signal in the end of the processor outage condition, instead of an
MSU or FISU as in MTP2. The BSN value in this message is used by the 
peer to resynchronize its sequence numbers, if this is required by
MTP2.

What does M2PA draft 09 with BSN in this context mean, is it (assume, MTP3 has send Continue Command)

(1) the FSN of the last received message before LPO or 
(2) the FSN of the last message in the incoming messages that were queued and unacknowledged
during the processor outage (with that the remote side can release the messages in its RTB) ? 



thanks
best regards, Franz 





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



From exim@www1.ietf.org  Mon Jul 21 10:04:49 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA27698
	for <sigtran-archive@odin.ietf.org>; Mon, 21 Jul 2003 10:04:49 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ebH5-0005x5-UO
	for sigtran-archive@odin.ietf.org; Mon, 21 Jul 2003 10:04:23 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6LE4Nob022873
	for sigtran-archive@odin.ietf.org; Mon, 21 Jul 2003 10:04:23 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ebGi-0005wa-Vp; Mon, 21 Jul 2003 10:04:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19ebGI-0005wF-6J
	for sigtran@optimus.ietf.org; Mon, 21 Jul 2003 10:03:34 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA27567
	for <sigtran@ietf.org>; Mon, 21 Jul 2003 10:03:29 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19ebGG-00037S-00
	for sigtran@ietf.org; Mon, 21 Jul 2003 10:03:32 -0400
Received: from web41808.mail.yahoo.com ([66.218.93.142])
	by ietf-mx with smtp (Exim 4.12)
	id 19ebG5-000371-00
	for sigtran@ietf.org; Mon, 21 Jul 2003 10:03:21 -0400
Message-ID: <20030721140222.94759.qmail@web41808.mail.yahoo.com>
Received: from [65.198.40.51] by web41808.mail.yahoo.com via HTTP; Mon, 21 Jul 2003 09:02:22 CDT
Date: Mon, 21 Jul 2003 09:02:22 -0500 (CDT)
From: =?iso-8859-1?q?Pablo=20Salinas?= <pablopesb@yahoo.com>
To: sigtran@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="0-2033814148-1058796142=:94374"
Subject: [Sigtran] More M2UA doubts
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

--0-2033814148-1058796142=:94374
Content-Type: text/plain; charset=iso-8859-1
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id KAA27568
Content-Transfer-Encoding: quoted-printable

Hi there,
         I have a question for you, people. Suppose, I have a SG and a MG=
C that have to communicate with each other. ASP #1 is attached to AS #1 a=
nd ASP #2 is attached to AS #2. Suppose both ASP's are
in the active state. I know the sctp association is a half-duplex channel=
, rigth? So, is it rigth for them to use the same IP:Port address to send=
 and receive M2UA messages. Example, for ASP #1 in the SG the 65.198.40.5=
0:3005 (IP:Port) address is used, and in the MGC the 65.198.40.52:3006 ad=
dress is used.
If that is the case, how is the communication synchronized?
=20
|----------- SG -------------------------------| |--------------------- M=
GC ---------------------|
ASP #1 --- 65.198.40.50:3005 <-----------> 65.198.40.52:3006 ----- ASP #1
ASP #2 --- 65.198.40.51:3005 <-----------> 65.198.40.53:3006 ----- ASP #2

Another question, can a SG support both, dynamic and static registration,=
 of ASP's. I mean, to have initially pre-configured ASP's and to support =
dynamic registration at the same time.

                  thanks,
                          Pablo S.



---------------------------------
Do You Yahoo!?
Todo lo que quieres saber de Estados Unidos, Am=E9rica Latina y el resto =
del Mundo.
Vis=EDta Yahoo! Noticias.

--0-2033814148-1058796142=:94374
Content-Type: text/html; charset=iso-8859-1
X-MIME-Autoconverted: from 8bit to quoted-printable by ietf.org id KAA27568
Content-Transfer-Encoding: quoted-printable

<DIV>Hi there,</DIV>
<DIV>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; I have a question f=
or you, people. Suppose, I have a SG and a MGC that have to communicate w=
ith each other. ASP #1 is attached to AS #1 and ASP #2 is attached to AS =
#2. Suppose both ASP's are<BR>in the active state. I know the sctp associ=
ation is a half-duplex channel, rigth? So, is it rigth for them to use th=
e same IP:Port address to send and receive M2UA messages. Example, for AS=
P #1 in the SG the 65.198.40.50:3005 (IP:Port) address is used, and in th=
e MGC the 65.198.40.52:3006 address is used.</DIV>
<DIV>If that is the case, how is the communication synchronized?</DIV>
<DIV>&nbsp;</DIV>
<DIV>|----------- SG -------------------------------| |------------------=
--- MGC ---------------------|</DIV>
<DIV>ASP #1 --- 65.198.40.50:3005 &lt;-----------&gt; 65.198.40.52:3006 -=
---- ASP #1</DIV>
<DIV>ASP #2 --- 65.198.40.51:3005 &lt;-----------&gt; 65.198.40.53:3006 -=
---- ASP #2</DIV>
<DIV><BR>Another question, can a SG support both, dynamic and static regi=
stration, of ASP's. I mean, to have initially pre-configured ASP's and to=
 support dynamic registration at the same time.</DIV>
<DIV><BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; thanks,<BR>&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; Pablo S.</DIV><p><=
br><hr size=3D1 noshade><b>Do You Yahoo!?</b><br>
<a href=3Dhttp://espanol.yahoo.com/mail_tagline/*http://espanol.news.yaho=
o.com><img src=3D"http://us.i1.yimg.com/us.yimg.com/i/not/dailynews.gif" =
width=3D30 height=3D26 border=3D0 align=3Dleft></a>
Todo lo que quieres saber de Estados Unidos, Am=E9rica Latina y el resto =
del Mundo.<br>
Vis=EDta <a href=3Dhttp://espanol.yahoo.com/mail_tagline/*http://espanol.=
news.yahoo.com>Yahoo! Noticias</a>.<br>
--0-2033814148-1058796142=:94374--

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



From exim@www1.ietf.org  Tue Jul 22 06:07:49 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA07843
	for <sigtran-archive@odin.ietf.org>; Tue, 22 Jul 2003 06:07:49 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19eu3I-0007ix-AU
	for sigtran-archive@odin.ietf.org; Tue, 22 Jul 2003 06:07:24 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6MA7Oee029690
	for sigtran-archive@odin.ietf.org; Tue, 22 Jul 2003 06:07:24 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19eu2v-0007bb-Kb; Tue, 22 Jul 2003 06:07:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19eu2l-0007b8-Vt
	for sigtran@optimus.ietf.org; Tue, 22 Jul 2003 06:06:52 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA07811
	for <sigtran@ietf.org>; Tue, 22 Jul 2003 06:06:46 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19eu2i-0003AJ-00
	for sigtran@ietf.org; Tue, 22 Jul 2003 06:06:48 -0400
Received: from albatross-ext.wise.edt.ericsson.se ([193.180.251.49] helo=albatross.tn.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19eu2X-0003A2-00
	for sigtran@ietf.org; Tue, 22 Jul 2003 06:06:37 -0400
Received: from esealnt611.al.sw.ericsson.se ([153.88.254.121])
	by albatross.tn.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h6MA5WG6003799;
	Tue, 22 Jul 2003 12:05:32 +0200 (MEST)
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 3S6KSS77; Tue, 22 Jul 2003 12:08:41 +0200
Received: by ESEALNT744.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <PLTBLMFZ>; Tue, 22 Jul 2003 12:05:33 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C043316DE@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
To: "'kakaul@hss.hns.com'" <kakaul@hss.hns.com>
Cc: sigtran@ietf.org
Subject: RE: [Sigtran] M3UA IG v04 Section 3.18
Date: Tue, 22 Jul 2003 12:04:59 +0200
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 Kamesh, 

Sorry for not answering before, but I was attending the IETF meeting and the access tool is not very good to track all the issues...

It seems to me that the timer referred in the Solution Description section has been misunderstood.

- The referred timer is said to be located at the ASP peer node, i.e. the SGP. And the reason for the timer is to solve the problem outlined in the Problem Description section: ASP indication of own congestion abatement. Nothing to do with the reception of SCON from the SGP.

- The timer is mentioned in the Solution Description section. Then if the timer is really seen as an important implementation clue we should move it to the Changes section, maybe as an IMPLEMENTATION NOTE since interop is not involved. Take into account that the Solution Description MAY disappear from the final RFC, if we get an RFC3332bis instead of a IG separate RFC (we'll have this discussion when the doc is ready for approval).

At this point I'd choose either:
 1- Remove the timer from the Solution Section.
 2- Include an IMPLEMENTOR'S NOTE in the changes with reference to the timer.

If you feel that mentioning the timer may help to some implementations we should go for option 2.

Cheers // Javier.



> -----Original Message-----
> From: kakaul@hss.hns.com [mailto:kakaul@hss.hns.com]
> Sent: martes, 15 de julio de 2003 6:05
> To: J.Javier Pastor (EE/EEM)
> Cc: sigtran@ietf.org
> Subject: RE: [Sigtran] M3UA IG v04 Section 3.18
> 
> 
> 
> 
> 
> 
> Hi Javier,
>                     This timer may be used by some implementations to
> supervise the congestion status of point code if SCON 
> messages send by SGP
> are lost in transit to ASP.It was discussed earlier and was 
> agreed upon by
> all to keep this timer.I disagree with removing the mention 
> of the timer.
> 
> Regards
> Kamesh Kaul
> www.hssworld.com
> Hughes Software Systems
> 
> 
> 
> 
> "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>@ietf.org on
> 07/14/2003 04:05:58 PM
> 
> Sent by:    sigtran-admin@ietf.org
> 
> 
> To:    "'Brian F. G. Bidulock '" <bidulock@openss7.org>
> cc:    "''sigtran@ietf.org' '" <sigtran@ietf.org>
> 
> Subject:    RE: [Sigtran] M3UA IG v04 Section 3.18
> 
> 
> Brian,
> 
> We'll remove the mention of the timer, if nobody disagrees...
> 
> Cheers // Javier.
> 
> -----Original Message-----
> From: Brian F. G. Bidulock
> To: J.Javier Pastor (EE/EEM)
> Cc: 'sigtran@ietf.org'
> Sent: 11/07/2003 11:07
> Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> 
> J.Javier,
> 
> J.Javier Pastor (EE/EEM) wrote:
> (Fri, 11 Jul 2003 09:56:59)
> > Hi Brian,
> >
> > Pls, see below...
> >
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: jueves, 10 de julio de 2003 15:34
> > > To: J.Javier Pastor (EE/EEM)
> > > Cc: 'sigtran@ietf.org'
> > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > >
> > >
> > > Javier,
> > >
> > > 3.18.3 Solution Description
> > >
> > >    Clarify that the ASP needs to indicate when the congestion
> > > level has
> > >    changed (including abatement).  Further, the ASP peer can
> maintain
> > >    a timer, if desired, in case the ASP fails to indicate 
> congestion
> > >    abatement.
> > >
> > > Where are the procedures for this timer described?  This 
> is the only
> place
> > > that the timer is mentioned.  Please describe the timer or remove
> the
> > > sentence saying that it has been described...
> > >
> >
> > Do you think is worth include the timer explanation into 
> the new text?
> > Otherwise, we may remove that sentence.
> 
> Please remove mention of the timer.
> 
> --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 exim@www1.ietf.org  Tue Jul 22 06:18:50 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA08078
	for <sigtran-archive@odin.ietf.org>; Tue, 22 Jul 2003 06:18:50 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19euDw-0008C0-Ga
	for sigtran-archive@odin.ietf.org; Tue, 22 Jul 2003 06:18:25 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6MAIOSJ031479
	for sigtran-archive@odin.ietf.org; Tue, 22 Jul 2003 06:18:24 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19euDZ-00087I-RD; Tue, 22 Jul 2003 06:18:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19euDN-000872-Av
	for sigtran@optimus.ietf.org; Tue, 22 Jul 2003 06:17:49 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA08043
	for <sigtran@ietf.org>; Tue, 22 Jul 2003 06:17:44 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19euDJ-0003Fn-00
	for sigtran@ietf.org; Tue, 22 Jul 2003 06:17:45 -0400
Received: from albatross-ext.wise.edt.ericsson.se ([193.180.251.49] helo=albatross.tn.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19euD8-0003Fj-00
	for sigtran@ietf.org; Tue, 22 Jul 2003 06:17:34 -0400
Received: from esealnt612.al.sw.ericsson.se ([153.88.254.118])
	by albatross.tn.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h6MAHUG6003200;
	Tue, 22 Jul 2003 12:17:30 +0200 (MEST)
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 NW4YBP4W; Tue, 22 Jul 2003 12:17:30 +0200
Received: by ESEALNT744.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <PLTBLMMF>; Tue, 22 Jul 2003 12:17:31 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C043316DF@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
To: "'Samuel Dur D. Jeyaseelan'" <sjeyasee@ssd.usa.alcatel.com>,
        "''sigtran@ietf.org' '" <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA IG v04  Section 3.14
Date: Tue, 22 Jul 2003 12:16:56 +0200
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>

Samuel, 

In this case, I'm afraid you are not ready to handle the traffic with other ASPs since all of them are in ASP_DOWN. If you want to have a fail-safe configuration, you need to have at least one ASP in ASP_INACTIVE state. Then you could NTFY...

Cheers // Javier.

> -----Original Message-----
> From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> Sent: martes, 15 de julio de 2003 6:06
> To: ''sigtran@ietf.org' '
> Subject: Re: [Sigtran] M3UA IG v04 Section 3.14
> 
> 
> 
> Brian,
> 
>  In n=5(example) case, if 4 ASPes are in DOWN state and 1 ASP 
> is in ACTIVE
>  state and which handles the traffic.Suppose ACTIVE ASP moves 
> to INACTIVE 
>  or DOWN state so SGP tries sending NTFY(AS-DOWN) to other ASPes.
> 
>  How does an ASP process NTFY message in DOWN state?
> 
> 
> --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 Mon, 14 Jul 2003, Brian F. G. Bidulock wrote:
> 
> > Samuel,
> > 
> > To what end?
> > 
> > --brian
> > 
> > Samuel Dur D. Jeyaseelan wrote:                             
>                         (Mon, 14 Jul 2003 09:06:36)
> > > 
> > > Javier,Brian,
> > > 
> > > 
> > >  (i was bit late to enter into this conversation)
> > > 
> > >  Including a new status type 'AS-DOWN' in NTFY messages 
> is a good idea,
> > >  but section 4.3.4.5 in RFC says an ASP in DOWN state 
> will not process
> > >  NTFY message....?
> > > 
> > > 
> > >  Let me take an example of n=5, SGP needs 5 ASPes in ACTIVE state
> > >  to declare the AS as ACTIVE(true-only for start case).
> > > 
> > >  Start-case
> > >  ----------
> > > 
> > >  ACTIVE ASPes    INACTIVE ASPes  DOWN ASPes    AS state   
>      case-#
> > >  
> --------------------------------------------------------------------
> > >     0             0               5            DOWN       
>      case-1
> > >     0             1               4            DOWN       
>      case-2
> > >     0             2               3            DOWN       
>      case-3
> > >     .
> > >     .
> > >     .
> > >     0             5               0            INACTIVE   
>      case-4
> > >     1             4               0            INACTIVE   
>      case-5
> > >     .
> > >     .
> > >     .
> > >     5             0               0            ACTIVE     
>      case-6
> > > 
> > >    (assume traffic flows)
> > > 
> > >     Non-start case
> > >     --------------
> > >     4             1               0            ACTIVE     
>      case-7
> > >     3             2               0            ACTIVE     
>      case-8
> > >     .
> > >     .
> > >     .
> > >     1             4               0            ACTIVE     
>      case-8
> > >     0             5               0            INACTIVE   
>      case-9
> > >     0             4               1            still 
> INACTIVE  case-10
> > >     0             3               2            still 
> INACTIVE  case-11
> > >     .
> > >     .
> > >     .
> > >     0             1               4            still 
> INACTIVE  case-12
> > >     0             0               5            DOWN       
>      case-13
> > > 
> > > 
> > >  I think we should define the following state-conversion 
> definition.
> > > 
> > >  for Start-case
> > >  --------------
> > > 
> > >  DN2IA : One ASP moves to ASP-INACTIVE causing that n ASPs are in
> > >          ASP-INACTIVE states.
> > > 
> > >  IA2DN : One ASP moves to ASP-DOWN causing that n ASPs are in
> > >          ASP-DOWN state.
> > > 
> > >  IA2AC : one ASP moves to ASP-ACTIVE, causing number of 
> ASPs are in 
> > >          ASP-ACTIVE state.
> > > 
> > > 
> > > Waiting for comments to go ahead.
> > > 
> > > 
> > > --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 Mon, 14 Jul 2003, J.Javier Pastor (EE/EEM) wrote:
> > > 
> > > > Brian, 
> > > > 
> > > > What you say it's right, only 1 'n' transition would be 
> kept in the proposal: the IA2AC transition.
> > > > 
> > > > I think it makes sense since "n" refers only to the 
> capacity of processing traffic and in the AS-INACTIVE state 
> no traffic is processed.
> > > > 
> > > > Does anyone have any objection to this change?
> > > > 
> > > > 
> > > > Thanks // Javier.
> > > > 
> > > > -----Original Message-----
> > > > From: Brian F. G. Bidulock
> > > > To: J.Javier Pastor (EE/EEM)
> > > > Cc: 'sigtran@ietf.org'
> > > > Sent: 11/07/2003 11:13
> > > > Subject: Re: [Sigtran] M3UA IG v04  Section 3.14
> > > > 
> > > > J.Javier,
> > > > 
> > > > J.Javier Pastor (EE/EEM) wrote:
> > > > (Fri, 11 Jul 2003 10:35:57)
> > > > > Hi Brian, 
> > > > > 
> > > > > You are right saying that there is currently no way 
> of notifying
> > > > AS-DOWN. In
> > > > > fact "n" only applies to the traffic case.
> > > > > 
> > > > > Then I see two options:
> > > > > 	1- include AS-DOWN in NTFY
> > > > > 	2- The AS-DOWN state is only got when all the 
> ASPs are in
> > > > ASP-DOWN state.
> > > > > 
> > > > > I would addopt 2 since it is inline what we are 
> having in the other
> > > > UALs. 
> > > > 
> > > > Yes, please go with 2 as it is also in line with the RFC.
> > > > 
> > > > That will also leave only 1 'n' transition in the 
> proposal: the IA2AC
> > > > transition, correct?
> > > > 
> > > > --brian
> > > > 
> > > > > 
> > > > > Thanks // Javier.
> > > > > 
> > > > > 
> > > > > > -----Original Message-----
> > > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > > Sent: jueves, 10 de julio de 2003 17:15
> > > > > > To: J.Javier Pastor (EE/EEM)
> > > > > > Cc: 'sigtran@ietf.org'
> > > > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.14
> > > > > > 
> > > > > > 
> > > > > > Javier,
> > > > > > 
> > > > > > Section 3.14 says:
> > > > > > 
> > > > > >    IA2DN: One ASP moves to ASP-DOWN causing that 
> the number of ASPs
> > > > in
> > > > > >    either ASP-ACTIVE or ASP-INACTIVE drops below n.
> > > > > > 
> > > > > > Am I correct to say that if n=5 and there are 4 ASPs in the 
> > > > > > ASP-INACTIVE
> > > > > > and 1 ASP in the ASP-ACTIVE state, and then the 1 
> active ASP moves
> > > > to
> > > > > > the ASP-DOWN state and T(r) expires that the AS is 
> to move to 
> > > > > > the AS-DOWN
> > > > > > state?  As follows:
> > > > > > 
> > > > > >   If T(r) expires before an ASP becomes ASP-ACTIVE, 
> the SGP MAY stop
> > > > > >   queuing messages and discards all
> > > > > >   previously queued messages. The AS will move to the 
> > > > > > AS-INACTIVE state
> > > > > >   if at least the number of ASPs in ASP-INACTIVE 
> sum n, otherwise it
> > > > > >   will move to AS-DOWN state.
> > > > > > 
> > > > > > Now, that is an AS state change and we must send a NTFY to 
> > > > > > all the INACTIVE
> > > > > > ASPs (the 4 left inactive) of the state change.  But from 
> > > > > > Section 3.8.2 of
> > > > > > the RFC, there is no message to notify of an AS-DOWN state:
> > > > > > 
> > > > > >          1    reserved
> > > > > >          2    Application Server Inactive (AS-INACTIVE)
> > > > > >          3    Application Server Active (AS-ACTIVE)
> > > > > >          4    Application Server Pending (AS-PENDING)
> > > > > > 
> > > > > > You see, we removed it because there was never an 
> instance in which
> > > > it
> > > > > > would be sent.  You have inserted an instance again.
> > > > > > 
> > > > > > Aren't you going to have to define a Notify message (the 
> > > > > > Reserved value)
> > > > > > for "Application Server Down (AS-DOWN)" so that the 
> INACTIVE ASPs
> > > > can
> > > > > > be properly notified by the SGP?
> > > > > > 
> > > > > > Particularly when the RFC says in Section 4.3.4.5: (Notify
> > > > Procedures)
> > > > > > 
> > > > > >    A Notify message reflecting a change in the AS 
> state MUST 
> > > > > > be sent to
> > > > > >    all ASPs in the AS, except those in the ASP-DOWN 
> state, with
> > > > > >    appropriate Status Information and any ASP 
> Identifier of the
> > > > failed
> > > > > >    ASP.  At the ASP, Layer Management is informed 
> with an M-NOTIFY
> > > > > >    indication primitive.  The Notify message must be sent 
> > > > > > whether the AS
> > > > > >    state change was a result of an ASP failure or 
> reception of an
> > > > ASP
> > > > > >    State management (ASPSM) / ASP Traffic Management (ASPTM)
> > > > message.
> > > > > >    In the second case, the Notify message MUST be 
> sent after 
> > > > > > any related
> > > > > >    acknowledgement messages  (e.g., ASP Up Ack, ASP 
> Down Ack, 
> > > > > > ASP Active
> > > > > >    Ack, or ASP Inactive Ack).
> > > > > > 
> > > > > > The above tells me that I MUST send a Notify 
> message becase of the
> > > > AS
> > > > > > state change, but there is no message to inform 
> about AS-DOWN state
> > > > > > transition.
> > > > > > 
> > > > > > What is the purpose of doing this?  What is the protocol 
> > > > > > difference between
> > > > > > the AS-DOWN state and the AS-INACTIVE state?  What 
> is it that the
> > > > SGP
> > > > > > does or doesn't do in the one state that it doesn't or does 
> > > > > > do in the other?
> > > > > > 
> > > > > > I don't recall this being discussed at all.
> > > > > > 
> > > > > > Please clarify.
> > > > > > 
> > > > > > -- 
> > > > > > 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
> 

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



From exim@www1.ietf.org  Tue Jul 22 06:47:13 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA08898
	for <sigtran-archive@odin.ietf.org>; Tue, 22 Jul 2003 06:47:13 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19euf0-0000j3-Rw
	for sigtran-archive@odin.ietf.org; Tue, 22 Jul 2003 06:46:48 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6MAkMQv002777
	for sigtran-archive@odin.ietf.org; Tue, 22 Jul 2003 06:46:22 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19euee-0000eX-UV; Tue, 22 Jul 2003 06:46:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19eucS-0000bv-C5
	for sigtran@optimus.ietf.org; Tue, 22 Jul 2003 06:45:15 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA08828
	for <sigtran@ietf.org>; Tue, 22 Jul 2003 06:43:38 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19eucO-0003Rk-00
	for sigtran@ietf.org; Tue, 22 Jul 2003 06:43:40 -0400
Received: from penguin-ext.wise.edt.ericsson.se ([193.180.251.47] helo=penguin.al.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19eucD-0003Ra-00
	for sigtran@ietf.org; Tue, 22 Jul 2003 06:43:29 -0400
Received: from esealnt612.al.sw.ericsson.se ([153.88.254.118])
	by penguin.al.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h6MAgejG000724;
	Tue, 22 Jul 2003 12:42:41 +0200 (MEST)
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 NW4YB42A; Tue, 22 Jul 2003 12:42:40 +0200
Received: by ESEALNT744.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <PLTBLMZB>; Tue, 22 Jul 2003 12:42:41 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C043316E1@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA IG v04 AS State Machine
Date: Tue, 22 Jul 2003 12:42:10 +0200
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,

Pls, see below...

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: viernes, 11 de julio de 2003 10:38
> To: J.Javier Pastor (EE/EEM)
> Cc: 'sigtran@ietf.org'
> Subject: Re: [Sigtran] M3UA IG v04 AS State Machine
> 
> 
> J.Javier,
> 
> That doesn't really answer my questions below.  Could you please
> clarify the proposal by answering the questions?
> 
> --brian
> 
> J.Javier Pastor (EE/EEM) wrote:                               
>                     (Fri, 11 Jul 2003 09:21:14)
> > Hi Brian, 
> > 
> > In the consensus text we have not add or remove anything 
> regarding the AS
> > state scope. Then it remains the same.
> > 
> > Cheers // Javier.
> > 
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: jueves, 10 de julio de 2003 15:16
> > > To: 'sigtran@ietf.org'
> > > Subject: Re: [Sigtran] M3UA IG v04 AS State Machine
> > > 
> > > 
> > > Javier,
> > > 
> > > The AS State Machine illustrated in 3.14 is per-SGP.  Correct?

Yes
> > > 
> > > SGP can communicate with each other in an SG.  Correct?

It's an implementation option. I think there is nothing in the RFC that mandates or forbids this...
> > > 
> > > Why should an SGP pend when the last ASP goes ASP-INACTIVE for
> > > an AS with a given SGP?

Because hopefully another ASP may take over from it. And you don't need to stop all the traffic.
> > > 
> > > Cannot the SGP redirect traffic via another SGP that has ASPs
> > > active for the AS?

Implementation option. I think there is nothing in the RFC that mandates or forbids this option...
> > > 
> > > Example:
> > > 
> > >         _______                    _______
> > >        |       |______x    x______|       |
> > >        | SGP1  |         _________| ASP1  |
> > >        |_______|        /         |_______|
> > >         _______        /           _______
> > >        |       |______/           |       |
> > >        | SGP2  |__________________| ASP2  |
> > >        |_______|                  |_______|
> > > 
> > > 
> > > When ASP1 goes inactive (or down) w.r.t SGP1, cannot
> > > SGP1 direct traffic via SGP2?

M3UA draft defines SGP-ASP communication. Although you may use a proprietary communication between SGPs belonging to the same SG to distribute traffic. It basically depends on your implementation choice.
> > > 
> > > What is the state of the AS at SGP1?
> > > 
> > > In section 3.14 AC2PN says that:
> > > 
> > >   AC2PN: the last ASP in the ASP-ACTIVE state moves to
> > >   ASP-INACTIVE or ASP-DOWN states, causing the number of ASPs
> > >   in the ASP-ACTIVE [sic] drop below 1.
> > > 
> > > Is that the number of ASPs for SGP1?  for SGP2? for the SG?
> > > for all (both) SGs?

It depends on which SGP receives the ASPIA or ASPDN. Remember that the AS state is per SGP. Since nothing has been change on that regard.
> > > 
> > > In the RFC the AS state machine was per-SGP.  Is the 3.14 state
> > > machine per-SGP?

Yes.
> > > 
> > >    IA2AC: one ASP moves to [the] ASP-ACTIVE [state], causing
> > >    [the] number of ASPs in the ASP-ACTIVE state to be n.
> > > 
> > > When SGP1 receives an ASP Active and sends an ASP Active Ack
> > > to ASP1, how many ASPs are active for the AS?   1?  2?

This question is already answered saying that there is no change in the AS state scope: per SGP. If SGP1 didn't receive any other request from ASP2 then you only have one ASP active in SGP1 for that AS.

Cheers // Javier.

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



From exim@www1.ietf.org  Tue Jul 22 06:48:50 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA09006
	for <sigtran-archive@odin.ietf.org>; Tue, 22 Jul 2003 06:48:50 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19eugz-0000sI-O3
	for sigtran-archive@odin.ietf.org; Tue, 22 Jul 2003 06:48:25 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6MAmPop003350
	for sigtran-archive@odin.ietf.org; Tue, 22 Jul 2003 06:48:25 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19euga-0000lv-O3; Tue, 22 Jul 2003 06:48:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19eugP-0000lb-1b
	for sigtran@optimus.ietf.org; Tue, 22 Jul 2003 06:47:49 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA08909
	for <sigtran@ietf.org>; Tue, 22 Jul 2003 06:47:43 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19eugK-0003TT-00
	for sigtran@ietf.org; Tue, 22 Jul 2003 06:47:44 -0400
Received: from auds952.usa.alcatel.com ([143.209.238.7])
	by ietf-mx with esmtp (Exim 4.12)
	id 19eugA-0003So-00
	for sigtran@ietf.org; Tue, 22 Jul 2003 06:47:34 -0400
Received: from ssd.usa.alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6MAjuB3014701;
	Tue, 22 Jul 2003 05:45:56 -0500 (CDT)
Received: from sun2685.ssd.usa.alcatel.com (sun2685.ssd.usa.alcatel.com [143.209.151.167])
	by ssd.usa.alcatel.com (8.12.8/8.12.8) with ESMTP id h6MAjG1C024948;
	Tue, 22 Jul 2003 05:45:16 -0500 (CDT)
Date: Tue, 22 Jul 2003 05:45:16 -0500 (CDT)
From: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
cc: "''sigtran@ietf.org' '" <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA IG v04  Section 3.14
In-Reply-To: <1AB3D30B989BF141BBD5C70057B2EF7C043316DF@eestqnt105.es.eu.ericsson.se>
Message-ID: <Pine.SOL.4.10.10307220530000.15922-100000@sun2685.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>


Pastor,
 I think,it has been decided to add AS-DOWN in NTFY message.
 in n+k reduntancy case, When do we move the AS to DOWN state?
 (normally an AS is in DOWN state when all its ASPes are in DOWN state)

--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, 22 Jul 2003, J.Javier Pastor (EE/EEM) wrote:

> Samuel, 
> 
> In this case, I'm afraid you are not ready to handle the traffic with other ASPs since all of them are in ASP_DOWN. If you want to have a fail-safe configuration, you need to have at least one ASP in ASP_INACTIVE state. Then you could NTFY...
> 
> Cheers // Javier.
> 
> > -----Original Message-----
> > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > Sent: martes, 15 de julio de 2003 6:06
> > To: ''sigtran@ietf.org' '
> > Subject: Re: [Sigtran] M3UA IG v04 Section 3.14
> > 
> > 
> > 
> > Brian,
> > 
> >  In n=5(example) case, if 4 ASPes are in DOWN state and 1 ASP 
> > is in ACTIVE
> >  state and which handles the traffic.Suppose ACTIVE ASP moves 
> > to INACTIVE 
> >  or DOWN state so SGP tries sending NTFY(AS-DOWN) to other ASPes.
> > 
> >  How does an ASP process NTFY message in DOWN state?
> > 
> > 
> > --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 Mon, 14 Jul 2003, Brian F. G. Bidulock wrote:
> > 
> > > Samuel,
> > > 
> > > To what end?
> > > 
> > > --brian
> > > 
> > > Samuel Dur D. Jeyaseelan wrote:                             
> >                         (Mon, 14 Jul 2003 09:06:36)
> > > > 
> > > > Javier,Brian,
> > > > 
> > > > 
> > > >  (i was bit late to enter into this conversation)
> > > > 
> > > >  Including a new status type 'AS-DOWN' in NTFY messages 
> > is a good idea,
> > > >  but section 4.3.4.5 in RFC says an ASP in DOWN state 
> > will not process
> > > >  NTFY message....?
> > > > 
> > > > 
> > > >  Let me take an example of n=5, SGP needs 5 ASPes in ACTIVE state
> > > >  to declare the AS as ACTIVE(true-only for start case).
> > > > 
> > > >  Start-case
> > > >  ----------
> > > > 
> > > >  ACTIVE ASPes    INACTIVE ASPes  DOWN ASPes    AS state   
> >      case-#
> > > >  
> > --------------------------------------------------------------------
> > > >     0             0               5            DOWN       
> >      case-1
> > > >     0             1               4            DOWN       
> >      case-2
> > > >     0             2               3            DOWN       
> >      case-3
> > > >     .
> > > >     .
> > > >     .
> > > >     0             5               0            INACTIVE   
> >      case-4
> > > >     1             4               0            INACTIVE   
> >      case-5
> > > >     .
> > > >     .
> > > >     .
> > > >     5             0               0            ACTIVE     
> >      case-6
> > > > 
> > > >    (assume traffic flows)
> > > > 
> > > >     Non-start case
> > > >     --------------
> > > >     4             1               0            ACTIVE     
> >      case-7
> > > >     3             2               0            ACTIVE     
> >      case-8
> > > >     .
> > > >     .
> > > >     .
> > > >     1             4               0            ACTIVE     
> >      case-8
> > > >     0             5               0            INACTIVE   
> >      case-9
> > > >     0             4               1            still 
> > INACTIVE  case-10
> > > >     0             3               2            still 
> > INACTIVE  case-11
> > > >     .
> > > >     .
> > > >     .
> > > >     0             1               4            still 
> > INACTIVE  case-12
> > > >     0             0               5            DOWN       
> >      case-13
> > > > 
> > > > 
> > > >  I think we should define the following state-conversion 
> > definition.
> > > > 
> > > >  for Start-case
> > > >  --------------
> > > > 
> > > >  DN2IA : One ASP moves to ASP-INACTIVE causing that n ASPs are in
> > > >          ASP-INACTIVE states.
> > > > 
> > > >  IA2DN : One ASP moves to ASP-DOWN causing that n ASPs are in
> > > >          ASP-DOWN state.
> > > > 
> > > >  IA2AC : one ASP moves to ASP-ACTIVE, causing number of 
> > ASPs are in 
> > > >          ASP-ACTIVE state.
> > > > 
> > > > 
> > > > Waiting for comments to go ahead.
> > > > 
> > > > 
> > > > --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 Mon, 14 Jul 2003, J.Javier Pastor (EE/EEM) wrote:
> > > > 
> > > > > Brian, 
> > > > > 
> > > > > What you say it's right, only 1 'n' transition would be 
> > kept in the proposal: the IA2AC transition.
> > > > > 
> > > > > I think it makes sense since "n" refers only to the 
> > capacity of processing traffic and in the AS-INACTIVE state 
> > no traffic is processed.
> > > > > 
> > > > > Does anyone have any objection to this change?
> > > > > 
> > > > > 
> > > > > Thanks // Javier.
> > > > > 
> > > > > -----Original Message-----
> > > > > From: Brian F. G. Bidulock
> > > > > To: J.Javier Pastor (EE/EEM)
> > > > > Cc: 'sigtran@ietf.org'
> > > > > Sent: 11/07/2003 11:13
> > > > > Subject: Re: [Sigtran] M3UA IG v04  Section 3.14
> > > > > 
> > > > > J.Javier,
> > > > > 
> > > > > J.Javier Pastor (EE/EEM) wrote:
> > > > > (Fri, 11 Jul 2003 10:35:57)
> > > > > > Hi Brian, 
> > > > > > 
> > > > > > You are right saying that there is currently no way 
> > of notifying
> > > > > AS-DOWN. In
> > > > > > fact "n" only applies to the traffic case.
> > > > > > 
> > > > > > Then I see two options:
> > > > > > 	1- include AS-DOWN in NTFY
> > > > > > 	2- The AS-DOWN state is only got when all the 
> > ASPs are in
> > > > > ASP-DOWN state.
> > > > > > 
> > > > > > I would addopt 2 since it is inline what we are 
> > having in the other
> > > > > UALs. 
> > > > > 
> > > > > Yes, please go with 2 as it is also in line with the RFC.
> > > > > 
> > > > > That will also leave only 1 'n' transition in the 
> > proposal: the IA2AC
> > > > > transition, correct?
> > > > > 
> > > > > --brian
> > > > > 
> > > > > > 
> > > > > > Thanks // Javier.
> > > > > > 
> > > > > > 
> > > > > > > -----Original Message-----
> > > > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > > > Sent: jueves, 10 de julio de 2003 17:15
> > > > > > > To: J.Javier Pastor (EE/EEM)
> > > > > > > Cc: 'sigtran@ietf.org'
> > > > > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.14
> > > > > > > 
> > > > > > > 
> > > > > > > Javier,
> > > > > > > 
> > > > > > > Section 3.14 says:
> > > > > > > 
> > > > > > >    IA2DN: One ASP moves to ASP-DOWN causing that 
> > the number of ASPs
> > > > > in
> > > > > > >    either ASP-ACTIVE or ASP-INACTIVE drops below n.
> > > > > > > 
> > > > > > > Am I correct to say that if n=5 and there are 4 ASPs in the 
> > > > > > > ASP-INACTIVE
> > > > > > > and 1 ASP in the ASP-ACTIVE state, and then the 1 
> > active ASP moves
> > > > > to
> > > > > > > the ASP-DOWN state and T(r) expires that the AS is 
> > to move to 
> > > > > > > the AS-DOWN
> > > > > > > state?  As follows:
> > > > > > > 
> > > > > > >   If T(r) expires before an ASP becomes ASP-ACTIVE, 
> > the SGP MAY stop
> > > > > > >   queuing messages and discards all
> > > > > > >   previously queued messages. The AS will move to the 
> > > > > > > AS-INACTIVE state
> > > > > > >   if at least the number of ASPs in ASP-INACTIVE 
> > sum n, otherwise it
> > > > > > >   will move to AS-DOWN state.
> > > > > > > 
> > > > > > > Now, that is an AS state change and we must send a NTFY to 
> > > > > > > all the INACTIVE
> > > > > > > ASPs (the 4 left inactive) of the state change.  But from 
> > > > > > > Section 3.8.2 of
> > > > > > > the RFC, there is no message to notify of an AS-DOWN state:
> > > > > > > 
> > > > > > >          1    reserved
> > > > > > >          2    Application Server Inactive (AS-INACTIVE)
> > > > > > >          3    Application Server Active (AS-ACTIVE)
> > > > > > >          4    Application Server Pending (AS-PENDING)
> > > > > > > 
> > > > > > > You see, we removed it because there was never an 
> > instance in which
> > > > > it
> > > > > > > would be sent.  You have inserted an instance again.
> > > > > > > 
> > > > > > > Aren't you going to have to define a Notify message (the 
> > > > > > > Reserved value)
> > > > > > > for "Application Server Down (AS-DOWN)" so that the 
> > INACTIVE ASPs
> > > > > can
> > > > > > > be properly notified by the SGP?
> > > > > > > 
> > > > > > > Particularly when the RFC says in Section 4.3.4.5: (Notify
> > > > > Procedures)
> > > > > > > 
> > > > > > >    A Notify message reflecting a change in the AS 
> > state MUST 
> > > > > > > be sent to
> > > > > > >    all ASPs in the AS, except those in the ASP-DOWN 
> > state, with
> > > > > > >    appropriate Status Information and any ASP 
> > Identifier of the
> > > > > failed
> > > > > > >    ASP.  At the ASP, Layer Management is informed 
> > with an M-NOTIFY
> > > > > > >    indication primitive.  The Notify message must be sent 
> > > > > > > whether the AS
> > > > > > >    state change was a result of an ASP failure or 
> > reception of an
> > > > > ASP
> > > > > > >    State management (ASPSM) / ASP Traffic Management (ASPTM)
> > > > > message.
> > > > > > >    In the second case, the Notify message MUST be 
> > sent after 
> > > > > > > any related
> > > > > > >    acknowledgement messages  (e.g., ASP Up Ack, ASP 
> > Down Ack, 
> > > > > > > ASP Active
> > > > > > >    Ack, or ASP Inactive Ack).
> > > > > > > 
> > > > > > > The above tells me that I MUST send a Notify 
> > message becase of the
> > > > > AS
> > > > > > > state change, but there is no message to inform 
> > about AS-DOWN state
> > > > > > > transition.
> > > > > > > 
> > > > > > > What is the purpose of doing this?  What is the protocol 
> > > > > > > difference between
> > > > > > > the AS-DOWN state and the AS-INACTIVE state?  What 
> > is it that the
> > > > > SGP
> > > > > > > does or doesn't do in the one state that it doesn't or does 
> > > > > > > do in the other?
> > > > > > > 
> > > > > > > I don't recall this being discussed at all.
> > > > > > > 
> > > > > > > Please clarify.
> > > > > > > 
> > > > > > > -- 
> > > > > > > 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
> > 
> 
> _______________________________________________
> 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 exim@www1.ietf.org  Tue Jul 22 06:55:49 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA09175
	for <sigtran-archive@odin.ietf.org>; Tue, 22 Jul 2003 06:55:49 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19eunk-00016H-QE
	for sigtran-archive@odin.ietf.org; Tue, 22 Jul 2003 06:55:24 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6MAtOET004221
	for sigtran-archive@odin.ietf.org; Tue, 22 Jul 2003 06:55:24 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19eunO-00013o-2x; Tue, 22 Jul 2003 06:55:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19eunJ-00013F-2B
	for sigtran@optimus.ietf.org; Tue, 22 Jul 2003 06:54:57 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA09159
	for <sigtran@ietf.org>; Tue, 22 Jul 2003 06:54:51 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19eunE-0003Vd-00
	for sigtran@ietf.org; Tue, 22 Jul 2003 06:54:52 -0400
Received: from penguin-ext.wise.edt.ericsson.se ([193.180.251.47] helo=penguin.al.sw.ericsson.se)
	by ietf-mx with esmtp (Exim 4.12)
	id 19eun3-0003VW-00
	for sigtran@ietf.org; Tue, 22 Jul 2003 06:54:41 -0400
Received: from esealnt612.al.sw.ericsson.se ([153.88.254.118])
	by penguin.al.sw.ericsson.se (8.12.9/8.12.9/WIREfire-1.6b) with ESMTP id h6MAsbjG002768;
	Tue, 22 Jul 2003 12:54:37 +0200 (MEST)
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 NW4YBXA4; Tue, 22 Jul 2003 12:54:37 +0200
Received: by ESEALNT745.al.sw.ericsson.se with Internet Mail Service (5.5.2653.19)
	id <PLS0LZ9V>; Tue, 22 Jul 2003 12:54:01 +0200
Message-ID: <1AB3D30B989BF141BBD5C70057B2EF7C043316E2@eestqnt105.es.eu.ericsson.se>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
To: "'Samuel Dur D. Jeyaseelan'" <sjeyasee@ssd.usa.alcatel.com>
Cc: "''sigtran@ietf.org' '" <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA IG v04  Section 3.14
Date: Tue, 22 Jul 2003 12:54:07 +0200
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>

Samuel, 

I agree, an AS is in DOWN state when all its ASPes are in DOWN state. As soon as the first ASP gets the ASP-INACTIVE state, the AS moves to ACTIVE. As n just refers to traffic processing we'll rely on it only for the transition IA2AC...

Then AS-DOWN is not needed for NTFY messages since this state is reached when all ASPs are in ASP-DOWN. Right?

Cheers // Javier.

> -----Original Message-----
> From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> Sent: martes, 22 de julio de 2003 12:45
> To: J.Javier Pastor (EE/EEM)
> Cc: ''sigtran@ietf.org' '
> Subject: RE: [Sigtran] M3UA IG v04 Section 3.14
> 
> 
> 
> Pastor,
>  I think,it has been decided to add AS-DOWN in NTFY message.
>  in n+k reduntancy case, When do we move the AS to DOWN state?
>  (normally an AS is in DOWN state when all its ASPes are in 
> DOWN state)
> 
> --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, 22 Jul 2003, J.Javier Pastor (EE/EEM) wrote:
> 
> > Samuel, 
> > 
> > In this case, I'm afraid you are not ready to handle the 
> traffic with other ASPs since all of them are in ASP_DOWN. If 
> you want to have a fail-safe configuration, you need to have 
> at least one ASP in ASP_INACTIVE state. Then you could NTFY...
> > 
> > Cheers // Javier.
> > 
> > > -----Original Message-----
> > > From: Samuel Dur D. Jeyaseelan 
> [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > Sent: martes, 15 de julio de 2003 6:06
> > > To: ''sigtran@ietf.org' '
> > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.14
> > > 
> > > 
> > > 
> > > Brian,
> > > 
> > >  In n=5(example) case, if 4 ASPes are in DOWN state and 1 ASP 
> > > is in ACTIVE
> > >  state and which handles the traffic.Suppose ACTIVE ASP moves 
> > > to INACTIVE 
> > >  or DOWN state so SGP tries sending NTFY(AS-DOWN) to other ASPes.
> > > 
> > >  How does an ASP process NTFY message in DOWN state?
> > > 
> > > 
> > > --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 Mon, 14 Jul 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Samuel,
> > > > 
> > > > To what end?
> > > > 
> > > > --brian
> > > > 
> > > > Samuel Dur D. Jeyaseelan wrote:                             
> > >                         (Mon, 14 Jul 2003 09:06:36)
> > > > > 
> > > > > Javier,Brian,
> > > > > 
> > > > > 
> > > > >  (i was bit late to enter into this conversation)
> > > > > 
> > > > >  Including a new status type 'AS-DOWN' in NTFY messages 
> > > is a good idea,
> > > > >  but section 4.3.4.5 in RFC says an ASP in DOWN state 
> > > will not process
> > > > >  NTFY message....?
> > > > > 
> > > > > 
> > > > >  Let me take an example of n=5, SGP needs 5 ASPes in 
> ACTIVE state
> > > > >  to declare the AS as ACTIVE(true-only for start case).
> > > > > 
> > > > >  Start-case
> > > > >  ----------
> > > > > 
> > > > >  ACTIVE ASPes    INACTIVE ASPes  DOWN ASPes    AS state   
> > >      case-#
> > > > >  
> > > 
> --------------------------------------------------------------------
> > > > >     0             0               5            DOWN       
> > >      case-1
> > > > >     0             1               4            DOWN       
> > >      case-2
> > > > >     0             2               3            DOWN       
> > >      case-3
> > > > >     .
> > > > >     .
> > > > >     .
> > > > >     0             5               0            INACTIVE   
> > >      case-4
> > > > >     1             4               0            INACTIVE   
> > >      case-5
> > > > >     .
> > > > >     .
> > > > >     .
> > > > >     5             0               0            ACTIVE     
> > >      case-6
> > > > > 
> > > > >    (assume traffic flows)
> > > > > 
> > > > >     Non-start case
> > > > >     --------------
> > > > >     4             1               0            ACTIVE     
> > >      case-7
> > > > >     3             2               0            ACTIVE     
> > >      case-8
> > > > >     .
> > > > >     .
> > > > >     .
> > > > >     1             4               0            ACTIVE     
> > >      case-8
> > > > >     0             5               0            INACTIVE   
> > >      case-9
> > > > >     0             4               1            still 
> > > INACTIVE  case-10
> > > > >     0             3               2            still 
> > > INACTIVE  case-11
> > > > >     .
> > > > >     .
> > > > >     .
> > > > >     0             1               4            still 
> > > INACTIVE  case-12
> > > > >     0             0               5            DOWN       
> > >      case-13
> > > > > 
> > > > > 
> > > > >  I think we should define the following state-conversion 
> > > definition.
> > > > > 
> > > > >  for Start-case
> > > > >  --------------
> > > > > 
> > > > >  DN2IA : One ASP moves to ASP-INACTIVE causing that n 
> ASPs are in
> > > > >          ASP-INACTIVE states.
> > > > > 
> > > > >  IA2DN : One ASP moves to ASP-DOWN causing that n ASPs are in
> > > > >          ASP-DOWN state.
> > > > > 
> > > > >  IA2AC : one ASP moves to ASP-ACTIVE, causing number of 
> > > ASPs are in 
> > > > >          ASP-ACTIVE state.
> > > > > 
> > > > > 
> > > > > Waiting for comments to go ahead.
> > > > > 
> > > > > 
> > > > > --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 Mon, 14 Jul 2003, J.Javier Pastor (EE/EEM) wrote:
> > > > > 
> > > > > > Brian, 
> > > > > > 
> > > > > > What you say it's right, only 1 'n' transition would be 
> > > kept in the proposal: the IA2AC transition.
> > > > > > 
> > > > > > I think it makes sense since "n" refers only to the 
> > > capacity of processing traffic and in the AS-INACTIVE state 
> > > no traffic is processed.
> > > > > > 
> > > > > > Does anyone have any objection to this change?
> > > > > > 
> > > > > > 
> > > > > > Thanks // Javier.
> > > > > > 
> > > > > > -----Original Message-----
> > > > > > From: Brian F. G. Bidulock
> > > > > > To: J.Javier Pastor (EE/EEM)
> > > > > > Cc: 'sigtran@ietf.org'
> > > > > > Sent: 11/07/2003 11:13
> > > > > > Subject: Re: [Sigtran] M3UA IG v04  Section 3.14
> > > > > > 
> > > > > > J.Javier,
> > > > > > 
> > > > > > J.Javier Pastor (EE/EEM) wrote:
> > > > > > (Fri, 11 Jul 2003 10:35:57)
> > > > > > > Hi Brian, 
> > > > > > > 
> > > > > > > You are right saying that there is currently no way 
> > > of notifying
> > > > > > AS-DOWN. In
> > > > > > > fact "n" only applies to the traffic case.
> > > > > > > 
> > > > > > > Then I see two options:
> > > > > > > 	1- include AS-DOWN in NTFY
> > > > > > > 	2- The AS-DOWN state is only got when all the 
> > > ASPs are in
> > > > > > ASP-DOWN state.
> > > > > > > 
> > > > > > > I would addopt 2 since it is inline what we are 
> > > having in the other
> > > > > > UALs. 
> > > > > > 
> > > > > > Yes, please go with 2 as it is also in line with the RFC.
> > > > > > 
> > > > > > That will also leave only 1 'n' transition in the 
> > > proposal: the IA2AC
> > > > > > transition, correct?
> > > > > > 
> > > > > > --brian
> > > > > > 
> > > > > > > 
> > > > > > > Thanks // Javier.
> > > > > > > 
> > > > > > > 
> > > > > > > > -----Original Message-----
> > > > > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > > > > Sent: jueves, 10 de julio de 2003 17:15
> > > > > > > > To: J.Javier Pastor (EE/EEM)
> > > > > > > > Cc: 'sigtran@ietf.org'
> > > > > > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.14
> > > > > > > > 
> > > > > > > > 
> > > > > > > > Javier,
> > > > > > > > 
> > > > > > > > Section 3.14 says:
> > > > > > > > 
> > > > > > > >    IA2DN: One ASP moves to ASP-DOWN causing that 
> > > the number of ASPs
> > > > > > in
> > > > > > > >    either ASP-ACTIVE or ASP-INACTIVE drops below n.
> > > > > > > > 
> > > > > > > > Am I correct to say that if n=5 and there are 4 
> ASPs in the 
> > > > > > > > ASP-INACTIVE
> > > > > > > > and 1 ASP in the ASP-ACTIVE state, and then the 1 
> > > active ASP moves
> > > > > > to
> > > > > > > > the ASP-DOWN state and T(r) expires that the AS is 
> > > to move to 
> > > > > > > > the AS-DOWN
> > > > > > > > state?  As follows:
> > > > > > > > 
> > > > > > > >   If T(r) expires before an ASP becomes ASP-ACTIVE, 
> > > the SGP MAY stop
> > > > > > > >   queuing messages and discards all
> > > > > > > >   previously queued messages. The AS will move to the 
> > > > > > > > AS-INACTIVE state
> > > > > > > >   if at least the number of ASPs in ASP-INACTIVE 
> > > sum n, otherwise it
> > > > > > > >   will move to AS-DOWN state.
> > > > > > > > 
> > > > > > > > Now, that is an AS state change and we must 
> send a NTFY to 
> > > > > > > > all the INACTIVE
> > > > > > > > ASPs (the 4 left inactive) of the state change. 
>  But from 
> > > > > > > > Section 3.8.2 of
> > > > > > > > the RFC, there is no message to notify of an 
> AS-DOWN state:
> > > > > > > > 
> > > > > > > >          1    reserved
> > > > > > > >          2    Application Server Inactive (AS-INACTIVE)
> > > > > > > >          3    Application Server Active (AS-ACTIVE)
> > > > > > > >          4    Application Server Pending (AS-PENDING)
> > > > > > > > 
> > > > > > > > You see, we removed it because there was never an 
> > > instance in which
> > > > > > it
> > > > > > > > would be sent.  You have inserted an instance again.
> > > > > > > > 
> > > > > > > > Aren't you going to have to define a Notify 
> message (the 
> > > > > > > > Reserved value)
> > > > > > > > for "Application Server Down (AS-DOWN)" so that the 
> > > INACTIVE ASPs
> > > > > > can
> > > > > > > > be properly notified by the SGP?
> > > > > > > > 
> > > > > > > > Particularly when the RFC says in Section 
> 4.3.4.5: (Notify
> > > > > > Procedures)
> > > > > > > > 
> > > > > > > >    A Notify message reflecting a change in the AS 
> > > state MUST 
> > > > > > > > be sent to
> > > > > > > >    all ASPs in the AS, except those in the ASP-DOWN 
> > > state, with
> > > > > > > >    appropriate Status Information and any ASP 
> > > Identifier of the
> > > > > > failed
> > > > > > > >    ASP.  At the ASP, Layer Management is informed 
> > > with an M-NOTIFY
> > > > > > > >    indication primitive.  The Notify message 
> must be sent 
> > > > > > > > whether the AS
> > > > > > > >    state change was a result of an ASP failure or 
> > > reception of an
> > > > > > ASP
> > > > > > > >    State management (ASPSM) / ASP Traffic 
> Management (ASPTM)
> > > > > > message.
> > > > > > > >    In the second case, the Notify message MUST be 
> > > sent after 
> > > > > > > > any related
> > > > > > > >    acknowledgement messages  (e.g., ASP Up Ack, ASP 
> > > Down Ack, 
> > > > > > > > ASP Active
> > > > > > > >    Ack, or ASP Inactive Ack).
> > > > > > > > 
> > > > > > > > The above tells me that I MUST send a Notify 
> > > message becase of the
> > > > > > AS
> > > > > > > > state change, but there is no message to inform 
> > > about AS-DOWN state
> > > > > > > > transition.
> > > > > > > > 
> > > > > > > > What is the purpose of doing this?  What is the 
> protocol 
> > > > > > > > difference between
> > > > > > > > the AS-DOWN state and the AS-INACTIVE state?  What 
> > > is it that the
> > > > > > SGP
> > > > > > > > does or doesn't do in the one state that it 
> doesn't or does 
> > > > > > > > do in the other?
> > > > > > > > 
> > > > > > > > I don't recall this being discussed at all.
> > > > > > > > 
> > > > > > > > Please clarify.
> > > > > > > > 
> > > > > > > > -- 
> > > > > > > > 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
> > > 
> > 
> > _______________________________________________
> > 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 exim@www1.ietf.org  Tue Jul 22 15:18:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA24167
	for <sigtran-archive@odin.ietf.org>; Tue, 22 Jul 2003 15:18:32 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19f2eD-0004sV-6t
	for sigtran-archive@odin.ietf.org; Tue, 22 Jul 2003 15:18:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6MJI50C018740
	for sigtran-archive@odin.ietf.org; Tue, 22 Jul 2003 15:18:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19f2e9-0004rV-Q1; Tue, 22 Jul 2003 15:18:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19f2di-0004qz-CF
	for sigtran@optimus.ietf.org; Tue, 22 Jul 2003 15:17:34 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA24130
	for <sigtran@ietf.org>; Tue, 22 Jul 2003 15:17:31 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19f2dh-0006on-00
	for sigtran@ietf.org; Tue, 22 Jul 2003 15:17:33 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[RNnRTTFCoYiqxx1GpwF8Su5eZnrBKssj])
	by ietf-mx with esmtp (Exim 4.12)
	id 19f2dQ-0006oZ-00
	for sigtran@ietf.org; Tue, 22 Jul 2003 15:17:16 -0400
Received: from ns.pigworks.openss7.net (IDENT:8wjqK11QWyXb3h5Nzr+JQzsSiBkeXRxV@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6MJGk109884;
	Tue, 22 Jul 2003 13:16:46 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6MJGkB03715;
	Tue, 22 Jul 2003 13:16:46 -0600
Date: Tue, 22 Jul 2003 13:16:46 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>
Cc: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
        "''sigtran@ietf.org' '" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04  Section 3.14
Message-ID: <20030722131646.A3222@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "Samuel Dur D. Jeyaseelan" <sjeyasee@ssd.usa.alcatel.com>,
	"J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
	"''sigtran@ietf.org' '" <sigtran@ietf.org>
References: <1AB3D30B989BF141BBD5C70057B2EF7C043316DF@eestqnt105.es.eu.ericsson.se> <Pine.SOL.4.10.10307220530000.15922-100000@sun2685.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.4.10.10307220530000.15922-100000@sun2685.ssd.usa.alcatel.com>; from sjeyasee@ssd.usa.alcatel.com on Tue, Jul 22, 2003 at 05:45:16AM -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>

Samuel,

I think it hasn't been decided.

--brian

Samuel Dur D. Jeyaseelan wrote:                                                     (Tue, 22 Jul 2003 05:45:16)
> 
> Pastor,
>  I think,it has been decided to add AS-DOWN in NTFY message.
>  in n+k reduntancy case, When do we move the AS to DOWN state?
>  (normally an AS is in DOWN state when all its ASPes are in DOWN state)
> 
> --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, 22 Jul 2003, J.Javier Pastor (EE/EEM) wrote:
> 
> > Samuel, 
> > 
> > In this case, I'm afraid you are not ready to handle the traffic with other ASPs since all of them are in ASP_DOWN. If you want to have a fail-safe configuration, you need to have at least one ASP in ASP_INACTIVE state. Then you could NTFY...
> > 
> > Cheers // Javier.
> > 
> > > -----Original Message-----
> > > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > Sent: martes, 15 de julio de 2003 6:06
> > > To: ''sigtran@ietf.org' '
> > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.14
> > > 
> > > 
> > > 
> > > Brian,
> > > 
> > >  In n=5(example) case, if 4 ASPes are in DOWN state and 1 ASP 
> > > is in ACTIVE
> > >  state and which handles the traffic.Suppose ACTIVE ASP moves 
> > > to INACTIVE 
> > >  or DOWN state so SGP tries sending NTFY(AS-DOWN) to other ASPes.
> > > 
> > >  How does an ASP process NTFY message in DOWN state?
> > > 
> > > 
> > > --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 Mon, 14 Jul 2003, Brian F. G. Bidulock wrote:
> > > 
> > > > Samuel,
> > > > 
> > > > To what end?
> > > > 
> > > > --brian
> > > > 
> > > > Samuel Dur D. Jeyaseelan wrote:                             
> > >                         (Mon, 14 Jul 2003 09:06:36)
> > > > > 
> > > > > Javier,Brian,
> > > > > 
> > > > > 
> > > > >  (i was bit late to enter into this conversation)
> > > > > 
> > > > >  Including a new status type 'AS-DOWN' in NTFY messages 
> > > is a good idea,
> > > > >  but section 4.3.4.5 in RFC says an ASP in DOWN state 
> > > will not process
> > > > >  NTFY message....?
> > > > > 
> > > > > 
> > > > >  Let me take an example of n=5, SGP needs 5 ASPes in ACTIVE state
> > > > >  to declare the AS as ACTIVE(true-only for start case).
> > > > > 
> > > > >  Start-case
> > > > >  ----------
> > > > > 
> > > > >  ACTIVE ASPes    INACTIVE ASPes  DOWN ASPes    AS state   
> > >      case-#
> > > > >  
> > > --------------------------------------------------------------------
> > > > >     0             0               5            DOWN       
> > >      case-1
> > > > >     0             1               4            DOWN       
> > >      case-2
> > > > >     0             2               3            DOWN       
> > >      case-3
> > > > >     .
> > > > >     .
> > > > >     .
> > > > >     0             5               0            INACTIVE   
> > >      case-4
> > > > >     1             4               0            INACTIVE   
> > >      case-5
> > > > >     .
> > > > >     .
> > > > >     .
> > > > >     5             0               0            ACTIVE     
> > >      case-6
> > > > > 
> > > > >    (assume traffic flows)
> > > > > 
> > > > >     Non-start case
> > > > >     --------------
> > > > >     4             1               0            ACTIVE     
> > >      case-7
> > > > >     3             2               0            ACTIVE     
> > >      case-8
> > > > >     .
> > > > >     .
> > > > >     .
> > > > >     1             4               0            ACTIVE     
> > >      case-8
> > > > >     0             5               0            INACTIVE   
> > >      case-9
> > > > >     0             4               1            still 
> > > INACTIVE  case-10
> > > > >     0             3               2            still 
> > > INACTIVE  case-11
> > > > >     .
> > > > >     .
> > > > >     .
> > > > >     0             1               4            still 
> > > INACTIVE  case-12
> > > > >     0             0               5            DOWN       
> > >      case-13
> > > > > 
> > > > > 
> > > > >  I think we should define the following state-conversion 
> > > definition.
> > > > > 
> > > > >  for Start-case
> > > > >  --------------
> > > > > 
> > > > >  DN2IA : One ASP moves to ASP-INACTIVE causing that n ASPs are in
> > > > >          ASP-INACTIVE states.
> > > > > 
> > > > >  IA2DN : One ASP moves to ASP-DOWN causing that n ASPs are in
> > > > >          ASP-DOWN state.
> > > > > 
> > > > >  IA2AC : one ASP moves to ASP-ACTIVE, causing number of 
> > > ASPs are in 
> > > > >          ASP-ACTIVE state.
> > > > > 
> > > > > 
> > > > > Waiting for comments to go ahead.
> > > > > 
> > > > > 
> > > > > --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 Mon, 14 Jul 2003, J.Javier Pastor (EE/EEM) wrote:
> > > > > 
> > > > > > Brian, 
> > > > > > 
> > > > > > What you say it's right, only 1 'n' transition would be 
> > > kept in the proposal: the IA2AC transition.
> > > > > > 
> > > > > > I think it makes sense since "n" refers only to the 
> > > capacity of processing traffic and in the AS-INACTIVE state 
> > > no traffic is processed.
> > > > > > 
> > > > > > Does anyone have any objection to this change?
> > > > > > 
> > > > > > 
> > > > > > Thanks // Javier.
> > > > > > 
> > > > > > -----Original Message-----
> > > > > > From: Brian F. G. Bidulock
> > > > > > To: J.Javier Pastor (EE/EEM)
> > > > > > Cc: 'sigtran@ietf.org'
> > > > > > Sent: 11/07/2003 11:13
> > > > > > Subject: Re: [Sigtran] M3UA IG v04  Section 3.14
> > > > > > 
> > > > > > J.Javier,
> > > > > > 
> > > > > > J.Javier Pastor (EE/EEM) wrote:
> > > > > > (Fri, 11 Jul 2003 10:35:57)
> > > > > > > Hi Brian, 
> > > > > > > 
> > > > > > > You are right saying that there is currently no way 
> > > of notifying
> > > > > > AS-DOWN. In
> > > > > > > fact "n" only applies to the traffic case.
> > > > > > > 
> > > > > > > Then I see two options:
> > > > > > > 	1- include AS-DOWN in NTFY
> > > > > > > 	2- The AS-DOWN state is only got when all the 
> > > ASPs are in
> > > > > > ASP-DOWN state.
> > > > > > > 
> > > > > > > I would addopt 2 since it is inline what we are 
> > > having in the other
> > > > > > UALs. 
> > > > > > 
> > > > > > Yes, please go with 2 as it is also in line with the RFC.
> > > > > > 
> > > > > > That will also leave only 1 'n' transition in the 
> > > proposal: the IA2AC
> > > > > > transition, correct?
> > > > > > 
> > > > > > --brian
> > > > > > 
> > > > > > > 
> > > > > > > Thanks // Javier.
> > > > > > > 
> > > > > > > 
> > > > > > > > -----Original Message-----
> > > > > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > > > > Sent: jueves, 10 de julio de 2003 17:15
> > > > > > > > To: J.Javier Pastor (EE/EEM)
> > > > > > > > Cc: 'sigtran@ietf.org'
> > > > > > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.14
> > > > > > > > 
> > > > > > > > 
> > > > > > > > Javier,
> > > > > > > > 
> > > > > > > > Section 3.14 says:
> > > > > > > > 
> > > > > > > >    IA2DN: One ASP moves to ASP-DOWN causing that 
> > > the number of ASPs
> > > > > > in
> > > > > > > >    either ASP-ACTIVE or ASP-INACTIVE drops below n.
> > > > > > > > 
> > > > > > > > Am I correct to say that if n=5 and there are 4 ASPs in the 
> > > > > > > > ASP-INACTIVE
> > > > > > > > and 1 ASP in the ASP-ACTIVE state, and then the 1 
> > > active ASP moves
> > > > > > to
> > > > > > > > the ASP-DOWN state and T(r) expires that the AS is 
> > > to move to 
> > > > > > > > the AS-DOWN
> > > > > > > > state?  As follows:
> > > > > > > > 
> > > > > > > >   If T(r) expires before an ASP becomes ASP-ACTIVE, 
> > > the SGP MAY stop
> > > > > > > >   queuing messages and discards all
> > > > > > > >   previously queued messages. The AS will move to the 
> > > > > > > > AS-INACTIVE state
> > > > > > > >   if at least the number of ASPs in ASP-INACTIVE 
> > > sum n, otherwise it
> > > > > > > >   will move to AS-DOWN state.
> > > > > > > > 
> > > > > > > > Now, that is an AS state change and we must send a NTFY to 
> > > > > > > > all the INACTIVE
> > > > > > > > ASPs (the 4 left inactive) of the state change.  But from 
> > > > > > > > Section 3.8.2 of
> > > > > > > > the RFC, there is no message to notify of an AS-DOWN state:
> > > > > > > > 
> > > > > > > >          1    reserved
> > > > > > > >          2    Application Server Inactive (AS-INACTIVE)
> > > > > > > >          3    Application Server Active (AS-ACTIVE)
> > > > > > > >          4    Application Server Pending (AS-PENDING)
> > > > > > > > 
> > > > > > > > You see, we removed it because there was never an 
> > > instance in which
> > > > > > it
> > > > > > > > would be sent.  You have inserted an instance again.
> > > > > > > > 
> > > > > > > > Aren't you going to have to define a Notify message (the 
> > > > > > > > Reserved value)
> > > > > > > > for "Application Server Down (AS-DOWN)" so that the 
> > > INACTIVE ASPs
> > > > > > can
> > > > > > > > be properly notified by the SGP?
> > > > > > > > 
> > > > > > > > Particularly when the RFC says in Section 4.3.4.5: (Notify
> > > > > > Procedures)
> > > > > > > > 
> > > > > > > >    A Notify message reflecting a change in the AS 
> > > state MUST 
> > > > > > > > be sent to
> > > > > > > >    all ASPs in the AS, except those in the ASP-DOWN 
> > > state, with
> > > > > > > >    appropriate Status Information and any ASP 
> > > Identifier of the
> > > > > > failed
> > > > > > > >    ASP.  At the ASP, Layer Management is informed 
> > > with an M-NOTIFY
> > > > > > > >    indication primitive.  The Notify message must be sent 
> > > > > > > > whether the AS
> > > > > > > >    state change was a result of an ASP failure or 
> > > reception of an
> > > > > > ASP
> > > > > > > >    State management (ASPSM) / ASP Traffic Management (ASPTM)
> > > > > > message.
> > > > > > > >    In the second case, the Notify message MUST be 
> > > sent after 
> > > > > > > > any related
> > > > > > > >    acknowledgement messages  (e.g., ASP Up Ack, ASP 
> > > Down Ack, 
> > > > > > > > ASP Active
> > > > > > > >    Ack, or ASP Inactive Ack).
> > > > > > > > 
> > > > > > > > The above tells me that I MUST send a Notify 
> > > message becase of the
> > > > > > AS
> > > > > > > > state change, but there is no message to inform 
> > > about AS-DOWN state
> > > > > > > > transition.
> > > > > > > > 
> > > > > > > > What is the purpose of doing this?  What is the protocol 
> > > > > > > > difference between
> > > > > > > > the AS-DOWN state and the AS-INACTIVE state?  What 
> > > is it that the
> > > > > > SGP
> > > > > > > > does or doesn't do in the one state that it doesn't or does 
> > > > > > > > do in the other?
> > > > > > > > 
> > > > > > > > I don't recall this being discussed at all.
> > > > > > > > 
> > > > > > > > Please clarify.
> > > > > > > > 
> > > > > > > > -- 
> > > > > > > > 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
> > > 
> > 
> > _______________________________________________
> > 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 exim@www1.ietf.org  Tue Jul 22 16:08:02 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA24166
	for <sigtran-archive@odin.ietf.org>; Tue, 22 Jul 2003 15:18:32 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19f2eD-0004sW-72
	for sigtran-archive@odin.ietf.org; Tue, 22 Jul 2003 15:18:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6MJI5T1018739
	for sigtran-archive@odin.ietf.org; Tue, 22 Jul 2003 15:18:05 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19f2eA-0004rf-AQ; Tue, 22 Jul 2003 15:18:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19f2e7-0004rF-Nw
	for sigtran@optimus.ietf.org; Tue, 22 Jul 2003 15:17:59 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA24152
	for <sigtran@ietf.org>; Tue, 22 Jul 2003 15:17:56 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19f2e6-0006oy-00
	for sigtran@ietf.org; Tue, 22 Jul 2003 15:17:58 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[4i9b9PVPL83QQDQlRIUvR7vNTaqenvTf])
	by ietf-mx with esmtp (Exim 4.12)
	id 19f2dp-0006or-00
	for sigtran@ietf.org; Tue, 22 Jul 2003 15:17:41 -0400
Received: from ns.pigworks.openss7.net (IDENT:zsyK2jsCu9iy2XJPVzAK2Mv3dOfl7s6k@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6MJHU109895;
	Tue, 22 Jul 2003 13:17:30 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6MJHUE03722;
	Tue, 22 Jul 2003 13:17:30 -0600
Date: Tue, 22 Jul 2003 13:17:29 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
Cc: "'Samuel Dur D. Jeyaseelan'" <sjeyasee@ssd.usa.alcatel.com>,
        "''sigtran@ietf.org' '" <sigtran@ietf.org>
Subject: Re: [Sigtran] M3UA IG v04  Section 3.14
Message-ID: <20030722131729.B3222@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
	"'Samuel Dur D. Jeyaseelan'" <sjeyasee@ssd.usa.alcatel.com>,
	"''sigtran@ietf.org' '" <sigtran@ietf.org>
References: <1AB3D30B989BF141BBD5C70057B2EF7C043316E2@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: <1AB3D30B989BF141BBD5C70057B2EF7C043316E2@eestqnt105.es.eu.ericsson.se>; from j.javier.pastor@ericsson.com on Tue, Jul 22, 2003 at 12:54:07PM +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>

J.Javier,

Right!

--brian

J.Javier Pastor (EE/EEM) wrote:                                                   (Tue, 22 Jul 2003 12:54:07)
> Samuel, 
> 
> I agree, an AS is in DOWN state when all its ASPes are in DOWN state. As soon as the first ASP gets the ASP-INACTIVE state, the AS moves to ACTIVE. As n just refers to traffic processing we'll rely on it only for the transition IA2AC...
> 
> Then AS-DOWN is not needed for NTFY messages since this state is reached when all ASPs are in ASP-DOWN. Right?
> 
> Cheers // Javier.
> 
> > -----Original Message-----
> > From: Samuel Dur D. Jeyaseelan [mailto:sjeyasee@ssd.usa.alcatel.com]
> > Sent: martes, 22 de julio de 2003 12:45
> > To: J.Javier Pastor (EE/EEM)
> > Cc: ''sigtran@ietf.org' '
> > Subject: RE: [Sigtran] M3UA IG v04 Section 3.14
> > 
> > 
> > 
> > Pastor,
> >  I think,it has been decided to add AS-DOWN in NTFY message.
> >  in n+k reduntancy case, When do we move the AS to DOWN state?
> >  (normally an AS is in DOWN state when all its ASPes are in 
> > DOWN state)
> > 
> > --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, 22 Jul 2003, J.Javier Pastor (EE/EEM) wrote:
> > 
> > > Samuel, 
> > > 
> > > In this case, I'm afraid you are not ready to handle the 
> > traffic with other ASPs since all of them are in ASP_DOWN. If 
> > you want to have a fail-safe configuration, you need to have 
> > at least one ASP in ASP_INACTIVE state. Then you could NTFY...
> > > 
> > > Cheers // Javier.
> > > 
> > > > -----Original Message-----
> > > > From: Samuel Dur D. Jeyaseelan 
> > [mailto:sjeyasee@ssd.usa.alcatel.com]
> > > > Sent: martes, 15 de julio de 2003 6:06
> > > > To: ''sigtran@ietf.org' '
> > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.14
> > > > 
> > > > 
> > > > 
> > > > Brian,
> > > > 
> > > >  In n=5(example) case, if 4 ASPes are in DOWN state and 1 ASP 
> > > > is in ACTIVE
> > > >  state and which handles the traffic.Suppose ACTIVE ASP moves 
> > > > to INACTIVE 
> > > >  or DOWN state so SGP tries sending NTFY(AS-DOWN) to other ASPes.
> > > > 
> > > >  How does an ASP process NTFY message in DOWN state?
> > > > 
> > > > 
> > > > --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 Mon, 14 Jul 2003, Brian F. G. Bidulock wrote:
> > > > 
> > > > > Samuel,
> > > > > 
> > > > > To what end?
> > > > > 
> > > > > --brian
> > > > > 
> > > > > Samuel Dur D. Jeyaseelan wrote:                             
> > > >                         (Mon, 14 Jul 2003 09:06:36)
> > > > > > 
> > > > > > Javier,Brian,
> > > > > > 
> > > > > > 
> > > > > >  (i was bit late to enter into this conversation)
> > > > > > 
> > > > > >  Including a new status type 'AS-DOWN' in NTFY messages 
> > > > is a good idea,
> > > > > >  but section 4.3.4.5 in RFC says an ASP in DOWN state 
> > > > will not process
> > > > > >  NTFY message....?
> > > > > > 
> > > > > > 
> > > > > >  Let me take an example of n=5, SGP needs 5 ASPes in 
> > ACTIVE state
> > > > > >  to declare the AS as ACTIVE(true-only for start case).
> > > > > > 
> > > > > >  Start-case
> > > > > >  ----------
> > > > > > 
> > > > > >  ACTIVE ASPes    INACTIVE ASPes  DOWN ASPes    AS state   
> > > >      case-#
> > > > > >  
> > > > 
> > --------------------------------------------------------------------
> > > > > >     0             0               5            DOWN       
> > > >      case-1
> > > > > >     0             1               4            DOWN       
> > > >      case-2
> > > > > >     0             2               3            DOWN       
> > > >      case-3
> > > > > >     .
> > > > > >     .
> > > > > >     .
> > > > > >     0             5               0            INACTIVE   
> > > >      case-4
> > > > > >     1             4               0            INACTIVE   
> > > >      case-5
> > > > > >     .
> > > > > >     .
> > > > > >     .
> > > > > >     5             0               0            ACTIVE     
> > > >      case-6
> > > > > > 
> > > > > >    (assume traffic flows)
> > > > > > 
> > > > > >     Non-start case
> > > > > >     --------------
> > > > > >     4             1               0            ACTIVE     
> > > >      case-7
> > > > > >     3             2               0            ACTIVE     
> > > >      case-8
> > > > > >     .
> > > > > >     .
> > > > > >     .
> > > > > >     1             4               0            ACTIVE     
> > > >      case-8
> > > > > >     0             5               0            INACTIVE   
> > > >      case-9
> > > > > >     0             4               1            still 
> > > > INACTIVE  case-10
> > > > > >     0             3               2            still 
> > > > INACTIVE  case-11
> > > > > >     .
> > > > > >     .
> > > > > >     .
> > > > > >     0             1               4            still 
> > > > INACTIVE  case-12
> > > > > >     0             0               5            DOWN       
> > > >      case-13
> > > > > > 
> > > > > > 
> > > > > >  I think we should define the following state-conversion 
> > > > definition.
> > > > > > 
> > > > > >  for Start-case
> > > > > >  --------------
> > > > > > 
> > > > > >  DN2IA : One ASP moves to ASP-INACTIVE causing that n 
> > ASPs are in
> > > > > >          ASP-INACTIVE states.
> > > > > > 
> > > > > >  IA2DN : One ASP moves to ASP-DOWN causing that n ASPs are in
> > > > > >          ASP-DOWN state.
> > > > > > 
> > > > > >  IA2AC : one ASP moves to ASP-ACTIVE, causing number of 
> > > > ASPs are in 
> > > > > >          ASP-ACTIVE state.
> > > > > > 
> > > > > > 
> > > > > > Waiting for comments to go ahead.
> > > > > > 
> > > > > > 
> > > > > > --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 Mon, 14 Jul 2003, J.Javier Pastor (EE/EEM) wrote:
> > > > > > 
> > > > > > > Brian, 
> > > > > > > 
> > > > > > > What you say it's right, only 1 'n' transition would be 
> > > > kept in the proposal: the IA2AC transition.
> > > > > > > 
> > > > > > > I think it makes sense since "n" refers only to the 
> > > > capacity of processing traffic and in the AS-INACTIVE state 
> > > > no traffic is processed.
> > > > > > > 
> > > > > > > Does anyone have any objection to this change?
> > > > > > > 
> > > > > > > 
> > > > > > > Thanks // Javier.
> > > > > > > 
> > > > > > > -----Original Message-----
> > > > > > > From: Brian F. G. Bidulock
> > > > > > > To: J.Javier Pastor (EE/EEM)
> > > > > > > Cc: 'sigtran@ietf.org'
> > > > > > > Sent: 11/07/2003 11:13
> > > > > > > Subject: Re: [Sigtran] M3UA IG v04  Section 3.14
> > > > > > > 
> > > > > > > J.Javier,
> > > > > > > 
> > > > > > > J.Javier Pastor (EE/EEM) wrote:
> > > > > > > (Fri, 11 Jul 2003 10:35:57)
> > > > > > > > Hi Brian, 
> > > > > > > > 
> > > > > > > > You are right saying that there is currently no way 
> > > > of notifying
> > > > > > > AS-DOWN. In
> > > > > > > > fact "n" only applies to the traffic case.
> > > > > > > > 
> > > > > > > > Then I see two options:
> > > > > > > > 	1- include AS-DOWN in NTFY
> > > > > > > > 	2- The AS-DOWN state is only got when all the 
> > > > ASPs are in
> > > > > > > ASP-DOWN state.
> > > > > > > > 
> > > > > > > > I would addopt 2 since it is inline what we are 
> > > > having in the other
> > > > > > > UALs. 
> > > > > > > 
> > > > > > > Yes, please go with 2 as it is also in line with the RFC.
> > > > > > > 
> > > > > > > That will also leave only 1 'n' transition in the 
> > > > proposal: the IA2AC
> > > > > > > transition, correct?
> > > > > > > 
> > > > > > > --brian
> > > > > > > 
> > > > > > > > 
> > > > > > > > Thanks // Javier.
> > > > > > > > 
> > > > > > > > 
> > > > > > > > > -----Original Message-----
> > > > > > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > > > > > Sent: jueves, 10 de julio de 2003 17:15
> > > > > > > > > To: J.Javier Pastor (EE/EEM)
> > > > > > > > > Cc: 'sigtran@ietf.org'
> > > > > > > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.14
> > > > > > > > > 
> > > > > > > > > 
> > > > > > > > > Javier,
> > > > > > > > > 
> > > > > > > > > Section 3.14 says:
> > > > > > > > > 
> > > > > > > > >    IA2DN: One ASP moves to ASP-DOWN causing that 
> > > > the number of ASPs
> > > > > > > in
> > > > > > > > >    either ASP-ACTIVE or ASP-INACTIVE drops below n.
> > > > > > > > > 
> > > > > > > > > Am I correct to say that if n=5 and there are 4 
> > ASPs in the 
> > > > > > > > > ASP-INACTIVE
> > > > > > > > > and 1 ASP in the ASP-ACTIVE state, and then the 1 
> > > > active ASP moves
> > > > > > > to
> > > > > > > > > the ASP-DOWN state and T(r) expires that the AS is 
> > > > to move to 
> > > > > > > > > the AS-DOWN
> > > > > > > > > state?  As follows:
> > > > > > > > > 
> > > > > > > > >   If T(r) expires before an ASP becomes ASP-ACTIVE, 
> > > > the SGP MAY stop
> > > > > > > > >   queuing messages and discards all
> > > > > > > > >   previously queued messages. The AS will move to the 
> > > > > > > > > AS-INACTIVE state
> > > > > > > > >   if at least the number of ASPs in ASP-INACTIVE 
> > > > sum n, otherwise it
> > > > > > > > >   will move to AS-DOWN state.
> > > > > > > > > 
> > > > > > > > > Now, that is an AS state change and we must 
> > send a NTFY to 
> > > > > > > > > all the INACTIVE
> > > > > > > > > ASPs (the 4 left inactive) of the state change. 
> >  But from 
> > > > > > > > > Section 3.8.2 of
> > > > > > > > > the RFC, there is no message to notify of an 
> > AS-DOWN state:
> > > > > > > > > 
> > > > > > > > >          1    reserved
> > > > > > > > >          2    Application Server Inactive (AS-INACTIVE)
> > > > > > > > >          3    Application Server Active (AS-ACTIVE)
> > > > > > > > >          4    Application Server Pending (AS-PENDING)
> > > > > > > > > 
> > > > > > > > > You see, we removed it because there was never an 
> > > > instance in which
> > > > > > > it
> > > > > > > > > would be sent.  You have inserted an instance again.
> > > > > > > > > 
> > > > > > > > > Aren't you going to have to define a Notify 
> > message (the 
> > > > > > > > > Reserved value)
> > > > > > > > > for "Application Server Down (AS-DOWN)" so that the 
> > > > INACTIVE ASPs
> > > > > > > can
> > > > > > > > > be properly notified by the SGP?
> > > > > > > > > 
> > > > > > > > > Particularly when the RFC says in Section 
> > 4.3.4.5: (Notify
> > > > > > > Procedures)
> > > > > > > > > 
> > > > > > > > >    A Notify message reflecting a change in the AS 
> > > > state MUST 
> > > > > > > > > be sent to
> > > > > > > > >    all ASPs in the AS, except those in the ASP-DOWN 
> > > > state, with
> > > > > > > > >    appropriate Status Information and any ASP 
> > > > Identifier of the
> > > > > > > failed
> > > > > > > > >    ASP.  At the ASP, Layer Management is informed 
> > > > with an M-NOTIFY
> > > > > > > > >    indication primitive.  The Notify message 
> > must be sent 
> > > > > > > > > whether the AS
> > > > > > > > >    state change was a result of an ASP failure or 
> > > > reception of an
> > > > > > > ASP
> > > > > > > > >    State management (ASPSM) / ASP Traffic 
> > Management (ASPTM)
> > > > > > > message.
> > > > > > > > >    In the second case, the Notify message MUST be 
> > > > sent after 
> > > > > > > > > any related
> > > > > > > > >    acknowledgement messages  (e.g., ASP Up Ack, ASP 
> > > > Down Ack, 
> > > > > > > > > ASP Active
> > > > > > > > >    Ack, or ASP Inactive Ack).
> > > > > > > > > 
> > > > > > > > > The above tells me that I MUST send a Notify 
> > > > message becase of the
> > > > > > > AS
> > > > > > > > > state change, but there is no message to inform 
> > > > about AS-DOWN state
> > > > > > > > > transition.
> > > > > > > > > 
> > > > > > > > > What is the purpose of doing this?  What is the 
> > protocol 
> > > > > > > > > difference between
> > > > > > > > > the AS-DOWN state and the AS-INACTIVE state?  What 
> > > > is it that the
> > > > > > > SGP
> > > > > > > > > does or doesn't do in the one state that it 
> > doesn't or does 
> > > > > > > > > do in the other?
> > > > > > > > > 
> > > > > > > > > I don't recall this being discussed at all.
> > > > > > > > > 
> > > > > > > > > Please clarify.
> > > > > > > > > 
> > > > > > > > > -- 
> > > > > > > > > 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
> > > > 
> > > 
> > > _______________________________________________
> > > 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 exim@www1.ietf.org  Wed Jul 23 00:10:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA05149
	for <sigtran-archive@odin.ietf.org>; Wed, 23 Jul 2003 00:10:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fAx1-0005Nw-Rv
	for sigtran-archive@odin.ietf.org; Wed, 23 Jul 2003 00:10:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6N4A36f020696
	for sigtran-archive@odin.ietf.org; Wed, 23 Jul 2003 00:10:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fAwz-0005NT-SP; Wed, 23 Jul 2003 00:10:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fAwW-0005Mx-VS
	for sigtran@optimus.ietf.org; Wed, 23 Jul 2003 00:09:33 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA05135
	for <sigtran@ietf.org>; Wed, 23 Jul 2003 00:09:28 -0400 (EDT)
From: kakaul@hss.hns.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fAwU-0001Um-00
	for sigtran@ietf.org; Wed, 23 Jul 2003 00:09:30 -0400
Received: from [202.54.26.202] (helo=hindon.hss.co.in)
	by ietf-mx with esmtp (Exim 4.12)
	id 19fAwJ-0001UX-00
	for sigtran@ietf.org; Wed, 23 Jul 2003 00:09:19 -0400
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h6N44Xs23139;
	Wed, 23 Jul 2003 09:34:33 +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 ESMTP id h6N48KD29259;
	Wed, 23 Jul 2003 09:38:20 +0530 (IST)
Subject: RE: [Sigtran] M3UA IG v04 Section 3.18
To: "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>
Cc: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.5  September 22, 2000
Message-ID: <OFF9D7F690.2E7F88F8-ON65256D6C.0014A9CF@hss.hns.com>
Date: Wed, 23 Jul 2003 09:32:55 +0530
X-MIMETrack: Serialize by Router on Sandesh/HSS(Release 6.0|September 26, 2002) at 23/07/2003
 09:31:40 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>





Hi Javier,

Yes the solution description section does not give full intent of the
timer.Basically congestion supervision using timer is applicable at both SG
and ASP.Its
use is clear that if you are not getting updates about current congestion
status of affected point code through SCON message some implementors would
like to supervise the congestion status of affected point code using some
timers and they should be given flexibility to do so.

For now i agree with your option 2

Kamesh Kaul
www.hssworld.com
Hughes Software Systems




"J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>@ietf.org on
07/22/2003 03:34:59 PM

Sent by:    sigtran-admin@ietf.org


To:    Kamesh Kaul/HSS@HSS
cc:    sigtran@ietf.org

Subject:    RE: [Sigtran] M3UA IG v04 Section 3.18


Hi Kamesh,

Sorry for not answering before, but I was attending the IETF meeting and
the access tool is not very good to track all the issues...

It seems to me that the timer referred in the Solution Description section
has been misunderstood.

- The referred timer is said to be located at the ASP peer node, i.e. the
SGP. And the reason for the timer is to solve the problem outlined in the
Problem Description section: ASP indication of own congestion abatement.
Nothing to do with the reception of SCON from the SGP.

- The timer is mentioned in the Solution Description section. Then if the
timer is really seen as an important implementation clue we should move it
to the Changes section, maybe as an IMPLEMENTATION NOTE since interop is
not involved. Take into account that the Solution Description MAY disappear
from the final RFC, if we get an RFC3332bis instead of a IG separate RFC
(we'll have this discussion when the doc is ready for approval).

At this point I'd choose either:
 1- Remove the timer from the Solution Section.
 2- Include an IMPLEMENTOR'S NOTE in the changes with reference to the
 timer.

If you feel that mentioning the timer may help to some implementations we
should go for option 2.

Cheers // Javier.



> -----Original Message-----
> From: kakaul@hss.hns.com [mailto:kakaul@hss.hns.com]
> Sent: martes, 15 de julio de 2003 6:05
> To: J.Javier Pastor (EE/EEM)
> Cc: sigtran@ietf.org
> Subject: RE: [Sigtran] M3UA IG v04 Section 3.18
>
>
>
>
>
>
> Hi Javier,
>                     This timer may be used by some implementations to
> supervise the congestion status of point code if SCON
> messages send by SGP
> are lost in transit to ASP.It was discussed earlier and was
> agreed upon by
> all to keep this timer.I disagree with removing the mention
> of the timer.
>
> Regards
> Kamesh Kaul
> www.hssworld.com
> Hughes Software Systems
>
>
>
>
> "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>@ietf.org on
> 07/14/2003 04:05:58 PM
>
> Sent by:    sigtran-admin@ietf.org
>
>
> To:    "'Brian F. G. Bidulock '" <bidulock@openss7.org>
> cc:    "''sigtran@ietf.org' '" <sigtran@ietf.org>
>
> Subject:    RE: [Sigtran] M3UA IG v04 Section 3.18
>
>
> Brian,
>
> We'll remove the mention of the timer, if nobody disagrees...
>
> Cheers // Javier.
>
> -----Original Message-----
> From: Brian F. G. Bidulock
> To: J.Javier Pastor (EE/EEM)
> Cc: 'sigtran@ietf.org'
> Sent: 11/07/2003 11:07
> Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
>
> J.Javier,
>
> J.Javier Pastor (EE/EEM) wrote:
> (Fri, 11 Jul 2003 09:56:59)
> > Hi Brian,
> >
> > Pls, see below...
> >
> > > -----Original Message-----
> > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > Sent: jueves, 10 de julio de 2003 15:34
> > > To: J.Javier Pastor (EE/EEM)
> > > Cc: 'sigtran@ietf.org'
> > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > >
> > >
> > > Javier,
> > >
> > > 3.18.3 Solution Description
> > >
> > >    Clarify that the ASP needs to indicate when the congestion
> > > level has
> > >    changed (including abatement).  Further, the ASP peer can
> maintain
> > >    a timer, if desired, in case the ASP fails to indicate
> congestion
> > >    abatement.
> > >
> > > Where are the procedures for this timer described?  This
> is the only
> place
> > > that the timer is mentioned.  Please describe the timer or remove
> the
> > > sentence saying that it has been described...
> > >
> >
> > Do you think is worth include the timer explanation into
> the new text?
> > Otherwise, we may remove that sentence.
>
> Please remove mention of the timer.
>
> --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 exim@www1.ietf.org  Wed Jul 23 16:34:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA01654
	for <sigtran-archive@odin.ietf.org>; Wed, 23 Jul 2003 16:34:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fQJH-0005AP-4n
	for sigtran-archive@odin.ietf.org; Wed, 23 Jul 2003 16:34:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6NKY3U2019855
	for sigtran-archive@odin.ietf.org; Wed, 23 Jul 2003 16:34:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fQJE-0005A0-Tu; Wed, 23 Jul 2003 16:34:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fQJA-00059o-3T
	for sigtran@optimus.ietf.org; Wed, 23 Jul 2003 16:33:56 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA01643
	for <SIGTRAN@ietf.org>; Wed, 23 Jul 2003 16:33:52 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fQJ8-0000Yv-00
	for SIGTRAN@ietf.org; Wed, 23 Jul 2003 16:33:54 -0400
Received: from auds951.usa.alcatel.com ([143.209.238.80])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fQIx-0000Yd-00
	for SIGTRAN@ietf.org; Wed, 23 Jul 2003 16:33:43 -0400
Received: from alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6NKWiwA002193;
	Wed, 23 Jul 2003 15:32:45 -0500 (CDT)
Message-ID: <3F1EF0EC.4090409@alcatel.com>
Date: Wed, 23 Jul 2003 15:32:44 -0500
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
User-Agent: Mozilla/5.0 (X11; U; SunOS sun4u; en-US; rv:1.0.1) Gecko/20020920 Netscape/7.0
X-Accept-Language: en-us, en
MIME-Version: 1.0
CC: SIGTRAN e-mail list <SIGTRAN@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - T1 Editorial
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030711224130.A20278@openss7.org> <3F171334.6070308@alcatel.com> <20030717175223.A29221@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

Brian,

Maybe this?

=====
The Link Status Ready message replaces the FISU of MTP2 that is sent
at the end of the proving period. The Link Status Ready message is
used to verify that both ends have completed proving. When M2PA starts
timer T1, it SHALL send a Link Status Ready message to its peer in the
case where MTP2 would send a FISU after proving is complete. If the
Link Status Ready message is sent, then M2PA MAY send additional Link
Status Ready messages while timer T1 is running.

In the case that MTP2 sends an MSU or SIPO message at the end of
proving, M2PA SHALL send (respectively) a User Data or Link Status
Processor Outage message.
=====

tomG.

Brian F. G. Bidulock wrote:
> Tom,
> 
> Maybe change "would send" to "sends"???
> 
> Also, it may only send additional LSR where MTP2 sends additional FISU?
> 
> I think the intent is unchanged, maybe we can find clearer wording for the
> final.
> 
> --brian
> 
> On Thu, 17 Jul 2003, Tom George wrote:
> 
> 
>>Brian,
>>
>>I don't have a problem with your suggested paragraph, except that is a 
>>bit wordy. I put in the qualification "in the case where MTP2 would send 
>>a FISU after proving is complete" to cover the cases you mention. Send 
>>LS Ready when MTP2 would send the FISU. Elsewhere in the document it 
>>says to send LS PO where MTP2 sends SIPO, and User Data where MTP2 sends 
>>an MSU.
>>
>>But the judge of readability is the reader. Let me know if you still 
>>think I should change it.
>>
>>tomG.
>>
>>Brian F. G. Bidulock wrote:
>>
>>>I have been updating the M2PA-TEST spec and have noticed a few things in
>>>M2PA-09...
>>>
>>>M2PA-09 Says: (top of Page 27):
>>>
>>>  The Link Status Ready message replaces the FISU of MTP2 that is sent
>>>  at the end of the proving period. The Link Status Ready message is
>>>  used to verify that both ends have completed proving. When M2PA starts
>>>  timer T1, it SHALL send a Link Status Ready message to its peer in the
>>>  case where MTP2 would send a FISU after proving is complete. M2PA MAY
>>>  send additional Link Status Ready messages while timer T1 is running.
>>>
>>>This is all that is mention on T1.
>>>
>>>In MTP2 T1 is started when the state machine enters the "Aligned Ready"
>>>or "Aligned Not Ready" state.  Only from the "Aligned Ready" state does
>>>MTP2 send FISU.  From "Aligned Not Ready" state, it sends SIPO.  Also,
>>>if data is pending transmission, MTP2 can enter the "Aligned Ready" state
>>>by sending an MSU rather than a FISU.
>>>
>>>The problem is that M2PA-09 says "When M2PA starts timer T1, it SHALL send
>>>a Link Status message...", when in fact if M2PA follows Q.703 could start
>>>T1 and send a DATA message, or could start T1 and send a PROCESSOR-OUTAGE
>>>message.
>>>
>>>Perhaps we could say:
>>>
>>>  The Link Status Ready message replaces the FISU, the Data message replaces
>>>  the MSU, and the Link Status Processor Outage message replaces the SIPO,
>>>  that would be sent at the end of the proving period.  The Link Status Ready
>>>  Data or Link Status Processor Outage message is used to verify that both
>>>  ends have completed proving.  Whem M2PA starts timer T1, it SHALL send a
>>>  Link Status Ready, Data or Link Status Processor Outage message to is peer
>>>  in the case where MTP2 would send a FISU, MSU or SIPO ater proving is
>>>  complete.  M2PA MAY send additional Link Status Ready, Data or Link Status
>>>  Processor Outage messages while timer T1 is running.
>>>
>>>Ok?
>>>
>>>--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.  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.  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 exim@www1.ietf.org  Wed Jul 23 18:39:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA06534
	for <sigtran-archive@odin.ietf.org>; Wed, 23 Jul 2003 18:39:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fSGE-0003fE-Q0
	for sigtran-archive@odin.ietf.org; Wed, 23 Jul 2003 18:39:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6NMd2At014078
	for sigtran-archive@odin.ietf.org; Wed, 23 Jul 2003 18:39:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fSGC-0003eq-Rk; Wed, 23 Jul 2003 18:39:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fSFf-0003d5-NV
	for sigtran@optimus.ietf.org; Wed, 23 Jul 2003 18:38:27 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA06492
	for <SIGTRAN@ietf.org>; Wed, 23 Jul 2003 18:38:21 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fSFb-0001QL-00
	for SIGTRAN@ietf.org; Wed, 23 Jul 2003 18:38:23 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[JiYYfdPjWCJgWv59r/7tFGI7qghbwi+A])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fSFL-0001Q7-00
	for SIGTRAN@ietf.org; Wed, 23 Jul 2003 18:38:07 -0400
Received: from ns.pigworks.openss7.net (IDENT:q86uu0Z44KmjOTNzLmXizMM8QlAYCXlr@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6NMbf104441;
	Wed, 23 Jul 2003 16:37:41 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6NMbf026919;
	Wed, 23 Jul 2003 16:37:41 -0600
Date: Wed, 23 Jul 2003 16:37:41 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tom George <Tom.George@alcatel.com>
Cc: SIGTRAN e-mail list <SIGTRAN@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - T1 Editorial
Message-ID: <20030723163741.A26533@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tom George <Tom.George@alcatel.com>,
	SIGTRAN e-mail list <SIGTRAN@ietf.org>
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030711224130.A20278@openss7.org> <3F171334.6070308@alcatel.com> <20030717175223.A29221@openss7.org> <3F1EF0EC.4090409@alcatel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3F1EF0EC.4090409@alcatel.com>; from Tom.George@alcatel.com on Wed, Jul 23, 2003 at 03:32:44PM -0500
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6NMbf104441
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Tom,

Poifect!

--brian


On Wed, 23 Jul 2003, Tom George wrote:

> Brian,
>=20
> Maybe this?
>=20
> =3D=3D=3D=3D=3D
> The Link Status Ready message replaces the FISU of MTP2 that is sent
> at the end of the proving period. The Link Status Ready message is
> used to verify that both ends have completed proving. When M2PA starts
> timer T1, it SHALL send a Link Status Ready message to its peer in the
> case where MTP2 would send a FISU after proving is complete. If the
> Link Status Ready message is sent, then M2PA MAY send additional Link
> Status Ready messages while timer T1 is running.
>=20
> In the case that MTP2 sends an MSU or SIPO message at the end of
> proving, M2PA SHALL send (respectively) a User Data or Link Status
> Processor Outage message.
> =3D=3D=3D=3D=3D
>=20
> tomG.
>=20
> Brian F. G. Bidulock wrote:
> > Tom,
> >=20
> > Maybe change "would send" to "sends"???
> >=20
> > Also, it may only send additional LSR where MTP2 sends additional FIS=
U?
> >=20
> > I think the intent is unchanged, maybe we can find clearer wording fo=
r the
> > final.
> >=20
> > --brian
> >=20
> > On Thu, 17 Jul 2003, Tom George wrote:
> >=20
> >=20
> >>Brian,
> >>
> >>I don't have a problem with your suggested paragraph, except that is =
a=20
> >>bit wordy. I put in the qualification "in the case where MTP2 would s=
end=20
> >>a FISU after proving is complete" to cover the cases you mention. Sen=
d=20
> >>LS Ready when MTP2 would send the FISU. Elsewhere in the document it=20
> >>says to send LS PO where MTP2 sends SIPO, and User Data where MTP2 se=
nds=20
> >>an MSU.
> >>
> >>But the judge of readability is the reader. Let me know if you still=20
> >>think I should change it.
> >>
> >>tomG.
> >>
> >>Brian F. G. Bidulock wrote:
> >>
> >>>I have been updating the M2PA-TEST spec and have noticed a few thing=
s in
> >>>M2PA-09...
> >>>
> >>>M2PA-09 Says: (top of Page 27):
> >>>
> >>>  The Link Status Ready message replaces the FISU of MTP2 that is se=
nt
> >>>  at the end of the proving period. The Link Status Ready message is
> >>>  used to verify that both ends have completed proving. When M2PA st=
arts
> >>>  timer T1, it SHALL send a Link Status Ready message to its peer in=
 the
> >>>  case where MTP2 would send a FISU after proving is complete. M2PA =
MAY
> >>>  send additional Link Status Ready messages while timer T1 is runni=
ng.
> >>>
> >>>This is all that is mention on T1.
> >>>
> >>>In MTP2 T1 is started when the state machine enters the "Aligned Rea=
dy"
> >>>or "Aligned Not Ready" state.  Only from the "Aligned Ready" state d=
oes
> >>>MTP2 send FISU.  From "Aligned Not Ready" state, it sends SIPO.  Als=
o,
> >>>if data is pending transmission, MTP2 can enter the "Aligned Ready" =
state
> >>>by sending an MSU rather than a FISU.
> >>>
> >>>The problem is that M2PA-09 says "When M2PA starts timer T1, it SHAL=
L send
> >>>a Link Status message...", when in fact if M2PA follows Q.703 could =
start
> >>>T1 and send a DATA message, or could start T1 and send a PROCESSOR-O=
UTAGE
> >>>message.
> >>>
> >>>Perhaps we could say:
> >>>
> >>>  The Link Status Ready message replaces the FISU, the Data message =
replaces
> >>>  the MSU, and the Link Status Processor Outage message replaces the=
 SIPO,
> >>>  that would be sent at the end of the proving period.  The Link Sta=
tus Ready
> >>>  Data or Link Status Processor Outage message is used to verify tha=
t both
> >>>  ends have completed proving.  Whem M2PA starts timer T1, it SHALL =
send a
> >>>  Link Status Ready, Data or Link Status Processor Outage message to=
 is peer
> >>>  in the case where MTP2 would send a FISU, MSU or SIPO ater proving=
 is
> >>>  complete.  M2PA MAY send additional Link Status Ready, Data or Lin=
k Status
> >>>  Processor Outage messages while timer T1 is running.
> >>>
> >>>Ok?
> >>>
> >>>--brian
> >>>
> >>
> >>
> >>--=20
> >>* 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.  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
> >=20
> >=20
>=20
>=20
> --=20
> * 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.  I do not purport to    *
> *   advertise or represent its position, operations, or activities.   *
>=20
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--=20
Brian F. G. Bidulock    =A6 The reasonable man adapts himself to the =A6
bidulock@openss7.org    =A6 world; the unreasonable one persists in  =A6
http://www.openss7.org/ =A6 trying  to adapt the  world  to himself. =A6
                        =A6 Therefore  all  progress  depends on the =A6
                        =A6 unreasonable man. -- George Bernard Shaw =A6

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



From exim@www1.ietf.org  Wed Jul 23 18:45:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA06617
	for <sigtran-archive@odin.ietf.org>; Wed, 23 Jul 2003 18:45:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fSM1-0003pA-Rg
	for sigtran-archive@odin.ietf.org; Wed, 23 Jul 2003 18:45:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6NMj1Wh014693
	for sigtran-archive@odin.ietf.org; Wed, 23 Jul 2003 18:45:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fSM0-0003oj-UN; Wed, 23 Jul 2003 18:45:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fSL6-0003oO-SR
	for sigtran@optimus.ietf.org; Wed, 23 Jul 2003 18:44:04 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA06592
	for <sigtran@ietf.org>; Wed, 23 Jul 2003 18:43:59 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fSL3-0001Rt-00
	for sigtran@ietf.org; Wed, 23 Jul 2003 18:44:01 -0400
Received: from auds951.usa.alcatel.com ([143.209.238.80])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fSKs-0001Rg-00
	for sigtran@ietf.org; Wed, 23 Jul 2003 18:43:50 -0400
Received: from alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6NMgtwA025661;
	Wed, 23 Jul 2003 17:42:55 -0500 (CDT)
Message-ID: <3F1F0F6F.9020005@alcatel.com>
Date: Wed, 23 Jul 2003 17:42:55 -0500
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
User-Agent: Mozilla/5.0 (X11; U; SunOS sun4u; en-US; rv:1.0.1) Gecko/20020920 Netscape/7.0
X-Accept-Language: en-us, en
MIME-Version: 1.0
CC: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - TTC
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030714040938.B13583@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

My JT-Q703 (that reference needs to be changed, too) has both in Figure 
13-6/JT-Q703 (sheet 4 of 4).

The "Retrieval Request and FSNC" erases the messages in the RTB up to 
where FSN = FSNC. The The "Retrieval Request" skips that part and just 
retrieves everything in the RTB.

I should update the text to note that both kinds of requests are 
allowed, instead of saying that all of the messages are always retrieved.

tomG.

Brian F. G. Bidulock wrote:
> M2PA-09 says:
> 
>   Note: For the Japanese version of MTP defined in [JT-Q704], MTP3
>   retrieves both unsent and unacknowleged messages for transmission on
>   the alternate link(s). In this version of MTP, if M2PA receives a
>   Retrieval Request and FSNC request with no FSNC value, or with an
>   invalid FSNC, then M2PA SHALL retrieve from its buffers and deliver to
>   MTP3 in order:
> 
>      (a) any transmitted but unacknowledged User Data messages.
> 
>      (b) any untransmitted User Data messages.
> 
> JT-Q703 does not have a Retrieval Request and FSNC request.  Only a
> Retrieval Request.
> 
> Change M2PA to say:
> 
>   Note: For the Japanese version of MTP defined in [JT-Q704], MTP3
>   retrieves both unsent and unacknowleged messages for transmission on
>   the alternate link(s). In this version of MTP, if M2PA receives a
>   Retrieval Request then M2PA SHALL retrieve from its buffers and
>   deliver to MTP3 in order:
> 
>      (a) any transmitted but unacknowledged User Data messages.
> 
>      (b) any untransmitted User Data messages.
> 
> --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.  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 exim@www1.ietf.org  Wed Jul 23 19:00:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA06969
	for <sigtran-archive@odin.ietf.org>; Wed, 23 Jul 2003 19:00:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fSaY-0004Jh-Fv
	for sigtran-archive@odin.ietf.org; Wed, 23 Jul 2003 19:00:02 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6NN02qA016576
	for sigtran-archive@odin.ietf.org; Wed, 23 Jul 2003 19:00:02 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fSaX-0004JE-EN; Wed, 23 Jul 2003 19:00:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fSZf-0004Id-Py
	for sigtran@optimus.ietf.org; Wed, 23 Jul 2003 18:59:07 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA06923
	for <sigtran@ietf.org>; Wed, 23 Jul 2003 18:59:02 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fSZc-0001XF-00
	for sigtran@ietf.org; Wed, 23 Jul 2003 18:59:04 -0400
Received: from auds951.usa.alcatel.com ([143.209.238.80])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fSZR-0001X0-00
	for sigtran@ietf.org; Wed, 23 Jul 2003 18:58:53 -0400
Received: from alcatel.com (localhost [127.0.0.1])
	by auds951.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6NMwIwA027124;
	Wed, 23 Jul 2003 17:58:18 -0500 (CDT)
Message-ID: <3F1F130A.4030606@alcatel.com>
Date: Wed, 23 Jul 2003 17:58:18 -0500
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
User-Agent: Mozilla/5.0 (X11; U; SunOS sun4u; en-US; rv:1.0.1) Gecko/20020920 Netscape/7.0
X-Accept-Language: en-us, en
MIME-Version: 1.0
CC: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - JT-Q703 nit pick
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030714042516.D13583@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

Brian,

Correct. The values are defined in JT-Q704, but the format is in 
JT-Q703. The new text says:

=====
It is not necessary to put the message length in the LI octet as in
MTP2. The LI octet is included because the two spare bits in the LI
octet are used by MTP3 in at least one national version of SS7 to
carry MTP3 information. For example, the Japanese TTC standard uses
these spare bits as an MTP3 Message Priority field. See [JT-Q703],
section 2.3.10 "Priority Indicator (PRI:Priority Indicator)". The
values of this field are in [JT-Q704], section 14 "Common
Characteristics of message signal unit formats", section 14.2 (A)
"Priority Indicator (PRI)". For versions of MTP that do not use these
two bits, the entire octet is spare.
=====

The new reference says:
=====
   [JT-Q703]
        TTC, "Message Transfer Part Signalling Link," TTC Standard
        JT-Q703, Telecommunication Technology Committee (TTC), version
        3 (April 27, 1994).
=====



Brian F. G. Bidulock wrote:
> M2PA-09 says:
> 
>  It is not necessary to put the message length in the LI octet as in
>  MTP2. The LI octet is included because the two spare bits in the LI
>  octet are used by MTP3 in at least one national version of SS7 to
>  carry MTP3 information. For example, the Japanese TTC standard uses
>  these spare bits as an MTP3 Message Priority field. See [JT-Q704],
>  section 14 "Common Characteristics of message signal unit formats",
>  section 14.2 (A) "Priority Indicator (PRI)". For versions of MTP that
>  do not use these two bits, the entire octet is spare.
> 
> and
> 
>    PRI - Priority used only in national MTP defined in [JT-Q704].
>          Spare for other MTP versions.
> 
> These bits are defined in JT-Q703, not JT-Q704.
> 
> Change JT-Q704 to JT-Q703 above and add [JT-Q703] to the references.
> 
> --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.  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 exim@www1.ietf.org  Wed Jul 23 19:02:26 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA07020
	for <sigtran-archive@odin.ietf.org>; Wed, 23 Jul 2003 19:02:26 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fScT-0004Rj-Kl
	for sigtran-archive@odin.ietf.org; Wed, 23 Jul 2003 19:02:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6NN21MW017062
	for sigtran-archive@odin.ietf.org; Wed, 23 Jul 2003 19:02:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fScT-0004R5-Cy; Wed, 23 Jul 2003 19:02:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fSbj-0004Q9-FH
	for sigtran@optimus.ietf.org; Wed, 23 Jul 2003 19:01:15 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA06995
	for <sigtran@ietf.org>; Wed, 23 Jul 2003 19:01:10 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fSbg-0001Yl-00
	for sigtran@ietf.org; Wed, 23 Jul 2003 19:01:12 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[HT6QvHEOy9GSnErxOT5iHpj18aNlF4x+])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fSbQ-0001Yf-00
	for sigtran@ietf.org; Wed, 23 Jul 2003 19:00:56 -0400
Received: from ns.pigworks.openss7.net (IDENT:/P8gyWZ1wtHic5RMfPE3SXt2Ce9k6Jfc@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6NN0k104820;
	Wed, 23 Jul 2003 17:00:46 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6NN0jH27329;
	Wed, 23 Jul 2003 17:00:45 -0600
Date: Wed, 23 Jul 2003 17:00:45 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tom George <Tom.George@alcatel.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - TTC
Message-ID: <20030723170045.A27159@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tom George <Tom.George@alcatel.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030714040938.B13583@openss7.org> <3F1F0F6F.9020005@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: <3F1F0F6F.9020005@alcatel.com>; from Tom.George@alcatel.com on Wed, Jul 23, 2003 at 05:42:55PM -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>

Tom,

The way the text sounded to me was that it was saying that when a
"Retrieval Request and FSNC" was received without an FSNC rather
than when a plain "Retrieval Request" was received...

Just a nit, really.

--brian

Tom George wrote:                                                                      (Wed, 23 Jul 2003 17:42:55)
> My JT-Q703 (that reference needs to be changed, too) has both in Figure 
> 13-6/JT-Q703 (sheet 4 of 4).
> 
> The "Retrieval Request and FSNC" erases the messages in the RTB up to 
> where FSN = FSNC. The The "Retrieval Request" skips that part and just 
> retrieves everything in the RTB.
> 
> I should update the text to note that both kinds of requests are 
> allowed, instead of saying that all of the messages are always retrieved.
> 
> tomG.
> 
> Brian F. G. Bidulock wrote:
> > M2PA-09 says:
> > 
> >   Note: For the Japanese version of MTP defined in [JT-Q704], MTP3
> >   retrieves both unsent and unacknowleged messages for transmission on
> >   the alternate link(s). In this version of MTP, if M2PA receives a
> >   Retrieval Request and FSNC request with no FSNC value, or with an
> >   invalid FSNC, then M2PA SHALL retrieve from its buffers and deliver to
> >   MTP3 in order:
> > 
> >      (a) any transmitted but unacknowledged User Data messages.
> > 
> >      (b) any untransmitted User Data messages.
> > 
> > JT-Q703 does not have a Retrieval Request and FSNC request.  Only a
> > Retrieval Request.
> > 
> > Change M2PA to say:
> > 
> >   Note: For the Japanese version of MTP defined in [JT-Q704], MTP3
> >   retrieves both unsent and unacknowleged messages for transmission on
> >   the alternate link(s). In this version of MTP, if M2PA receives a
> >   Retrieval Request then M2PA SHALL retrieve from its buffers and
> >   deliver to MTP3 in order:
> > 
> >      (a) any transmitted but unacknowledged User Data messages.
> > 
> >      (b) any untransmitted User Data messages.
> > 
> > --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.  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 exim@www1.ietf.org  Wed Jul 23 22:00:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA10191
	for <sigtran-archive@odin.ietf.org>; Wed, 23 Jul 2003 22:00:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fVOl-0003VM-Iw
	for sigtran-archive@odin.ietf.org; Wed, 23 Jul 2003 22:00:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6O203TS013466
	for sigtran-archive@odin.ietf.org; Wed, 23 Jul 2003 22:00:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fVOj-0003Uv-LG; Wed, 23 Jul 2003 22:00:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fVO4-0003Ty-Au
	for sigtran@optimus.ietf.org; Wed, 23 Jul 2003 21:59:20 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA10100
	for <sigtran@ietf.org>; Wed, 23 Jul 2003 21:59:16 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fVO1-0002MZ-00
	for sigtran@ietf.org; Wed, 23 Jul 2003 21:59:17 -0400
Received: from ottgw.pt.com ([206.191.2.130] helo=pt.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19fVNq-0002Lu-00
	for sigtran@ietf.org; Wed, 23 Jul 2003 21:59:06 -0400
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id VAA10188;
	Wed, 23 Jul 2003 21:58:02 -0400
Message-ID: <3F1F3D2F.300@pt.com>
Date: Wed, 23 Jul 2003 21:58:07 -0400
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: Gaiser Franz <franz.gaiser@siemens.com>
CC: "'sigtran@ietf.org'" <sigtran@ietf.org>, tom.george@alcatel.com,
        "Brian F. G. Bidulock" <bidulock@openss7.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - LPO
References: <99522F0FD7EFD611A47D0002A58EDAB7C84A86@mchh2a8e.mchh.siemens.de>
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 Franz,

Before answering your question I'll ask another: do we really need to 
resynchronize sequence numbers in M2PA, or can we just mandate that the 
sender continues to increment the FSN on every transmitted MSU?
- This seems simpler than resynchronizing.
- Since M2PA does not retransmit, the motivation for resynchronization 
doesn't seem to be present as it does in MTP2.
- Some of the examples below give examples where resequencing can go 
wrong.  Although these situations are unlikely, the consequences are 
unpleasant.

Now for my best answer to your question: Option (2), but if there are no 
unacknowledged messages then the BSN should be set to the FSN of the 
last acknowledged message.
Note also that if the receiver of the LS PR uses the BSN to free 
messages in the RTB (as you indicated below) then they have to be extra 
careful when validating BSNs in ensuing Data messages that may have been 
sent before the LS PR. I don't think this is a good idea.

Some potential issues with resynchronizing sequence numbers:

Note that the sender of LS PR cannot know if the receiver will use the 
BSN to resynchronize sequence numbers or not, since it is the MTP in RPO 
that decides when RPO is "long duration" via the MTP3 T1 timer.
This leads to race conditions in which one side flushes and the other 
does not, though they are unlikely.  For example, the following could 
occur if an MSU was delayed longer than MTP3 T1 on the IP network:


     MTP3                  M2PA                        M2PA
       |                     |                          |
       |                     |<--Data (FSN 0)-----------|
       |                     |                          |
       |                     |                     +----|
       |----LPO------------->|                    /     |
       |                     |----------LS PO----/----->|
       |                     |                  /       |
       |                     |                 /        |<--Flush
       |----LPR + Continue-->|                /         |
       |                     |----LS PR------/--------->| resync FSN
       |                     |    (BSN 0)   /           |
       |                     |             /            |
       |                     |<--Data-----+             |
       |                     |   (FSN 1)                |
       |                     |                          |
       |             Discard |<--Data-------------------|
       |           (bad FSN) |   (FSN 0)                |

The following example could occur if the side in LPO flushes while the 
side in RPO does not (perhaps they have different T1 timer settings), 
and if the side in LPO sends a BSN different from the last received by M2PA:

     MTP3                  M2PA                        M2PA
       |                     |                          |
       |                     |<--Data (FSN 0)-----------|
       |                     |                          |
       |                     |                     +----|
       |----LPO------------->|                    /     |
       |                     |----------LS PO----/----->|
       |                     |                  /       |
       |                     |                 /        |
       |               To RB |<--Data---------+         |
       |                     |   (FSN 1)                |
       |                     |                          |
       |----LPR + Flush----->|                          |
       |                     |----LS PR---------------->|
       |                     |    (BSN 0)               |<- Continue,
       |                     |                          |   don't resync FSN
       |                     |                          |
       |                     |                          |
       |             Discard |<--Data (FSN 2)-----------|
       |    (expected FSN 1) |                          |
       |                     |                          |
       |             Discard |<--Data (FSN 3)-----------|
       |    (expected FSN 1) |                          |

I think this possible condition provides reasonable incentive to always 
set the BSN in the LS Processor Recovered to the last message received 
from the M2PA peer.  In that case, there doesn't seem to be any point in 
resynching sequence numbers, just let the peer increment the FSN on 
every MSU.

In summary, these examples are unlikely but why complicate things? 
 Unless there is a good reason to keep the MTP2 resynching procedures 
lets get rid of them in favour of the simpler approach: the sender 
always increments the FSN by 1 when sending MSUs.

Thoughts?
Andrew

Gaiser Franz wrote:

>Hi All,
>
>implementing Processor Outage Handling of M2PA draft 09 the following questions came up:
>
>The Processor Outage chapter 4.1.4 says
>
>When the local processor outage conditions ends, M2PA shall send a Link
>Status Processor Recovered message to its peer. This message is used
>to signal in the end of the processor outage condition, instead of an
>MSU or FISU as in MTP2. The BSN value in this message is used by the 
>peer to resynchronize its sequence numbers, if this is required by
>MTP2.
>
>What does M2PA draft 09 with BSN in this context mean, is it (assume, MTP3 has send Continue Command)
>
>(1) the FSN of the last received message before LPO or 
>(2) the FSN of the last message in the incoming messages that were queued and unacknowledged
>during the processor outage (with that the remote side can release the messages in its RTB) ? 
>
>
>
>thanks
>best regards, Franz 
>
>
>
>
>
>_______________________________________________
>Sigtran mailing list
>Sigtran@ietf.org
>https://www1.ietf.org/mailman/listinfo/sigtran
>
>  
>

-- 
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 exim@www1.ietf.org  Thu Jul 24 06:11:05 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA01158
	for <sigtran-archive@odin.ietf.org>; Thu, 24 Jul 2003 06:11:05 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fd3Y-0001Zc-1R
	for sigtran-archive@odin.ietf.org; Thu, 24 Jul 2003 06:10:40 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6OAAesS006042
	for sigtran-archive@odin.ietf.org; Thu, 24 Jul 2003 06:10:40 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fd2v-0001Yy-Bf; Thu, 24 Jul 2003 06:10:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fd23-0001Sg-Cv
	for sigtran@optimus.ietf.org; Thu, 24 Jul 2003 06:09:07 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA01120
	for <sigtran@ietf.org>; Thu, 24 Jul 2003 06:09:01 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fd1z-0004yD-00
	for sigtran@ietf.org; Thu, 24 Jul 2003 06:09:03 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[3zuCl16DnQPNYHq8w/Wy4V3SrqVfomkJ])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fd1j-0004y9-00
	for sigtran@ietf.org; Thu, 24 Jul 2003 06:08:47 -0400
Received: from ns.pigworks.openss7.net (IDENT:679EHK04v9OgcMfeK6KjQvA6/mvyQABI@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6OA8K114087;
	Thu, 24 Jul 2003 04:08:21 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6OA8JD03786;
	Thu, 24 Jul 2003 04:08:19 -0600
Date: Thu, 24 Jul 2003 04:08:19 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Andrew Booth <abooth@pt.com>
Cc: Gaiser Franz <franz.gaiser@siemens.com>,
        "'sigtran@ietf.org'" <sigtran@ietf.org>, tom.george@alcatel.com
Subject: Re: [Sigtran] M2PA Last Call Comment - LPO
Message-ID: <20030724040819.A30771@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Andrew Booth <abooth@pt.com>,
	Gaiser Franz <franz.gaiser@siemens.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>, tom.george@alcatel.com
References: <99522F0FD7EFD611A47D0002A58EDAB7C84A86@mchh2a8e.mchh.siemens.de> <3F1F3D2F.300@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: <3F1F3D2F.300@pt.com>; from abooth@pt.com on Wed, Jul 23, 2003 at 09:58:07PM -0400
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,

Please see comments below...

Andrew Booth wrote:                                                                 (Wed, 23 Jul 2003 21:58:07)
> Hi Franz,
> 
> Before answering your question I'll ask another: do we really need to 
> resynchronize sequence numbers in M2PA, or can we just mandate that the 
> sender continues to increment the FSN on every transmitted MSU?
> - This seems simpler than resynchronizing.
> - Since M2PA does not retransmit, the motivation for resynchronization 
> doesn't seem to be present as it does in MTP2.
> - Some of the examples below give examples where resequencing can go 
> wrong.  Although these situations are unlikely, the consequences are 
> unpleasant.
> 
> Now for my best answer to your question: Option (2), but if there are no 
> unacknowledged messages then the BSN should be set to the FSN of the 
> last acknowledged message.
> Note also that if the receiver of the LS PR uses the BSN to free 
> messages in the RTB (as you indicated below) then they have to be extra 
> careful when validating BSNs in ensuing Data messages that may have been 
> sent before the LS PR. I don't think this is a good idea.
> 
> Some potential issues with resynchronizing sequence numbers:
> 
> Note that the sender of LS PR cannot know if the receiver will use the 
> BSN to resynchronize sequence numbers or not, since it is the MTP in RPO 
> that decides when RPO is "long duration" via the MTP3 T1 timer.
> This leads to race conditions in which one side flushes and the other 
> does not, though they are unlikely.  For example, the following could 
> occur if an MSU was delayed longer than MTP3 T1 on the IP network:
> 
> 
>      MTP3                  M2PA                        M2PA
>        |                     |                          |
>        |                     |<--Data (FSN 0)-----------|
>        |                     |                          |
>        |                     |                     +----|
>        |----LPO------------->|                    /     |
>        |                     |----------LS PO----/----->|
>        |                     |                  /       |
>        |                     |                 /        |<--Flush
>        |----LPR + Continue-->|                /         |
>        |                     |----LS PR------/--------->| resync FSN
>        |                     |    (BSN 0)   /           |
>        |                     |             /            |
>        |                     |<--Data-----+             |
>        |                     |   (FSN 1)                |
>        |                     |                          |
>        |             Discard |<--Data-------------------|
>        |           (bad FSN) |   (FSN 0)                |

This is not really probable.  This would mean that the MTP RPO timer
expires when the LPO timer hasn't (a very narrow window) and that the
Data (FSN 1) was delayed longer than the RPO timer (which shouldn't
really happen).

Set your SCTP parameters (or lifetimes) so that messages not
acknowledged within the RPO timeout will drop the SCTP association and
you won't have this problem.

Do you see?  The "long duration" of the procedure (T1 or about 1.0 to
2.0 seconds) is used so that old messages stored in buffers are not sent
later.  RTB clear is performed immediately before changeover (after T1
expiry).  One doesn't want a recovering processor to transfer old
messages after changeover has occured or a lot of missequencing is going
to happen in the SS7 network.

Now, if you have SCTP parameters set to allow messages to be buffered IN
THE NETWORK for longer than T1, you have a real big problem: both the
time-controlled changeover and controlled rerouting procedures will
result in missequenced messages, even without processor outage.

So, if you have a problem above, becuase your SCTP will not drop the
link before delaying a data message longer than T1, you have the greater
problem that every time-controlled changeover or controlled rerouting
procedure for those M2PA links are going to missequence messagses.

Therefore, you need to tighten your SCTP parameters.  If you cannot
(e.g. the outher end is a really bad implemetnation), you really need to
bring the link down when the duplicate FSN arrives.

I think you have an error here too: BSN 0 acknowledges the receipt of
data with FSN 0.  After the resync, the next data sent should have BSN
1.


> 
> The following example could occur if the side in LPO flushes while the 
> side in RPO does not (perhaps they have different T1 timer settings), 
> and if the side in LPO sends a BSN different from the last received by M2PA:
> 
>      MTP3                  M2PA                        M2PA
>        |                     |                          |
>        |                     |<--Data (FSN 0)-----------|
>        |                     |                          |
>        |                     |                     +----|
>        |----LPO------------->|                    /     |
>        |                     |----------LS PO----/----->|
>        |                     |                  /       |
>        |                     |                 /        |
>        |               To RB |<--Data---------+         |
>        |                     |   (FSN 1)                |
>        |                     |                          |
>        |----LPR + Flush----->|                          |
>        |                     |----LS PR---------------->|
>        |                     |    (BSN 0)               |<- Continue,
>        |                     |                          |   don't resync FSN
>        |                     |                          |
>        |                     |                          |
>        |             Discard |<--Data (FSN 2)-----------|
>        |    (expected FSN 1) |                          |
>        |                     |                          |
>        |             Discard |<--Data (FSN 3)-----------|
>        |    (expected FSN 1) |                          |

Again, having one side flush without the other is a small time window,
however, why are you sending LS PR (BSN 0)?  I think you should send
LS PR (BSN 1) in this instance.  Then there is no problem.

MTP2 would send BSN 0 here, but it is also discarding messages with the
expectation of retransmission.  M2PA buffers but does not acknowledge
messages.  The purpose of delaying acknowledgement is so that the peer
M2PA keeps these messages in the retransmission buffer for changeover.
(Remember than some variants collect from the RTB instead of clearing
it, unlike ANSI).  But, for M2PA, the BSN should be incremented when
the data is buffered to accomplish what happens here.

So this one is not a problem either.

--brian

> 
> I think this possible condition provides reasonable incentive to always 
> set the BSN in the LS Processor Recovered to the last message received 
> from the M2PA peer.  In that case, there doesn't seem to be any point in 
> resynching sequence numbers, just let the peer increment the FSN on 
> every MSU.
> 
> In summary, these examples are unlikely but why complicate things? 
>  Unless there is a good reason to keep the MTP2 resynching procedures 
> lets get rid of them in favour of the simpler approach: the sender 
> always increments the FSN by 1 when sending MSUs.
> 
> Thoughts?
> Andrew
> 
> Gaiser Franz wrote:
> 
> >Hi All,
> >
> >implementing Processor Outage Handling of M2PA draft 09 the following questions came up:
> >
> >The Processor Outage chapter 4.1.4 says
> >
> >When the local processor outage conditions ends, M2PA shall send a Link
> >Status Processor Recovered message to its peer. This message is used
> >to signal in the end of the processor outage condition, instead of an
> >MSU or FISU as in MTP2. The BSN value in this message is used by the 
> >peer to resynchronize its sequence numbers, if this is required by
> >MTP2.
> >
> >What does M2PA draft 09 with BSN in this context mean, is it (assume, MTP3 has send Continue Command)
> >
> >(1) the FSN of the last received message before LPO or 
> >(2) the FSN of the last message in the incoming messages that were queued and unacknowledged
> >during the processor outage (with that the remote side can release the messages in its RTB) ? 
> >
> >
> >
> >thanks
> >best regards, Franz 
> >
> >
> >
> >
> >
> >_______________________________________________
> >Sigtran mailing list
> >Sigtran@ietf.org
> >https://www1.ietf.org/mailman/listinfo/sigtran
> >
> >  
> >
> 
> -- 
> 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.
> 

-- 
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 exim@www1.ietf.org  Thu Jul 24 11:04:12 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA10099
	for <sigtran-archive@odin.ietf.org>; Thu, 24 Jul 2003 11:04:12 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fhdD-0002cb-EX
	for sigtran-archive@odin.ietf.org; Thu, 24 Jul 2003 11:03:47 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6OF3lQT010073
	for sigtran-archive@odin.ietf.org; Thu, 24 Jul 2003 11:03:47 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fhcT-0002aS-Ec; Thu, 24 Jul 2003 11:03:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fhbz-0002ZT-CW
	for sigtran@optimus.ietf.org; Thu, 24 Jul 2003 11:02:32 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA10042
	for <sigtran@ietf.org>; Thu, 24 Jul 2003 11:02:25 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fhbw-0006i2-00
	for sigtran@ietf.org; Thu, 24 Jul 2003 11:02:28 -0400
Received: from ottgw.pt.com ([206.191.2.130] helo=pt.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19fhbl-0006hZ-00
	for sigtran@ietf.org; Thu, 24 Jul 2003 11:02:18 -0400
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id LAA17632;
	Thu, 24 Jul 2003 11:01:00 -0400
Message-ID: <3F1FF4AE.2060305@pt.com>
Date: Thu, 24 Jul 2003 11:01:02 -0400
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: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - LPO
References: <99522F0FD7EFD611A47D0002A58EDAB7C84A86@mchh2a8e.mchh.siemens.de> <3F1F3D2F.300@pt.com> <20030724040819.A30771@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

Hi Brian,

I don't think synchronization of sequence numbers after PO is necessary 
or desirable in M2PA.  Can you see any advantage to it?

I'd like to highlight this point by bringing up your comment after 
example 2 below: in order to avoid potential race conditions the side 
sending LS Processor Recovered should set the BSN to the last FSN 
received by M2PA.  By setting the BSN in this way we expect is that the 
peer will continue sending FSNs "in sequence" i.e. that the 
resynchronization will not actually change the next FSN the peer will 
send.  If that's the behaviour we want, and I think it is, then I 
propose that we specify it directly rather than carry over an 
unnecessary procedure from MTP2.

More comments below.

Brian F. G. Bidulock wrote:

>Andrew,
>
>Please see comments below...
>
>Andrew Booth wrote:                                                                 (Wed, 23 Jul 2003 21:58:07)
>  
>
>>Hi Franz,
>>
>>Before answering your question I'll ask another: do we really need to 
>>resynchronize sequence numbers in M2PA, or can we just mandate that the 
>>sender continues to increment the FSN on every transmitted MSU?
>>- This seems simpler than resynchronizing.
>>- Since M2PA does not retransmit, the motivation for resynchronization 
>>doesn't seem to be present as it does in MTP2.
>>- Some of the examples below give examples where resequencing can go 
>>wrong.  Although these situations are unlikely, the consequences are 
>>unpleasant.
>>
>>Now for my best answer to your question: Option (2), but if there are no 
>>unacknowledged messages then the BSN should be set to the FSN of the 
>>last acknowledged message.
>>Note also that if the receiver of the LS PR uses the BSN to free 
>>messages in the RTB (as you indicated below) then they have to be extra 
>>careful when validating BSNs in ensuing Data messages that may have been 
>>sent before the LS PR. I don't think this is a good idea.
>>
>>Some potential issues with resynchronizing sequence numbers:
>>
>>Note that the sender of LS PR cannot know if the receiver will use the 
>>BSN to resynchronize sequence numbers or not, since it is the MTP in RPO 
>>that decides when RPO is "long duration" via the MTP3 T1 timer.
>>This leads to race conditions in which one side flushes and the other 
>>does not, though they are unlikely.  For example, the following could 
>>occur if an MSU was delayed longer than MTP3 T1 on the IP network:
>>
>>
>>     MTP3                  M2PA                        M2PA
>>       |                     |                          |
>>       |                     |<--Data (FSN 0)-----------|
>>       |                     |                          |
>>       |                     |                     +----|
>>       |----LPO------------->|                    /     |
>>       |                     |----------LS PO----/----->|
>>       |                     |                  /       |
>>       |                     |                 /        |<--Flush
>>       |----LPR + Continue-->|                /         |
>>       |                     |----LS PR------/--------->| resync FSN
>>       |                     |    (BSN 0)   /           |
>>       |                     |             /            |
>>       |                     |<--Data-----+             |
>>       |                     |   (FSN 1)                |
>>       |                     |                          |
>>       |             Discard |<--Data-------------------|
>>       |           (bad FSN) |   (FSN 0)                |
>>    
>>
>
>This is not really probable.  This would mean that the MTP RPO timer
>expires when the LPO timer hasn't (a very narrow window) and that the
>Data (FSN 1) was delayed longer than the RPO timer (which shouldn't
>really happen).
>
Agreed, it is unlikely.  If SCTP parameters are set correctly it might 
even be impossible.

>
>Set your SCTP parameters (or lifetimes) so that messages not
>acknowledged within the RPO timeout will drop the SCTP association and
>you won't have this problem.
>
>Do you see?  The "long duration" of the procedure (T1 or about 1.0 to
>2.0 seconds) is used so that old messages stored in buffers are not sent
>later.  RTB clear is performed immediately before changeover (after T1
>expiry).  One doesn't want a recovering processor to transfer old
>messages after changeover has occured or a lot of missequencing is going
>to happen in the SS7 network.
>
Yes I understand, though I think MTP3 T1 is 0.5 to 1.2 seconds.

>
>Now, if you have SCTP parameters set to allow messages to be buffered IN
>THE NETWORK for longer than T1, you have a real big problem: both the
>time-controlled changeover and controlled rerouting procedures will
>result in missequenced messages, even without processor outage.
>
Agreed.  Should this be stated explicitly in M2PA?

>
>So, if you have a problem above, becuase your SCTP will not drop the
>link before delaying a data message longer than T1, you have the greater
>problem that every time-controlled changeover or controlled rerouting
>procedure for those M2PA links are going to missequence messagses.
>
>Therefore, you need to tighten your SCTP parameters.  If you cannot
>(e.g. the outher end is a really bad implemetnation), you really need to
>bring the link down when the duplicate FSN arrives.
>
>I think you have an error here too: BSN 0 acknowledges the receipt of
>data with FSN 0.  After the resync, the next data sent should have BSN
>1.
>
Agreed, the last FSN should be 1 (still invalid).

>
>
>  
>
>>The following example could occur if the side in LPO flushes while the 
>>side in RPO does not (perhaps they have different T1 timer settings), 
>>and if the side in LPO sends a BSN different from the last received by M2PA:
>>
>>     MTP3                  M2PA                        M2PA
>>       |                     |                          |
>>       |                     |<--Data (FSN 0)-----------|
>>       |                     |                          |
>>       |                     |                     +----|
>>       |----LPO------------->|                    /     |
>>       |                     |----------LS PO----/----->|
>>       |                     |                  /       |
>>       |                     |                 /        |
>>       |               To RB |<--Data---------+         |
>>       |                     |   (FSN 1)                |
>>       |                     |                          |
>>       |----LPR + Flush----->|                          |
>>       |                     |----LS PR---------------->|
>>       |                     |    (BSN 0)               |<- Continue,
>>       |                     |                          |   don't resync FSN
>>       |                     |                          |
>>       |                     |                          |
>>       |             Discard |<--Data (FSN 2)-----------|
>>       |    (expected FSN 1) |                          |
>>       |                     |                          |
>>       |             Discard |<--Data (FSN 3)-----------|
>>       |    (expected FSN 1) |                          |
>>    
>>
>
>Again, having one side flush without the other is a small time window,
>however, why are you sending LS PR (BSN 0)?  I think you should send
>LS PR (BSN 1) in this instance.  Then there is no problem.
>
I agree this is unlikely, but I don't think the time window is 
necessarily that small: if one side sets T1 to 0.5s and the other to 
1.2s you have 0.7s to hit this.

>
>MTP2 would send BSN 0 here, but it is also discarding messages with the
>expectation of retransmission.  M2PA buffers but does not acknowledge
>messages.  The purpose of delaying acknowledgement is so that the peer
>M2PA keeps these messages in the retransmission buffer for changeover.
>(Remember than some variants collect from the RTB instead of clearing
>it, unlike ANSI).  But, for M2PA, the BSN should be incremented when
>the data is buffered to accomplish what happens here.
>
>So this one is not a problem either.
>
Agreed, it's not a problem, but it does highlight the point I made 
above: we should set the BSN so that when resynchronizing sequence 
numbers the peer doesn't actually have to change anything.  It seems 
strange to have an exchange of BSN values for this purpose.

Thoughts?
Andrew

>
>--brian
>
>  
>
>>I think this possible condition provides reasonable incentive to always 
>>set the BSN in the LS Processor Recovered to the last message received 
>>from the M2PA peer.  In that case, there doesn't seem to be any point in 
>>resynching sequence numbers, just let the peer increment the FSN on 
>>every MSU.
>>
>>In summary, these examples are unlikely but why complicate things? 
>> Unless there is a good reason to keep the MTP2 resynching procedures 
>>lets get rid of them in favour of the simpler approach: the sender 
>>always increments the FSN by 1 when sending MSUs.
>>
>>Thoughts?
>>Andrew
>>
>>Gaiser Franz wrote:
>>
>>    
>>
>>>Hi All,
>>>
>>>implementing Processor Outage Handling of M2PA draft 09 the following questions came up:
>>>
>>>The Processor Outage chapter 4.1.4 says
>>>
>>>When the local processor outage conditions ends, M2PA shall send a Link
>>>Status Processor Recovered message to its peer. This message is used
>>>to signal in the end of the processor outage condition, instead of an
>>>MSU or FISU as in MTP2. The BSN value in this message is used by the 
>>>peer to resynchronize its sequence numbers, if this is required by
>>>MTP2.
>>>
>>>What does M2PA draft 09 with BSN in this context mean, is it (assume, MTP3 has send Continue Command)
>>>
>>>(1) the FSN of the last received message before LPO or 
>>>(2) the FSN of the last message in the incoming messages that were queued and unacknowledged
>>>during the processor outage (with that the remote side can release the messages in its RTB) ? 
>>>
>>>
>>>
>>>thanks
>>>best regards, Franz 
>>>
>>>
>>>
>>>
>>>
>>>_______________________________________________
>>>Sigtran mailing list
>>>Sigtran@ietf.org
>>>https://www1.ietf.org/mailman/listinfo/sigtran
>>>
>>> 
>>>
>>>      
>>>
>>-- 
>>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.
>>
>>    
>>
>
>  
>

-- 
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 exim@www1.ietf.org  Thu Jul 24 12:13:14 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA12206
	for <sigtran-archive@odin.ietf.org>; Thu, 24 Jul 2003 12:13:13 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fii0-00068z-NF
	for sigtran-archive@odin.ietf.org; Thu, 24 Jul 2003 12:12:48 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6OGCm3q023618
	for sigtran-archive@odin.ietf.org; Thu, 24 Jul 2003 12:12:48 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fihG-00068b-25; Thu, 24 Jul 2003 12:12:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19figw-00068N-GE
	for sigtran@optimus.ietf.org; Thu, 24 Jul 2003 12:11:42 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA12173
	for <sigtran@ietf.org>; Thu, 24 Jul 2003 12:11:37 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19figv-0007BQ-00
	for sigtran@ietf.org; Thu, 24 Jul 2003 12:11:41 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[308+0bPBYXkQgBthsK5KSoEYbNSJZ/8N])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fige-0007BJ-00
	for sigtran@ietf.org; Thu, 24 Jul 2003 12:11:24 -0400
Received: from ns.pigworks.openss7.net (IDENT:LHV1hJNMIbkMoPBOTOa7augyfFgA8rpq@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6OGAp120016;
	Thu, 24 Jul 2003 10:10:51 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6OGAoM09403;
	Thu, 24 Jul 2003 10:10:50 -0600
Date: Thu, 24 Jul 2003 10:10:50 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Andrew Booth <abooth@pt.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - LPO
Message-ID: <20030724101050.C8372@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Andrew Booth <abooth@pt.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <99522F0FD7EFD611A47D0002A58EDAB7C84A86@mchh2a8e.mchh.siemens.de> <3F1F3D2F.300@pt.com> <20030724040819.A30771@openss7.org> <3F1FF4AE.2060305@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: <3F1FF4AE.2060305@pt.com>; from abooth@pt.com on Thu, Jul 24, 2003 at 11:01:02AM -0400
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,

On the first point, performance related issues are outside the scope
of the protocol.

On the second point, nowhere in M2PA-09 does it say to stop incrementing
BSN.  Just to withhold acknowledgements and buffer during PO.  The
peer is welcome, as it says in M2PA-09, to resync off of the BSN in the
LPR.  The fact that it doesn't change the FSN for ANSI does not mean that
it is not performing the resynchronization.

Look at the resync wording in M2PA-09 again.

I don't think it need changing.

--brian

Andrew Booth wrote:                                                                 (Thu, 24 Jul 2003 11:01:02)
> Hi Brian,
> 
> I don't think synchronization of sequence numbers after PO is necessary 
> or desirable in M2PA.  Can you see any advantage to it?
> 
> I'd like to highlight this point by bringing up your comment after 
> example 2 below: in order to avoid potential race conditions the side 
> sending LS Processor Recovered should set the BSN to the last FSN 
> received by M2PA.  By setting the BSN in this way we expect is that the 
> peer will continue sending FSNs "in sequence" i.e. that the 
> resynchronization will not actually change the next FSN the peer will 
> send.  If that's the behaviour we want, and I think it is, then I 
> propose that we specify it directly rather than carry over an 
> unnecessary procedure from MTP2.
> 
> More comments below.
> 
> Brian F. G. Bidulock wrote:
> 
> >Andrew,
> >
> >Please see comments below...
> >
> >Andrew Booth wrote:                                                                 (Wed, 23 Jul 2003 21:58:07)
> >  
> >
> >>Hi Franz,
> >>
> >>Before answering your question I'll ask another: do we really need to 
> >>resynchronize sequence numbers in M2PA, or can we just mandate that the 
> >>sender continues to increment the FSN on every transmitted MSU?
> >>- This seems simpler than resynchronizing.
> >>- Since M2PA does not retransmit, the motivation for resynchronization 
> >>doesn't seem to be present as it does in MTP2.
> >>- Some of the examples below give examples where resequencing can go 
> >>wrong.  Although these situations are unlikely, the consequences are 
> >>unpleasant.
> >>
> >>Now for my best answer to your question: Option (2), but if there are no 
> >>unacknowledged messages then the BSN should be set to the FSN of the 
> >>last acknowledged message.
> >>Note also that if the receiver of the LS PR uses the BSN to free 
> >>messages in the RTB (as you indicated below) then they have to be extra 
> >>careful when validating BSNs in ensuing Data messages that may have been 
> >>sent before the LS PR. I don't think this is a good idea.
> >>
> >>Some potential issues with resynchronizing sequence numbers:
> >>
> >>Note that the sender of LS PR cannot know if the receiver will use the 
> >>BSN to resynchronize sequence numbers or not, since it is the MTP in RPO 
> >>that decides when RPO is "long duration" via the MTP3 T1 timer.
> >>This leads to race conditions in which one side flushes and the other 
> >>does not, though they are unlikely.  For example, the following could 
> >>occur if an MSU was delayed longer than MTP3 T1 on the IP network:
> >>
> >>
> >>     MTP3                  M2PA                        M2PA
> >>       |                     |                          |
> >>       |                     |<--Data (FSN 0)-----------|
> >>       |                     |                          |
> >>       |                     |                     +----|
> >>       |----LPO------------->|                    /     |
> >>       |                     |----------LS PO----/----->|
> >>       |                     |                  /       |
> >>       |                     |                 /        |<--Flush
> >>       |----LPR + Continue-->|                /         |
> >>       |                     |----LS PR------/--------->| resync FSN
> >>       |                     |    (BSN 0)   /           |
> >>       |                     |             /            |
> >>       |                     |<--Data-----+             |
> >>       |                     |   (FSN 1)                |
> >>       |                     |                          |
> >>       |             Discard |<--Data-------------------|
> >>       |           (bad FSN) |   (FSN 0)                |
> >>    
> >>
> >
> >This is not really probable.  This would mean that the MTP RPO timer
> >expires when the LPO timer hasn't (a very narrow window) and that the
> >Data (FSN 1) was delayed longer than the RPO timer (which shouldn't
> >really happen).
> >
> Agreed, it is unlikely.  If SCTP parameters are set correctly it might 
> even be impossible.
> 
> >
> >Set your SCTP parameters (or lifetimes) so that messages not
> >acknowledged within the RPO timeout will drop the SCTP association and
> >you won't have this problem.
> >
> >Do you see?  The "long duration" of the procedure (T1 or about 1.0 to
> >2.0 seconds) is used so that old messages stored in buffers are not sent
> >later.  RTB clear is performed immediately before changeover (after T1
> >expiry).  One doesn't want a recovering processor to transfer old
> >messages after changeover has occured or a lot of missequencing is going
> >to happen in the SS7 network.
> >
> Yes I understand, though I think MTP3 T1 is 0.5 to 1.2 seconds.
> 
> >
> >Now, if you have SCTP parameters set to allow messages to be buffered IN
> >THE NETWORK for longer than T1, you have a real big problem: both the
> >time-controlled changeover and controlled rerouting procedures will
> >result in missequenced messages, even without processor outage.
> >
> Agreed.  Should this be stated explicitly in M2PA?
> 
> >
> >So, if you have a problem above, becuase your SCTP will not drop the
> >link before delaying a data message longer than T1, you have the greater
> >problem that every time-controlled changeover or controlled rerouting
> >procedure for those M2PA links are going to missequence messagses.
> >
> >Therefore, you need to tighten your SCTP parameters.  If you cannot
> >(e.g. the outher end is a really bad implemetnation), you really need to
> >bring the link down when the duplicate FSN arrives.
> >
> >I think you have an error here too: BSN 0 acknowledges the receipt of
> >data with FSN 0.  After the resync, the next data sent should have BSN
> >1.
> >
> Agreed, the last FSN should be 1 (still invalid).
> 
> >
> >
> >  
> >
> >>The following example could occur if the side in LPO flushes while the 
> >>side in RPO does not (perhaps they have different T1 timer settings), 
> >>and if the side in LPO sends a BSN different from the last received by M2PA:
> >>
> >>     MTP3                  M2PA                        M2PA
> >>       |                     |                          |
> >>       |                     |<--Data (FSN 0)-----------|
> >>       |                     |                          |
> >>       |                     |                     +----|
> >>       |----LPO------------->|                    /     |
> >>       |                     |----------LS PO----/----->|
> >>       |                     |                  /       |
> >>       |                     |                 /        |
> >>       |               To RB |<--Data---------+         |
> >>       |                     |   (FSN 1)                |
> >>       |                     |                          |
> >>       |----LPR + Flush----->|                          |
> >>       |                     |----LS PR---------------->|
> >>       |                     |    (BSN 0)               |<- Continue,
> >>       |                     |                          |   don't resync FSN
> >>       |                     |                          |
> >>       |                     |                          |
> >>       |             Discard |<--Data (FSN 2)-----------|
> >>       |    (expected FSN 1) |                          |
> >>       |                     |                          |
> >>       |             Discard |<--Data (FSN 3)-----------|
> >>       |    (expected FSN 1) |                          |
> >>    
> >>
> >
> >Again, having one side flush without the other is a small time window,
> >however, why are you sending LS PR (BSN 0)?  I think you should send
> >LS PR (BSN 1) in this instance.  Then there is no problem.
> >
> I agree this is unlikely, but I don't think the time window is 
> necessarily that small: if one side sets T1 to 0.5s and the other to 
> 1.2s you have 0.7s to hit this.
> 
> >
> >MTP2 would send BSN 0 here, but it is also discarding messages with the
> >expectation of retransmission.  M2PA buffers but does not acknowledge
> >messages.  The purpose of delaying acknowledgement is so that the peer
> >M2PA keeps these messages in the retransmission buffer for changeover.
> >(Remember than some variants collect from the RTB instead of clearing
> >it, unlike ANSI).  But, for M2PA, the BSN should be incremented when
> >the data is buffered to accomplish what happens here.
> >
> >So this one is not a problem either.
> >
> Agreed, it's not a problem, but it does highlight the point I made 
> above: we should set the BSN so that when resynchronizing sequence 
> numbers the peer doesn't actually have to change anything.  It seems 
> strange to have an exchange of BSN values for this purpose.
> 
> Thoughts?
> Andrew
> 
> >
> >--brian
> >
> >  
> >
> >>I think this possible condition provides reasonable incentive to always 
> >>set the BSN in the LS Processor Recovered to the last message received 
> >>from the M2PA peer.  In that case, there doesn't seem to be any point in 
> >>resynching sequence numbers, just let the peer increment the FSN on 
> >>every MSU.
> >>
> >>In summary, these examples are unlikely but why complicate things? 
> >> Unless there is a good reason to keep the MTP2 resynching procedures 
> >>lets get rid of them in favour of the simpler approach: the sender 
> >>always increments the FSN by 1 when sending MSUs.
> >>
> >>Thoughts?
> >>Andrew
> >>
> >>Gaiser Franz wrote:
> >>
> >>    
> >>
> >>>Hi All,
> >>>
> >>>implementing Processor Outage Handling of M2PA draft 09 the following questions came up:
> >>>
> >>>The Processor Outage chapter 4.1.4 says
> >>>
> >>>When the local processor outage conditions ends, M2PA shall send a Link
> >>>Status Processor Recovered message to its peer. This message is used
> >>>to signal in the end of the processor outage condition, instead of an
> >>>MSU or FISU as in MTP2. The BSN value in this message is used by the 
> >>>peer to resynchronize its sequence numbers, if this is required by
> >>>MTP2.
> >>>
> >>>What does M2PA draft 09 with BSN in this context mean, is it (assume, MTP3 has send Continue Command)
> >>>
> >>>(1) the FSN of the last received message before LPO or 
> >>>(2) the FSN of the last message in the incoming messages that were queued and unacknowledged
> >>>during the processor outage (with that the remote side can release the messages in its RTB) ? 
> >>>
> >>>
> >>>
> >>>thanks
> >>>best regards, Franz 
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>_______________________________________________
> >>>Sigtran mailing list
> >>>Sigtran@ietf.org
> >>>https://www1.ietf.org/mailman/listinfo/sigtran
> >>>
> >>> 
> >>>
> >>>      
> >>>
> >>-- 
> >>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.
> >>
> >>    
> >>
> >
> >  
> >
> 
> -- 
> 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.
> 

-- 
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 exim@www1.ietf.org  Thu Jul 24 15:04:17 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA22464
	for <sigtran-archive@odin.ietf.org>; Thu, 24 Jul 2003 15:04:17 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19flNX-0006d9-PS
	for sigtran-archive@odin.ietf.org; Thu, 24 Jul 2003 15:03:52 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6OJ3pLX025487
	for sigtran-archive@odin.ietf.org; Thu, 24 Jul 2003 15:03:51 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19flMj-0006aA-IF; Thu, 24 Jul 2003 15:03:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19flMJ-0006Zo-R9
	for sigtran@optimus.ietf.org; Thu, 24 Jul 2003 15:02:35 -0400
Received: from auds952.usa.alcatel.com (auds952.usa.alcatel.com [143.209.238.7])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA22070
	for <sigtran@ietf.org>; Thu, 24 Jul 2003 15:02:27 -0400 (EDT)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6OIx5B3013679;
	Thu, 24 Jul 2003 13:59:05 -0500 (CDT)
Message-ID: <3F202C78.8060109@alcatel.com>
Date: Thu, 24 Jul 2003 13:59:04 -0500
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
User-Agent: Mozilla/5.0 (X11; U; SunOS sun4u; en-US; rv:1.0.1) Gecko/20020920 Netscape/7.0
X-Accept-Language: en-us, en
MIME-Version: 1.0
CC: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - TTC
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030714040938.B13583@openss7.org> <3F1F0F6F.9020005@alcatel.com> <20030723170045.A27159@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

Brian,

It is a nit, but the way I had the text washed over the fact that there 
were two different requests. Here is my suggested new wording:

=====
The Japanese TTC version of MTP defined in [JT-Q703] has a Retrieval
Request (as well as Retrieval Request and FSNC). The Retrieval Request
allows MTP3 to retrieve both unsent and unacknowleged messages for
transmission on the alternate link(s). In this version of MTP, if M2PA
receives a Retrieval Request, then M2PA SHALL retrieve from its
buffers and deliver to MTP3 in order:

    (a) any transmitted but unacknowledged User Data messages.

    (b) any untransmitted User Data messages.
=====

tomG.

Brian F. G. Bidulock wrote:
> Tom,
> 
> The way the text sounded to me was that it was saying that when a
> "Retrieval Request and FSNC" was received without an FSNC rather
> than when a plain "Retrieval Request" was received...
> 
> Just a nit, really.
> 
> --brian
> 
> Tom George wrote:                                                                      (Wed, 23 Jul 2003 17:42:55)
> 
>>My JT-Q703 (that reference needs to be changed, too) has both in Figure 
>>13-6/JT-Q703 (sheet 4 of 4).
>>
>>The "Retrieval Request and FSNC" erases the messages in the RTB up to 
>>where FSN = FSNC. The The "Retrieval Request" skips that part and just 
>>retrieves everything in the RTB.
>>
>>I should update the text to note that both kinds of requests are 
>>allowed, instead of saying that all of the messages are always retrieved.
>>
>>tomG.
>>
>>Brian F. G. Bidulock wrote:
>>
>>>M2PA-09 says:
>>>
>>>  Note: For the Japanese version of MTP defined in [JT-Q704], MTP3
>>>  retrieves both unsent and unacknowleged messages for transmission on
>>>  the alternate link(s). In this version of MTP, if M2PA receives a
>>>  Retrieval Request and FSNC request with no FSNC value, or with an
>>>  invalid FSNC, then M2PA SHALL retrieve from its buffers and deliver to
>>>  MTP3 in order:
>>>
>>>     (a) any transmitted but unacknowledged User Data messages.
>>>
>>>     (b) any untransmitted User Data messages.
>>>
>>>JT-Q703 does not have a Retrieval Request and FSNC request.  Only a
>>>Retrieval Request.
>>>
>>>Change M2PA to say:
>>>
>>>  Note: For the Japanese version of MTP defined in [JT-Q704], MTP3
>>>  retrieves both unsent and unacknowleged messages for transmission on
>>>  the alternate link(s). In this version of MTP, if M2PA receives a
>>>  Retrieval Request then M2PA SHALL retrieve from its buffers and
>>>  deliver to MTP3 in order:
>>>
>>>     (a) any transmitted but unacknowledged User Data messages.
>>>
>>>     (b) any untransmitted User Data messages.
>>>
>>>--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.  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.  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 exim@www1.ietf.org  Thu Jul 24 15:31:11 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA26330
	for <sigtran-archive@odin.ietf.org>; Thu, 24 Jul 2003 15:31:11 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19flnY-0008K2-U3
	for sigtran-archive@odin.ietf.org; Thu, 24 Jul 2003 15:30:45 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6OJUi0q031990
	for sigtran-archive@odin.ietf.org; Thu, 24 Jul 2003 15:30:44 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19flmr-0008FE-Kk; Thu, 24 Jul 2003 15:30:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19flme-0008EY-7k
	for sigtran@optimus.ietf.org; Thu, 24 Jul 2003 15:29:48 -0400
Received: from auds952.usa.alcatel.com (auds952.usa.alcatel.com [143.209.238.7])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA26228
	for <sigtran@ietf.org>; Thu, 24 Jul 2003 15:29:43 -0400 (EDT)
Received: from alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6OJTEB3017676;
	Thu, 24 Jul 2003 14:29:15 -0500 (CDT)
Message-ID: <3F20338A.9020000@alcatel.com>
Date: Thu, 24 Jul 2003 14:29:14 -0500
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
User-Agent: Mozilla/5.0 (X11; U; SunOS sun4u; en-US; rv:1.0.1) Gecko/20020920 Netscape/7.0
X-Accept-Language: en-us, en
MIME-Version: 1.0
CC: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - LI nit pick
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030714045624.F13583@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

Brian,

Your explanation is better. I will change my wording to yours.

tomG.

Brian F. G. Bidulock wrote:
> M2PA-09 says:
> 
>   The Data field contains the following fields of the MTP Message Signal
>   Unit (MSU):
>   
>      - Length Indicator (LI), including the two undefined bits between
>        the SIO and LI fields.
>      - Service Information Octet (SIO)
>      - Signaling Information Field (SIF)
>   
>   The MTP MSU is described in Q.703 [Q.703], section 2.2 "Signal Unit
>   Format", and T1.111.3 [T1.111] section 2.2 "Signal Unit Format".
>   
>   M2PA SHALL NOT add padding to the MTP3 message. 
>   
>   Note that the Data field SHALL NOT contain other components of the MTP
>   MSU format:
>   
>      - Flag
>      - Backward Sequence Number (BSN)
>      - Backward Indicator Bit (BIB)
>      - Forward Sequence Number (FSN)
>      - Forward Indicator Bit (FIB)
>      - Check bits (CK) 
> 
>   The Data field SHALL be transmitted in the byte order as defined by
>   MTP3. 
> 
>   It is not necessary to put the message length in the LI octet as in
>   MTP2. The LI octet is included because the two spare bits in the LI
>   octet are used by MTP3 in at least one national version of SS7 to
>   carry MTP3 information. For example, the Japanese TTC standard uses
>   these spare bits as an MTP3 Message Priority field. See [JT-Q704],
>   section 14 "Common Characteristics of message signal unit formats",
>   section 14.2 (A) "Priority Indicator (PRI)". For versions of MTP that
>   do not use these two bits, the entire octet is spare.
> 
>   Therefore in M2PA the format of the LI octet is:
> 
>       0               
>       0 1 2 3 4 5 6 7 
>      +-+-+-+-+-+-+-+-+
>      |PRI|   spare   | (followed by SIO, SIF) 
>      +-+-+-+-+-+-+-+-+
>   
>      PRI - Priority used only in national MTP defined in [JT-Q704].
>            Spare for other MTP versions.
> 
>   Since the LI octet is not used for a message length, there is no need
>   to support the expanded LI field in Q.703 [Q.703] Annex A. Therefore
>   the LI field in M2PA is always one octet.
> 
> ------
> 
> LI is not an octet, it is a field.  One that happens to not be included
> in the M2PA Data field.
> 
> Change to:
> 
> ------
> 
>   The Data field contains the following fields of the MTP Message Signal
>   Unit (MSU):
>   
>      - the Message Priority field (PRI)
>      - Service Information Octet (SIO)
>      - Signaling Information Field (SIF)
>   
>   The MTP MSU is described in Q.703 [Q.703], section 2.2 "Signal Unit
>   Format", and T1.111.3 [T1.111] section 2.2 "Signal Unit Format".
>   The Japanese TTC standard uses the PRI field as an MTP3 Message
>   Priority field [JT-Q703].  For versions of MTP that do not use these
>   two bits, the entire first octet of the Data field is spare.
> 
>   The format of the first octet of the Data field is:
> 
>       0               
>       0 1 2 3 4 5 6 7 
>      +-+-+-+-+-+-+-+-+
>      |PRI|   spare   | (followed by SIO, SIF) 
>      +-+-+-+-+-+-+-+-+
>   
>      PRI - Priority used only in national MTP defined in [JT-Q703].
>            Spare for other MTP versions.
> 
>   Note that the Data field SHALL NOT contain other components of the MTP
>   MSU format:
>   
>      - Flag
>      - Backward Sequence Number (BSN)
>      - Backward Indicator Bit (BIB)
>      - Forward Sequence Number (FSN)
>      - Forward Indicator Bit (FIB)
>      - Length Indicator (LI)
>      - Check bits (CK) 
> 
>   The Data field SHALL be transmitted in the byte order as defined by
>   MTP3. 
> 
>   M2PA SHALL NOT add padding to the MTP3 message. 
> 
> 
> ------
> 
> This could be less confusing that saying on the onset that the LI field
> is part of the Data field and then going on to explain how it isn't.
> 
> --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.  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 exim@www1.ietf.org  Thu Jul 24 15:47:08 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA27006
	for <sigtran-archive@odin.ietf.org>; Thu, 24 Jul 2003 15:47:08 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fm30-0000de-0r
	for sigtran-archive@odin.ietf.org; Thu, 24 Jul 2003 15:46:42 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6OJkfGY002448
	for sigtran-archive@odin.ietf.org; Thu, 24 Jul 2003 15:46:42 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fm2L-0000ZS-CO; Thu, 24 Jul 2003 15:46:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fm28-0000ZC-E8
	for sigtran@optimus.ietf.org; Thu, 24 Jul 2003 15:45:48 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA26960
	for <sigtran@ietf.org>; Thu, 24 Jul 2003 15:45:43 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fm26-0001Vq-00
	for sigtran@ietf.org; Thu, 24 Jul 2003 15:45:46 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[GDcOiQBs/46EUXycj23vbeeYsEYI30Mv])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fm1q-0001VY-00
	for sigtran@ietf.org; Thu, 24 Jul 2003 15:45:30 -0400
Received: from ns.pigworks.openss7.net (IDENT:5Yra5E/CrslBCfwnf4TOyIXgMSgViDRJ@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6OJj3W23449;
	Thu, 24 Jul 2003 13:45:03 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6OJj2E12924;
	Thu, 24 Jul 2003 13:45:02 -0600
Date: Thu, 24 Jul 2003 13:45:01 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tom George <Tom.George@alcatel.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - TTC
Message-ID: <20030724134501.C11627@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tom George <Tom.George@alcatel.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030714040938.B13583@openss7.org> <3F1F0F6F.9020005@alcatel.com> <20030723170045.A27159@openss7.org> <3F202C78.8060109@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: <3F202C78.8060109@alcatel.com>; from Tom.George@alcatel.com on Thu, Jul 24, 2003 at 01:59:04PM -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>

Tom,

Sounds good.

--brian

Tom George wrote:                                                                      (Thu, 24 Jul 2003 13:59:04)
> Brian,
> 
> It is a nit, but the way I had the text washed over the fact that there 
> were two different requests. Here is my suggested new wording:
> 
> =====
> The Japanese TTC version of MTP defined in [JT-Q703] has a Retrieval
> Request (as well as Retrieval Request and FSNC). The Retrieval Request
> allows MTP3 to retrieve both unsent and unacknowleged messages for
> transmission on the alternate link(s). In this version of MTP, if M2PA
> receives a Retrieval Request, then M2PA SHALL retrieve from its
> buffers and deliver to MTP3 in order:
> 
>     (a) any transmitted but unacknowledged User Data messages.
> 
>     (b) any untransmitted User Data messages.
> =====
> 
> tomG.
> 
> Brian F. G. Bidulock wrote:
> > Tom,
> > 
> > The way the text sounded to me was that it was saying that when a
> > "Retrieval Request and FSNC" was received without an FSNC rather
> > than when a plain "Retrieval Request" was received...
> > 
> > Just a nit, really.
> > 
> > --brian
> > 
> > Tom George wrote:                                                                      (Wed, 23 Jul 2003 17:42:55)
> > 
> >>My JT-Q703 (that reference needs to be changed, too) has both in Figure 
> >>13-6/JT-Q703 (sheet 4 of 4).
> >>
> >>The "Retrieval Request and FSNC" erases the messages in the RTB up to 
> >>where FSN = FSNC. The The "Retrieval Request" skips that part and just 
> >>retrieves everything in the RTB.
> >>
> >>I should update the text to note that both kinds of requests are 
> >>allowed, instead of saying that all of the messages are always retrieved.
> >>
> >>tomG.
> >>
> >>Brian F. G. Bidulock wrote:
> >>
> >>>M2PA-09 says:
> >>>
> >>>  Note: For the Japanese version of MTP defined in [JT-Q704], MTP3
> >>>  retrieves both unsent and unacknowleged messages for transmission on
> >>>  the alternate link(s). In this version of MTP, if M2PA receives a
> >>>  Retrieval Request and FSNC request with no FSNC value, or with an
> >>>  invalid FSNC, then M2PA SHALL retrieve from its buffers and deliver to
> >>>  MTP3 in order:
> >>>
> >>>     (a) any transmitted but unacknowledged User Data messages.
> >>>
> >>>     (b) any untransmitted User Data messages.
> >>>
> >>>JT-Q703 does not have a Retrieval Request and FSNC request.  Only a
> >>>Retrieval Request.
> >>>
> >>>Change M2PA to say:
> >>>
> >>>  Note: For the Japanese version of MTP defined in [JT-Q704], MTP3
> >>>  retrieves both unsent and unacknowleged messages for transmission on
> >>>  the alternate link(s). In this version of MTP, if M2PA receives a
> >>>  Retrieval Request then M2PA SHALL retrieve from its buffers and
> >>>  deliver to MTP3 in order:
> >>>
> >>>     (a) any transmitted but unacknowledged User Data messages.
> >>>
> >>>     (b) any untransmitted User Data messages.
> >>>
> >>>--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.  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.  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 exim@www1.ietf.org  Thu Jul 24 15:51:09 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA27105
	for <sigtran-archive@odin.ietf.org>; Thu, 24 Jul 2003 15:51:09 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fm6s-0000pJ-TD
	for sigtran-archive@odin.ietf.org; Thu, 24 Jul 2003 15:50:43 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6OJogIq003173
	for sigtran-archive@odin.ietf.org; Thu, 24 Jul 2003 15:50:42 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fm6D-0000kI-Mi; Thu, 24 Jul 2003 15:50:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fm5T-0000iZ-Ra
	for sigtran@optimus.ietf.org; Thu, 24 Jul 2003 15:49:16 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA27057
	for <sigtran@ietf.org>; Thu, 24 Jul 2003 15:49:11 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fm5S-0001XE-00
	for sigtran@ietf.org; Thu, 24 Jul 2003 15:49:14 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[bLVswCzq0OX+MQUQUpHXOnTvilHohAvy])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fm5C-0001X8-00
	for sigtran@ietf.org; Thu, 24 Jul 2003 15:48:58 -0400
Received: from ns.pigworks.openss7.net (IDENT:9PEbzIWel19lpxXyxoIYY7HDpiSEGLzh@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6OJmjW23479;
	Thu, 24 Jul 2003 13:48:45 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6OJmjA12953;
	Thu, 24 Jul 2003 13:48:45 -0600
Date: Thu, 24 Jul 2003 13:48:45 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Andrew Booth <abooth@pt.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>, tom.george@alcatel.com
Subject: Re: [Sigtran] M2PA Last Call Comment - RPO
Message-ID: <20030724134845.D11627@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Andrew Booth <abooth@pt.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>, tom.george@alcatel.com
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030712170744.A11077@openss7.org> <3F15C7F2.6070704@pt.com> <20030716191143.F6725@openss7.org> <3F18328D.1080706@pt.com> <20030718120053.C14760@openss7.org> <3F184959.2040103@pt.com> <20030718133957.A16698@openss7.org> <3F1867CB.6060108@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: <3F1867CB.6060108@pt.com>; from abooth@pt.com on Fri, Jul 18, 2003 at 05:34:03PM -0400
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,

No that doesn't sound right.  The side in LPO must be able to drain it's TB to
the peer and the peer needs to acknowledge messsages received after LPO but
before LPR if its MTP3 is able to do so.  That's what started this thread off
in the first place.

--brian

Andrew Booth wrote:                                                                 (Fri, 18 Jul 2003 17:34:03)
> Hi Brian,
> 
> 
> Brian F. G. Bidulock wrote:
> 
> >[snip]
> >
> >Sure.  Just as long as the sender will generate acknowledgements in RPO
> >I don't have a problem.  Also, T6 does not always run when in RPO.
> >Did you want to propose some alternate text to the text I proposed below?
> >
> >--brian
> >
> >  
> >
> What do you think of the following proposal, does it sum up both our points?
> As a miscellaneous note I changed the LS PR to LS PO in the final 
> paragraph; the motivation is that I think the FSN in the LSPO and LSPR 
> should be the same (the side in LPO shouldn't be sending data) and 
> because the Flush could arrive before the LS PR in which case the old 
> text is clearly wrong. I'm not sure though if that change was the right 
> one. Please let me know what you think.
> 
> Thanks,
> Andrew
> 
> 4.1.4 Processor Outage
> 
> The Link Status Processor Outage message replaces the SIPO message of
> MTP2. Unlike MTP2, the message SHOULD NOT be transmitted
> continuously. M2PA SHALL send a Link Status Processor Outage message
> to its peer at the beginning of a processor outage condition where
> MTP2 would send SIPO. M2PA MAY send additional Link Status Processor
> Outage messages as long as that condition persists.
> 
> When the local processor outage condition ends, M2PA SHALL send a Link
> Status Processor Recovered message to its peer. This message is used
> to signal in the end of the processor outage condition, instead of an
> MSU or FISU as in MTP2. The BSN value in this message is used by the
> peer to resynchronize its sequence numbers, if this is required by
> MTP2.
> 
> Some MTP3 implementations may be able to accept received User Data
> while in Remote Processor Outage while others may not. While there
> is a remote processor outage condition:
> 
> (a) M2PA SHOULD continue to acknowledge User Data Message received and
> accepted by MTP3 regardless of the remote processor outage.
> 
> (b) Any User Data messages received from the peer after
> the Link Status Processor Outage that may not be passed to MTP3
> MUST NOT be acknowledged and MUST be buffered.
> 
> If M2PA receives a Flush command from MTP3, M2PA SHALL
> discard the incoming messages that were queued and unacknowledged during the
> processor outage condition. Furthermore, M2PA SHALL discard messages in the
> transmit and retransmit queues as required by MTP2. Similar behaviour
> is appropriate in response to corresponding primitives in other SS7 variants
> such as the ANSI Clear RTB primitive.
> 
> Note that because the Link Status and User Data messages are sent on 
> separate
> streams, it is possible for a User Data message sent after the Link Status
> Processor Recovered message to arrive before it. Therefore, when flushing
> buffers, M2PA must take care not to flush messages sent after the remote
> processor recovered. M2PA SHOULD NOT flush User Data messages with FSN 
> greater
> than the FSN in the last received Link Status Processor Outage message.
> Note that similar conditions could occur if User Data is received after Link
> Status Processor Recovered but before the Flush command from MTP3. Similar
> behaviour is appropriate in response to comparable primitives in other 
> SS7 variants,
> such as the ANSI Clear RTB primitive.
> 

-- 
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 exim@www1.ietf.org  Thu Jul 24 16:31:08 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA29031
	for <sigtran-archive@odin.ietf.org>; Thu, 24 Jul 2003 16:31:08 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fmja-00034s-Rf
	for sigtran-archive@odin.ietf.org; Thu, 24 Jul 2003 16:30:42 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6OKUg4L011828
	for sigtran-archive@odin.ietf.org; Thu, 24 Jul 2003 16:30:42 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fmiv-000333-0X; Thu, 24 Jul 2003 16:30:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fmig-00032J-0r
	for sigtran@optimus.ietf.org; Thu, 24 Jul 2003 16:29:46 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA28957
	for <sigtran@ietf.org>; Thu, 24 Jul 2003 16:29:40 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fmid-0001ws-00
	for sigtran@ietf.org; Thu, 24 Jul 2003 16:29:43 -0400
Received: from auds952.usa.alcatel.com ([143.209.238.7])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fmiT-0001w1-00
	for sigtran@ietf.org; Thu, 24 Jul 2003 16:29:33 -0400
Received: from alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6OKSTB3027879;
	Thu, 24 Jul 2003 15:28:29 -0500 (CDT)
Message-ID: <3F20416D.7060305@alcatel.com>
Date: Thu, 24 Jul 2003 15:28:29 -0500
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
User-Agent: Mozilla/5.0 (X11; U; SunOS sun4u; en-US; rv:1.0.1) Gecko/20020920 Netscape/7.0
X-Accept-Language: en-us, en
MIME-Version: 1.0
CC: IETF sigtran <sigtran@ietf.org>
Subject: Re: FW: [Sigtran] M2PA Last Call Comment
References: <D73559B38D51D4119CB800508B9527B505AC1965@mchh266e.mchh.siemens.de> <20030714034015.A13132@openss7.org> <3F172056.9080909@alcatel.com> <20030717175456.B29221@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

That does not seem like a good situation. I see the point now. Instead 
of your suggested text:

   The proving period MAY be ommitted if this is allowed by the
   applicable standard (e.g., [Q.2140]); however, the same standard MUST
   be followed on both sides of the M2PA link.

would it be better to make a more general statement, in order to avoid 
this kind of problem that might exist in other procedures? Maybe I could 
  add to the 2nd paragraph in section 4:

=====
Except as modified in this document, M2PA SHOULD follow the
requirements of the applicable MTP2 specification. These may include
[Q.703] or [T1.111]. The same standard MUST be followed on both ends of 
the M2PA link.
=====

tomG.



Brian F. G. Bidulock wrote:
> Tom,
> 
> Q.703 does not give the option to talk to a Q.2140 link.  But it
> appears that M2PA allows a Q.703-state-machine-based M2PA link to
> talk to a Q.2140-state-machine-based M2PA link.  That is were the
> problem ensues.
> 
> --brian
> 
> Tom George wrote:                                                                      (Thu, 17 Jul 2003 17:16:54)
> 
>>Brian,
>>
>>In the non-IP realm, is it possible for a Q.703 to talk to a Q.2140? 
>>What would be done in the scenario you described?
>>
>>tomG.
>>
>>Brian F. G. Bidulock wrote:
>>
>>>Hanns Juergen,
>>>
>>>Consider that one side follows Q.703 and requires proving while the
>>>other side follows Q.2140 and foregoes proving.  The following will
>>>occur:
>>
   [...]


-- 
* 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.  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 exim@www1.ietf.org  Fri Jul 25 01:19:50 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA12110
	for <sigtran-archive@odin.ietf.org>; Fri, 25 Jul 2003 01:19:50 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fuzD-0007yB-Up
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 01:19:24 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6P5JNiZ030635
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 01:19:23 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fuxt-0007tf-IE; Fri, 25 Jul 2003 01:18:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fuww-0007qz-20
	for sigtran@optimus.ietf.org; Fri, 25 Jul 2003 01:17:02 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA12069
	for <sigtran@ietf.org>; Fri, 25 Jul 2003 01:16:58 -0400 (EDT)
From: kakaul@hss.hns.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fuwt-00053I-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 01:16:59 -0400
Received: from [202.54.26.202] (helo=hindon.hss.co.in)
	by ietf-mx with esmtp (Exim 4.12)
	id 19fuwh-00052V-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 01:16:47 -0400
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h6P5BQs15579;
	Fri, 25 Jul 2003 10:41:27 +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 ESMTP id h6P5FDB26518;
	Fri, 25 Jul 2003 10:45:13 +0530 (IST)
Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
To: bidulock@openss7.org
Cc: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.5  September 22, 2000
Message-ID: <OFC835ED71.D1087FE3-ON65256D6E.001C305D@hss.hns.com>
Date: Fri, 25 Jul 2003 10:39:47 +0530
X-MIMETrack: Serialize by Router on Sandesh/HSS(Release 6.0|September 26, 2002) at 25/07/2003
 10:38:29 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,
             You have not replied to my earlier mail




Kamesh Kaul/HSS@HSS@ietf.org on 07/18/2003 08:38:57 PM

Sent by:    sigtran-admin@ietf.org


To:    bidulock@openss7.org
cc:    sigtran@ietf.org

Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18






Brian,
   I agree with running congestion timer per affected route-set(point code)
basis as you have mentioned. But now the point is that in SS7, SEP runs the
RCT congestion supervisory timers after sending RCT message with current
congestion status of the affected route-set so that if the TFC for that
message is not received congestion abatement procedure will be supervised
by timers.So it means that in SS7, SEP RCT message checks the clearance of
path (for congestion) to affected point code and then monitors the
congestion abatement by timers.But in ASP there is no RCT message so we in
a way have to depend on DAUD message per SG to simulate RCT message and
monitor congestion by one timer per affected point code .But for that we
need to know to which SG DAUD message should be sent. SS7, SEP need not
bother about sending RCT per route because it sends RCT message as if it
has got dummy message from user part for affected point code with current
routeset congestion set in priority.For ASP to figure out to which SG DAUD
has to be send either congestion status or flag(signifying occurance or
abatement) has to be maintained per SG.Please comment.

Do you agree with not striking of timer mention in IG.

Regards
Kamesh Kaul
www.hssworld.com
Hughes Software Systems




"Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/18/2003
01:10:22 PM

Please respond to bidulock@openss7.org

Sent by:    sigtran-admin@ietf.org


To:    Kamesh Kaul/HSS@HSS
cc:    sigtran@ietf.org

Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18


kakaul,

I don't think the ASP needs to do all that.  Instead of DAUD, it can simply
lower the congestion status of the routeset after a time interval and
indicate
abatement to the National MTP-User.  The next message sent by the MTP-User
will illicit SCON if the abatement was premature (without sending any
message
on the SS7 network) and the timer can be reset.  Only one timer per
routeset
is necessary.

In the multiple SG scenario SCON is comparable to TFC, not MTP-STATUS.
SCON(0) is not necessary in the multiple SG scenario, the ASP can abate
after a time interval and rely on responsive SCON from the SG.  If no
SCON is delivered, then abatement to no congestion sticks.

Also, understand that the Nation MTP (e.g. ANSI) does not indicate
abatement
in all situations: only when the congestion is remote.  When the congestion
is a result of congestion of a signalling link, abatement is not indicated.
(e.g. see ANSI T1.111.4/2000 SDLs).

--brian


kakaul@hss.hns.com wrote:
(Fri, 18 Jul 2003 10:15:37)
>
>
>
>
> Brian ,
>    I agree with you that in SS7 SEP congestion status is only maintained
> per routeset basis and not per route basis, beacause due to RCT procedure
> there is no need of maintaining congestion status  on per route basis.Now
> putting ASP as SEP you have no RCT procedure defined, but as you say SG's
> connected with ASP perform  RCT on the behalf of ASP.Assuming affected
> point code x-y-z is reachable through two SG's SG1 and SG2.Now ASP
receives
> SCON(1) from SG1 and update the point code (X-Y-Z) congestion status  to
1
> and MTP3 users are informed with MTP-STATUS with congestion status 1.Now
> ASP has to start DAUD procedure for affected point code x-y-z per SG and
> ASP will continue to send DAUD messages with affected point code X-Y-Z to
> SG1 untill ASP is sure that point code(x-y-z) congestion status  is 0
> (no-congestion) through SG1.Mean- while SG1 might be doing RCT procedure
on
> the behalf of ASP for affected point code X-Y-Z.There are actually two
> options for ASP to stop sending DAUD messages to SG1  with affected point
> code x-y-z. Either ASP maintains congestion status per SG route or mark
the
> SG1 route for affected point code x-y-z congested by using some flag so
> that ASP can know when to stop DAUD procedure for affected point code per
> SG.Roueset (x-y-z) congestion status is just there to indicate MTP-STATUS
> primitives.ASP will AUDIT affected point code per SG till congestion
> abates, for that ASP has to know when to start and stop sending DAUD
> messages for affected congested point code to a given SG which in a way
> implies ASP has to maintain either congestion status or flag (indicationg
> occurance and abatement of congestion).Now for abatement of affected
point
> code congestion status there are two approaches either SCON(0) received
> from SG for affected point code or timer supervision because once you
have
> indicated congestion status greater than 0 to MTP3 user you have to
> indicate congestion abatement with congestion status 0 to MTP3 user this
> statement is only true if the network appearance associated with affected
> point code is of national option with congestion priority.I have taken
two
> SG's in my above example to explain that affected point code  X-Y-Z
> congestion status can get updated to 3 on receiving SCON(3) from SG2.Now
> you have to send DAUD messages to both SG1 and SG2 till abatement of
> congestion occures. Please comment
>
> Regards
> Kamesh Kaul
> www.hssworld.com
> Hughes Software Systems
>
>
>
>
>
> "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/17/2003
> 06:46:00 AM
>
> Please respond to bidulock@openss7.org
>
> Sent by:    sigtran-admin@ietf.org
>
>
> To:    Kamesh Kaul/HSS@HSS
> cc:    sigtran@ietf.org
>
> Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
>
>
> kakaul,
>
> You are confusing congestion at an STP due to the conestion of signalling
> links on a route with routeset congestion.  Although the congestion of a
> signalling link can trigger sending TFC, the route upon which the TFC is
> received has nothing to do with congestion.
>
> Put an SEP where the ASP is and tell me where in the SS7 specs it says
> that a TFC received from an STP is correlated to the route upon which it
> was received.  A: it doesn't.
>
> The passages you quote correspond to transmit congestion on the SCTP
> associations between the ASP and the SG.  (SCON is not involved there.)
>
> --brian
>
> kakaul@hss.hns.com wrote:
> (Wed, 16 Jul 2003 21:57:11)
> >
> >
> >
> >
> > Brian,
> >    Comments inline
> >
> >
> >
> >
> > "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/16/2003
> > 12:41:40 PM
> >
> > Please respond to bidulock@openss7.org
> >
> > Sent by:    sigtran-admin@ietf.org
> >
> >
> > To:    Kamesh Kaul/HSS@HSS
> > cc:    sigtran@ietf.org
> >
> > Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> >
> >
> > kakaul,
> >
> > An SS7 SEP at the position of the ASP, where the SGs are STPs would
> > not maintain congestion state on a per-route (per-SG) basis.  So
> > what makes you think that the ASP needs to?
> > [Kamesh]It would because route-set congestion status will change
> depending
> > on the derived congestion status of individual routes I am attaching
> > section 11.2.4 from Q.704
> >
> >
> > 11.2.4 Signalling route set congestion (National option with congestion
> >        priorities)
> > In the case when the congestion status of a signalling route set
changes
> as
> > a result of either the receipt of a transfer controlled message
relating
> to
> > a particular destination (see 13.7) or an indication of local
signalling
> > link congestion, or due to the signalling route-set-congestion-test
> > procedure (see 13.9) an indication is given from the Message Transfer
> Part
> > to the local level 4 informing it about the current congestion status
of
> > the signalling route set. Each user then takes appropriate actions in
> order
> > to stop generation of signalling messages destined for the affected
> > signalling point with congestion priorities lower than the specified
> > congestion status. Messages received from the local level 4 with
> congestion
> > priorities lower than the current signalling route set congestion
status
> > are discarded by the Message Transfer Part.
> > Also, SCON() in this configuration corresponds to TFC (not MTP-STATUS)
> > and therefore cannot signal abatement.  (TFC does not signal abatement,
> > the lack of a TFC does.)  In the same ways, the ASP can use the lack
> > of a SCON to indicate abatement.
> > [Kamesh]Yes they do and derived congestion status leads to MTP-STATUS
> >
> > So, a) SCON(1) and SCON(3) in your example are congestion signals
> > (neither is an abatement signal); b) there is no need to maintain
> > congestion
> > status on a per-SG basis at the ASP ('cause the SS7 SEP doesn't).
> > [Kamesh]Yes there is need to maintain congestion status on the per-SG
> basis
> > at ASP because SG1 doing RCT procedure on the behalf of ASP may return
> > SCON(0) when it receives audit for the affected point code and if
> affected
> > point code is reachable through SG1 and SG2 as shown in previous figure
> SG2
> > returns SCON(2) then congestion status sent in MTP-STATUS is 2 because
> > affected point code is still congested.
> >
> > This might lead back to the discustion why ASPs don't send RCT, but the
> > simple answer is that each SG as STP can send RCT on all the ASP's
behalf
> > as it would for an "Alias" point code (ANSI T1.111/2000).
> >
> > Because this is getting into a narrower and narrower corner of the
> > specification, I think we are safe to strike the mention of the timer
> > from the IG.
> > [Kamesh]I dont agree with striking timer
> > --brian
> >
> > kakaul@hss.hns.com wrote:
> > (Tue, 15 Jul 2003 18:59:37)
> > >
> > >
> > >
> > >
> > > Brian
> > >    some comments inline
> > >
> > >
> > >
> > >
> > > "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/15/2003
> > > 12:16:38 PM
> > >
> > > Please respond to bidulock@openss7.org
> > >
> > > Sent by:    sigtran-admin@ietf.org
> > >
> > >
> > > To:    Kamesh Kaul/HSS@HSS
> > > cc:    sigtran@ietf.org
> > >
> > > Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> > >
> > >
> > > kakaul,
> > >
> > > If there is no interoperablity issue then it is implementation
> dependent.
> > > A description of a procedure for the timer could be added as an
> > > IMPLEMENTATION
> > > NOTE making a suggestion to the implementor of one way to address
local
> > > issues
> > > at the ASP, but it cannot be required.
> > >
> > > But looking at your description, this mechanism would be a
duplication
> of
> > > the
> > > mechanism used by SCCP, ISUP, BICC, which use a dual-timer approach
to
> > > detecting congestion abatement.  Why would it be necessary for the
ASP
> to
> > > do
> > > this as well?
> > > >> It is also done in MTP3 for nation options with priority do you
mean
> > to
> > > say that RCT mechanism is not required in MTP3.
> > >
> > > Also, you should not that SCON received from one SG is no different
> than
> > > SCON
> > > received from another.  The SCON represents a statistical sampling of
> > > congestion in the SS7 network between the ASP and the destination,
and
> is
> > > not
> > > referenced to an SG.  Therefore, the ASP should never perform this
> > > procedure
> > > on a per-SG basis.  The SCON received from either SG will generate an
> > > MTP-STATUS primitive to the AS, and the AS will run suitable timers
at
> > the
> > > MTP-User (where the responsibility lies to avoid sending messages of
a
> > > priority less that the congestion state).
> > > >> In national option with priority MTP3 user is given MTP-STATUS
> > > indication only if point code (routeset )congestion status
> > > changes.Similarly at ASP MTP-STATUS indication shall be given to the
> user
> > > part only if derived status of affected point code changes as shown
in
> > > figure below :
> > >
> > >
> > >                               SCON (1)+--+----++
> > >                              <------- |        |
> > >                           +-----------+  SG1   +-----+
> > >                           |           |        |     |
> > >                           |           +--------+     |      SS7 NODE
> > >            +--+----++     |                          |
+----------+
> > >            |        |     |                          |    |   PC
|
> > >            |  ASP   +-----+                          +----+  X-Y-Z
|
> > >            |        |     |                          |    |
|
> > >            +--------+     |                          |
+----------+
> > >                           |          +--+----++      |
> > >              AS:A-B-C     |          |        |      |
> > >                           +----------+  SG2   +------+
> > >                             <------- |        |
> > >                              SCON(3) +--------+
> > >
> > > MTP-User will be given only MTP-STATUS indication with congestion
> status
> > 3
> > > if SCON(3) reaches before SCON(1).So for this you have to maintain
> > > congestion status of affected point code through each SG and
ultimately
> > > given the Derived congestion status to MTP-User.
> > >
> > > Understand that SCON is not an M3UA congestion indication: it is an
SS7
> > > network congestion indication that is directed at the MTP-User.  The
> ASP
> > > need
> > > only translate SCON messages to MTP-STATUS primitives to the AS to
> which
> > > they
> > > correspond.  A AS that conforms to MTP-User standards specifications
> with
> > > deal
> > > with the MTP-STATUS primitives appropriately.
> > > >>It is only true in international and national networks without
> priority
> > > option.
> > >
> > > It is also important to understand that the ASP should not change the
> > > routing
> > > of messages on the basis of a received SCON message.  Where
applicable
> > and
> > > for
> > > applicable SS7 standards specifications, the SG will send DRST under
> very
> > > specific and standardized congestion situations.  The ASP is already
> > > expected
> > > to perform some functions of MTP routing in response to DRST.
> > > >> Here i definitely agree with you that rerouting of messages should
> not
> > > be done at ASP to avoid missequencing.
> > >
> > > So, now I don't understand why you expect the ASP to run this timer
at
> > all.
> > > >> Some implementations may want to provision congestion of affected
> > point
> > > code through supervisory congestion timer, they should be allowed to
do
> > so.
> > >
> > > Please explain.
> > >
> > > --brian
> > >
> > > kakaul@hss.hns.com wrote:
> > > (Tue, 15 Jul 2003 10:46:03)
> > > >
> > > >
> > > >
> > > >
> > > > Brian,
> > > >    There is no interoperability issue involved in it and also i
dont
> > see
> > > > any need of peer knowing the existence of this timer.It is local to
> ASP
> > > and
> > > > is used by ASP to supervise the congestion status of the affected
> point
> > > > code reachable through a given SG.ASP after receiving an SCON
message
> > > with
> > > > congestion status greater than 0 can start this timer and if no
SCON
> > > > message is received from peer till the expiry of congestion timer
ASP
> > can
> > > > decrement affected point code  congestion status by 1 and start the
 > > timer
> > > > provisioning of congestion again if congestion status of affected
> point
> > > > code has not reached 0 or no congestion.
> > > >
> > > > Regards
> > > > Kamesh Kaul
> > > > www.hssworld.com
> > > > Hughes Software Systems
> > > >
> > > >
> > > >
> > > >
> > > > "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on
07/15/2003
> > > > 10:28:40 AM
> > > >
> > > > Please respond to bidulock@openss7.org
> > > >
> > > > Sent by:    sigtran-admin@ietf.org
> > > >
> > > >
> > > > To:    Kamesh Kaul/HSS@HSS
> > > > cc:    "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
> > > >        sigtran@ietf.org
> > > >
> > > > Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> > > >
> > > >
> > > > kakaul,
> > > >
> > > > Then add text describing how to use the timer in an interoperable
> way.
> > > > Also, please describe why the peer needs to know about the
existence
> > > > of this timer to remain interoperable.  Otherwise, there would not
be
> > > > any need to describe it.
> > > >
> > > > --brian
> > > >
> > > >
> > > > kakaul@hss.hns.com wrote:
> > > > (Tue, 15 Jul 2003 09:35:27)
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > Hi Javier,
> > > > >                     This timer may be used by some
implementations
> to
> > > > > supervise the congestion status of point code if SCON messages
send
> > by
> > > > SGP
> > > > > are lost in transit to ASP.It was discussed earlier and was
agreed
> > upon
> > > > by
> > > > > all to keep this timer.I disagree with removing the mention of
the
> > > timer.
 > > > > >
> > > > > Regards
> > > > > Kamesh Kaul
> > > > > www.hssworld.com
> > > > > Hughes Software Systems
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > "J.Javier Pastor (EE/EEM)"
<j.javier.pastor@ericsson.com>@ietf.org
> on
> > > > > 07/14/2003 04:05:58 PM
> > > > >
> > > > > Sent by:    sigtran-admin@ietf.org
> > > > >
> > > > >
> > > > > To:    "'Brian F. G. Bidulock '" <bidulock@openss7.org>
> > > > > cc:    "''sigtran@ietf.org' '" <sigtran@ietf.org>
> > > > >
> > > > > Subject:    RE: [Sigtran] M3UA IG v04 Section 3.18
> > > > >
> > > > >
> > > > > Brian,
> > > > >
> > > > > We'll remove the mention of the timer, if nobody disagrees...
> > > > >
> > > > > Cheers // Javier.
> > > > >
> > > > > -----Original Message-----
> > > > > From: Brian F. G. Bidulock
> > > > > To: J.Javier Pastor (EE/EEM)
> > > > > Cc: 'sigtran@ietf.org'
> > > > > Sent: 11/07/2003 11:07
> > > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > > > >
> > > > > J.Javier,
> > > > >
> > > > > J.Javier Pastor (EE/EEM) wrote:
> > > > > (Fri, 11 Jul 2003 09:56:59)
> > > > > > Hi Brian,
> > > > > >
> > > > > > Pls, see below...
> > > > > >
> > > > > > > -----Original Message-----
> > > > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > > > Sent: jueves, 10 de julio de 2003 15:34
> > > > > > > To: J.Javier Pastor (EE/EEM)
> > > > > > > Cc: 'sigtran@ietf.org'
> > > > > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > > > > > >
> > > > > > >
> > > > > > > Javier,
> > > > > > >
> > > > > > > 3.18.3 Solution Description
> > > > > > >
> > > > > > >    Clarify that the ASP needs to indicate when the congestion
> > > > > > > level has
> > > > > > >    changed (including abatement).  Further, the ASP peer can
> > > > > maintain
> > > > > > >    a timer, if desired, in case the ASP fails to indicate
> > > congestion
> > > > > > >    abatement.
> > > > > > >
> > > > > > > Where are the procedures for this timer described?  This is
the
> > > only
> > > > > place
> > > > > > > that the timer is mentioned.  Please describe the timer or
> remove
> > > > > the
> > > > > > > sentence saying that it has been described...
> > > > > > >
> > > > > >
> > > > > > Do you think is worth include the timer explanation into the
new
> > > text?
> > > > > > Otherwise, we may remove that sentence.
> > > > >
> > > > > Please remove mention of the timer.
> > > > >
> > > > > --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
> > > >
> > >
> > > --
> > > 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



_______________________________________________
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 exim@www1.ietf.org  Fri Jul 25 03:44:05 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA27559
	for <sigtran-archive@odin.ietf.org>; Fri, 25 Jul 2003 03:44:05 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fxEp-0006xC-39
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 03:43:39 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6P7hcr6026724
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 03:43:38 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fxEC-0006wB-LN; Fri, 25 Jul 2003 03:43:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fxE5-0006w0-2w
	for sigtran@optimus.ietf.org; Fri, 25 Jul 2003 03:42:53 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA27523
	for <sigtran@ietf.org>; Fri, 25 Jul 2003 03:42:48 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fxE2-00068a-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 03:42:50 -0400
Received: from david.siemens.de ([192.35.17.14])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fxDr-00068Q-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 03:42:40 -0400
Received: from mail3.siemens.de (mail3.siemens.de [139.25.208.14])
	by david.siemens.de (8.11.7/8.11.7) with ESMTP id h6P7gFP24941;
	Fri, 25 Jul 2003 09:42:15 +0200 (MEST)
Received: from moody.mchh.siemens.de (moody.mchh.siemens.de [139.21.205.85])
	by mail3.siemens.de (8.11.7/8.11.7) with ESMTP id h6P7gFG17074;
	Fri, 25 Jul 2003 09:42:15 +0200 (MEST)
Received: from mchh248e.mchh.siemens.de (mchh248e.mchh.siemens.de [139.21.200.58])
	by moody.mchh.siemens.de (8.9.3/8.9.1) with ESMTP id JAA27151;
	Fri, 25 Jul 2003 09:42:14 +0200 (MET DST)
Received: by mchh248e.mchh.siemens.de with Internet Mail Service (5.5.2653.19)
	id <PLMJBK30>; Fri, 25 Jul 2003 09:41:38 +0200
Message-ID: <D73559B38D51D4119CB800508B9527B505AC19D8@mchh266e.mchh.siemens.de>
From: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>
To: "'Tom George'" <Tom.George@alcatel.com>
Cc: IETF sigtran <sigtran@ietf.org>
Subject: RE: FW: [Sigtran] M2PA Last Call Comment
Date: Fri, 25 Jul 2003 09:41:38 +0200
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>

Tom,

I sent an answer to your mail friday last week, albeit I received it back from the exploder, I never got a response on it, so I'm repeating parts of that mail.

========================
 Question: does STP A care about the fact that STPB has links according to Q.2140? 
 Answer: NO, NOT AT ALL, STP A operator might not even know that STP B has Q.2140 links.  

        +-----------+                 +-----------+
   Q.703|           |                 |           |   Q.2140
  ------+           |                 |           +--------
        |           |       Q.703     |           |
        | STP A     +-----------------+   STP B   |
        |           |                 |           |
   Q.703|           |                 |           |   Q.2140
  ------+           |                 |           +--------
        |           |                 |           |
        +-----------+                 +-----------+

Let's go to an IP based link between STP A and STP B

Question: should STP A care about the fact that STP B has links according to Q.2140?
Answer: I see no necessity to deviate from ANSI and ITU-T.


        +-----------+                 +-----------+
   Q.703|           |                 |           |   Q.2140
  ------+           |                 |           +--------
        |           |       M2PA      |           |
        | STP A     +-----------------+   STP B   |
        |           |                 |           |
   Q.703|           |                 |           |   Q.2140
  ------+           |                 |           +--------
        |           |                 |           |
        +-----------+                 +-----------+

If STP A wants to prove, he should do that and STP B has to support STP A in that procedure and vice versa.
But I see no necessity why both sides should perform similiar. 

In all other cases STP A would have to be informed if e.g. STP-B migrates from pure Q.703 to pure Q.2140 and that's not acceptable.
=========================

My point is that using or not using the proving of M2PA is independent from the flavour of MTP level 2 procedure used on the TDM side of the SG. According to Figure 2 in draft 9 there is an MTP 3 on top, effectively decoupling both sides.


greetings 

HannsJuergen


> -----Original Message-----
> From: Tom George [mailto:Tom.George@alcatel.com]
> Sent: Thursday, July 24, 2003 10:28 PM
> Cc: IETF sigtran
> Subject: Re: FW: [Sigtran] M2PA Last Call Comment
> 
> 
> That does not seem like a good situation. I see the point 
> now. Instead 
> of your suggested text:
> 
>    The proving period MAY be ommitted if this is allowed by the
>    applicable standard (e.g., [Q.2140]); however, the same 
> standard MUST
>    be followed on both sides of the M2PA link.
> 
> would it be better to make a more general statement, in order 
> to avoid 
> this kind of problem that might exist in other procedures? 
> Maybe I could 
>   add to the 2nd paragraph in section 4:
> 
> =====
> Except as modified in this document, M2PA SHOULD follow the
> requirements of the applicable MTP2 specification. These may include
> [Q.703] or [T1.111]. The same standard MUST be followed on 
> both ends of 
> the M2PA link.
> =====
> 
> tomG.
> 
> 
> 
> Brian F. G. Bidulock wrote:
> > Tom,
> > 
> > Q.703 does not give the option to talk to a Q.2140 link.  But it
> > appears that M2PA allows a Q.703-state-machine-based M2PA link to
> > talk to a Q.2140-state-machine-based M2PA link.  That is were the
> > problem ensues.
> > 
> > --brian
> > 
> > Tom George wrote:                                           
>                            (Thu, 17 Jul 2003 17:16:54)
> > 
> >>Brian,
> >>
> >>In the non-IP realm, is it possible for a Q.703 to talk to 
> a Q.2140? 
> >>What would be done in the scenario you described?
> >>
> >>tomG.
> >>
> >>Brian F. G. Bidulock wrote:
> >>
> >>>Hanns Juergen,
> >>>
> >>>Consider that one side follows Q.703 and requires proving while the
> >>>other side follows Q.2140 and foregoes proving.  The following will
> >>>occur:
> >>
>    [...]
> 
> 
> -- 
> * 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.  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 exim@www1.ietf.org  Fri Jul 25 04:08:10 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA28133
	for <sigtran-archive@odin.ietf.org>; Fri, 25 Jul 2003 04:08:10 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fxc7-0008RO-Lx
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 04:07:44 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6P87h37032442
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 04:07:43 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fxbS-0008Cd-Pq; Fri, 25 Jul 2003 04:07:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fxan-0008BU-Qw
	for sigtran@optimus.ietf.org; Fri, 25 Jul 2003 04:06:21 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA28075
	for <sigtran@ietf.org>; Fri, 25 Jul 2003 04:06:17 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fxal-0006Hw-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 04:06:19 -0400
Received: from goliath.siemens.de ([192.35.17.28])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fxaa-0006Ht-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 04:06:08 -0400
Received: from mail2.siemens.de (mail2.siemens.de [139.25.208.11])
	by goliath.siemens.de (8.11.7/8.11.7) with ESMTP id h6P860H11709;
	Fri, 25 Jul 2003 10:06:01 +0200 (MEST)
Received: from moody.mchh.siemens.de (moody.mchh.siemens.de [139.21.205.85])
	by mail2.siemens.de (8.11.7/8.11.7) with ESMTP id h6P860Z11028;
	Fri, 25 Jul 2003 10:06:00 +0200 (MEST)
Received: from mchh273e.demchh201e.icn.siemens.de (mchh273e.mchh.siemens.de [139.21.200.83])
	by moody.mchh.siemens.de (8.9.3/8.9.1) with ESMTP id KAA02485;
	Fri, 25 Jul 2003 10:06:00 +0200 (MET DST)
Received: by mchh273e.mchh.siemens.de with Internet Mail Service (5.5.2656.59)
	id <PLMKPTGC>; Fri, 25 Jul 2003 10:05:24 +0200
Message-ID: <D73559B38D51D4119CB800508B9527B505AC19D9@mchh266e.mchh.siemens.de>
From: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>
To: "'Andrew Booth'" <abooth@pt.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: RE: [Sigtran] M2PA Last Call Comment - LPO
Date: Fri, 25 Jul 2003 10:05:19 +0200
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 Andrew,

sorry for the late response.

just one remark on the figures shown below just for sake of correctness

ITU-T Recommendation Q.703 Figure 8 (sheet 3 of 14) shows

LPO comes from the "management layer" and not from MTP L3. 
According to "note 2" MTP 3 might also send LPO to MTP2 in case of using the national option of latching processor outage. 

But neither ANSI T1.111 nor Telcordia GR-246 indicate the use of this national option.

greetings

HannsJuergen

> -----Original Message-----
> From: Andrew Booth [mailto:abooth@pt.com]
> Sent: Thursday, July 24, 2003 3:58 AM
> To: Gaiser Franz
> Cc: 'sigtran@ietf.org'; tom.george@alcatel.com; Brian F. G. Bidulock
> Subject: Re: [Sigtran] M2PA Last Call Comment - LPO
> 
> 
> Hi Franz,
> 
> Before answering your question I'll ask another: do we really need to 
> resynchronize sequence numbers in M2PA, or can we just 
> mandate that the 
> sender continues to increment the FSN on every transmitted MSU?
> - This seems simpler than resynchronizing.
> - Since M2PA does not retransmit, the motivation for 
> resynchronization 
> doesn't seem to be present as it does in MTP2.
> - Some of the examples below give examples where resequencing can go 
> wrong.  Although these situations are unlikely, the consequences are 
> unpleasant.
> 
> Now for my best answer to your question: Option (2), but if 
> there are no 
> unacknowledged messages then the BSN should be set to the FSN of the 
> last acknowledged message.
> Note also that if the receiver of the LS PR uses the BSN to free 
> messages in the RTB (as you indicated below) then they have 
> to be extra 
> careful when validating BSNs in ensuing Data messages that 
> may have been 
> sent before the LS PR. I don't think this is a good idea.
> 
> Some potential issues with resynchronizing sequence numbers:
> 
> Note that the sender of LS PR cannot know if the receiver 
> will use the 
> BSN to resynchronize sequence numbers or not, since it is the 
> MTP in RPO 
> that decides when RPO is "long duration" via the MTP3 T1 timer.
> This leads to race conditions in which one side flushes and the other 
> does not, though they are unlikely.  For example, the following could 
> occur if an MSU was delayed longer than MTP3 T1 on the IP network:
> 
> 
>      MTP3                  M2PA                        M2PA
>        |                     |                          |
>        |                     |<--Data (FSN 0)-----------|
>        |                     |                          |
>        |                     |                     +----|
>        |----LPO------------->|                    /     |
>        |                     |----------LS PO----/----->|
>        |                     |                  /       |
>        |                     |                 /        |<--Flush
>        |----LPR + Continue-->|                /         |
>        |                     |----LS PR------/--------->| resync FSN
>        |                     |    (BSN 0)   /           |
>        |                     |             /            |
>        |                     |<--Data-----+             |
>        |                     |   (FSN 1)                |
>        |                     |                          |
>        |             Discard |<--Data-------------------|
>        |           (bad FSN) |   (FSN 0)                |
> 
> The following example could occur if the side in LPO flushes 
> while the 
> side in RPO does not (perhaps they have different T1 timer settings), 
> and if the side in LPO sends a BSN different from the last 
> received by M2PA:
> 
>      MTP3                  M2PA                        M2PA
>        |                     |                          |
>        |                     |<--Data (FSN 0)-----------|
>        |                     |                          |
>        |                     |                     +----|
>        |----LPO------------->|                    /     |
>        |                     |----------LS PO----/----->|
>        |                     |                  /       |
>        |                     |                 /        |
>        |               To RB |<--Data---------+         |
>        |                     |   (FSN 1)                |
>        |                     |                          |
>        |----LPR + Flush----->|                          |
>        |                     |----LS PR---------------->|
>        |                     |    (BSN 0)               |<- Continue,
>        |                     |                          |   
> don't resync FSN
>        |                     |                          |
>        |                     |                          |
>        |             Discard |<--Data (FSN 2)-----------|
>        |    (expected FSN 1) |                          |
>        |                     |                          |
>        |             Discard |<--Data (FSN 3)-----------|
>        |    (expected FSN 1) |                          |
> 
> I think this possible condition provides reasonable incentive 
> to always 
> set the BSN in the LS Processor Recovered to the last message 
> received 
> from the M2PA peer.  In that case, there doesn't seem to be 
> any point in 
> resynching sequence numbers, just let the peer increment the FSN on 
> every MSU.
> 
> In summary, these examples are unlikely but why complicate things? 
>  Unless there is a good reason to keep the MTP2 resynching procedures 
> lets get rid of them in favour of the simpler approach: the sender 
> always increments the FSN by 1 when sending MSUs.
> 
> Thoughts?
> Andrew
> 
> Gaiser Franz wrote:
> 
> >Hi All,
> >
> >implementing Processor Outage Handling of M2PA draft 09 the 
> following questions came up:
> >
> >The Processor Outage chapter 4.1.4 says
> >
> >When the local processor outage conditions ends, M2PA shall 
> send a Link
> >Status Processor Recovered message to its peer. This message is used
> >to signal in the end of the processor outage condition, instead of an
> >MSU or FISU as in MTP2. The BSN value in this message is used by the 
> >peer to resynchronize its sequence numbers, if this is required by
> >MTP2.
> >
> >What does M2PA draft 09 with BSN in this context mean, is it 
> (assume, MTP3 has send Continue Command)
> >
> >(1) the FSN of the last received message before LPO or 
> >(2) the FSN of the last message in the incoming messages 
> that were queued and unacknowledged
> >during the processor outage (with that the remote side can 
> release the messages in its RTB) ? 
> >
> >
> >
> >thanks
> >best regards, Franz 
> >
> >
> >
> >
> >
> >_______________________________________________
> >Sigtran mailing list
> >Sigtran@ietf.org
> >https://www1.ietf.org/mailman/listinfo/sigtran
> >
> >  
> >
> 
> -- 
> 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
> 

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



From exim@www1.ietf.org  Fri Jul 25 05:51:08 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA00381
	for <sigtran-archive@odin.ietf.org>; Fri, 25 Jul 2003 05:51:07 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fzDn-0003sd-BA
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 05:50:43 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6P9ohsV014911
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 05:50:43 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fzD7-0003s7-3K; Fri, 25 Jul 2003 05:50:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fzC8-0003mo-6G
	for sigtran@optimus.ietf.org; Fri, 25 Jul 2003 05:49:00 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA00352
	for <sigtran@ietf.org>; Fri, 25 Jul 2003 05:48:54 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fzC4-0006ov-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 05:48:56 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[1jkMD/V2C4Yr4UUYac8Wq0xSetWBYUNj])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fzBj-0006oh-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 05:48:35 -0400
Received: from ns.pigworks.openss7.net (IDENT:ocEfZ4kzcTbqKbuVltlYEJzWF7/ro0Fm@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6P9loE01883;
	Fri, 25 Jul 2003 03:47:50 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6P9lnm20741;
	Fri, 25 Jul 2003 03:47:49 -0600
Date: Fri, 25 Jul 2003 03:47:49 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: kakaul@hss.hns.com
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
Message-ID: <20030725034749.A20716@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: kakaul@hss.hns.com, sigtran@ietf.org
References: <OFC835ED71.D1087FE3-ON65256D6E.001C305D@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: <OFC835ED71.D1087FE3-ON65256D6E.001C305D@hss.hns.com>; from kakaul@hss.hns.com on Fri, Jul 25, 2003 at 10:39:47AM +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>

kakaul,

I originally asked to either strike mention in the IG or describe the
timer.  Javier's (1) or (2).  Mentioning a timer that has no description
is of little use.

--brian

kakaul@hss.hns.com wrote:                                                           (Fri, 25 Jul 2003 10:39:47)
> 
> 
> 
> 
> Brian,
>              You have not replied to my earlier mail
> 
> 
> 
> 
> Kamesh Kaul/HSS@HSS@ietf.org on 07/18/2003 08:38:57 PM
> 
> Sent by:    sigtran-admin@ietf.org
> 
> 
> To:    bidulock@openss7.org
> cc:    sigtran@ietf.org
> 
> Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> 
> 
> 
> 
> 
> 
> Brian,
>    I agree with running congestion timer per affected route-set(point code)
> basis as you have mentioned. But now the point is that in SS7, SEP runs the
> RCT congestion supervisory timers after sending RCT message with current
> congestion status of the affected route-set so that if the TFC for that
> message is not received congestion abatement procedure will be supervised
> by timers.So it means that in SS7, SEP RCT message checks the clearance of
> path (for congestion) to affected point code and then monitors the
> congestion abatement by timers.But in ASP there is no RCT message so we in
> a way have to depend on DAUD message per SG to simulate RCT message and
> monitor congestion by one timer per affected point code .But for that we
> need to know to which SG DAUD message should be sent. SS7, SEP need not
> bother about sending RCT per route because it sends RCT message as if it
> has got dummy message from user part for affected point code with current
> routeset congestion set in priority.For ASP to figure out to which SG DAUD
> has to be send either congestion status or flag(signifying occurance or
> abatement) has to be maintained per SG.Please comment.
> 
> Do you agree with not striking of timer mention in IG.
> 
> Regards
> Kamesh Kaul
> www.hssworld.com
> Hughes Software Systems
> 
> 
> 
> 
> "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/18/2003
> 01:10:22 PM
> 
> Please respond to bidulock@openss7.org
> 
> Sent by:    sigtran-admin@ietf.org
> 
> 
> To:    Kamesh Kaul/HSS@HSS
> cc:    sigtran@ietf.org
> 
> Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> 
> 
> kakaul,
> 
> I don't think the ASP needs to do all that.  Instead of DAUD, it can simply
> lower the congestion status of the routeset after a time interval and
> indicate
> abatement to the National MTP-User.  The next message sent by the MTP-User
> will illicit SCON if the abatement was premature (without sending any
> message
> on the SS7 network) and the timer can be reset.  Only one timer per
> routeset
> is necessary.
> 
> In the multiple SG scenario SCON is comparable to TFC, not MTP-STATUS.
> SCON(0) is not necessary in the multiple SG scenario, the ASP can abate
> after a time interval and rely on responsive SCON from the SG.  If no
> SCON is delivered, then abatement to no congestion sticks.
> 
> Also, understand that the Nation MTP (e.g. ANSI) does not indicate
> abatement
> in all situations: only when the congestion is remote.  When the congestion
> is a result of congestion of a signalling link, abatement is not indicated.
> (e.g. see ANSI T1.111.4/2000 SDLs).
> 
> --brian
> 
> 
> kakaul@hss.hns.com wrote:
> (Fri, 18 Jul 2003 10:15:37)
> >
> >
> >
> >
> > Brian ,
> >    I agree with you that in SS7 SEP congestion status is only maintained
> > per routeset basis and not per route basis, beacause due to RCT procedure
> > there is no need of maintaining congestion status  on per route basis.Now
> > putting ASP as SEP you have no RCT procedure defined, but as you say SG's
> > connected with ASP perform  RCT on the behalf of ASP.Assuming affected
> > point code x-y-z is reachable through two SG's SG1 and SG2.Now ASP
> receives
> > SCON(1) from SG1 and update the point code (X-Y-Z) congestion status  to
> 1
> > and MTP3 users are informed with MTP-STATUS with congestion status 1.Now
> > ASP has to start DAUD procedure for affected point code x-y-z per SG and
> > ASP will continue to send DAUD messages with affected point code X-Y-Z to
> > SG1 untill ASP is sure that point code(x-y-z) congestion status  is 0
> > (no-congestion) through SG1.Mean- while SG1 might be doing RCT procedure
> on
> > the behalf of ASP for affected point code X-Y-Z.There are actually two
> > options for ASP to stop sending DAUD messages to SG1  with affected point
> > code x-y-z. Either ASP maintains congestion status per SG route or mark
> the
> > SG1 route for affected point code x-y-z congested by using some flag so
> > that ASP can know when to stop DAUD procedure for affected point code per
> > SG.Roueset (x-y-z) congestion status is just there to indicate MTP-STATUS
> > primitives.ASP will AUDIT affected point code per SG till congestion
> > abates, for that ASP has to know when to start and stop sending DAUD
> > messages for affected congested point code to a given SG which in a way
> > implies ASP has to maintain either congestion status or flag (indicationg
> > occurance and abatement of congestion).Now for abatement of affected
> point
> > code congestion status there are two approaches either SCON(0) received
> > from SG for affected point code or timer supervision because once you
> have
> > indicated congestion status greater than 0 to MTP3 user you have to
> > indicate congestion abatement with congestion status 0 to MTP3 user this
> > statement is only true if the network appearance associated with affected
> > point code is of national option with congestion priority.I have taken
> two
> > SG's in my above example to explain that affected point code  X-Y-Z
> > congestion status can get updated to 3 on receiving SCON(3) from SG2.Now
> > you have to send DAUD messages to both SG1 and SG2 till abatement of
> > congestion occures. Please comment
> >
> > Regards
> > Kamesh Kaul
> > www.hssworld.com
> > Hughes Software Systems
> >
> >
> >
> >
> >
> > "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/17/2003
> > 06:46:00 AM
> >
> > Please respond to bidulock@openss7.org
> >
> > Sent by:    sigtran-admin@ietf.org
> >
> >
> > To:    Kamesh Kaul/HSS@HSS
> > cc:    sigtran@ietf.org
> >
> > Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> >
> >
> > kakaul,
> >
> > You are confusing congestion at an STP due to the conestion of signalling
> > links on a route with routeset congestion.  Although the congestion of a
> > signalling link can trigger sending TFC, the route upon which the TFC is
> > received has nothing to do with congestion.
> >
> > Put an SEP where the ASP is and tell me where in the SS7 specs it says
> > that a TFC received from an STP is correlated to the route upon which it
> > was received.  A: it doesn't.
> >
> > The passages you quote correspond to transmit congestion on the SCTP
> > associations between the ASP and the SG.  (SCON is not involved there.)
> >
> > --brian
> >
> > kakaul@hss.hns.com wrote:
> > (Wed, 16 Jul 2003 21:57:11)
> > >
> > >
> > >
> > >
> > > Brian,
> > >    Comments inline
> > >
> > >
> > >
> > >
> > > "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/16/2003
> > > 12:41:40 PM
> > >
> > > Please respond to bidulock@openss7.org
> > >
> > > Sent by:    sigtran-admin@ietf.org
> > >
> > >
> > > To:    Kamesh Kaul/HSS@HSS
> > > cc:    sigtran@ietf.org
> > >
> > > Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> > >
> > >
> > > kakaul,
> > >
> > > An SS7 SEP at the position of the ASP, where the SGs are STPs would
> > > not maintain congestion state on a per-route (per-SG) basis.  So
> > > what makes you think that the ASP needs to?
> > > [Kamesh]It would because route-set congestion status will change
> > depending
> > > on the derived congestion status of individual routes I am attaching
> > > section 11.2.4 from Q.704
> > >
> > >
> > > 11.2.4 Signalling route set congestion (National option with congestion
> > >        priorities)
> > > In the case when the congestion status of a signalling route set
> changes
> > as
> > > a result of either the receipt of a transfer controlled message
> relating
> > to
> > > a particular destination (see 13.7) or an indication of local
> signalling
> > > link congestion, or due to the signalling route-set-congestion-test
> > > procedure (see 13.9) an indication is given from the Message Transfer
> > Part
> > > to the local level 4 informing it about the current congestion status
> of
> > > the signalling route set. Each user then takes appropriate actions in
> > order
> > > to stop generation of signalling messages destined for the affected
> > > signalling point with congestion priorities lower than the specified
> > > congestion status. Messages received from the local level 4 with
> > congestion
> > > priorities lower than the current signalling route set congestion
> status
> > > are discarded by the Message Transfer Part.
> > > Also, SCON() in this configuration corresponds to TFC (not MTP-STATUS)
> > > and therefore cannot signal abatement.  (TFC does not signal abatement,
> > > the lack of a TFC does.)  In the same ways, the ASP can use the lack
> > > of a SCON to indicate abatement.
> > > [Kamesh]Yes they do and derived congestion status leads to MTP-STATUS
> > >
> > > So, a) SCON(1) and SCON(3) in your example are congestion signals
> > > (neither is an abatement signal); b) there is no need to maintain
> > > congestion
> > > status on a per-SG basis at the ASP ('cause the SS7 SEP doesn't).
> > > [Kamesh]Yes there is need to maintain congestion status on the per-SG
> > basis
> > > at ASP because SG1 doing RCT procedure on the behalf of ASP may return
> > > SCON(0) when it receives audit for the affected point code and if
> > affected
> > > point code is reachable through SG1 and SG2 as shown in previous figure
> > SG2
> > > returns SCON(2) then congestion status sent in MTP-STATUS is 2 because
> > > affected point code is still congested.
> > >
> > > This might lead back to the discustion why ASPs don't send RCT, but the
> > > simple answer is that each SG as STP can send RCT on all the ASP's
> behalf
> > > as it would for an "Alias" point code (ANSI T1.111/2000).
> > >
> > > Because this is getting into a narrower and narrower corner of the
> > > specification, I think we are safe to strike the mention of the timer
> > > from the IG.
> > > [Kamesh]I dont agree with striking timer
> > > --brian
> > >
> > > kakaul@hss.hns.com wrote:
> > > (Tue, 15 Jul 2003 18:59:37)
> > > >
> > > >
> > > >
> > > >
> > > > Brian
> > > >    some comments inline
> > > >
> > > >
> > > >
> > > >
> > > > "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on 07/15/2003
> > > > 12:16:38 PM
> > > >
> > > > Please respond to bidulock@openss7.org
> > > >
> > > > Sent by:    sigtran-admin@ietf.org
> > > >
> > > >
> > > > To:    Kamesh Kaul/HSS@HSS
> > > > cc:    sigtran@ietf.org
> > > >
> > > > Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> > > >
> > > >
> > > > kakaul,
> > > >
> > > > If there is no interoperablity issue then it is implementation
> > dependent.
> > > > A description of a procedure for the timer could be added as an
> > > > IMPLEMENTATION
> > > > NOTE making a suggestion to the implementor of one way to address
> local
> > > > issues
> > > > at the ASP, but it cannot be required.
> > > >
> > > > But looking at your description, this mechanism would be a
> duplication
> > of
> > > > the
> > > > mechanism used by SCCP, ISUP, BICC, which use a dual-timer approach
> to
> > > > detecting congestion abatement.  Why would it be necessary for the
> ASP
> > to
> > > > do
> > > > this as well?
> > > > >> It is also done in MTP3 for nation options with priority do you
> mean
> > > to
> > > > say that RCT mechanism is not required in MTP3.
> > > >
> > > > Also, you should not that SCON received from one SG is no different
> > than
> > > > SCON
> > > > received from another.  The SCON represents a statistical sampling of
> > > > congestion in the SS7 network between the ASP and the destination,
> and
> > is
> > > > not
> > > > referenced to an SG.  Therefore, the ASP should never perform this
> > > > procedure
> > > > on a per-SG basis.  The SCON received from either SG will generate an
> > > > MTP-STATUS primitive to the AS, and the AS will run suitable timers
> at
> > > the
> > > > MTP-User (where the responsibility lies to avoid sending messages of
> a
> > > > priority less that the congestion state).
> > > > >> In national option with priority MTP3 user is given MTP-STATUS
> > > > indication only if point code (routeset )congestion status
> > > > changes.Similarly at ASP MTP-STATUS indication shall be given to the
> > user
> > > > part only if derived status of affected point code changes as shown
> in
> > > > figure below :
> > > >
> > > >
> > > >                               SCON (1)+--+----++
> > > >                              <------- |        |
> > > >                           +-----------+  SG1   +-----+
> > > >                           |           |        |     |
> > > >                           |           +--------+     |      SS7 NODE
> > > >            +--+----++     |                          |
> +----------+
> > > >            |        |     |                          |    |   PC
> |
> > > >            |  ASP   +-----+                          +----+  X-Y-Z
> |
> > > >            |        |     |                          |    |
> |
> > > >            +--------+     |                          |
> +----------+
> > > >                           |          +--+----++      |
> > > >              AS:A-B-C     |          |        |      |
> > > >                           +----------+  SG2   +------+
> > > >                             <------- |        |
> > > >                              SCON(3) +--------+
> > > >
> > > > MTP-User will be given only MTP-STATUS indication with congestion
> > status
> > > 3
> > > > if SCON(3) reaches before SCON(1).So for this you have to maintain
> > > > congestion status of affected point code through each SG and
> ultimately
> > > > given the Derived congestion status to MTP-User.
> > > >
> > > > Understand that SCON is not an M3UA congestion indication: it is an
> SS7
> > > > network congestion indication that is directed at the MTP-User.  The
> > ASP
> > > > need
> > > > only translate SCON messages to MTP-STATUS primitives to the AS to
> > which
> > > > they
> > > > correspond.  A AS that conforms to MTP-User standards specifications
> > with
> > > > deal
> > > > with the MTP-STATUS primitives appropriately.
> > > > >>It is only true in international and national networks without
> > priority
> > > > option.
> > > >
> > > > It is also important to understand that the ASP should not change the
> > > > routing
> > > > of messages on the basis of a received SCON message.  Where
> applicable
> > > and
> > > > for
> > > > applicable SS7 standards specifications, the SG will send DRST under
> > very
> > > > specific and standardized congestion situations.  The ASP is already
> > > > expected
> > > > to perform some functions of MTP routing in response to DRST.
> > > > >> Here i definitely agree with you that rerouting of messages should
> > not
> > > > be done at ASP to avoid missequencing.
> > > >
> > > > So, now I don't understand why you expect the ASP to run this timer
> at
> > > all.
> > > > >> Some implementations may want to provision congestion of affected
> > > point
> > > > code through supervisory congestion timer, they should be allowed to
> do
> > > so.
> > > >
> > > > Please explain.
> > > >
> > > > --brian
> > > >
> > > > kakaul@hss.hns.com wrote:
> > > > (Tue, 15 Jul 2003 10:46:03)
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > Brian,
> > > > >    There is no interoperability issue involved in it and also i
> dont
> > > see
> > > > > any need of peer knowing the existence of this timer.It is local to
> > ASP
> > > > and
> > > > > is used by ASP to supervise the congestion status of the affected
> > point
> > > > > code reachable through a given SG.ASP after receiving an SCON
> message
> > > > with
> > > > > congestion status greater than 0 can start this timer and if no
> SCON
> > > > > message is received from peer till the expiry of congestion timer
> ASP
> > > can
> > > > > decrement affected point code  congestion status by 1 and start the
>  > > timer
> > > > > provisioning of congestion again if congestion status of affected
> > point
> > > > > code has not reached 0 or no congestion.
> > > > >
> > > > > Regards
> > > > > Kamesh Kaul
> > > > > www.hssworld.com
> > > > > Hughes Software Systems
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > "Brian F. G. Bidulock" <bidulock@openss7.org>@ietf.org on
> 07/15/2003
> > > > > 10:28:40 AM
> > > > >
> > > > > Please respond to bidulock@openss7.org
> > > > >
> > > > > Sent by:    sigtran-admin@ietf.org
> > > > >
> > > > >
> > > > > To:    Kamesh Kaul/HSS@HSS
> > > > > cc:    "J.Javier Pastor (EE/EEM)" <j.javier.pastor@ericsson.com>,
> > > > >        sigtran@ietf.org
> > > > >
> > > > > Subject:    Re: [Sigtran] M3UA IG v04 Section 3.18
> > > > >
> > > > >
> > > > > kakaul,
> > > > >
> > > > > Then add text describing how to use the timer in an interoperable
> > way.
> > > > > Also, please describe why the peer needs to know about the
> existence
> > > > > of this timer to remain interoperable.  Otherwise, there would not
> be
> > > > > any need to describe it.
> > > > >
> > > > > --brian
> > > > >
> > > > >
> > > > > kakaul@hss.hns.com wrote:
> > > > > (Tue, 15 Jul 2003 09:35:27)
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > Hi Javier,
> > > > > >                     This timer may be used by some
> implementations
> > to
> > > > > > supervise the congestion status of point code if SCON messages
> send
> > > by
> > > > > SGP
> > > > > > are lost in transit to ASP.It was discussed earlier and was
> agreed
> > > upon
> > > > > by
> > > > > > all to keep this timer.I disagree with removing the mention of
> the
> > > > timer.
>  > > > > >
> > > > > > Regards
> > > > > > Kamesh Kaul
> > > > > > www.hssworld.com
> > > > > > Hughes Software Systems
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > "J.Javier Pastor (EE/EEM)"
> <j.javier.pastor@ericsson.com>@ietf.org
> > on
> > > > > > 07/14/2003 04:05:58 PM
> > > > > >
> > > > > > Sent by:    sigtran-admin@ietf.org
> > > > > >
> > > > > >
> > > > > > To:    "'Brian F. G. Bidulock '" <bidulock@openss7.org>
> > > > > > cc:    "''sigtran@ietf.org' '" <sigtran@ietf.org>
> > > > > >
> > > > > > Subject:    RE: [Sigtran] M3UA IG v04 Section 3.18
> > > > > >
> > > > > >
> > > > > > Brian,
> > > > > >
> > > > > > We'll remove the mention of the timer, if nobody disagrees...
> > > > > >
> > > > > > Cheers // Javier.
> > > > > >
> > > > > > -----Original Message-----
> > > > > > From: Brian F. G. Bidulock
> > > > > > To: J.Javier Pastor (EE/EEM)
> > > > > > Cc: 'sigtran@ietf.org'
> > > > > > Sent: 11/07/2003 11:07
> > > > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > > > > >
> > > > > > J.Javier,
> > > > > >
> > > > > > J.Javier Pastor (EE/EEM) wrote:
> > > > > > (Fri, 11 Jul 2003 09:56:59)
> > > > > > > Hi Brian,
> > > > > > >
> > > > > > > Pls, see below...
> > > > > > >
> > > > > > > > -----Original Message-----
> > > > > > > > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > > > > > > > Sent: jueves, 10 de julio de 2003 15:34
> > > > > > > > To: J.Javier Pastor (EE/EEM)
> > > > > > > > Cc: 'sigtran@ietf.org'
> > > > > > > > Subject: Re: [Sigtran] M3UA IG v04 Section 3.18
> > > > > > > >
> > > > > > > >
> > > > > > > > Javier,
> > > > > > > >
> > > > > > > > 3.18.3 Solution Description
> > > > > > > >
> > > > > > > >    Clarify that the ASP needs to indicate when the congestion
> > > > > > > > level has
> > > > > > > >    changed (including abatement).  Further, the ASP peer can
> > > > > > maintain
> > > > > > > >    a timer, if desired, in case the ASP fails to indicate
> > > > congestion
> > > > > > > >    abatement.
> > > > > > > >
> > > > > > > > Where are the procedures for this timer described?  This is
> the
> > > > only
> > > > > > place
> > > > > > > > that the timer is mentioned.  Please describe the timer or
> > remove
> > > > > > the
> > > > > > > > sentence saying that it has been described...
> > > > > > > >
> > > > > > >
> > > > > > > Do you think is worth include the timer explanation into the
> new
> > > > text?
> > > > > > > Otherwise, we may remove that sentence.
> > > > > >
> > > > > > Please remove mention of the timer.
> > > > > >
> > > > > > --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
> > > > >
> > > >
> > > > --
> > > > 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
> 
> 
> 
> _______________________________________________
> 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 exim@www1.ietf.org  Fri Jul 25 07:37:12 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA02386
	for <sigtran-archive@odin.ietf.org>; Fri, 25 Jul 2003 07:37:12 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19g0sP-0007UT-0L
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 07:36:45 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6PBaiW1028794
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 07:36:44 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19g0ri-0007Tx-Aw; Fri, 25 Jul 2003 07:36:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19g0rA-0007TT-VL
	for sigtran@optimus.ietf.org; Fri, 25 Jul 2003 07:35:28 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA02380
	for <sigtran@ietf.org>; Fri, 25 Jul 2003 07:35:25 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19g0rA-0007Oz-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 07:35:28 -0400
Received: from thoth.sbs.de ([192.35.17.2])
	by ietf-mx with esmtp (Exim 4.12)
	id 19g0qz-0007Ok-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 07:35:17 -0400
Received: from mail1.siemens.de (mail1.siemens.de [139.23.33.14])
	by thoth.sbs.de (8.11.7/8.11.7) with ESMTP id h6PBYXK16065
	for <sigtran@ietf.org>; Fri, 25 Jul 2003 13:34:33 +0200 (MEST)
Received: from moody.mchh.siemens.de (moody.mchh.siemens.de [139.21.205.85])
	by mail1.siemens.de (8.11.7/8.11.7) with ESMTP id h6PBYWL13026
	for <sigtran@ietf.org>; Fri, 25 Jul 2003 13:34:32 +0200 (MEST)
Received: from mchh246e.demchh201e.icn.siemens.de (mchh246e.mchh.siemens.de [139.21.200.56])
	by moody.mchh.siemens.de (8.9.3/8.9.1) with ESMTP id NAA09480
	for <sigtran@ietf.org>; Fri, 25 Jul 2003 13:34:32 +0200 (MET DST)
Received: by mchh246e.mchh.siemens.de with Internet Mail Service (5.5.2656.59)
	id <PLM2B1Z8>; Fri, 25 Jul 2003 13:33:57 +0200
Message-ID: <99522F0FD7EFD611A47D0002A58EDAB7C84A8F@mchh2a8e.mchh.siemens.de>
From: Gaiser Franz <franz.gaiser@siemens.com>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: AW: [Sigtran] M2PA Last Call Comment - LPO
Date: Fri, 25 Jul 2003 13:33:55 +0200
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2656.59)
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 All,

From a developer point of view I would prefer that the statements about resync are described more precisely and not in a way that the developer has to interpret it especially when he was not involved in the development process of M2PA.
 
Why don't complement it with the additional remarks i.e. made by Andrew that 
(1) the BSN in the LS PR is set to the last message received from the M2PA. 
(2) the sender always increments the FSN by 1 when sending MSUs. 

best regards, Franz





-----Ursprungliche Nachricht-----
Von: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
Gesendet: Donnerstag, 24. Juli 2003 18:11
An: Andrew Booth
Cc: 'sigtran@ietf.org'
Betreff: Re: [Sigtran] M2PA Last Call Comment - LPO


Andrew,

On the first point, performance related issues are outside the scope
of the protocol.

On the second point, nowhere in M2PA-09 does it say to stop incrementing
BSN.  Just to withhold acknowledgements and buffer during PO.  The
peer is welcome, as it says in M2PA-09, to resync off of the BSN in the
LPR.  The fact that it doesn't change the FSN for ANSI does not mean that
it is not performing the resynchronization.

Look at the resync wording in M2PA-09 again.

I don't think it need changing.

--brian

Andrew Booth wrote:                                                                 (Thu, 24 Jul 2003 11:01:02)
> Hi Brian,
> 
> I don't think synchronization of sequence numbers after PO is necessary 
> or desirable in M2PA.  Can you see any advantage to it?
> 
> I'd like to highlight this point by bringing up your comment after 
> example 2 below: in order to avoid potential race conditions the side 
> sending LS Processor Recovered should set the BSN to the last FSN 
> received by M2PA.  By setting the BSN in this way we expect is that the 
> peer will continue sending FSNs "in sequence" i.e. that the 
> resynchronization will not actually change the next FSN the peer will 
> send.  If that's the behaviour we want, and I think it is, then I 
> propose that we specify it directly rather than carry over an 
> unnecessary procedure from MTP2.
> 
> More comments below.
> 
> Brian F. G. Bidulock wrote:
> 
> >Andrew,
> >
> >Please see comments below...
> >
> >Andrew Booth wrote:                                                                 (Wed, 23 Jul 2003 21:58:07)
> >  
> >
> >>Hi Franz,
> >>
> >>Before answering your question I'll ask another: do we really need to 
> >>resynchronize sequence numbers in M2PA, or can we just mandate that the 
> >>sender continues to increment the FSN on every transmitted MSU?
> >>- This seems simpler than resynchronizing.
> >>- Since M2PA does not retransmit, the motivation for resynchronization 
> >>doesn't seem to be present as it does in MTP2.
> >>- Some of the examples below give examples where resequencing can go 
> >>wrong.  Although these situations are unlikely, the consequences are 
> >>unpleasant.
> >>
> >>Now for my best answer to your question: Option (2), but if there are no 
> >>unacknowledged messages then the BSN should be set to the FSN of the 
> >>last acknowledged message.
> >>Note also that if the receiver of the LS PR uses the BSN to free 
> >>messages in the RTB (as you indicated below) then they have to be extra 
> >>careful when validating BSNs in ensuing Data messages that may have been 
> >>sent before the LS PR. I don't think this is a good idea.
> >>
> >>Some potential issues with resynchronizing sequence numbers:
> >>
> >>Note that the sender of LS PR cannot know if the receiver will use the 
> >>BSN to resynchronize sequence numbers or not, since it is the MTP in RPO 
> >>that decides when RPO is "long duration" via the MTP3 T1 timer.
> >>This leads to race conditions in which one side flushes and the other 
> >>does not, though they are unlikely.  For example, the following could 
> >>occur if an MSU was delayed longer than MTP3 T1 on the IP network:
> >>
> >>
> >>     MTP3                  M2PA                        M2PA
> >>       |                     |                          |
> >>       |                     |<--Data (FSN 0)-----------|
> >>       |                     |                          |
> >>       |                     |                     +----|
> >>       |----LPO------------->|                    /     |
> >>       |                     |----------LS PO----/----->|
> >>       |                     |                  /       |
> >>       |                     |                 /        |<--Flush
> >>       |----LPR + Continue-->|                /         |
> >>       |                     |----LS PR------/--------->| resync FSN
> >>       |                     |    (BSN 0)   /           |
> >>       |                     |             /            |
> >>       |                     |<--Data-----+             |
> >>       |                     |   (FSN 1)                |
> >>       |                     |                          |
> >>       |             Discard |<--Data-------------------|
> >>       |           (bad FSN) |   (FSN 0)                |
> >>    
> >>
> >
> >This is not really probable.  This would mean that the MTP RPO timer
> >expires when the LPO timer hasn't (a very narrow window) and that the
> >Data (FSN 1) was delayed longer than the RPO timer (which shouldn't
> >really happen).
> >
> Agreed, it is unlikely.  If SCTP parameters are set correctly it might 
> even be impossible.
> 
> >
> >Set your SCTP parameters (or lifetimes) so that messages not
> >acknowledged within the RPO timeout will drop the SCTP association and
> >you won't have this problem.
> >
> >Do you see?  The "long duration" of the procedure (T1 or about 1.0 to
> >2.0 seconds) is used so that old messages stored in buffers are not sent
> >later.  RTB clear is performed immediately before changeover (after T1
> >expiry).  One doesn't want a recovering processor to transfer old
> >messages after changeover has occured or a lot of missequencing is going
> >to happen in the SS7 network.
> >
> Yes I understand, though I think MTP3 T1 is 0.5 to 1.2 seconds.
> 
> >
> >Now, if you have SCTP parameters set to allow messages to be buffered IN
> >THE NETWORK for longer than T1, you have a real big problem: both the
> >time-controlled changeover and controlled rerouting procedures will
> >result in missequenced messages, even without processor outage.
> >
> Agreed.  Should this be stated explicitly in M2PA?
> 
> >
> >So, if you have a problem above, becuase your SCTP will not drop the
> >link before delaying a data message longer than T1, you have the greater
> >problem that every time-controlled changeover or controlled rerouting
> >procedure for those M2PA links are going to missequence messagses.
> >
> >Therefore, you need to tighten your SCTP parameters.  If you cannot
> >(e.g. the outher end is a really bad implemetnation), you really need to
> >bring the link down when the duplicate FSN arrives.
> >
> >I think you have an error here too: BSN 0 acknowledges the receipt of
> >data with FSN 0.  After the resync, the next data sent should have BSN
> >1.
> >
> Agreed, the last FSN should be 1 (still invalid).
> 
> >
> >
> >  
> >
> >>The following example could occur if the side in LPO flushes while the 
> >>side in RPO does not (perhaps they have different T1 timer settings), 
> >>and if the side in LPO sends a BSN different from the last received by M2PA:
> >>
> >>     MTP3                  M2PA                        M2PA
> >>       |                     |                          |
> >>       |                     |<--Data (FSN 0)-----------|
> >>       |                     |                          |
> >>       |                     |                     +----|
> >>       |----LPO------------->|                    /     |
> >>       |                     |----------LS PO----/----->|
> >>       |                     |                  /       |
> >>       |                     |                 /        |
> >>       |               To RB |<--Data---------+         |
> >>       |                     |   (FSN 1)                |
> >>       |                     |                          |
> >>       |----LPR + Flush----->|                          |
> >>       |                     |----LS PR---------------->|
> >>       |                     |    (BSN 0)               |<- Continue,
> >>       |                     |                          |   don't resync FSN
> >>       |                     |                          |
> >>       |                     |                          |
> >>       |             Discard |<--Data (FSN 2)-----------|
> >>       |    (expected FSN 1) |                          |
> >>       |                     |                          |
> >>       |             Discard |<--Data (FSN 3)-----------|
> >>       |    (expected FSN 1) |                          |
> >>    
> >>
> >
> >Again, having one side flush without the other is a small time window,
> >however, why are you sending LS PR (BSN 0)?  I think you should send
> >LS PR (BSN 1) in this instance.  Then there is no problem.
> >
> I agree this is unlikely, but I don't think the time window is 
> necessarily that small: if one side sets T1 to 0.5s and the other to 
> 1.2s you have 0.7s to hit this.
> 
> >
> >MTP2 would send BSN 0 here, but it is also discarding messages with the
> >expectation of retransmission.  M2PA buffers but does not acknowledge
> >messages.  The purpose of delaying acknowledgement is so that the peer
> >M2PA keeps these messages in the retransmission buffer for changeover.
> >(Remember than some variants collect from the RTB instead of clearing
> >it, unlike ANSI).  But, for M2PA, the BSN should be incremented when
> >the data is buffered to accomplish what happens here.
> >
> >So this one is not a problem either.
> >
> Agreed, it's not a problem, but it does highlight the point I made 
> above: we should set the BSN so that when resynchronizing sequence 
> numbers the peer doesn't actually have to change anything.  It seems 
> strange to have an exchange of BSN values for this purpose.
> 
> Thoughts?
> Andrew
> 
> >
> >--brian
> >
> >  
> >
> >>I think this possible condition provides reasonable incentive to always 
> >>set the BSN in the LS Processor Recovered to the last message received 
> >>from the M2PA peer.  In that case, there doesn't seem to be any point in 
> >>resynching sequence numbers, just let the peer increment the FSN on 
> >>every MSU.
> >>
> >>In summary, these examples are unlikely but why complicate things? 
> >> Unless there is a good reason to keep the MTP2 resynching procedures 
> >>lets get rid of them in favour of the simpler approach: the sender 
> >>always increments the FSN by 1 when sending MSUs.
> >>
> >>Thoughts?
> >>Andrew
> >>
> >>Gaiser Franz wrote:
> >>
> >>    
> >>
> >>>Hi All,
> >>>
> >>>implementing Processor Outage Handling of M2PA draft 09 the following questions came up:
> >>>
> >>>The Processor Outage chapter 4.1.4 says
> >>>
> >>>When the local processor outage conditions ends, M2PA shall send a Link
> >>>Status Processor Recovered message to its peer. This message is used
> >>>to signal in the end of the processor outage condition, instead of an
> >>>MSU or FISU as in MTP2. The BSN value in this message is used by the 
> >>>peer to resynchronize its sequence numbers, if this is required by
> >>>MTP2.
> >>>
> >>>What does M2PA draft 09 with BSN in this context mean, is it (assume, MTP3 has send Continue Command)
> >>>
> >>>(1) the FSN of the last received message before LPO or 
> >>>(2) the FSN of the last message in the incoming messages that were queued and unacknowledged
> >>>during the processor outage (with that the remote side can release the messages in its RTB) ? 
> >>>
> >>>
> >>>
> >>>thanks
> >>>best regards, Franz 
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>_______________________________________________
> >>>Sigtran mailing list
> >>>Sigtran@ietf.org
> >>>https://www1.ietf.org/mailman/listinfo/sigtran
> >>>
> >>> 
> >>>
> >>>      
> >>>
> >>-- 
> >>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.
> >>
> >>    
> >>
> >
> >  
> >
> 
> -- 
> 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.
> 

-- 
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 exim@www1.ietf.org  Fri Jul 25 11:24:09 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA09047
	for <sigtran-archive@odin.ietf.org>; Fri, 25 Jul 2003 11:24:09 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19g4Q3-0000z1-66
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 11:23:43 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6PFNh9Y003776
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 11:23:43 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19g4PM-0000qa-Tx; Fri, 25 Jul 2003 11:23:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19g4Op-0000iU-Kk
	for sigtran@optimus.ietf.org; Fri, 25 Jul 2003 11:22:28 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA08977
	for <sigtran@ietf.org>; Fri, 25 Jul 2003 11:22:22 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19g4Oo-00013g-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 11:22:26 -0400
Received: from ottgw.pt.com ([206.191.2.130] helo=pt.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19g4OY-000130-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 11:22:10 -0400
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id LAA18181;
	Fri, 25 Jul 2003 11:21:08 -0400
Message-ID: <3F214AE7.7040105@pt.com>
Date: Fri, 25 Jul 2003 11:21:11 -0400
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: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>
CC: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - LPO
References: <D73559B38D51D4119CB800508B9527B505AC19D9@mchh266e.mchh.siemens.de>
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

Hi HannsJuergen,

Thanks for the correction, my appologies for the innacuracy.

Andrew

Schwarzbauer Hanns Juergen wrote:

>Hi Andrew,
>
>sorry for the late response.
>
>just one remark on the figures shown below just for sake of correctness
>
>ITU-T Recommendation Q.703 Figure 8 (sheet 3 of 14) shows
>
>LPO comes from the "management layer" and not from MTP L3. 
>According to "note 2" MTP 3 might also send LPO to MTP2 in case of using the national option of latching processor outage. 
>
>But neither ANSI T1.111 nor Telcordia GR-246 indicate the use of this national option.
>
>greetings
>
>HannsJuergen
>
>  
>
>>-----Original Message-----
>>From: Andrew Booth [mailto:abooth@pt.com]
>>Sent: Thursday, July 24, 2003 3:58 AM
>>To: Gaiser Franz
>>Cc: 'sigtran@ietf.org'; tom.george@alcatel.com; Brian F. G. Bidulock
>>Subject: Re: [Sigtran] M2PA Last Call Comment - LPO
>>
>>
>>Hi Franz,
>>
>>Before answering your question I'll ask another: do we really need to 
>>resynchronize sequence numbers in M2PA, or can we just 
>>mandate that the 
>>sender continues to increment the FSN on every transmitted MSU?
>>- This seems simpler than resynchronizing.
>>- Since M2PA does not retransmit, the motivation for 
>>resynchronization 
>>doesn't seem to be present as it does in MTP2.
>>- Some of the examples below give examples where resequencing can go 
>>wrong.  Although these situations are unlikely, the consequences are 
>>unpleasant.
>>
>>Now for my best answer to your question: Option (2), but if 
>>there are no 
>>unacknowledged messages then the BSN should be set to the FSN of the 
>>last acknowledged message.
>>Note also that if the receiver of the LS PR uses the BSN to free 
>>messages in the RTB (as you indicated below) then they have 
>>to be extra 
>>careful when validating BSNs in ensuing Data messages that 
>>may have been 
>>sent before the LS PR. I don't think this is a good idea.
>>
>>Some potential issues with resynchronizing sequence numbers:
>>
>>Note that the sender of LS PR cannot know if the receiver 
>>will use the 
>>BSN to resynchronize sequence numbers or not, since it is the 
>>MTP in RPO 
>>that decides when RPO is "long duration" via the MTP3 T1 timer.
>>This leads to race conditions in which one side flushes and the other 
>>does not, though they are unlikely.  For example, the following could 
>>occur if an MSU was delayed longer than MTP3 T1 on the IP network:
>>
>>
>>     MTP3                  M2PA                        M2PA
>>       |                     |                          |
>>       |                     |<--Data (FSN 0)-----------|
>>       |                     |                          |
>>       |                     |                     +----|
>>       |----LPO------------->|                    /     |
>>       |                     |----------LS PO----/----->|
>>       |                     |                  /       |
>>       |                     |                 /        |<--Flush
>>       |----LPR + Continue-->|                /         |
>>       |                     |----LS PR------/--------->| resync FSN
>>       |                     |    (BSN 0)   /           |
>>       |                     |             /            |
>>       |                     |<--Data-----+             |
>>       |                     |   (FSN 1)                |
>>       |                     |                          |
>>       |             Discard |<--Data-------------------|
>>       |           (bad FSN) |   (FSN 0)                |
>>
>>The following example could occur if the side in LPO flushes 
>>while the 
>>side in RPO does not (perhaps they have different T1 timer settings), 
>>and if the side in LPO sends a BSN different from the last 
>>received by M2PA:
>>
>>     MTP3                  M2PA                        M2PA
>>       |                     |                          |
>>       |                     |<--Data (FSN 0)-----------|
>>       |                     |                          |
>>       |                     |                     +----|
>>       |----LPO------------->|                    /     |
>>       |                     |----------LS PO----/----->|
>>       |                     |                  /       |
>>       |                     |                 /        |
>>       |               To RB |<--Data---------+         |
>>       |                     |   (FSN 1)                |
>>       |                     |                          |
>>       |----LPR + Flush----->|                          |
>>       |                     |----LS PR---------------->|
>>       |                     |    (BSN 0)               |<- Continue,
>>       |                     |                          |   
>>don't resync FSN
>>       |                     |                          |
>>       |                     |                          |
>>       |             Discard |<--Data (FSN 2)-----------|
>>       |    (expected FSN 1) |                          |
>>       |                     |                          |
>>       |             Discard |<--Data (FSN 3)-----------|
>>       |    (expected FSN 1) |                          |
>>
>>I think this possible condition provides reasonable incentive 
>>to always 
>>set the BSN in the LS Processor Recovered to the last message 
>>received 
>>from the M2PA peer.  In that case, there doesn't seem to be 
>>any point in 
>>resynching sequence numbers, just let the peer increment the FSN on 
>>every MSU.
>>
>>In summary, these examples are unlikely but why complicate things? 
>> Unless there is a good reason to keep the MTP2 resynching procedures 
>>lets get rid of them in favour of the simpler approach: the sender 
>>always increments the FSN by 1 when sending MSUs.
>>
>>Thoughts?
>>Andrew
>>
>>Gaiser Franz wrote:
>>
>>    
>>
>>>Hi All,
>>>
>>>implementing Processor Outage Handling of M2PA draft 09 the 
>>>      
>>>
>>following questions came up:
>>    
>>
>>>The Processor Outage chapter 4.1.4 says
>>>
>>>When the local processor outage conditions ends, M2PA shall 
>>>      
>>>
>>send a Link
>>    
>>
>>>Status Processor Recovered message to its peer. This message is used
>>>to signal in the end of the processor outage condition, instead of an
>>>MSU or FISU as in MTP2. The BSN value in this message is used by the 
>>>peer to resynchronize its sequence numbers, if this is required by
>>>MTP2.
>>>
>>>What does M2PA draft 09 with BSN in this context mean, is it 
>>>      
>>>
>>(assume, MTP3 has send Continue Command)
>>    
>>
>>>(1) the FSN of the last received message before LPO or 
>>>(2) the FSN of the last message in the incoming messages 
>>>      
>>>
>>that were queued and unacknowledged
>>    
>>
>>>during the processor outage (with that the remote side can 
>>>      
>>>
>>release the messages in its RTB) ? 
>>    
>>
>>>
>>>thanks
>>>best regards, Franz 
>>>
>>>
>>>
>>>
>>>
>>>_______________________________________________
>>>Sigtran mailing list
>>>Sigtran@ietf.org
>>>https://www1.ietf.org/mailman/listinfo/sigtran
>>>
>>> 
>>>
>>>      
>>>
>>-- 
>>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
>>
>>    
>>
>
>_______________________________________________
>Sigtran mailing list
>Sigtran@ietf.org
>https://www1.ietf.org/mailman/listinfo/sigtran
>
>  
>

-- 
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 exim@www1.ietf.org  Fri Jul 25 11:31:08 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA09273
	for <sigtran-archive@odin.ietf.org>; Fri, 25 Jul 2003 11:31:08 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19g4Wo-0001T8-Ha
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 11:30:42 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6PFUgkk005641
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 11:30:42 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19g4WA-0001OX-0B; Fri, 25 Jul 2003 11:30:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19g4Vu-0001O5-16
	for sigtran@optimus.ietf.org; Fri, 25 Jul 2003 11:29:46 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA09227
	for <sigtran@ietf.org>; Fri, 25 Jul 2003 11:29:41 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19g4Vt-00018y-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 11:29:45 -0400
Received: from auds953.usa.alcatel.com ([143.209.238.6])
	by ietf-mx with esmtp (Exim 4.12)
	id 19g4Vd-000186-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 11:29:29 -0400
Received: from alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6PFSNLX015366;
	Fri, 25 Jul 2003 10:28:23 -0500 (CDT)
Message-ID: <3F214C97.8010003@alcatel.com>
Date: Fri, 25 Jul 2003 10:28:23 -0500
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
User-Agent: Mozilla/5.0 (X11; U; SunOS sun4u; en-US; rv:1.0.1) Gecko/20020920 Netscape/7.0
X-Accept-Language: en-us, en
MIME-Version: 1.0
CC: IETF sigtran <sigtran@ietf.org>
Subject: Re: FW: [Sigtran] M2PA Last Call Comment
References: <D73559B38D51D4119CB800508B9527B505AC19D8@mchh266e.mchh.siemens.de>
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

Hanns,

In the 1st diagram, there is no problem.

In the 2nd diagram, the problem is this: What procedures are the 2 ends 
of the M2PA link following? If one is following a procedure that omits 
proving, and the other is following a procedure that requires proving, 
there is a problem.

You are right that what is happening on the TDM side is not an issue 
here. And I think we agree that both ends of the M2PA link must have 
procedures that are consistent with each other. The text I proposed is 
the best way I can think of to do that.

tomG.

Schwarzbauer Hanns Juergen wrote:
> Tom,
> 
> I sent an answer to your mail friday last week, albeit I received it back from the exploder, I never got a response on it, so I'm repeating parts of that mail.
> 
> ========================
>  Question: does STP A care about the fact that STPB has links according to Q.2140? 
>  Answer: NO, NOT AT ALL, STP A operator might not even know that STP B has Q.2140 links.  
> 
>         +-----------+                 +-----------+
>    Q.703|           |                 |           |   Q.2140
>   ------+           |                 |           +--------
>         |           |       Q.703     |           |
>         | STP A     +-----------------+   STP B   |
>         |           |                 |           |
>    Q.703|           |                 |           |   Q.2140
>   ------+           |                 |           +--------
>         |           |                 |           |
>         +-----------+                 +-----------+
> 
> Let's go to an IP based link between STP A and STP B
> 
> Question: should STP A care about the fact that STP B has links according to Q.2140?
> Answer: I see no necessity to deviate from ANSI and ITU-T.
> 
> 
>         +-----------+                 +-----------+
>    Q.703|           |                 |           |   Q.2140
>   ------+           |                 |           +--------
>         |           |       M2PA      |           |
>         | STP A     +-----------------+   STP B   |
>         |           |                 |           |
>    Q.703|           |                 |           |   Q.2140
>   ------+           |                 |           +--------
>         |           |                 |           |
>         +-----------+                 +-----------+
> 
> If STP A wants to prove, he should do that and STP B has to support STP A in that procedure and vice versa.
> But I see no necessity why both sides should perform similiar. 
> 
> In all other cases STP A would have to be informed if e.g. STP-B migrates from pure Q.703 to pure Q.2140 and that's not acceptable.
> =========================
> 
> My point is that using or not using the proving of M2PA is independent from the flavour of MTP level 2 procedure used on the TDM side of the SG. According to Figure 2 in draft 9 there is an MTP 3 on top, effectively decoupling both sides.
> 
> 
> greetings 
> 
> HannsJuergen
> 
> 
> 
>>-----Original Message-----
>>From: Tom George [mailto:Tom.George@alcatel.com]
>>Sent: Thursday, July 24, 2003 10:28 PM
>>Cc: IETF sigtran
>>Subject: Re: FW: [Sigtran] M2PA Last Call Comment
>>
>>
>>That does not seem like a good situation. I see the point 
>>now. Instead 
>>of your suggested text:
>>
>>   The proving period MAY be ommitted if this is allowed by the
>>   applicable standard (e.g., [Q.2140]); however, the same 
>>standard MUST
>>   be followed on both sides of the M2PA link.
>>
>>would it be better to make a more general statement, in order 
>>to avoid 
>>this kind of problem that might exist in other procedures? 
>>Maybe I could 
>>  add to the 2nd paragraph in section 4:
>>
>>=====
>>Except as modified in this document, M2PA SHOULD follow the
>>requirements of the applicable MTP2 specification. These may include
>>[Q.703] or [T1.111]. The same standard MUST be followed on 
>>both ends of 
>>the M2PA link.
>>=====
>>
>>tomG.
   [...]

-- 
* 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.  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 exim@www1.ietf.org  Fri Jul 25 13:00:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA12321
	for <sigtran-archive@odin.ietf.org>; Fri, 25 Jul 2003 13:00:33 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19g5vN-00070F-5T
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 13:00:09 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6PH09o1026898
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 13:00:09 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19g5vF-0006zQ-Ja; Fri, 25 Jul 2003 13:00:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19fxo0-0000he-Hg
	for sigtran@optimus.ietf.org; Fri, 25 Jul 2003 04:20:00 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA28455
	for <sigtran@ietf.org>; Fri, 25 Jul 2003 04:19:55 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19fxnx-0006Nj-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 04:19:57 -0400
Received: from web8107.mail.in.yahoo.com ([203.199.70.121])
	by ietf-mx with smtp (Exim 4.12)
	id 19fxnm-0006Mu-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 04:19:46 -0400
Message-ID: <20030725081449.806.qmail@web8107.mail.in.yahoo.com>
Received: from [202.134.200.4] by web8107.mail.in.yahoo.com via HTTP; Fri, 25 Jul 2003 09:14:49 BST
Date: Fri, 25 Jul 2003 09:14:49 +0100 (BST)
From: =?iso-8859-1?q?andrew=20hall?= <andrew_pot2000@yahoo.co.in>
To: sigtran@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="0-1422672782-1059120889=:246"
Content-Transfer-Encoding: 8bit
Subject: [Sigtran] M3UA M2PA 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>

--0-1422672782-1059120889=:246
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit


Hi all,

I want to ask you why M3UA protocol is at all required if you have M2PA protocol.Because using MTP3 over M2PA will give full IP to SS7 connectivity.So please explain in good detail when specifically M3UA should be used and when M2PA is required.What will be the network scenarios in which M3UA will used instead of M2PA. Please clear the doubt

Best Regards

andrew 

 

SMS using the Yahoo! Messenger;Download latest version.
--0-1422672782-1059120889=:246
Content-Type: text/html; charset=iso-8859-1
Content-Transfer-Encoding: 8bit

<P>Hi all,</P>
<P>I want to ask you why M3UA protocol is at all required if you have M2PA protocol.Because using MTP3 over M2PA will give full IP to SS7 connectivity.So please explain in&nbsp;good detail when specifically M3UA should be used and when M2PA is required.What will be the network scenarios in which M3UA&nbsp;will used instead of M2PA. Please clear the doubt</P>
<P>Best&nbsp;Regards</P>
<P>andrew&nbsp;</P>
<P>&nbsp;</P><p><font face=arial size=-1><img src="http://in.yimg.com/i/in/mobile/pctomob.gif" width=28 height=22 border=0 alt="Yahoo! India Mobile"></a>
<a href="http://in.mobile.yahoo.com/new/pc/" target="_blank"><b>SMS</b></a> using the Yahoo! 
<a href="http://in.mobile.yahoo.com/new/pc/"><b>Messenger</a>;</b></font>
<font face=arial size=-1><b>
<a href="http://in.messenger.yahoo.com/">Download</b></a> latest version.</font>
--0-1422672782-1059120889=:246--

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



From exim@www1.ietf.org  Fri Jul 25 16:00:32 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20785
	for <sigtran-archive@odin.ietf.org>; Fri, 25 Jul 2003 16:00:32 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19g8jW-0007CV-7L
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 16:00:06 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6PK060M027659
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 16:00:06 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19g8jR-0007Bj-IK; Fri, 25 Jul 2003 16:00:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19g8iv-00079F-VB
	for sigtran@optimus.ietf.org; Fri, 25 Jul 2003 15:59:29 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA20706
	for <sigtran@ietf.org>; Fri, 25 Jul 2003 15:59:25 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19g8iu-0003bv-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 15:59:28 -0400
Received: from auds953.usa.alcatel.com ([143.209.238.6])
	by ietf-mx with esmtp (Exim 4.12)
	id 19g8it-0003be-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 15:59:27 -0400
Received: from alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6PJwtLX019382;
	Fri, 25 Jul 2003 14:58:56 -0500 (CDT)
Message-ID: <3F218BFF.5020409@alcatel.com>
Date: Fri, 25 Jul 2003 14:58:55 -0500
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
User-Agent: Mozilla/5.0 (X11; U; SunOS sun4u; en-US; rv:1.0.1) Gecko/20020920 Netscape/7.0
X-Accept-Language: en-us, en
MIME-Version: 1.0
CC: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - FSN
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030714040430.A13583@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

Brian,

Done.

tomG.

Brian F. G. Bidulock wrote:
> M2PA-09 says:
> 
>   If M2PA receives a User Data message with an FSN that is out of order,
>   M2PA SHALL place the link out of service (see section 4.1.6 Link Out
>   of Service).
> 
> This conflicts with Q.703 that ignores an MSU that contains an FSN that
> is other than the next expected FSN (i.e, out of order).  This makes it
> difficult to implement (and is probably the last thing standing in the
> way of implementing) something like the following (that was requested by
> ITU):
> 
>                            _________________
>                           |                 |
>                           |       IWF       |
>                           |_________________|
>                           |        |        |
>                           |  MTP2  |  M2PA  |
>                           |________|________|
>                                |       |
>                            ____|       |___
> 
> If M2PA discards the message as Q.703 does, and the other end of the
> link was actually broken, the link will be go out of service in due
> course anyway.
> 
> Change to:
> 
>   If M2PA receives a User Data message with an FSN that is out of order,
>   M2PA SHALL discard the message.
> 
> --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.  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 exim@www1.ietf.org  Fri Jul 25 16:07:27 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA21064
	for <sigtran-archive@odin.ietf.org>; Fri, 25 Jul 2003 16:07:27 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19g8qD-0007a7-P5
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 16:07:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6PK71Fj029131
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 16:07:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19g8qD-0007ZL-3Z; Fri, 25 Jul 2003 16:07:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19g8pg-0007Y8-D5
	for sigtran@optimus.ietf.org; Fri, 25 Jul 2003 16:06:28 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20930
	for <sigtran@ietf.org>; Fri, 25 Jul 2003 16:06:18 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19g8pZ-0003fB-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 16:06:21 -0400
Received: from auds953.usa.alcatel.com ([143.209.238.6])
	by ietf-mx with esmtp (Exim 4.12)
	id 19g8pY-0003f8-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 16:06:20 -0400
Received: from alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6PK5oLX020196;
	Fri, 25 Jul 2003 15:05:50 -0500 (CDT)
Message-ID: <3F218D9E.2060202@alcatel.com>
Date: Fri, 25 Jul 2003 15:05:50 -0500
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
User-Agent: Mozilla/5.0 (X11; U; SunOS sun4u; en-US; rv:1.0.1) Gecko/20020920 Netscape/7.0
X-Accept-Language: en-us, en
MIME-Version: 1.0
CC: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA last call comment - ANSI primitives for LPO
References: <3F15CB44.1010204@pt.com> <20030716191748.I6725@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

I had put this in section 4 to try to cover for this:

=====
Since M2PA provides MTP3 with an interface and functionality like
MTP2, its internal functioning is similar to that of MTP2.

Except as modified in this document, M2PA SHOULD follow the
requirements of the applicable MTP2 specification. These may include
[Q.703] or [T1.111].

When referring to MTP2 terminology in this document, the terminology
of [Q.703] is used. This does not imply that the requirements of
[Q.703] are to be followed.
=====

I did not think it helpful to have to describe all of the differences in 
terminology between the different brands of MTP2. So I tried to say in 
general that M2PA works like the appropriate MTP2 standard. When I had 
to say something specific about MTP2 terminology, I stayed with Q.703, 
figuring that the reader could substitute the corresponding term if he 
was dealing with ANSI or TTC.

tomG.

Brian F. G. Bidulock wrote:
> Andrew,
> 
> Either that or say "such as" or "e.g." in front of the ITU primitives.
> Or, say something more generic instead of Continue or Resume.
> 
> --brian
> 
> Andrew Booth wrote:                                                                 (Wed, 16 Jul 2003 18:01:40)
> 
>>Hi Tom,
>>
>>The ANSI primitives for exiting processor outage seem to be Resume, 
>>Clear Buffers and Clear RTB.  The corresponding ITU primitives 
>>(continue, flush, ... ) are described explicitly in M2PA draft 9. 
>> Should the ANSI primitives also be described explicitly?
>>
>>Andrew
>>
>>
>>
>>_______________________________________________
>>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.  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 exim@www1.ietf.org  Fri Jul 25 16:39:29 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA23613
	for <sigtran-archive@odin.ietf.org>; Fri, 25 Jul 2003 16:39:29 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19g9LD-0000ne-VU
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 16:39:04 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6PKd3b1003053
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 16:39:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19g9LB-0000n0-6N; Fri, 25 Jul 2003 16:39:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19g9KD-0000mb-1B
	for sigtran@optimus.ietf.org; Fri, 25 Jul 2003 16:38:01 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA23555
	for <sigtran@ietf.org>; Fri, 25 Jul 2003 16:37:55 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19g9KB-0004NZ-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 16:37:59 -0400
Received: from sjmail.ciena.com ([12.7.169.25] helo=w2ksjexg01.ciena.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19g9KA-0004NW-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 16:37:58 -0400
Received: by w2ksjexg01.ciena.com with Internet Mail Service (5.5.2653.19)
	id <PDWTPBKL>; Fri, 25 Jul 2003 13:37:54 -0700
Message-ID: <829F074A10F98943A218DC363D09C92AAE5EFC@w2ksjexg06.oni.com>
From: "Ong, Lyndon" <LyOng@Ciena.com>
To: "'sigtran@ietf.org'" <sigtran@ietf.org>
Date: Fri, 25 Jul 2003 13:41:46 -0700
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Sigtran] m2pa status
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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,

From the drift of the comments on M2PA Last Call it looks like there are places
that the text can be clarified but no critical problems.  Assuming that's the case,
we can move this on to review by our A-Ds after Tom has put out an updated
draft incorporating comments.

If people as for it, we can hold a 1 week LC on the updated draft, otherwise we
can just ask the A-Ds to start their review once the draft comes out.

Cheers,

Lyndon

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



From exim@www1.ietf.org  Fri Jul 25 17:26:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA26763
	for <sigtran-archive@odin.ietf.org>; Fri, 25 Jul 2003 17:26:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19gA4h-0003A1-M0
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 17:26:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6PLQ3US012128
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 17:26:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19gA4g-00039J-0E; Fri, 25 Jul 2003 17:26:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19gA4F-00037g-76
	for sigtran@optimus.ietf.org; Fri, 25 Jul 2003 17:25:35 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA26715
	for <sigtran@ietf.org>; Fri, 25 Jul 2003 17:25:29 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19gA4C-0005HJ-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 17:25:32 -0400
Received: from rainier.illuminet.com ([12.41.110.52])
	by ietf-mx with esmtp (Exim 4.12)
	id 19gA4C-0005GM-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 17:25:32 -0400
Received: from oly1wnexc02.corp.illuminet.com (oly1wnexc02.corp.illuminet.com [10.55.13.13]) by rainier.illuminet.com (8.8.8/8.8.8) with ESMTP id OAA24637; Fri, 25 Jul 2003 14:24:14 -0700 (PDT)
Received: by oly1wnexc02.corp.illuminet.com with Internet Mail Service (5.5.2653.19)
	id <N99YLM4R>; Fri, 25 Jul 2003 14:24:13 -0700
Message-ID: <318266A4431BC549B4C4D6FD3EA6E24D012CE681@ove1wnexm01.corp.illuminet.com>
From: "McCandless, Kevin" <KMcCandless@verisign.com>
To: "'Ong, Lyndon'" <LyOng@Ciena.com>,
        "'sigtran@ietf.org'"
	 <sigtran@ietf.org>
Subject: RE: [Sigtran] m2pa status
Date: Fri, 25 Jul 2003 14:24:12 -0700
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>

Lyndon:

There have been so many comments on v9 that we probably need a chance to review the v10 for a least a week with a LC process before the ADs even see it.

Trust me we are ready for M2PA but we want it right!

K....

> -----Original Message-----
> From: Ong, Lyndon [mailto:LyOng@Ciena.com]
> Sent: Friday, July 25, 2003 3:42 PM
> To: 'sigtran@ietf.org'
> Subject: [Sigtran] m2pa status
> 
> 
> Hi Folks,
> 
> From the drift of the comments on M2PA Last Call it looks 
> like there are places
> that the text can be clarified but no critical problems.  
> Assuming that's the case,
> we can move this on to review by our A-Ds after Tom has put 
> out an updated
> draft incorporating comments.
> 
> If people as for it, we can hold a 1 week LC on the updated 
> draft, otherwise we
> can just ask the A-Ds to start their review once the draft comes out.
> 
> Cheers,
> 
> Lyndon
> 
> _______________________________________________
> 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 exim@www1.ietf.org  Fri Jul 25 17:48:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA27838
	for <sigtran-archive@odin.ietf.org>; Fri, 25 Jul 2003 17:48:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19gAPz-00040q-HU
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 17:48:03 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6PLm32p015406
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 17:48:03 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19gAPx-00040G-IR; Fri, 25 Jul 2003 17:48:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19gAPa-0003zp-8S
	for sigtran@optimus.ietf.org; Fri, 25 Jul 2003 17:47:38 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA27825
	for <sigtran@ietf.org>; Fri, 25 Jul 2003 17:47:32 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19gAPX-0005Qz-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 17:47:35 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19gAPX-0005Qw-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 17:47:35 -0400
Received: from [172.16.1.203] (helo=CCPU00186)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19gAPO-0006qn-D4; Fri, 25 Jul 2003 14:47:26 -0700
From: "VIvek Toky" <vivektoky@ccpu.com>
To: <sigtran@ietf.org>
Cc: <Sebastien.Flaux@etsi.org>
Subject: RE: [Sigtran] Sigtran Plugtest at ETSI 9/03
Date: Fri, 25 Jul 2003 14:52:15 -0700
Message-ID: <HKEBKFBNPIDALDOKLGKDEEJLCDAA.vivektoky@ccpu.com>
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 IMO, Build 9.0.6604 (9.0.2911.0)
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
In-Reply-To: <BE462396B4A022449E0EB2DD4753D05402F26906@vbeexc02.emea.cpqcorp.net>
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,
   can anyone tell, when shall we get the test specifications for UA's ?
Currently I see the first draft of M3UA test specifications on ETSI website.
Is this the one which is going to be used for September'03 M3UA bakeoff ?

thanks,
Vivek.

-----Original Message-----
From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf Of
Anzile, Christophe
Sent: Tuesday, July 08, 2003 3:57 AM
To: sigtran@ietf.org
Subject: RE: [Sigtran] Sigtran Plugtest at ETSI 9/03



All,
same remark applies for SUA (v14 stated @etsi.org) versus v15 out last week.
Christophe.


-----Original Message-----
From: Ken Morneault [mailto:kmorneau@cisco.com]
Sent: Monday, July 07, 2003 4:32 PM
To: sigtran@ietf.org
Subject: [Sigtran] Sigtran Plugtest at ETSI 9/03




All:

Which version of M2PA should be tested at the upcoming
Sigtran plugtest in 9/03?  Currently, the ETSI Plugtest site
indicates version 8, but version 9 was just released.  Version
6 was tested at the last Plugtest.

Please email your preference.

Regards,

Ken


_______________________________________________
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 exim@www1.ietf.org  Fri Jul 25 19:42:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA02104
	for <sigtran-archive@odin.ietf.org>; Fri, 25 Jul 2003 19:42:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19gCCK-0000Vd-SV
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 19:42:05 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6PNg4K2001953
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 19:42:04 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19gCCH-0000VH-Gq; Fri, 25 Jul 2003 19:42:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19gCBX-0000Pj-I8
	for sigtran@optimus.ietf.org; Fri, 25 Jul 2003 19:41:15 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA02075
	for <sigtran@ietf.org>; Fri, 25 Jul 2003 19:41:11 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19gCBV-0006hn-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 19:41:13 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[ZWrG+UBsahqK46uOe2EcWk4vIAZR95R7])
	by ietf-mx with esmtp (Exim 4.12)
	id 19gCBU-0006hV-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 19:41:12 -0400
Received: from ns.pigworks.openss7.net (IDENT:MDhu3M3ymjGdNBmiVC625XkqAJif4U4b@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6PNf3E13492;
	Fri, 25 Jul 2003 17:41:03 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6PNf2Y29525;
	Fri, 25 Jul 2003 17:41:02 -0600
Date: Fri, 25 Jul 2003 17:41:02 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tom George <Tom.George@alcatel.com>
Cc: IETF sigtran <sigtran@ietf.org>
Subject: Re: FW: [Sigtran] M2PA Last Call Comment
Message-ID: <20030725174102.A29483@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tom George <Tom.George@alcatel.com>,
	IETF sigtran <sigtran@ietf.org>
References: <D73559B38D51D4119CB800508B9527B505AC19D8@mchh266e.mchh.siemens.de> <3F214C97.8010003@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: <3F214C97.8010003@alcatel.com>; from Tom.George@alcatel.com on Fri, Jul 25, 2003 at 10:28:23AM -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>

Tom,

Sorry, I missed that note.  Your proposed text sounds fine to me.

--brian

Tom George wrote:                                                                      (Fri, 25 Jul 2003 10:28:23)

[snip]
> >>-----Original Message-----
> >>From: Tom George [mailto:Tom.George@alcatel.com]
> >>Sent: Thursday, July 24, 2003 10:28 PM
> >>Cc: IETF sigtran
> >>Subject: Re: FW: [Sigtran] M2PA Last Call Comment
> >>
> >>
> >>That does not seem like a good situation. I see the point 
> >>now. Instead 
> >>of your suggested text:
> >>
> >>   The proving period MAY be ommitted if this is allowed by the
> >>   applicable standard (e.g., [Q.2140]); however, the same 
> >>standard MUST
> >>   be followed on both sides of the M2PA link.
> >>
> >>would it be better to make a more general statement, in order 
> >>to avoid 
> >>this kind of problem that might exist in other procedures? 
> >>Maybe I could 
> >>  add to the 2nd paragraph in section 4:
> >>
> >>=====
> >>Except as modified in this document, M2PA SHOULD follow the
> >>requirements of the applicable MTP2 specification. These may include
> >>[Q.703] or [T1.111]. The same standard MUST be followed on 
> >>both ends of 
> >>the M2PA link.
> >>=====
> >>
> >>tomG.
>    [...]
> 
> -- 
> * 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.  I do not purport to    *
> *   advertise or represent its position, operations, or activities.   *
> 
-- 
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 exim@www1.ietf.org  Fri Jul 25 19:45:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA02221
	for <sigtran-archive@odin.ietf.org>; Fri, 25 Jul 2003 19:45:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19gCFB-0000dC-Ph
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 19:45:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6PNj122002411
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 19:45:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19gCFB-0000cm-J4; Fri, 25 Jul 2003 19:45:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19gCED-0000bg-Qb
	for sigtran@optimus.ietf.org; Fri, 25 Jul 2003 19:44:01 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA02190
	for <sigtran@ietf.org>; Fri, 25 Jul 2003 19:43:58 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19gCEC-0006ke-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 19:44:00 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[K+nLFY5ItGQPCd+mHPXpAPCIUwdpiD4Y])
	by ietf-mx with esmtp (Exim 4.12)
	id 19gCEA-0006kb-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 19:43:59 -0400
Received: from ns.pigworks.openss7.net (IDENT:/sUgqY7nEXRU05hLVGr4UpZ9pUQtKANm@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6PNhxE13517;
	Fri, 25 Jul 2003 17:43:59 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6PNhwK29548;
	Fri, 25 Jul 2003 17:43:58 -0600
Date: Fri, 25 Jul 2003 17:43:58 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Gaiser Franz <franz.gaiser@siemens.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - LPO
Message-ID: <20030725174358.B29483@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Gaiser Franz <franz.gaiser@siemens.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <99522F0FD7EFD611A47D0002A58EDAB7C84A8F@mchh2a8e.mchh.siemens.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: <99522F0FD7EFD611A47D0002A58EDAB7C84A8F@mchh2a8e.mchh.siemens.de>; from franz.gaiser@siemens.com on Fri, Jul 25, 2003 at 01:33:55PM +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>

Franz,

(2) is already stated.  I have no problem with adding (1) to the description
of handling of buffers at the side experiencing processor outage.

--brian

Gaiser Franz wrote:                                                                 (Fri, 25 Jul 2003 13:33:55)
> Hi All,
> 
> From a developer point of view I would prefer that the statements about
> resync are described more precisely and not in a way that the developer has
> to interpret it especially when he was not involved in the development
> process of M2PA.
>  
> Why don't complement it with the additional remarks i.e. made by Andrew that 
> (1) the BSN in the LS PR is set to the last message received from the M2PA. 
> (2) the sender always increments the FSN by 1 when sending MSUs. 
> 
> best regards, Franz
> 
> 
> 
> 
> 
> -----Ursprungliche Nachricht-----
> Von: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Gesendet: Donnerstag, 24. Juli 2003 18:11
> An: Andrew Booth
> Cc: 'sigtran@ietf.org'
> Betreff: Re: [Sigtran] M2PA Last Call Comment - LPO
> 
> 
> Andrew,
> 
> On the first point, performance related issues are outside the scope
> of the protocol.
> 
> On the second point, nowhere in M2PA-09 does it say to stop incrementing
> BSN.  Just to withhold acknowledgements and buffer during PO.  The
> peer is welcome, as it says in M2PA-09, to resync off of the BSN in the
> LPR.  The fact that it doesn't change the FSN for ANSI does not mean that
> it is not performing the resynchronization.
> 
> Look at the resync wording in M2PA-09 again.
> 
> I don't think it need changing.
> 
> --brian
> 
> Andrew Booth wrote:                                                                 (Thu, 24 Jul 2003 11:01:02)
> > Hi Brian,
> > 
> > I don't think synchronization of sequence numbers after PO is necessary 
> > or desirable in M2PA.  Can you see any advantage to it?
> > 
> > I'd like to highlight this point by bringing up your comment after 
> > example 2 below: in order to avoid potential race conditions the side 
> > sending LS Processor Recovered should set the BSN to the last FSN 
> > received by M2PA.  By setting the BSN in this way we expect is that the 
> > peer will continue sending FSNs "in sequence" i.e. that the 
> > resynchronization will not actually change the next FSN the peer will 
> > send.  If that's the behaviour we want, and I think it is, then I 
> > propose that we specify it directly rather than carry over an 
> > unnecessary procedure from MTP2.
> > 
> > More comments below.
> > 
> > Brian F. G. Bidulock wrote:
> > 
> > >Andrew,
> > >
> > >Please see comments below...
> > >
> > >Andrew Booth wrote:                                                                 (Wed, 23 Jul 2003 21:58:07)
> > >  
> > >
> > >>Hi Franz,
> > >>
> > >>Before answering your question I'll ask another: do we really need to 
> > >>resynchronize sequence numbers in M2PA, or can we just mandate that the 
> > >>sender continues to increment the FSN on every transmitted MSU?
> > >>- This seems simpler than resynchronizing.
> > >>- Since M2PA does not retransmit, the motivation for resynchronization 
> > >>doesn't seem to be present as it does in MTP2.
> > >>- Some of the examples below give examples where resequencing can go 
> > >>wrong.  Although these situations are unlikely, the consequences are 
> > >>unpleasant.
> > >>
> > >>Now for my best answer to your question: Option (2), but if there are no 
> > >>unacknowledged messages then the BSN should be set to the FSN of the 
> > >>last acknowledged message.
> > >>Note also that if the receiver of the LS PR uses the BSN to free 
> > >>messages in the RTB (as you indicated below) then they have to be extra 
> > >>careful when validating BSNs in ensuing Data messages that may have been 
> > >>sent before the LS PR. I don't think this is a good idea.
> > >>
> > >>Some potential issues with resynchronizing sequence numbers:
> > >>
> > >>Note that the sender of LS PR cannot know if the receiver will use the 
> > >>BSN to resynchronize sequence numbers or not, since it is the MTP in RPO 
> > >>that decides when RPO is "long duration" via the MTP3 T1 timer.
> > >>This leads to race conditions in which one side flushes and the other 
> > >>does not, though they are unlikely.  For example, the following could 
> > >>occur if an MSU was delayed longer than MTP3 T1 on the IP network:
> > >>
> > >>
> > >>     MTP3                  M2PA                        M2PA
> > >>       |                     |                          |
> > >>       |                     |<--Data (FSN 0)-----------|
> > >>       |                     |                          |
> > >>       |                     |                     +----|
> > >>       |----LPO------------->|                    /     |
> > >>       |                     |----------LS PO----/----->|
> > >>       |                     |                  /       |
> > >>       |                     |                 /        |<--Flush
> > >>       |----LPR + Continue-->|                /         |
> > >>       |                     |----LS PR------/--------->| resync FSN
> > >>       |                     |    (BSN 0)   /           |
> > >>       |                     |             /            |
> > >>       |                     |<--Data-----+             |
> > >>       |                     |   (FSN 1)                |
> > >>       |                     |                          |
> > >>       |             Discard |<--Data-------------------|
> > >>       |           (bad FSN) |   (FSN 0)                |
> > >>    
> > >>
> > >
> > >This is not really probable.  This would mean that the MTP RPO timer
> > >expires when the LPO timer hasn't (a very narrow window) and that the
> > >Data (FSN 1) was delayed longer than the RPO timer (which shouldn't
> > >really happen).
> > >
> > Agreed, it is unlikely.  If SCTP parameters are set correctly it might 
> > even be impossible.
> > 
> > >
> > >Set your SCTP parameters (or lifetimes) so that messages not
> > >acknowledged within the RPO timeout will drop the SCTP association and
> > >you won't have this problem.
> > >
> > >Do you see?  The "long duration" of the procedure (T1 or about 1.0 to
> > >2.0 seconds) is used so that old messages stored in buffers are not sent
> > >later.  RTB clear is performed immediately before changeover (after T1
> > >expiry).  One doesn't want a recovering processor to transfer old
> > >messages after changeover has occured or a lot of missequencing is going
> > >to happen in the SS7 network.
> > >
> > Yes I understand, though I think MTP3 T1 is 0.5 to 1.2 seconds.
> > 
> > >
> > >Now, if you have SCTP parameters set to allow messages to be buffered IN
> > >THE NETWORK for longer than T1, you have a real big problem: both the
> > >time-controlled changeover and controlled rerouting procedures will
> > >result in missequenced messages, even without processor outage.
> > >
> > Agreed.  Should this be stated explicitly in M2PA?
> > 
> > >
> > >So, if you have a problem above, becuase your SCTP will not drop the
> > >link before delaying a data message longer than T1, you have the greater
> > >problem that every time-controlled changeover or controlled rerouting
> > >procedure for those M2PA links are going to missequence messagses.
> > >
> > >Therefore, you need to tighten your SCTP parameters.  If you cannot
> > >(e.g. the outher end is a really bad implemetnation), you really need to
> > >bring the link down when the duplicate FSN arrives.
> > >
> > >I think you have an error here too: BSN 0 acknowledges the receipt of
> > >data with FSN 0.  After the resync, the next data sent should have BSN
> > >1.
> > >
> > Agreed, the last FSN should be 1 (still invalid).
> > 
> > >
> > >
> > >  
> > >
> > >>The following example could occur if the side in LPO flushes while the 
> > >>side in RPO does not (perhaps they have different T1 timer settings), 
> > >>and if the side in LPO sends a BSN different from the last received by M2PA:
> > >>
> > >>     MTP3                  M2PA                        M2PA
> > >>       |                     |                          |
> > >>       |                     |<--Data (FSN 0)-----------|
> > >>       |                     |                          |
> > >>       |                     |                     +----|
> > >>       |----LPO------------->|                    /     |
> > >>       |                     |----------LS PO----/----->|
> > >>       |                     |                  /       |
> > >>       |                     |                 /        |
> > >>       |               To RB |<--Data---------+         |
> > >>       |                     |   (FSN 1)                |
> > >>       |                     |                          |
> > >>       |----LPR + Flush----->|                          |
> > >>       |                     |----LS PR---------------->|
> > >>       |                     |    (BSN 0)               |<- Continue,
> > >>       |                     |                          |   don't resync FSN
> > >>       |                     |                          |
> > >>       |                     |                          |
> > >>       |             Discard |<--Data (FSN 2)-----------|
> > >>       |    (expected FSN 1) |                          |
> > >>       |                     |                          |
> > >>       |             Discard |<--Data (FSN 3)-----------|
> > >>       |    (expected FSN 1) |                          |
> > >>    
> > >>
> > >
> > >Again, having one side flush without the other is a small time window,
> > >however, why are you sending LS PR (BSN 0)?  I think you should send
> > >LS PR (BSN 1) in this instance.  Then there is no problem.
> > >
> > I agree this is unlikely, but I don't think the time window is 
> > necessarily that small: if one side sets T1 to 0.5s and the other to 
> > 1.2s you have 0.7s to hit this.
> > 
> > >
> > >MTP2 would send BSN 0 here, but it is also discarding messages with the
> > >expectation of retransmission.  M2PA buffers but does not acknowledge
> > >messages.  The purpose of delaying acknowledgement is so that the peer
> > >M2PA keeps these messages in the retransmission buffer for changeover.
> > >(Remember than some variants collect from the RTB instead of clearing
> > >it, unlike ANSI).  But, for M2PA, the BSN should be incremented when
> > >the data is buffered to accomplish what happens here.
> > >
> > >So this one is not a problem either.
> > >
> > Agreed, it's not a problem, but it does highlight the point I made 
> > above: we should set the BSN so that when resynchronizing sequence 
> > numbers the peer doesn't actually have to change anything.  It seems 
> > strange to have an exchange of BSN values for this purpose.
> > 
> > Thoughts?
> > Andrew
> > 
> > >
> > >--brian
> > >
> > >  
> > >
> > >>I think this possible condition provides reasonable incentive to always 
> > >>set the BSN in the LS Processor Recovered to the last message received 
> > >>from the M2PA peer.  In that case, there doesn't seem to be any point in 
> > >>resynching sequence numbers, just let the peer increment the FSN on 
> > >>every MSU.
> > >>
> > >>In summary, these examples are unlikely but why complicate things? 
> > >> Unless there is a good reason to keep the MTP2 resynching procedures 
> > >>lets get rid of them in favour of the simpler approach: the sender 
> > >>always increments the FSN by 1 when sending MSUs.
> > >>
> > >>Thoughts?
> > >>Andrew
> > >>
> > >>Gaiser Franz wrote:
> > >>
> > >>    
> > >>
> > >>>Hi All,
> > >>>
> > >>>implementing Processor Outage Handling of M2PA draft 09 the following questions came up:
> > >>>
> > >>>The Processor Outage chapter 4.1.4 says
> > >>>
> > >>>When the local processor outage conditions ends, M2PA shall send a Link
> > >>>Status Processor Recovered message to its peer. This message is used
> > >>>to signal in the end of the processor outage condition, instead of an
> > >>>MSU or FISU as in MTP2. The BSN value in this message is used by the 
> > >>>peer to resynchronize its sequence numbers, if this is required by
> > >>>MTP2.
> > >>>
> > >>>What does M2PA draft 09 with BSN in this context mean, is it (assume, MTP3 has send Continue Command)
> > >>>
> > >>>(1) the FSN of the last received message before LPO or 
> > >>>(2) the FSN of the last message in the incoming messages that were queued and unacknowledged
> > >>>during the processor outage (with that the remote side can release the messages in its RTB) ? 
> > >>>
> > >>>
> > >>>
> > >>>thanks
> > >>>best regards, Franz 
> > >>>
> > >>>
> > >>>
> > >>>
> > >>>
> > >>>_______________________________________________
> > >>>Sigtran mailing list
> > >>>Sigtran@ietf.org
> > >>>https://www1.ietf.org/mailman/listinfo/sigtran
> > >>>
> > >>> 
> > >>>
> > >>>      
> > >>>
> > >>-- 
> > >>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.
> > >>
> > >>    
> > >>
> > >
> > >  
> > >
> > 
> > -- 
> > 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.
> > 
> 
> -- 
> 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 exim@www1.ietf.org  Fri Jul 25 19:48:28 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA02341
	for <sigtran-archive@odin.ietf.org>; Fri, 25 Jul 2003 19:48:28 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19gCI5-0000qf-V0
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 19:48:01 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6PNm1c3003240
	for sigtran-archive@odin.ietf.org; Fri, 25 Jul 2003 19:48:01 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19gCI5-0000q9-H3; Fri, 25 Jul 2003 19:48:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19gCHo-0000ps-Df
	for sigtran@optimus.ietf.org; Fri, 25 Jul 2003 19:47:44 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA02318
	for <sigtran@ietf.org>; Fri, 25 Jul 2003 19:47:40 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19gCHm-0006oF-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 19:47:42 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[orBF7QJOGtLzHfPlgPpKYdZfPW3qS99n])
	by ietf-mx with esmtp (Exim 4.12)
	id 19gCHl-0006o7-00
	for sigtran@ietf.org; Fri, 25 Jul 2003 19:47:41 -0400
Received: from ns.pigworks.openss7.net (IDENT:GQtn0L5S7xnInvfwgRdcRBaaWriXhD13@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6PNlfE13564;
	Fri, 25 Jul 2003 17:47:41 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6PNlf929597;
	Fri, 25 Jul 2003 17:47:41 -0600
Date: Fri, 25 Jul 2003 17:47:41 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: andrew hall <andrew_pot2000@yahoo.co.in>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA M2PA question
Message-ID: <20030725174741.C29483@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: andrew hall <andrew_pot2000@yahoo.co.in>,
	sigtran@ietf.org
References: <20030725081449.806.qmail@web8107.mail.in.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: <20030725081449.806.qmail@web8107.mail.in.yahoo.com>; from andrew_pot2000@yahoo.co.in on Fri, Jul 25, 2003 at 09:14:49AM +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>

Andrew,

None of the UA protocols are "required".

The rationale behind the various SIGTRAN protocols are
described, in detail, in RFC 2719.

--brian

andrew hall wrote:                                                                  (Fri, 25 Jul 2003 09:14:49)
> 
>    Hi all,
> 
>    I  want  to  ask  you why M3UA protocol is at all required if you have
>    M2PA  protocol.Because  using  MTP3 over M2PA will give full IP to SS7
>    connectivity.So  please  explain in good detail when specifically M3UA
>    should  be  used  and  when  M2PA is required.What will be the network
>    scenarios  in  which  M3UA will used instead of M2PA. Please clear the
>    doubt
> 
>    Best Regards
> 
>    andrew
> 
> 
>    Yahoo! India Mobile [1]SMS using the Yahoo! [2]Messenger;
>    [3]Download latest version.
> 
> References
> 
>    1. http://in.mobile.yahoo.com/new/pc/
>    2. http://in.mobile.yahoo.com/new/pc/
>    3. http://in.messenger.yahoo.com/

-- 
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 exim@www1.ietf.org  Sat Jul 26 10:03:36 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA08014
	for <sigtran-archive@odin.ietf.org>; Sat, 26 Jul 2003 10:03:36 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19gPde-00060c-SR
	for sigtran-archive@odin.ietf.org; Sat, 26 Jul 2003 10:03:11 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6QE3AhS023093
	for sigtran-archive@odin.ietf.org; Sat, 26 Jul 2003 10:03:10 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19gPdV-00060F-Hi; Sat, 26 Jul 2003 10:03:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19gPcv-0005zc-Vo
	for sigtran@optimus.ietf.org; Sat, 26 Jul 2003 10:02:26 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA07974
	for <sigtran@ietf.org>; Sat, 26 Jul 2003 10:02:15 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19gPcp-0001cl-00
	for sigtran@ietf.org; Sat, 26 Jul 2003 10:02:19 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[6XjyCaasS/d1GOsLsvmE3tMYWV9Q9zcJ])
	by ietf-mx with esmtp (Exim 4.12)
	id 19gPcn-0001ci-00
	for sigtran@ietf.org; Sat, 26 Jul 2003 10:02:17 -0400
Received: from ns.pigworks.openss7.net (IDENT:+jSlcARg4BUjPSw85+JHJF6SKS0wQb69@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6QE2GE24570;
	Sat, 26 Jul 2003 08:02:16 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6QE2Fr08976;
	Sat, 26 Jul 2003 08:02:15 -0600
Date: Sat, 26 Jul 2003 08:02:15 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: john.loughney@nokia.com
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Message-ID: <20030726080215.A8560@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: john.loughney@nokia.com, sigtran@ietf.org
References: <DADF50F5EC506B41A0F375ABEB32063658F061@esebe023.ntc.nokia.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <DADF50F5EC506B41A0F375ABEB32063658F061@esebe023.ntc.nokia.com>; from john.loughney@nokia.com on Thu, Jul 03, 2003 at 07:29:23PM +0300
Organization: http://www.openss7.org/
Dsn-Notification-To: <bidulock@openss7.org>
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by gw.openss7.com id h6QE2GE24570
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

John,

Actually, there are a lot of problems with the changes made from
sua-14 to sua-15 as follows:

+   SCTP - Stream Control Transport Protocol.=20

That would be Stream Control TRANSMISSION Protocol.  There are quite a
few places in the document where this error is made.  Even in the
references:

+   Signalling Connection Control Part-User signalling over IP using the=20
+   Stream Control Transport Protocol. The protocol is designed to be=20
+   modular and symmetric, to allow it to work in diverse architectures,=20
+   such as a Signalling Gateway to IP Signalling Endpoint architecture=20
+   as well as a peer-to-peer IP Signalling Endpoint architecture.  =20

-   [2960]         RFC 2960 "Stream Control Transport Protocol" R.=20
+   [2960]         RFC 2960 "Stream Control Transport Protocol", R.=20
                   Stewart, et al, November 2000.=20

Please use "Siganlling" instead of "Signaling".

+   PC - Signaling System no. 7 Point Code.=20

Network Appearance:

+   Network Appearance - The Network Appearance is an SUA local=20
+   reference (typically an integer) shared by SG and AS that together=20
+   with a Signalling Point Code or global title uniquely identifies an=20
+   SS7 node by indicating the specific SS7 network it belongs to.=20

Network appearance has no meaning in relationship to a Global Title.
Global Titles are independent of signalling point code and are only
meaningful to the node performing global title translation.  Strike
global title from the definition.

+   code (PC) and sub-system number (SSN).  The subsystem identified by=20

That should be "subsystem number," no hyphen.

 1.4.1 Support for the transport of SCCP-User Messages=20
=20
    The SUA supports the transfer of SCCP-user messages. The SUA layer=20
-   at the SG and at the ASP support the seamless transport the user=20
-   messages between the SG and the ASP.=20
+   at the signaling gateway and at the ASP support the seamless=20
+   transport the user messages between the signaling gateway and the=20
+   ASP.=20

Above should be "transport of user messages" instead of "transport the
user messages".

 1.4.2 SCCP Protocol Class Support=20
=20
-   Depending upon the SCCP-users supported, the SUA shall support the 4=20
+   Depending upon the SCCP-users supported, the SUA supports the 4=20
    possible SCCP protocol classes transparently.  The SCCP protocol=20
    classes are defined as follows:=20

Why the removal of SHALL?

 1.5.2 Address Mapping at the ASP=20

-   An ASP routes responses to the SGP that it received messages from;=20
-   within the routing context which it is currently active and=20
-   receiving traffic.  The routing context itself is used by the ASP to=20
-   select the SGP.=20
+   An ASP routes all responses to the SGP that it received messages=20
+   from; within the routing context which it is currently active and=20
+   receiving traffic.  The ASP uses the routing context to select the=20
+   SGP.=20

The second sentence is clearer, but the first sure is not.

 1.5.4 SCTP Stream Mapping=20
=20
-   The SUA supports SCTP streams. The SG/AS needs to maintain a list of=20
-   SCTP and SUA-users for mapping purposes.  SCCP-users requiring=20
-   sequenced message transfer need to be sent over a stream supporting=20
-   sequenced delivery.=20
+   The SUA supports SCTP streams. Signaling Gateway SG and Application=20
+   Servers need to maintain a list of SCTP and SUA-users for mapping=20
+   purposes.  SCCP-users requiring sequenced message transfer need to=20
+   be sent over a stream supporting sequenced delivery.=20

Above use "Signalling" instead of "Signaling".  Also, ALL SCTP streams
support sequenced delivery.  Please say "stream with sequenced delivery"
instead of "stream supporting sequenced delivery".

+   The Reserved field is set to 0 in messages sent, is not be examined=20
+   on reception.=20
+

Please make the above a proper sentence.  Something like: "The Reserved
field is set to 0 in messages sent and is not to be examined in messages
received."

    Implementation note: This message covers the following SCCP=20
    messages: unitdata (UDT), extended unitdata (XUDT), long unitdata=20
-   (LUDT).=20
+   (LUDT). Sequence Control is set to 0 for class 0 messages.=20

This change is incorrect.  Please remove this addition.  We discussed thi=
s
long ago and Sequence Control is required in class 0 messages for some
network configurations and for proper MTP relay accross a relay point.
It might not be provided by the SCCP/SUA-User for class 0, but then the
SUA layer (at ASP, IPSP or SGP) must ultimately select a loadsharing valu=
e.
That is why we made the parameter MANDATORY for all protocol classes.

    Note 1:    When the SSN is included, the DUNA message corresponds to=20
-              the SCCP N-STATE primitive. When SSN is not, the DUNA=20
-              message corresponds to the SCCP N-PCSTATE primitive.  =20
+              the SCCP N-STATE primitive, and contains just one point=20
+              code in the Affected Point Code parameter. When SSN is=20
+              not, the DUNA message corresponds to the SCCP N-PCSTATE=20
+              primitive.  =20

The DUNA can contain multiple affected point codes even if the SSN is
included.  This is particularly the case when the SG or IPSP supports
the same subsystem on multiple point codes.  Please put it back the
way it was.

-   Note 1:    When the SSN is included, the SCON message corresponds to=20
-              the SCCP N-STATE primitive. When SSN is not included, the=20
-              SCON message corresponds to the SCCP N-PCSTATE primitive.=20
+   Note 1:    When the SSN is included it must be equal to 1. If the=20
+              SSN is included the SCON corresponds to the N-PCSTATE=20
+              primitive used to convey the Restricted Importance Level=20
+              (RIL) to the SCCP user. When the SSN is not included, the=20
+              SCON message corresponds to the SCCP N-PCSTATE primitive=20
+              reporting signalling point or network congestion status.=20

No, the SSN doesn't have to be 1.  Where did you get that idea?

-   Note 1:    The DUPU corresponds to the SCCP N-PCSTATE primitive. =20
+   Note 1:    The DUPU corresponds to the SCCP N-PCSTATE primitive, and=20
+               contains in the Affected Point Code parameter just one=20
+              point code. =20

Particularly as User/Cause is always the SCCP SI number, multiple affecte=
d
point codes can occur.  Please put it back the way it was.

 3.4.6 Destination Restricted (DRST)=20
=20
    The DRST message is optionally sent from the SG to all concerned=20
    ASPs to indicate that the SG has determined that one or more=20
    destinations are now restricted from the point of view of the SG, or=20
    in response to a DAUD message if appropriate. The SUA layer at the=20
    ASP is expected to send traffic to the affected destination via an=20
    alternate SG of equal priority, but only if such an alternate route=20
-   exists and is available. If the affected destination is currently=20
-   considered unavailable by the ASP, the peer should be informed that=20
-   traffic to the affected destination can be resumed.  In this case,=20
-   the SUA layer should route the traffic through the SG initiating the=20
-   DRST message.=20
+   exists and is available. If the ASP currently considers the affected=20
+   destination unavailable, the peer should be informed that traffic to=20
+   the affected destination could be resumed.  In this case, the SUA=20
+   layer should route the traffic through the SG initiating the DRST=20
+   message.=20

This is only true if the DRST corresponds to TFR.  Note also that the
DRST cannot correspond to N-PCSTATE because there is no N-PCSTATE primiti=
ve
for PC restriction (it is an MTP concept only).  This message corresponds=
 to
N-COORD that has little to do with the description above, no matter how i=
s is
worded exactly.

-            0x17      Routing Key Change Refused=20
+            0x17      Not Used in SUA=20

Good, but you forgot to remove the statement later:

   The "Routing Key Change Refused" error is sent when the SG refuses a=20
   change in the Routing Key parameters.  =20


-   User/Cause                           0x010C=20
+   Cause / User                         0x010C=20

Everwhere else in the document it is referred to as User/Cause, execpt he=
re:

 3.10.12 Cause / User=20

Change both of these to "User/Cause".

     0010     This is most commonly used in North American networks.=20
               The Translation Type implicitly determines Nature of=20
               Address and Numbering Plan. This data can be configured=20
               in the SG. The number of digits is always even and=20
               determined by the SCCP address length. =20
-    0011                     Numbering Plan and Translation Type are tak
+    0011                     Numbering Plan and Translation Type are tak=
en over. It=20
               is implicitly assumed that the Nature of Address =3D=20
               Unknown. =20
-    0100                     This format is used in international networ
+    0100                     This format is used in international networ=
ks and most=20
               commonly in networks outside North America. All=20
               information to populate the source address is present in=20
               the SCCP Address.=20

Above is text extending past 72 columns that was truncated in sua-14.  Pl=
ease
shift the text over.

      0              Unknown=20
      1 - 63         International services=20
      64 - 127       Spare=20
-     128 > 254      National network specific=20
+     128 =FB 254      National network specific=20
      255            Reserved=20

Non ASCII character, again.

           Value     Description =20
            0x0      No special options=20
            0x1      Return message on error=20
=20
-                    Bits 3-7 are spare and SHOULD be coded zero, and MUS
+                    Bits 3-7 are spare and SHOULD be coded zero, and MUS=
T be=20
          ignored by the receiver.=20

Again text extending beyond 72 columns.

 4.3 AS and ASP State Maintenance=20
=20
-   The SUA layer on the SGP maintains the state of each remote ASP, in=20
-   each Application Server that the ASP is configured to receive=20
+   The SUA layer on on the SGP maintains the state of each remote ASP,=20
+   in each Application Server that the ASP is configured to receive=20
    traffic, as input to the SUA message distribution function. =20

All this change did was introduce a typo: "on on".

+      1 IPSP Single Exchange (SE) model. Only a single exchange of=20
+        ASPTM or ASPSM messages is needed to change the IPSP state.=20
+        This means that a set of request from one end and acknowledge=20
+        from the other will be enough. This configuration requires=20
+        static RK configuration. =20
+
+      2 IPSP Double Exchange (DE) model. Both IPSPs have to send=20
+        request messages and both IPSPs have to acknowledge the request=20
+        messages from the other end. This results in a double exchange=20
+        of ASPTM and ASPSM message, one from each end. This=20
+        configuration supports dynamic routing key configuration by=20
+        using RKM messages in the same way as ASP-SGP scenario. =20

Remove mention of RK configuration.  This has not been agreed.

+   In order to ensure interoperability, an M3UA implementation=20
+   supporting IPSP communication MUST support IPSP SE model and MAY=20
+   implement IPSP DE model. =20

Change M3UA to SUA above.

+   * Reception of messages from the peer M3UA layer at the ASP/IPSP; =20
+   * Reception of some messages from the peer M3UA layer at other=20
+   ASPs/IPSPs in the AS (e.g., ASP Active message indicating=20

Change M3UA to SUA above.

+   ASP-DOWN: The remote M3UA peer at the ASP/IPSP is unavailable and/or=20
+   the related SCTP association is down.  Initially all ASPs/IPSPs will=20
+   be in this state. An ASP/IPSP in this state SHOULD NOT be sent any=20
+   SUA messages, with the exception of Heartbeat, ASP Down Ack and Error=
=20

Change M3UA to SUA above.

+   ASP-INACTIVE: The remote M3UA peer at the ASP/IPSP is available (and=20
+   the related SCTP association is up) but application traffic is=20
+   stopped. In this state the ASP/IPSP SHOULD NOT be sent any DATA or=20
+   SSNM messages for the AS for which the ASP/IPSP is inactive. =20
+
+   ASP-ACTIVE: The remote M3UA peer at the ASP/IPSP is available and=20
+   application traffic is active (for a particular Routing Context or=20
+   set of Routing Contexts). =20

Change M3UA to SUA above.

+   SCTP CDI: The SCTP CDI denotes the local SCTP layer's Communication=20
+   Down Indication to the Upper Layer Protocol (M3UA) on an SGP.  The=20
+   local SCTP layer will send this indication when it detects the loss=20
+   of connectivity to the ASP's peer SCTP layer.  SCTP CDI is understood=
=20
+   as either a SHUTDOWN_COMPLETE notification or COMMUNICATION_LOST=20
+   notification from the SCTP layer. =20
+
+   SCTP RI: The local SCTP layer's Restart indication to the upper layer=
=20
+   protocol (M3UA) on an SG.  The local SCTP will send this indication=20
+   when it detects a restart from the ASP's/IPSP's peer SCTP layer. =20

You now have two SCTP CDI and SCTP RI paragraphs.  Those above and these
below:  Delete the ones above because they have M3UA instead of SUA.

   SCTP CDI: The SCTP CDI denotes the local SCTP layer's Communication=20
   Down Indication to the Upper Layer Protocol (SUA) on an SGP. The=20
   local SCTP layer will send this indication when it detects the loss=20
   of connectivity to the ASP's peer SCTP layer.  SCTP CDI is understood=20
   as either a SHUTDOWN_COMPLETE notification or COMMUNICATION_LOST=20
   notification from the SCTP layer.=20

   SCTP RI: The local SCTP layer's Restart indication to the upper layer=20
   protocol (SUA) on an SG.  The local SCTP will send this indication=20
   when it detects a restart from the ASP's peer SCTP layer.=20

The diagram now says "Other IPSP in AS Overrides".  Please say
"Other ASP/IPSP in AS Overrides".

                |      Other   +-------|              |=20
-               |   ASP in AS  |       +--------------+=20
+               |   IPSP in AS |       +--------------+ =20
                |   Overrides  |           ^     |=20

 4.3.4.2 ASP Down Procedures=20
=20
    The ASP will send an ASP Down message to an SGP when the ASP wishes=20
    to be removed from service in all Application Servers that it is a=20
-   member and no longer receive any DATA, SSNM or ASPTM messages.  This=20
-   action MAY be initiated at the ASP by an M-ASP_DOWN request primitive
-   from Layer Management or MAY be initiated automatically by an SUA=20
-   management function.   =20
+   member and no longer receive any Connectionless or Connection -=20
+   Oriented, SNM or ASPTM messages.  This action MAY be initiated at the=
=20
+   ASP by an M-ASP_DOWN request primitive from Layer Management or MAY=20
+   be initiated automatically by an SUA management function.=20

Above change SNM to SSNM.

    When an ASP wishes to withdraw from receiving traffic within an AS,=20
-   the ASP sends an ASP Inactive message to the SGP or IPSP.  This=20
-   action MAY be initiated at the ASP by an M-ASP_INACTIVE request=20
-   primitive from Layer Management or MAY be initiated automatically by=20
-   an SUA management function.  In the case where an ASP is processing=20
-   the traffic for more than one Application Server across a common SCTP
-   association, the ASP Inactive message contains one or more Routing=20
-   Contexts to indicate for which Application Servers the ASP Inactive=20
-   message applies.  In the case where an ASP Inactive message does not=20
-   contain a Routing Context parameter, the receiver must know, via=20
-   configuration data, which Application Servers the ASP is a member and
-   move the ASP to the ASP-INACTIVE state in each all Application=20
-   Servers.  In the case of an Override mode AS, where another ASP has=20
-   already taken over the traffic within the AS with an ASP Active=20
-   ("Override") message, the ASP that sends the ASP Inactive message is=20
-   already considered by the SGP to be in state ASP-INACTIVE.  An ASP=20
-   Inactive Ack message is sent to the ASP, after ensuring that all=20
-   traffic is stopped to the ASP. =20
+   or the ASP wants to initiate the process of activation, the ASP sends=
=20
+   an ASP Inactive message to the SGP or IPSP. =20

Change above to "initiate the process of deactivation" instead of "initia=
te
the process of activation".

Also, above you have delted the description of the ASP Inactive procedure=
 for
Override mode.  Please put back:

-                     In the case where an ASP Inactive message does not=20
-   contain a Routing Context parameter, the receiver must know, via=20
-   configuration data, which Application Servers the ASP is a member and
-   move the ASP to the ASP-INACTIVE state in each all Application=20
-   Servers.

and

-             In the case of an Override mode AS, where another ASP has=20
-   already taken over the traffic within the AS with an ASP Active=20
-   ("Override") message, the ASP that sends the ASP Inactive message is=20
-   already considered by the SGP to be in state ASP-INACTIVE.  An ASP=20
-   Inactive Ack message is sent to the ASP, after ensuring that all=20
-   traffic is stopped to the ASP. =20

The following addition:

+4.8 NIF Not Available on SGP =20
+
+   If the SG (all the SGPs) is isolated from the NIF, then all the users=
=20
+   are isolated from the SS7 network.  A DUNA(*) message MUST be sent=20
+   from the SGPs to all the ASPs.  =20
+
+   If only one SGP in the SG is isolated entirely from the NIF, the SGP=20
+   SHOULD abort its associations. An alternative would be for the SGP to=
=20
+   send ASP Down Ack.=20
+
+   If one or more SGP suffer a partial failure (where aborting the=20
+   association(s) would cause all active AS(es) to fail), then the SGP=20
+   MUST send DUNA messages for the affected SPC(es).  This is the case=20
+   where an SGP can continue to service one or more active AS(es), but=20
+   due to a partial failure it is unable to service other(s) active=20
+   AS(es).=20

Please remove the entire addition.  The M3UA IG has removed this.  That i=
s
the problem with adding stuff from the M3UA IG.  The M3UA IG has not gone
through Last Call.

+   [SIGSEC]       J. Loughney, M. Tuexen, J. Pastor-Balbas, "Security=20
+                  Considerations for SIGTRAN Protocols", Work in=20
                   Progress. =20

There is a Work In Progress as a normative reference.  This draft will be
held until this reference becomes an RFC.

--brian


John Loughney wrote:                                (Thu, 03 Jul 2003 19:=
29:23)
> Hi all,
>=20
> I have updated SUA, I have tried to incorporate some of the major issue=
s
> from the M3UA Implementation Guide.  All interested parties, please
> give the document a quick read, because I would like to send this to th=
e
> IESG for final approval.
>=20
> thanks,
> John
>=20
> > -----Original Message-----
> > From: ext Internet-Drafts@ietf.org [mailto:Internet-Drafts@ietf.org]
> > Sent: 03 July, 2003 18:33
> > Cc: sigtran@ietf.org
> > Subject: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
> >=20
> >=20
> > A New Internet-Draft is available from the on-line=20
> > Internet-Drafts directories.
> > This draft is a work item of the Signaling Transport Working=20
> > Group of the IETF.
> >=20
> > 	Title		: Signalling Connection Control Part=20
> > User Adaptation=20
> >                           Layer (SUA)
> > 	Author(s)	: J. Loughney, G. Sidebottom
> > 	Filename	: draft-ietf-sigtran-sua-15.txt
> > 	Pages		: 118
> > 	Date		: 2003-7-3
> > =09
> > This Internet Draft defines a protocol for the transport of=20
> > any         Signalling Connection Control Part-User=20
> > signalling over IP using the   Stream Control Transport=20
> > Protocol. The protocol is designed to be=20
> > modular and symmetric, to allow it to work in diverse architectures,=20
> > such as a Signalling Gateway to IP Signalling Endpoint architecture=20
> > as well as a peer-to-peer IP Signalling Endpoint architecture.
> >=20
> > A URL for this Internet-Draft is:
> > http://www.ietf.org/internet-drafts/draft-ietf-sigtran-sua-15.txt
> >=20
> > To remove yourself from the IETF Announcement list, send a message to=
=20
> > ietf-announce-request with the word unsubscribe in the body=20
> > of the message.
> >=20
> > Internet-Drafts are also available by anonymous FTP. Login=20
> > 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-sua-15.txt".
> >=20
> > A list of Internet-Drafts directories can be found in
> > http://www.ietf.org/shadow.html=20
> > or ftp://ftp.ietf.org/ietf/1shadow-sites.txt
> >=20
> >=20
> > Internet-Drafts can also be obtained by e-mail.
> >=20
> > Send a message to:
> > 	mailserv@ietf.org.
> > In the body type:
> > 	"FILE /internet-drafts/draft-ietf-sigtran-sua-15.txt".
> > =09
> > 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=20
> > 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.
> > 	=09
> > 	=09
> > Below is the data which will enable a MIME compliant mail reader
> > implementation to automatically retrieve the ASCII version of the
> > Internet-Draft.
> >=20
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran

--=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



From exim@www1.ietf.org  Mon Jul 28 03:59:35 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA22670
	for <sigtran-archive@odin.ietf.org>; Mon, 28 Jul 2003 03:59:35 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19h2uS-0003ll-7h
	for sigtran-archive@odin.ietf.org; Mon, 28 Jul 2003 03:59:08 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6S7x8KH014490
	for sigtran-archive@odin.ietf.org; Mon, 28 Jul 2003 03:59:08 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19h2uL-0003lJ-ID; Mon, 28 Jul 2003 03:59:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19h2tb-0003kK-H8
	for sigtran@optimus.ietf.org; Mon, 28 Jul 2003 03:58:15 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA22585
	for <sigtran@ietf.org>; Mon, 28 Jul 2003 03:58:11 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19h2sV-0006UP-00
	for sigtran@ietf.org; Mon, 28 Jul 2003 03:57:07 -0400
Received: from thoth.sbs.de ([192.35.17.2])
	by ietf-mx with esmtp (Exim 4.12)
	id 19h2sU-0006UM-00
	for sigtran@ietf.org; Mon, 28 Jul 2003 03:57:06 -0400
Received: from mail1.siemens.de (mail1.siemens.de [139.23.33.14])
	by thoth.sbs.de (8.11.7/8.11.7) with ESMTP id h6S7v5Z23003;
	Mon, 28 Jul 2003 09:57:05 +0200 (MEST)
Received: from moody.mchh.siemens.de (moody.mchh.siemens.de [139.21.205.85])
	by mail1.siemens.de (8.11.7/8.11.7) with ESMTP id h6S7v5K26791;
	Mon, 28 Jul 2003 09:57:05 +0200 (MEST)
Received: from mchh273e.demchh201e.icn.siemens.de (mchh273e.mchh.siemens.de [139.21.200.83])
	by moody.mchh.siemens.de (8.9.3/8.9.1) with ESMTP id JAA16769;
	Mon, 28 Jul 2003 09:57:04 +0200 (MET DST)
Received: by mchh273e.mchh.siemens.de with Internet Mail Service (5.5.2656.59)
	id <PLMKQ3P0>; Mon, 28 Jul 2003 09:56:28 +0200
Message-ID: <D73559B38D51D4119CB800508B9527B505AC19EB@mchh266e.mchh.siemens.de>
From: Schwarzbauer Hanns Juergen <hannsjuergen.schwarzbauer@siemens.com>
To: "'Tom George'" <Tom.George@alcatel.com>
Cc: IETF sigtran <sigtran@ietf.org>
Subject: RE: FW: [Sigtran] M2PA Last Call Comment
Date: Mon, 28 Jul 2003 09:56:21 +0200
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2656.59)
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>

Tom,

see my comments inline (soory for the late response)

greetings

HannsJuergen

> -----Original Message-----
> From: Tom George [mailto:Tom.George@alcatel.com]
> Sent: Friday, July 25, 2003 5:28 PM
> Cc: IETF sigtran
> Subject: Re: FW: [Sigtran] M2PA Last Call Comment
> 
> 
> Hanns,
> 
> In the 1st diagram, there is no problem.
> 
> In the 2nd diagram, the problem is this: What procedures are 
> the 2 ends 
> of the M2PA link following? If one is following a procedure 
> that omits 
> proving, and the other is following a procedure that requires 
> proving, 
> there is a problem.

[HJS] Let's think about this issue the other way round: 
both STP's only have Q.703 to the rest of the SS7 network but use Q.2140 between then. 
I don't remember any discussion in ITU-T about any dependence for proving on the SAAL links on the other Q.703 links on both sides. It was simply "another level 2 protocol". So there was no problem at all, especially as all data links are decoupled from each other by MTP L3!

> 
> You are right that what is happening on the TDM side is not an issue 
> here. And I think we agree that both ends of the M2PA link must have 
> procedures that are consistent with each other. The text I 
> proposed is 
> the best way I can think of to do that.

[HJS] I think the text is misleading: the words "if this is allowed by the applicable MTP2 standard" couple the M2PA procedure to the TDM side.
My proposal reads liks:

section 4.1.3
The proving period MAY be omitted upon bilateral agreement between the concerned peers.
                                     +------------+
                                     |            |  Q.2140
                            +--------+            +---------
                            |        |   STP B    |
        +----------+        |        |            |  Q.2140
 Q.703  |          | M2PA   |        |            +----------
 -------+          +--------+     +--+            |
        | STP A    |              |  +------------+
 Q.703  |          +--------+     |
 -------+          | M2PA   |     |
        |          |        |     |
        +----------+        |     |
                            |     |  +------------+
                            |     +--+            |  Q.703
                            |        |            +---------
                            +--------+  STP C     |
                                     |            |  Q.703
                                     |            +---------
                                     |            |
                                     +------------+
In that scenario I even doubt that your text allows STP A and STP B to omit the proving. 
The text could be interpreted to allow omission of proving only if this is applicable on both ends.
I consider this correlation as not necessary at all, as all MTP L2 have to be considered independent from each other (due to MTP3). M2PA from the beginning was considered a "new Level 2" and therefore should not be designed a way to have built in dependencies from existing level 2 protocols. 

> 
> tomG.
> 
> Schwarzbauer Hanns Juergen wrote:
> > Tom,
> > 
> > I sent an answer to your mail friday last week, albeit I 
> received it back from the exploder, I never got a response on 
> it, so I'm repeating parts of that mail.
> > 
> > ========================
> >  Question: does STP A care about the fact that STPB has 
> links according to Q.2140? 
> >  Answer: NO, NOT AT ALL, STP A operator might not even know 
> that STP B has Q.2140 links.  
> > 
> >         +-----------+                 +-----------+
> >    Q.703|           |                 |           |   Q.2140
> >   ------+           |                 |           +--------
> >         |           |       Q.703     |           |
> >         | STP A     +-----------------+   STP B   |
> >         |           |                 |           |
> >    Q.703|           |                 |           |   Q.2140
> >   ------+           |                 |           +--------
> >         |           |                 |           |
> >         +-----------+                 +-----------+
> > 
> > Let's go to an IP based link between STP A and STP B
> > 
> > Question: should STP A care about the fact that STP B has 
> links according to Q.2140?
> > Answer: I see no necessity to deviate from ANSI and ITU-T.
> > 
> > 
> >         +-----------+                 +-----------+
> >    Q.703|           |                 |           |   Q.2140
> >   ------+           |                 |           +--------
> >         |           |       M2PA      |           |
> >         | STP A     +-----------------+   STP B   |
> >         |           |                 |           |
> >    Q.703|           |                 |           |   Q.2140
> >   ------+           |                 |           +--------
> >         |           |                 |           |
> >         +-----------+                 +-----------+
> > 
> > If STP A wants to prove, he should do that and STP B has to 
> support STP A in that procedure and vice versa.
> > But I see no necessity why both sides should perform similiar. 
> > 
> > In all other cases STP A would have to be informed if e.g. 
> STP-B migrates from pure Q.703 to pure Q.2140 and that's not 
> acceptable.
> > =========================
> > 
> > My point is that using or not using the proving of M2PA is 
> independent from the flavour of MTP level 2 procedure used on 
> the TDM side of the SG. According to Figure 2 in draft 9 
> there is an MTP 3 on top, effectively decoupling both sides.
> > 
> > 
> > greetings 
> > 
> > HannsJuergen
> > 
> > 
> > 
> >>-----Original Message-----
> >>From: Tom George [mailto:Tom.George@alcatel.com]
> >>Sent: Thursday, July 24, 2003 10:28 PM
> >>Cc: IETF sigtran
> >>Subject: Re: FW: [Sigtran] M2PA Last Call Comment
> >>
> >>
> >>That does not seem like a good situation. I see the point 
> >>now. Instead 
> >>of your suggested text:
> >>
> >>   The proving period MAY be ommitted if this is allowed by the
> >>   applicable standard (e.g., [Q.2140]); however, the same 
> >>standard MUST
> >>   be followed on both sides of the M2PA link.
> >>
> >>would it be better to make a more general statement, in order 
> >>to avoid 
> >>this kind of problem that might exist in other procedures? 
> >>Maybe I could 
> >>  add to the 2nd paragraph in section 4:
> >>
> >>=====
> >>Except as modified in this document, M2PA SHOULD follow the
> >>requirements of the applicable MTP2 specification. These may include
> >>[Q.703] or [T1.111]. The same standard MUST be followed on 
> >>both ends of 
> >>the M2PA link.
> >>=====
> >>
> >>tomG.
>    [...]
> 
> -- 
> * 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.  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 exim@www1.ietf.org  Tue Jul 29 08:17:34 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA20068
	for <sigtran-archive@odin.ietf.org>; Tue, 29 Jul 2003 08:17:34 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hTPg-0000ON-Gt
	for sigtran-archive@odin.ietf.org; Tue, 29 Jul 2003 08:17:08 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6TCH83W001503
	for sigtran-archive@odin.ietf.org; Tue, 29 Jul 2003 08:17:08 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hTPZ-0000O0-Af; Tue, 29 Jul 2003 08:17:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hTOd-0000Ml-En
	for sigtran@optimus.ietf.org; Tue, 29 Jul 2003 08:16:03 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA20002
	for <sigtran@ietf.org>; Tue, 29 Jul 2003 08:15:59 -0400 (EDT)
From: john.loughney@nokia.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hTOc-0004nX-00
	for sigtran@ietf.org; Tue, 29 Jul 2003 08:16:02 -0400
Received: from mgw-x4.nokia.com ([131.228.20.27])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hTOb-0004nT-00
	for sigtran@ietf.org; Tue, 29 Jul 2003 08:16:01 -0400
Received: from esvir03nok.nokia.com (esvir03nokt.ntc.nokia.com [172.21.143.35])
	by mgw-x4.nokia.com (Switch-2.2.6/Switch-2.2.6) with ESMTP id h6TCFxJ01613
	for <sigtran@ietf.org>; Tue, 29 Jul 2003 15:15:59 +0300 (EET DST)
Received: from esebh003.NOE.Nokia.com (unverified) by esvir03nok.nokia.com
 (Content Technologies SMTPRS 4.2.5) with ESMTP id <T63bb0a0956ac158f23077@esvir03nok.nokia.com>;
 Tue, 29 Jul 2003 15:15:59 +0300
Received: from esebe001.NOE.Nokia.com ([172.21.138.30]) by esebh003.NOE.Nokia.com with Microsoft SMTPSVC(5.0.2195.6139);
	 Tue, 29 Jul 2003 15:15:59 +0300
Received: from esebe023.NOE.Nokia.com ([172.21.138.115]) by esebe001.NOE.Nokia.com with Microsoft SMTPSVC(5.0.2195.6139);
	 Tue, 29 Jul 2003 15:15:58 +0300
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"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Date: Tue, 29 Jul 2003 15:15:58 +0300
Message-ID: <DADF50F5EC506B41A0F375ABEB32063658F222@esebe023.ntc.nokia.com>
Thread-Topic: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
Thread-Index: AcNTfomp3fTsS/MDSQW7CU2AHJnPEgCTJBRQ
To: <bidulock@openss7.org>
Cc: <sigtran@ietf.org>
X-OriginalArrivalTime: 29 Jul 2003 12:15:58.0787 (UTC) FILETIME=[2B312930:01C355CB]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Hi Brian,

Thanks for the review, I will look at updating this as soon
as my vacation is over.

thanks,
John

> -----Original Message-----
> From: ext Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: 26 July, 2003 17:02
> To: Loughney John (NRC/Helsinki)
> Cc: sigtran@ietf.org
> Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
>=20
>=20
> John,
>=20
> Actually, there are a lot of problems with the changes made from
> sua-14 to sua-15 as follows:
>=20
> +   SCTP - Stream Control Transport Protocol.=20
>=20
> That would be Stream Control TRANSMISSION Protocol.  There are quite a
> few places in the document where this error is made.  Even in the
> references:
>=20
> +   Signalling Connection Control Part-User signalling over=20
> IP using the=20
> +   Stream Control Transport Protocol. The protocol is designed to be=20
> +   modular and symmetric, to allow it to work in diverse=20
> architectures,=20
> +   such as a Signalling Gateway to IP Signalling Endpoint=20
> architecture=20
> +   as well as a peer-to-peer IP Signalling Endpoint architecture.  =20
>=20
> -   [2960]         RFC 2960 "Stream Control Transport Protocol" R.=20
> +   [2960]         RFC 2960 "Stream Control Transport Protocol", R.=20
>                    Stewart, et al, November 2000.=20
>=20
> Please use "Siganlling" instead of "Signaling".
>=20
> +   PC - Signaling System no. 7 Point Code.=20
>=20
> Network Appearance:
>=20
> +   Network Appearance - The Network Appearance is an SUA local=20
> +   reference (typically an integer) shared by SG and AS that=20
> together=20
> +   with a Signalling Point Code or global title uniquely=20
> identifies an=20
> +   SS7 node by indicating the specific SS7 network it belongs to.=20
>=20
> Network appearance has no meaning in relationship to a Global Title.
> Global Titles are independent of signalling point code and are only
> meaningful to the node performing global title translation.  Strike
> global title from the definition.
>=20
> +   code (PC) and sub-system number (SSN).  The subsystem=20
> identified by=20
>=20
> That should be "subsystem number," no hyphen.
>=20
>  1.4.1 Support for the transport of SCCP-User Messages=20
> =20
>     The SUA supports the transfer of SCCP-user messages. The=20
> SUA layer=20
> -   at the SG and at the ASP support the seamless transport the user=20
> -   messages between the SG and the ASP.=20
> +   at the signaling gateway and at the ASP support the seamless=20
> +   transport the user messages between the signaling gateway and the=20
> +   ASP.=20
>=20
> Above should be "transport of user messages" instead of "transport the
> user messages".
>=20
>  1.4.2 SCCP Protocol Class Support=20
> =20
> -   Depending upon the SCCP-users supported, the SUA shall=20
> support the 4=20
> +   Depending upon the SCCP-users supported, the SUA supports the 4=20
>     possible SCCP protocol classes transparently.  The SCCP protocol=20
>     classes are defined as follows:=20
>=20
> Why the removal of SHALL?
>=20
>  1.5.2 Address Mapping at the ASP=20
>=20
> -   An ASP routes responses to the SGP that it received=20
> messages from;=20
> -   within the routing context which it is currently active and=20
> -   receiving traffic.  The routing context itself is used by=20
> the ASP to=20
> -   select the SGP.=20
> +   An ASP routes all responses to the SGP that it received messages=20
> +   from; within the routing context which it is currently active and=20
> +   receiving traffic.  The ASP uses the routing context to=20
> select the=20
> +   SGP.=20
>=20
> The second sentence is clearer, but the first sure is not.
>=20
>  1.5.4 SCTP Stream Mapping=20
> =20
> -   The SUA supports SCTP streams. The SG/AS needs to=20
> maintain a list of=20
> -   SCTP and SUA-users for mapping purposes.  SCCP-users requiring=20
> -   sequenced message transfer need to be sent over a stream=20
> supporting=20
> -   sequenced delivery.=20
> +   The SUA supports SCTP streams. Signaling Gateway SG and=20
> Application=20
> +   Servers need to maintain a list of SCTP and SUA-users for mapping=20
> +   purposes.  SCCP-users requiring sequenced message=20
> transfer need to=20
> +   be sent over a stream supporting sequenced delivery.=20
>=20
> Above use "Signalling" instead of "Signaling".  Also, ALL SCTP streams
> support sequenced delivery.  Please say "stream with=20
> sequenced delivery"
> instead of "stream supporting sequenced delivery".
>=20
> +   The Reserved field is set to 0 in messages sent, is not=20
> be examined=20
> +   on reception.=20
> +
>=20
> Please make the above a proper sentence.  Something like:=20
> "The Reserved
> field is set to 0 in messages sent and is not to be examined=20
> in messages
> received."
>=20
>     Implementation note: This message covers the following SCCP=20
>     messages: unitdata (UDT), extended unitdata (XUDT), long unitdata=20
> -   (LUDT).=20
> +   (LUDT). Sequence Control is set to 0 for class 0 messages.=20
>=20
> This change is incorrect.  Please remove this addition.  We=20
> discussed this
> long ago and Sequence Control is required in class 0 messages for some
> network configurations and for proper MTP relay accross a relay point.
> It might not be provided by the SCCP/SUA-User for class 0,=20
> but then the
> SUA layer (at ASP, IPSP or SGP) must ultimately select a=20
> loadsharing value.
> That is why we made the parameter MANDATORY for all protocol classes.
>=20
>     Note 1:    When the SSN is included, the DUNA message=20
> corresponds to=20
> -              the SCCP N-STATE primitive. When SSN is not, the DUNA=20
> -              message corresponds to the SCCP N-PCSTATE primitive.  =20
> +              the SCCP N-STATE primitive, and contains just=20
> one point=20
> +              code in the Affected Point Code parameter. When SSN is=20
> +              not, the DUNA message corresponds to the SCCP=20
> N-PCSTATE=20
> +              primitive.  =20
>=20
> The DUNA can contain multiple affected point codes even if the SSN is
> included.  This is particularly the case when the SG or IPSP supports
> the same subsystem on multiple point codes.  Please put it back the
> way it was.
>=20
> -   Note 1:    When the SSN is included, the SCON message=20
> corresponds to=20
> -              the SCCP N-STATE primitive. When SSN is not=20
> included, the=20
> -              SCON message corresponds to the SCCP N-PCSTATE=20
> primitive.=20
> +   Note 1:    When the SSN is included it must be equal to 1. If the=20
> +              SSN is included the SCON corresponds to the N-PCSTATE=20
> +              primitive used to convey the Restricted=20
> Importance Level=20
> +              (RIL) to the SCCP user. When the SSN is not=20
> included, the=20
> +              SCON message corresponds to the SCCP N-PCSTATE=20
> primitive=20
> +              reporting signalling point or network=20
> congestion status.=20
>=20
> No, the SSN doesn't have to be 1.  Where did you get that idea?
>=20
> -   Note 1:    The DUPU corresponds to the SCCP N-PCSTATE primitive. =20
> +   Note 1:    The DUPU corresponds to the SCCP N-PCSTATE=20
> primitive, and=20
> +               contains in the Affected Point Code parameter=20
> just one=20
> +              point code. =20
>=20
> Particularly as User/Cause is always the SCCP SI number,=20
> multiple affected
> point codes can occur.  Please put it back the way it was.
>=20
>  3.4.6 Destination Restricted (DRST)=20
> =20
>     The DRST message is optionally sent from the SG to all concerned=20
>     ASPs to indicate that the SG has determined that one or more=20
>     destinations are now restricted from the point of view of=20
> the SG, or=20
>     in response to a DAUD message if appropriate. The SUA=20
> layer at the=20
>     ASP is expected to send traffic to the affected=20
> destination via an=20
>     alternate SG of equal priority, but only if such an=20
> alternate route=20
> -   exists and is available. If the affected destination is currently=20
> -   considered unavailable by the ASP, the peer should be=20
> informed that=20
> -   traffic to the affected destination can be resumed.  In=20
> this case,=20
> -   the SUA layer should route the traffic through the SG=20
> initiating the=20
> -   DRST message.=20
> +   exists and is available. If the ASP currently considers=20
> the affected=20
> +   destination unavailable, the peer should be informed that=20
> traffic to=20
> +   the affected destination could be resumed.  In this case, the SUA=20
> +   layer should route the traffic through the SG initiating the DRST=20
> +   message.=20
>=20
> This is only true if the DRST corresponds to TFR.  Note also that the
> DRST cannot correspond to N-PCSTATE because there is no=20
> N-PCSTATE primitive
> for PC restriction (it is an MTP concept only).  This message=20
> corresponds to
> N-COORD that has little to do with the description above, no=20
> matter how is is
> worded exactly.
>=20
> -            0x17      Routing Key Change Refused=20
> +            0x17      Not Used in SUA=20
>=20
> Good, but you forgot to remove the statement later:
>=20
>    The "Routing Key Change Refused" error is sent when the SG=20
> refuses a=20
>    change in the Routing Key parameters.  =20
>=20
>=20
> -   User/Cause                           0x010C=20
> +   Cause / User                         0x010C=20
>=20
> Everwhere else in the document it is referred to as=20
> User/Cause, execpt here:
>=20
>  3.10.12 Cause / User=20
>=20
> Change both of these to "User/Cause".
>=20
>      0010     This is most commonly used in North American networks.=20
>                The Translation Type implicitly determines Nature of=20
>                Address and Numbering Plan. This data can be=20
> configured=20
>                in the SG. The number of digits is always even and=20
>                determined by the SCCP address length. =20
> -    0011                     Numbering Plan and Translation=20
> Type are tak
> +    0011                     Numbering Plan and Translation=20
> Type are taken over. It=20
>                is implicitly assumed that the Nature of Address =3D=20
>                Unknown. =20
> -    0100                     This format is used in=20
> international networ
> +    0100                     This format is used in=20
> international networks and most=20
>                commonly in networks outside North America. All=20
>                information to populate the source address is=20
> present in=20
>                the SCCP Address.=20
>=20
> Above is text extending past 72 columns that was truncated in=20
> sua-14.  Please
> shift the text over.
>=20
>       0              Unknown=20
>       1 - 63         International services=20
>       64 - 127       Spare=20
> -     128 > 254      National network specific=20
> +     128 =FB 254      National network specific=20
>       255            Reserved=20
>=20
> Non ASCII character, again.
>=20
>            Value     Description =20
>             0x0      No special options=20
>             0x1      Return message on error=20
> =20
> -                    Bits 3-7 are spare and SHOULD be coded=20
> zero, and MUS
> +                    Bits 3-7 are spare and SHOULD be coded=20
> zero, and MUST be=20
>           ignored by the receiver.=20
>=20
> Again text extending beyond 72 columns.
>=20
>  4.3 AS and ASP State Maintenance=20
> =20
> -   The SUA layer on the SGP maintains the state of each=20
> remote ASP, in=20
> -   each Application Server that the ASP is configured to receive=20
> +   The SUA layer on on the SGP maintains the state of each=20
> remote ASP,=20
> +   in each Application Server that the ASP is configured to receive=20
>     traffic, as input to the SUA message distribution function. =20
>=20
> All this change did was introduce a typo: "on on".
>=20
> +      1 IPSP Single Exchange (SE) model. Only a single exchange of=20
> +        ASPTM or ASPSM messages is needed to change the IPSP state.=20
> +        This means that a set of request from one end and=20
> acknowledge=20
> +        from the other will be enough. This configuration requires=20
> +        static RK configuration. =20
> +
> +      2 IPSP Double Exchange (DE) model. Both IPSPs have to send=20
> +        request messages and both IPSPs have to acknowledge=20
> the request=20
> +        messages from the other end. This results in a=20
> double exchange=20
> +        of ASPTM and ASPSM message, one from each end. This=20
> +        configuration supports dynamic routing key configuration by=20
> +        using RKM messages in the same way as ASP-SGP scenario. =20
>=20
> Remove mention of RK configuration.  This has not been agreed.
>=20
> +   In order to ensure interoperability, an M3UA implementation=20
> +   supporting IPSP communication MUST support IPSP SE model and MAY=20
> +   implement IPSP DE model. =20
>=20
> Change M3UA to SUA above.
>=20
> +   * Reception of messages from the peer M3UA layer at the=20
> ASP/IPSP; =20
> +   * Reception of some messages from the peer M3UA layer at other=20
> +   ASPs/IPSPs in the AS (e.g., ASP Active message indicating=20
>=20
> Change M3UA to SUA above.
>=20
> +   ASP-DOWN: The remote M3UA peer at the ASP/IPSP is=20
> unavailable and/or=20
> +   the related SCTP association is down.  Initially all=20
> ASPs/IPSPs will=20
> +   be in this state. An ASP/IPSP in this state SHOULD NOT be=20
> sent any=20
> +   SUA messages, with the exception of Heartbeat, ASP Down=20
> Ack and Error=20
>=20
> Change M3UA to SUA above.
>=20
> +   ASP-INACTIVE: The remote M3UA peer at the ASP/IPSP is=20
> available (and=20
> +   the related SCTP association is up) but application traffic is=20
> +   stopped. In this state the ASP/IPSP SHOULD NOT be sent=20
> any DATA or=20
> +   SSNM messages for the AS for which the ASP/IPSP is inactive. =20
> +
> +   ASP-ACTIVE: The remote M3UA peer at the ASP/IPSP is available and=20
> +   application traffic is active (for a particular Routing=20
> Context or=20
> +   set of Routing Contexts). =20
>=20
> Change M3UA to SUA above.
>=20
> +   SCTP CDI: The SCTP CDI denotes the local SCTP layer's=20
> Communication=20
> +   Down Indication to the Upper Layer Protocol (M3UA) on an=20
> SGP.  The=20
> +   local SCTP layer will send this indication when it=20
> detects the loss=20
> +   of connectivity to the ASP's peer SCTP layer.  SCTP CDI=20
> is understood=20
> +   as either a SHUTDOWN_COMPLETE notification or COMMUNICATION_LOST=20
> +   notification from the SCTP layer. =20
> +
> +   SCTP RI: The local SCTP layer's Restart indication to the=20
> upper layer=20
> +   protocol (M3UA) on an SG.  The local SCTP will send this=20
> indication=20
> +   when it detects a restart from the ASP's/IPSP's peer SCTP layer. =20
>=20
> You now have two SCTP CDI and SCTP RI paragraphs.  Those=20
> above and these
> below:  Delete the ones above because they have M3UA instead of SUA.
>=20
>    SCTP CDI: The SCTP CDI denotes the local SCTP layer's=20
> Communication=20
>    Down Indication to the Upper Layer Protocol (SUA) on an SGP. The=20
>    local SCTP layer will send this indication when it detects=20
> the loss=20
>    of connectivity to the ASP's peer SCTP layer.  SCTP CDI is=20
> understood=20
>    as either a SHUTDOWN_COMPLETE notification or COMMUNICATION_LOST=20
>    notification from the SCTP layer.=20
>=20
>    SCTP RI: The local SCTP layer's Restart indication to the=20
> upper layer=20
>    protocol (SUA) on an SG.  The local SCTP will send this indication=20
>    when it detects a restart from the ASP's peer SCTP layer.=20
>=20
> The diagram now says "Other IPSP in AS Overrides".  Please say
> "Other ASP/IPSP in AS Overrides".
>=20
>                 |      Other   +-------|              |=20
> -               |   ASP in AS  |       +--------------+=20
> +               |   IPSP in AS |       +--------------+ =20
>                 |   Overrides  |           ^     |=20
>=20
>  4.3.4.2 ASP Down Procedures=20
> =20
>     The ASP will send an ASP Down message to an SGP when the=20
> ASP wishes=20
>     to be removed from service in all Application Servers=20
> that it is a=20
> -   member and no longer receive any DATA, SSNM or ASPTM=20
> messages.  This=20
> -   action MAY be initiated at the ASP by an M-ASP_DOWN=20
> request primitive
> -   from Layer Management or MAY be initiated automatically by an SUA=20
> -   management function.   =20
> +   member and no longer receive any Connectionless or Connection -=20
> +   Oriented, SNM or ASPTM messages.  This action MAY be=20
> initiated at the=20
> +   ASP by an M-ASP_DOWN request primitive from Layer=20
> Management or MAY=20
> +   be initiated automatically by an SUA management function.=20
>=20
> Above change SNM to SSNM.
>=20
>     When an ASP wishes to withdraw from receiving traffic=20
> within an AS,=20
> -   the ASP sends an ASP Inactive message to the SGP or IPSP.  This=20
> -   action MAY be initiated at the ASP by an M-ASP_INACTIVE request=20
> -   primitive from Layer Management or MAY be initiated=20
> automatically by=20
> -   an SUA management function.  In the case where an ASP is=20
> processing=20
> -   the traffic for more than one Application Server across a=20
> common SCTP
> -   association, the ASP Inactive message contains one or=20
> more Routing=20
> -   Contexts to indicate for which Application Servers the=20
> ASP Inactive=20
> -   message applies.  In the case where an ASP Inactive=20
> message does not=20
> -   contain a Routing Context parameter, the receiver must know, via=20
> -   configuration data, which Application Servers the ASP is=20
> a member and
> -   move the ASP to the ASP-INACTIVE state in each all Application=20
> -   Servers.  In the case of an Override mode AS, where=20
> another ASP has=20
> -   already taken over the traffic within the AS with an ASP Active=20
> -   ("Override") message, the ASP that sends the ASP Inactive=20
> message is=20
> -   already considered by the SGP to be in state=20
> ASP-INACTIVE.  An ASP=20
> -   Inactive Ack message is sent to the ASP, after ensuring that all=20
> -   traffic is stopped to the ASP. =20
> +   or the ASP wants to initiate the process of activation,=20
> the ASP sends=20
> +   an ASP Inactive message to the SGP or IPSP. =20
>=20
> Change above to "initiate the process of deactivation"=20
> instead of "initiate
> the process of activation".
>=20
> Also, above you have delted the description of the ASP=20
> Inactive procedure for
> Override mode.  Please put back:
>=20
> -                     In the case where an ASP Inactive=20
> message does not=20
> -   contain a Routing Context parameter, the receiver must know, via=20
> -   configuration data, which Application Servers the ASP is=20
> a member and
> -   move the ASP to the ASP-INACTIVE state in each all Application=20
> -   Servers.
>=20
> and
>=20
> -             In the case of an Override mode AS, where=20
> another ASP has=20
> -   already taken over the traffic within the AS with an ASP Active=20
> -   ("Override") message, the ASP that sends the ASP Inactive=20
> message is=20
> -   already considered by the SGP to be in state=20
> ASP-INACTIVE.  An ASP=20
> -   Inactive Ack message is sent to the ASP, after ensuring that all=20
> -   traffic is stopped to the ASP. =20
>=20
> The following addition:
>=20
> +4.8 NIF Not Available on SGP =20
> +
> +   If the SG (all the SGPs) is isolated from the NIF, then=20
> all the users=20
> +   are isolated from the SS7 network.  A DUNA(*) message=20
> MUST be sent=20
> +   from the SGPs to all the ASPs.  =20
> +
> +   If only one SGP in the SG is isolated entirely from the=20
> NIF, the SGP=20
> +   SHOULD abort its associations. An alternative would be=20
> for the SGP to=20
> +   send ASP Down Ack.=20
> +
> +   If one or more SGP suffer a partial failure (where aborting the=20
> +   association(s) would cause all active AS(es) to fail),=20
> then the SGP=20
> +   MUST send DUNA messages for the affected SPC(es).  This=20
> is the case=20
> +   where an SGP can continue to service one or more active=20
> AS(es), but=20
> +   due to a partial failure it is unable to service other(s) active=20
> +   AS(es).=20
>=20
> Please remove the entire addition.  The M3UA IG has removed=20
> this.  That is
> the problem with adding stuff from the M3UA IG.  The M3UA IG=20
> has not gone
> through Last Call.
>=20
> +   [SIGSEC]       J. Loughney, M. Tuexen, J. Pastor-Balbas,=20
> "Security=20
> +                  Considerations for SIGTRAN Protocols", Work in=20
>                    Progress. =20
>=20
> There is a Work In Progress as a normative reference.  This=20
> draft will be
> held until this reference becomes an RFC.
>=20
> --brian
>=20
>=20
> John Loughney wrote:                                (Thu, 03=20
> Jul 2003 19:29:23)
> > Hi all,
> >=20
> > I have updated SUA, I have tried to incorporate some of the=20
> major issues
> > from the M3UA Implementation Guide.  All interested parties, please
> > give the document a quick read, because I would like to=20
> send this to the
> > IESG for final approval.
> >=20
> > thanks,
> > John
> >=20
> > > -----Original Message-----
> > > From: ext Internet-Drafts@ietf.org=20
> [mailto:Internet-Drafts@ietf.org]
> > > Sent: 03 July, 2003 18:33
> > > Cc: sigtran@ietf.org
> > > Subject: [Sigtran] I-D ACTION:draft-ietf-sigtran-sua-15.txt
> > >=20
> > >=20
> > > A New Internet-Draft is available from the on-line=20
> > > Internet-Drafts directories.
> > > This draft is a work item of the Signaling Transport Working=20
> > > Group of the IETF.
> > >=20
> > > 	Title		: Signalling Connection Control Part=20
> > > User Adaptation=20
> > >                           Layer (SUA)
> > > 	Author(s)	: J. Loughney, G. Sidebottom
> > > 	Filename	: draft-ietf-sigtran-sua-15.txt
> > > 	Pages		: 118
> > > 	Date		: 2003-7-3
> > > =09
> > > This Internet Draft defines a protocol for the transport of=20
> > > any         Signalling Connection Control Part-User=20
> > > signalling over IP using the   Stream Control Transport=20
> > > Protocol. The protocol is designed to be=20
> > > modular and symmetric, to allow it to work in diverse=20
> architectures,=20
> > > such as a Signalling Gateway to IP Signalling Endpoint=20
> architecture=20
> > > as well as a peer-to-peer IP Signalling Endpoint architecture.
> > >=20
> > > A URL for this Internet-Draft is:
> > > http://www.ietf.org/internet-drafts/draft-ietf-sigtran-sua-15.txt
> > >=20
> > > To remove yourself from the IETF Announcement list, send=20
> a message to=20
> > > ietf-announce-request with the word unsubscribe in the body=20
> > > of the message.
> > >=20
> > > Internet-Drafts are also available by anonymous FTP. Login=20
> > > with the username
> > > "anonymous" and a password of your e-mail address. After=20
> logging in,
> > > type "cd internet-drafts" and then
> > > 	"get draft-ietf-sigtran-sua-15.txt".
> > >=20
> > > A list of Internet-Drafts directories can be found in
> > > http://www.ietf.org/shadow.html=20
> > > or ftp://ftp.ietf.org/ietf/1shadow-sites.txt
> > >=20
> > >=20
> > > Internet-Drafts can also be obtained by e-mail.
> > >=20
> > > Send a message to:
> > > 	mailserv@ietf.org.
> > > In the body type:
> > > 	"FILE /internet-drafts/draft-ietf-sigtran-sua-15.txt".
> > > =09
> > > 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=20
> > > 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.
> > > 	=09
> > > 	=09
> > > Below is the data which will enable a MIME compliant mail reader
> > > implementation to automatically retrieve the ASCII version of the
> > > Internet-Draft.
> > >=20
> >=20
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
>=20
> --=20
> Brian F. G. Bidulock
> bidulock@openss7.org
> http://www.openss7.org/
>=20

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



From exim@www1.ietf.org  Tue Jul 29 08:25:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA20450
	for <sigtran-archive@odin.ietf.org>; Tue, 29 Jul 2003 08:25:33 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hTXP-0000gB-0v
	for sigtran-archive@odin.ietf.org; Tue, 29 Jul 2003 08:25:07 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6TCP7ER002590
	for sigtran-archive@odin.ietf.org; Tue, 29 Jul 2003 08:25:07 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hTXJ-0000e9-Fm; Tue, 29 Jul 2003 08:25:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hTWg-0000d0-C4
	for sigtran@optimus.ietf.org; Tue, 29 Jul 2003 08:24:22 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA20402
	for <sigtran@ietf.org>; Tue, 29 Jul 2003 08:24:17 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hTWf-0004sX-00
	for sigtran@ietf.org; Tue, 29 Jul 2003 08:24:21 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[6hziyUGxAfJRlsCWosU3QEX+HmYXOTLM])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hTWe-0004sT-00
	for sigtran@ietf.org; Tue, 29 Jul 2003 08:24:20 -0400
Received: from ns.pigworks.openss7.net (IDENT:TVi6y//YJqgXRe4R716/TfRlxhGfDye5@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6TCODE21991;
	Tue, 29 Jul 2003 06:24:13 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6TCOC415276;
	Tue, 29 Jul 2003 06:24:12 -0600
Date: Tue, 29 Jul 2003 06:24:12 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: john.loughney@nokia.com, JDoss@verisign.com, sigtran@ietf.org
Subject: Re: [Sigtran] SUA Question - Trasaction ID based Routing
Message-ID: <20030729062412.A15238@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: john.loughney@nokia.com, JDoss@verisign.com,
	sigtran@ietf.org
References: <DADF50F5EC506B41A0F375ABEB32063658F124@esebe023.ntc.nokia.com> <20030711032034.I1680@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: <20030711032034.I1680@openss7.org>; from bidulock@openss7.org on Fri, Jul 11, 2003 at 03:20:34AM -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>

John,

Little problem... A draft was already submitted last January:

 draft-bidulock-sigtran-regext-01.txt .ps .pdf

that describes extensions to the registration mechanism for SUA
(addition of DRN range to RK, addition of RK to DEREG REQ, and
necessary errors) to permit this.  It also applies to M3UA (CIC
ranges) M2UA (SDL/SDT pairs), TUA, ...

The draft is about to expire, so I will resumbit it.

--brian

Brian F. G. Bidulock wrote:                                                           (Fri, 11 Jul 2003 03:20:34)
> John,
> 
> Excellent suggestion.
> 
> --brian
> 
> John Loughney wrote:                                                                  (Fri, 11 Jul 2003 11:47:29)
> > Brian & James,
> > 
> > Yes, this has a long & twisted history. My original goal was to
> > try to get SUA finished - something that the IESG has caused
> > a lot of trouble with.  
> > 
> > The WG never could come to consensus on how to solve this.  My proposal
> > to you is to write a short Internet draft on how to solve this,
> > lets get consensus on it and publish it as a supplemental RFC.
> > 
> > John
> > 

-- 
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 exim@www1.ietf.org  Tue Jul 29 08:27:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA20534
	for <sigtran-archive@odin.ietf.org>; Tue, 29 Jul 2003 08:27:32 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hTZL-0000nF-0P
	for sigtran-archive@odin.ietf.org; Tue, 29 Jul 2003 08:27:07 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6TCR6VE003028
	for sigtran-archive@odin.ietf.org; Tue, 29 Jul 2003 08:27:06 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hTZF-0000kg-TU; Tue, 29 Jul 2003 08:27:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hTYh-0000hI-Ga
	for sigtran@optimus.ietf.org; Tue, 29 Jul 2003 08:26:27 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA20484
	for <sigtran@ietf.org>; Tue, 29 Jul 2003 08:26:23 -0400 (EDT)
From: john.loughney@nokia.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hTYg-0004tV-00
	for sigtran@ietf.org; Tue, 29 Jul 2003 08:26:26 -0400
Received: from mgw-x1.nokia.com ([131.228.20.21])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hTYf-0004tR-00
	for sigtran@ietf.org; Tue, 29 Jul 2003 08:26:25 -0400
Received: from esvir05nok.ntc.nokia.com (esvir05nokt.ntc.nokia.com [172.21.143.37])
	by mgw-x1.nokia.com (Switch-2.2.6/Switch-2.2.6) with ESMTP id h6TCQNB24477
	for <sigtran@ietf.org>; Tue, 29 Jul 2003 15:26:24 +0300 (EET DST)
Received: from esebh002.NOE.Nokia.com (unverified) by esvir05nok.ntc.nokia.com
 (Content Technologies SMTPRS 4.2.5) with ESMTP id <T63bb138cc6ac158f25a28@esvir05nok.ntc.nokia.com>;
 Tue, 29 Jul 2003 15:26:22 +0300
Received: from esebe014.NOE.Nokia.com ([172.21.138.53]) by esebh002.NOE.Nokia.com with Microsoft SMTPSVC(5.0.2195.6139);
	 Tue, 29 Jul 2003 15:26:22 +0300
Received: from esebe023.NOE.Nokia.com ([172.21.138.115]) by esebe014.NOE.Nokia.com with Microsoft SMTPSVC(5.0.2195.6139);
	 Tue, 29 Jul 2003 15:26:21 +0300
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"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [Sigtran] SUA Question - Trasaction ID based Routing
Date: Tue, 29 Jul 2003 15:26:21 +0300
Message-ID: <DADF50F5EC506B41A0F375ABEB32063658F223@esebe023.ntc.nokia.com>
Thread-Topic: [Sigtran] SUA Question - Trasaction ID based Routing
Thread-Index: AcNVzFri2q5DQhWzTc25gTRJCjJh5QAAAntg
To: <bidulock@openss7.org>, <JDoss@verisign.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 29 Jul 2003 12:26:22.0043 (UTC) FILETIME=[9EAE7AB0:01C355CC]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

Hi Brian & James,

> Little problem... A draft was already submitted last January:
>=20
>  draft-bidulock-sigtran-regext-01.txt .ps .pdf
>=20
> that describes extensions to the registration mechanism for SUA
> (addition of DRN range to RK, addition of RK to DEREG REQ, and
> necessary errors) to permit this.  It also applies to M3UA (CIC
> ranges) M2UA (SDL/SDT pairs), TUA, ...
>=20
> The draft is about to expire, so I will resumbit it.

Resubmit it, make sure it is consistant with SUA and lets see if
other folks are interested in this.  One option, if noone opposes
it, would be to ask the RFC editor to approve it as an individual
RFC.  Even better would be ask Lyndon to have a last call in
the SIGTRAN WG, even though it is an individual draft.

John

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



From exim@www1.ietf.org  Tue Jul 29 08:57:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA21575
	for <sigtran-archive@odin.ietf.org>; Tue, 29 Jul 2003 08:57:33 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hU2N-0001qw-Ae
	for sigtran-archive@odin.ietf.org; Tue, 29 Jul 2003 08:57:07 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6TCv76U007116
	for sigtran-archive@odin.ietf.org; Tue, 29 Jul 2003 08:57:07 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hU2H-0001qN-OJ; Tue, 29 Jul 2003 08:57:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hU1o-0001pk-6f
	for sigtran@optimus.ietf.org; Tue, 29 Jul 2003 08:56:32 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA21562
	for <sigtran@ietf.org>; Tue, 29 Jul 2003 08:56:27 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hU1m-0005Dk-00
	for sigtran@ietf.org; Tue, 29 Jul 2003 08:56:30 -0400
Received: from sj-iport-3-in.cisco.com ([171.71.176.72] helo=sj-iport-3.cisco.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19hU1m-0005DW-00
	for sigtran@ietf.org; Tue, 29 Jul 2003 08:56:30 -0400
Received: from cisco.com (171.71.177.237)
  by sj-iport-3.cisco.com with ESMTP; 29 Jul 2003 05:56:00 -0700
Received: from fruitpie.cisco.com (IDENT:mirapoint@fruitpie.cisco.com [64.102.16.27])
	by sj-core-1.cisco.com (8.12.9/8.12.6) with ESMTP id h6TCtvmn011047;
	Tue, 29 Jul 2003 05:55:57 -0700 (PDT)
Received: from kmorneau-w2k01.cisco.com (rtp-vpn2-32.cisco.com [10.82.240.32])
	by fruitpie.cisco.com (Mirapoint Messaging Server MOS 3.3.3-GR)
	with ESMTP id AIT54946;
	Tue, 29 Jul 2003 05:55:55 -0700 (PDT)
Message-Id: <4.3.2.7.2.20030729085058.04e56ad8@cia.cisco.com>
X-Sender: kmorneau@cia.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Tue, 29 Jul 2003 08:55:55 -0400
To: "VIvek Toky" <vivektoky@ccpu.com>, <sigtran@ietf.org>
From: Ken Morneault <kmorneau@cisco.com>
Subject: RE: [Sigtran] Sigtran Plugtest at ETSI 9/03
Cc: <Sebastien.Flaux@etsi.org>
In-Reply-To: <HKEBKFBNPIDALDOKLGKDEEJLCDAA.vivektoky@ccpu.com>
References: <BE462396B4A022449E0EB2DD4753D05402F26906@vbeexc02.emea.cpqcorp.net>
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>



Vivek,

In regards to M3UA, we should review the following I-D as it is
more recent than the Sept '03 test spec:

http://www.ietf.org/internet-drafts/draft-anshoo-test-spec-m3ua-01.txt

There is also a recent IUA test spec:

http://www.ietf.org/internet-drafts/draft-bhola-conformance-test-iua-01.txt

Information about potential updates to the other test specs will
be available soon.

Regards,

Ken

At 02:52 PM 7/25/2003 -0700, VIvek Toky wrote:
>All,
>   can anyone tell, when shall we get the test specifications for UA's ?
>Currently I see the first draft of M3UA test specifications on ETSI website.
>Is this the one which is going to be used for September'03 M3UA bakeoff ?
>
>thanks,
>Vivek.
>
>-----Original Message-----
>From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf Of
>Anzile, Christophe
>Sent: Tuesday, July 08, 2003 3:57 AM
>To: sigtran@ietf.org
>Subject: RE: [Sigtran] Sigtran Plugtest at ETSI 9/03
>
>
>
>All,
>same remark applies for SUA (v14 stated @etsi.org) versus v15 out last week.
>Christophe.
>
>
>-----Original Message-----
>From: Ken Morneault [mailto:kmorneau@cisco.com]
>Sent: Monday, July 07, 2003 4:32 PM
>To: sigtran@ietf.org
>Subject: [Sigtran] Sigtran Plugtest at ETSI 9/03
>
>
>
>
>All:
>
>Which version of M2PA should be tested at the upcoming
>Sigtran plugtest in 9/03?  Currently, the ETSI Plugtest site
>indicates version 8, but version 9 was just released.  Version
>6 was tested at the last Plugtest.
>
>Please email your preference.
>
>Regards,
>
>Ken
>
>
>_______________________________________________
>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 exim@www1.ietf.org  Tue Jul 29 13:53:35 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA02388
	for <sigtran-archive@odin.ietf.org>; Tue, 29 Jul 2003 13:53:35 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hYer-00053c-NW
	for sigtran-archive@odin.ietf.org; Tue, 29 Jul 2003 13:53:09 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6THr9Qs019426
	for sigtran-archive@odin.ietf.org; Tue, 29 Jul 2003 13:53:09 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hYej-00052p-6t; Tue, 29 Jul 2003 13:53:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hYdn-00051x-5v
	for sigtran@optimus.ietf.org; Tue, 29 Jul 2003 13:52:03 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA02344
	for <sigtran@ietf.org>; Tue, 29 Jul 2003 13:51:58 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hYdk-0000Qd-00
	for sigtran@ietf.org; Tue, 29 Jul 2003 13:52:01 -0400
Received: from rainier.illuminet.com ([12.41.110.52])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hYdk-0000QH-00
	for sigtran@ietf.org; Tue, 29 Jul 2003 13:52:00 -0400
Received: from oly1wnexc02.corp.illuminet.com (oly1wnexc02.corp.illuminet.com [10.55.13.13]) by rainier.illuminet.com (8.8.8/8.8.8) with ESMTP id KAA25719; Tue, 29 Jul 2003 10:50:21 -0700 (PDT)
Received: from oly1wnexc01.corp.illuminet.com ([10.55.13.12]) by oly1wnexc02.corp.illuminet.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2653.13)
	id P5DVAX3W; Tue, 29 Jul 2003 10:50:21 -0700
Received: by oly1wnexc01.corp.illuminet.com with Internet Mail Service (5.5.2653.19)
	id <P7HYKHZV>; Tue, 29 Jul 2003 10:50:21 -0700
Message-ID: <6B830E7F37A7CF4A9806AE9D98AF42B53324F5@atwinex01.corp.illuminet.com>
X-Sybari-Space: 00000000 00000000 00000000 00000000
From: "Doss, James" <JDoss@verisign.com>
To: "'john.loughney@nokia.com'" <john.loughney@nokia.com>,
        bidulock@openss7.org, "Doss, James" <JDoss@verisign.com>,
        sigtran@ietf.org
Subject: RE: [Sigtran] SUA Question - Trasaction ID based Routing
Date: Tue, 29 Jul 2003 10:50:11 -0700
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 John,
That should not be a problem.
I went thru the SUA-15 report and I found answer by using TID for routing in single mode with out redundancy and that will work fine. I am looking for redundancy and mated SG(STP - typically 2) pairs with ASs.(AS1 to N). I do not think that should hold the draft and can be a page of supplemental.
regards
James



-----Original Message-----
From: john.loughney@nokia.com [mailto:john.loughney@nokia.com]
Sent: Tuesday, July 29, 2003 8:26 AM
To: bidulock@openss7.org; JDoss@verisign.com; sigtran@ietf.org
Subject: RE: [Sigtran] SUA Question - Trasaction ID based Routing


Hi Brian & James,

> Little problem... A draft was already submitted last January:
> 
>  draft-bidulock-sigtran-regext-01.txt .ps .pdf
> 
> that describes extensions to the registration mechanism for SUA
> (addition of DRN range to RK, addition of RK to DEREG REQ, and
> necessary errors) to permit this.  It also applies to M3UA (CIC
> ranges) M2UA (SDL/SDT pairs), TUA, ...
> 
> The draft is about to expire, so I will resumbit it.

Resubmit it, make sure it is consistant with SUA and lets see if
other folks are interested in this.  One option, if noone opposes
it, would be to ask the RFC editor to approve it as an individual
RFC.  Even better would be ask Lyndon to have a last call in
the SIGTRAN WG, even though it is an individual draft.

John

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



From exim@www1.ietf.org  Tue Jul 29 14:21:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA03662
	for <sigtran-archive@odin.ietf.org>; Tue, 29 Jul 2003 14:21:32 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hZ5v-0006gn-AS
	for sigtran-archive@odin.ietf.org; Tue, 29 Jul 2003 14:21:07 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6TIL7Uw025712
	for sigtran-archive@odin.ietf.org; Tue, 29 Jul 2003 14:21:07 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hZ5o-0006gU-HY; Tue, 29 Jul 2003 14:21:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hZ4t-0006dd-RH
	for sigtran@optimus.ietf.org; Tue, 29 Jul 2003 14:20:03 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA03644
	for <sigtran@ietf.org>; Tue, 29 Jul 2003 14:19:58 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hZ4r-0000kz-00
	for sigtran@ietf.org; Tue, 29 Jul 2003 14:20:01 -0400
Received: from ottgw.pt.com ([206.191.2.130] helo=pt.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19hZ4q-0000kv-00
	for sigtran@ietf.org; Tue, 29 Jul 2003 14:20:00 -0400
Received: from pt.com (IDENT:root@localhost [127.0.0.1])
	by  pt.com (8.9.3/8.9.3) with ESMTP id OAA03580;
	Tue, 29 Jul 2003 14:19:10 -0400
Message-ID: <3F26BAA2.1080908@pt.com>
Date: Tue, 29 Jul 2003 14:19:14 -0400
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: "'sigtran@ietf.org'" <sigtran@ietf.org>, tom.george@alcatel.com
Subject: Re: [Sigtran] M2PA Last Call Comment - RPO
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030712170744.A11077@openss7.org> <3F15C7F2.6070704@pt.com> <20030716191143.F6725@openss7.org> <3F18328D.1080706@pt.com> <20030718120053.C14760@openss7.org> <3F184959.2040103@pt.com> <20030718133957.A16698@openss7.org> <3F1867CB.6060108@pt.com> <20030724134845.D11627@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

Hi Brian,

Sorry for the slow reply.

I was thinking mostly about letting the RTB drain and forgot about the 
TB.  So if you want to let the TB drain, use the BSN in the LS PR is 
used to resynchronize sequence numbers as in the original text.  Is the 
rest of the proposed text ok?

In any case we could allow the side in LPO to continue to send the TB 
and RTB to the peer.  This leaves the M2PA in RPO with the freedom to 
deal with the messages as their interface to MTP3 allows e.g. if the 
peer cannot deliver these messages to MTP3 then they will be flushed 
after MTP3 T1.   This is not explicitly mentioned below, but could be. 
 The only disadvantage would be increased bandwidth use if the messages 
end up getting flushed.

After thinking some more about it though I think there are some 
complications that come from passing messages to MTP3 while in RPO.  I 
wonder if the details of this interface should be left as implementation 
dependent i.e. don't include the possibility in the text and leave the 
implementer to forge ahead with their own MTP3/M2PA interface...  I 
think in this case the semantics of the MTP3/M2PA interface are slightly 
different that usual.  Anyone who will use such an interface will still 
be able to use the usual MTP3/MTP2 interface as a reference.

If we are going to mention the possibility of passing data to the MTP3 
while in RPO, should we also warn about some of the issues that must be 
dealt with?  In particular, the M2PA that is passing data to the MTP3 
while in RPO must continue to interact properly with MTP3 procedures 
that may be going on e.g. sequence, time controlled or emergency 
changeover.  I don't think this is well covered in the MTP2 specs, since 
the posibility of getting data in RPO is limited to messages that 
arrived before the first SIPO.  In particular:
- data must not be delivered to MTP3 if the remote end has been 
performing a time controlled changeover and its T1 has expired
- data must not be delivered to MTP3 after a BSNT request has been anwered
- data must not be delivered to MTP3 after the local MTP3 has responded 
to a changeover order with an emergency changeover response
I'd rather avoid the whole issue.

Thoughts?
Andrew

Brian F. G. Bidulock wrote:

>Andrew,
>
>No that doesn't sound right.  The side in LPO must be able to drain it's TB to
>the peer and the peer needs to acknowledge messsages received after LPO but
>before LPR if its MTP3 is able to do so.  That's what started this thread off
>in the first place.
>
>--brian
>
>Andrew Booth wrote:                                                                 (Fri, 18 Jul 2003 17:34:03)
>  
>
>>Hi Brian,
>>
>>
>>Brian F. G. Bidulock wrote:
>>
>>    
>>
>>>[snip]
>>>
>>>Sure.  Just as long as the sender will generate acknowledgements in RPO
>>>I don't have a problem.  Also, T6 does not always run when in RPO.
>>>Did you want to propose some alternate text to the text I proposed below?
>>>
>>>--brian
>>>
>>> 
>>>
>>>      
>>>
>>What do you think of the following proposal, does it sum up both our points?
>>As a miscellaneous note I changed the LS PR to LS PO in the final 
>>paragraph; the motivation is that I think the FSN in the LSPO and LSPR 
>>should be the same (the side in LPO shouldn't be sending data) and 
>>because the Flush could arrive before the LS PR in which case the old 
>>text is clearly wrong. I'm not sure though if that change was the right 
>>one. Please let me know what you think.
>>
>>Thanks,
>>Andrew
>>
>>4.1.4 Processor Outage
>>
>>The Link Status Processor Outage message replaces the SIPO message of
>>MTP2. Unlike MTP2, the message SHOULD NOT be transmitted
>>continuously. M2PA SHALL send a Link Status Processor Outage message
>>to its peer at the beginning of a processor outage condition where
>>MTP2 would send SIPO. M2PA MAY send additional Link Status Processor
>>Outage messages as long as that condition persists.
>>
>>When the local processor outage condition ends, M2PA SHALL send a Link
>>Status Processor Recovered message to its peer. This message is used
>>to signal in the end of the processor outage condition, instead of an
>>MSU or FISU as in MTP2. The BSN value in this message is used by the
>>peer to resynchronize its sequence numbers, if this is required by
>>MTP2.
>>
>>Some MTP3 implementations may be able to accept received User Data
>>while in Remote Processor Outage while others may not. While there
>>is a remote processor outage condition:
>>
>>(a) M2PA SHOULD continue to acknowledge User Data Message received and
>>accepted by MTP3 regardless of the remote processor outage.
>>
>>(b) Any User Data messages received from the peer after
>>the Link Status Processor Outage that may not be passed to MTP3
>>MUST NOT be acknowledged and MUST be buffered.
>>
>>If M2PA receives a Flush command from MTP3, M2PA SHALL
>>discard the incoming messages that were queued and unacknowledged during the
>>processor outage condition. Furthermore, M2PA SHALL discard messages in the
>>transmit and retransmit queues as required by MTP2. Similar behaviour
>>is appropriate in response to corresponding primitives in other SS7 variants
>>such as the ANSI Clear RTB primitive.
>>
>>Note that because the Link Status and User Data messages are sent on 
>>separate
>>streams, it is possible for a User Data message sent after the Link Status
>>Processor Recovered message to arrive before it. Therefore, when flushing
>>buffers, M2PA must take care not to flush messages sent after the remote
>>processor recovered. M2PA SHOULD NOT flush User Data messages with FSN 
>>greater
>>than the FSN in the last received Link Status Processor Outage message.
>>Note that similar conditions could occur if User Data is received after Link
>>Status Processor Recovered but before the Flush command from MTP3. Similar
>>behaviour is appropriate in response to comparable primitives in other 
>>SS7 variants,
>>such as the ANSI Clear RTB primitive.
>>
>>    
>>
>
>  
>

-- 
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 exim@www1.ietf.org  Tue Jul 29 18:04:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA11756
	for <sigtran-archive@odin.ietf.org>; Tue, 29 Jul 2003 18:04:31 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hcZj-0000YR-IK
	for sigtran-archive@odin.ietf.org; Tue, 29 Jul 2003 18:04:07 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6TM471s002121
	for sigtran-archive@odin.ietf.org; Tue, 29 Jul 2003 18:04:07 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hcZd-0000WR-0E; Tue, 29 Jul 2003 18:04:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hcYn-0000Ta-Ei
	for sigtran@optimus.ietf.org; Tue, 29 Jul 2003 18:03:09 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA11598
	for <sigtran@ietf.org>; Tue, 29 Jul 2003 18:03:02 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hcYk-0002M3-00
	for sigtran@ietf.org; Tue, 29 Jul 2003 18:03:06 -0400
Received: from auds952.usa.alcatel.com ([143.209.238.7])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hcYj-0002Lc-00
	for sigtran@ietf.org; Tue, 29 Jul 2003 18:03:05 -0400
Received: from alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6TM2YB3003297;
	Tue, 29 Jul 2003 17:02:34 -0500 (CDT)
Message-ID: <3F26EEFA.7040203@alcatel.com>
Date: Tue, 29 Jul 2003 17:02:34 -0500
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
User-Agent: Mozilla/5.0 (X11; U; SunOS sun4u; en-US; rv:1.0.1) Gecko/20020920 Netscape/7.0
X-Accept-Language: en-us, en
MIME-Version: 1.0
CC: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - T6
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030712002921.A21136@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

Brian,

The old text about ignoring the BSN in the LS messages has been taken 
out in version 9. We explicitly mention that the BSN in the LS PO 
message can be used, but there is no reason to restrict the use of the 
BSN in other LS messages.

Right?

tomG.

Brian F. G. Bidulock wrote:
   [...]
> Now, I know that we said that Link Status messages are not used to acknowledge
> DATA messages, but it makes sense to permit the Link Status Busy Ended message
> to acknowledge outstanding User Data messages, so that upon its receipt, T7
> might not need to be started if all messages in the retransmission buffer are
> acknolwedged by the Link Status Busy Ended message.
> 
> The above also kind of changes the idea of T7 metering the oldest message in
> the retransmission buffer.  It would now measure the oldest message in the
> retransmission buffer since the receipt of the last Link Status Busy Ended.
   [...]

-- 
* 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.  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 exim@www1.ietf.org  Tue Jul 29 18:06:31 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA11969
	for <sigtran-archive@odin.ietf.org>; Tue, 29 Jul 2003 18:06:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hcbe-0000jp-TN
	for sigtran-archive@odin.ietf.org; Tue, 29 Jul 2003 18:06:06 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6TM66eO002831
	for sigtran-archive@odin.ietf.org; Tue, 29 Jul 2003 18:06:06 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hcbZ-0000i6-Px; Tue, 29 Jul 2003 18:06:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hcb1-0000hp-3Z
	for sigtran@optimus.ietf.org; Tue, 29 Jul 2003 18:05:27 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA11863
	for <sigtran@ietf.org>; Tue, 29 Jul 2003 18:05:20 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hcay-0002Nm-00
	for sigtran@ietf.org; Tue, 29 Jul 2003 18:05:24 -0400
Received: from auds952.usa.alcatel.com ([143.209.238.7])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hcas-0002NS-00
	for sigtran@ietf.org; Tue, 29 Jul 2003 18:05:18 -0400
Received: from alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6TM4mB3003537;
	Tue, 29 Jul 2003 17:04:48 -0500 (CDT)
Message-ID: <3F26EF80.8020105@alcatel.com>
Date: Tue, 29 Jul 2003 17:04:48 -0500
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
User-Agent: Mozilla/5.0 (X11; U; SunOS sun4u; en-US; rv:1.0.1) Gecko/20020920 Netscape/7.0
X-Accept-Language: en-us, en
MIME-Version: 1.0
CC: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - T6
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030712002921.A21136@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

Brian,

I incorporated your recommendations.

Here is the new text for section 4.1.5:
=====
4.1.5 Level 2 Flow Control

The Link Status Busy message replaces the SIB message of MTP2. The
message SHOULD NOT be transmitted continuously. M2PA SHALL send a Link
Status Busy message to its peer at the beginning of a receive
congestion condition where MTP2 would send SIB. M2PA MAY send
additional Link Status Busy messages as long as that condition
persists. When the condition ends, M2PA SHALL send a Link Status Busy
Ended message to its peer.

M2PA SHALL continue transmitting messages while it is in receive
congestion, but MUST NOT acknowledge the message that triggered the
sending of the Link Status Busy message, nor any messages received
before the sending of Link Status Busy Ended.

When the peer M2PA receives the first Link Status Busy message, it
SHALL start the Remote Congestion timer T6 if there are messages in
the retransmission buffer awaiting acknowledgement (i.e., T7 is
running).  In addition, M2PA SHALL stop the T7 timer if it is running.
Additional Link Status Busy messages received while T6 is running do
not cause T6 to be reset and do not cause T7 to be started.  Also,
while T6 is running, T7 SHALL NOT be started.

When the peer M2PA receives the Link Status Busy Ended message and T6
has not expired, it SHALL stop T6 if T6 is running and start T7 if
there are messages awaiting acknowledgement in the retransmission
buffer.

The peer M2PA SHOULD continue receiving and acknowledging messages
while the other end is busy, but MUST NOT send User Data messages
after receiving Link Status Busy and before receiving Link Status Busy
Ended.
=====


Here is the old (version 9) text for reference:
=====
4.1.5 Level 2 Flow Control

The Link Status Busy message replaces the SIB message of MTP2. The
message SHOULD NOT be transmitted continuously. M2PA SHALL send a Link
Status Busy message to its peer at the beginning of a receive
congestion condition where MTP2 would send SIB. M2PA MAY send
additional Link Status Busy messages as long as that condition
persists. When the condition ends, M2PA SHALL send a Link Status Busy
Ended message to its peer.

M2PA SHALL continue transmitting and acknowledging messages while it
is in receive congestion.

When the peer M2PA receives the first Link Status Busy message, it
SHALL start the Remote Congestion timer T6. Additional Link Status
Busy messages received while T6 is running do not cause T6 to be
reset.

The peer M2PA SHOULD continue transmitting and acknowledging messages
while its T6 timer is running, i.e., while the other end is Busy.
=====




Brian F. G. Bidulock wrote:
> M2PA-09 says:
> 
> 
>>M2PA SHALL continue transmitting and acknowledging messages while it
>>is in receive congestion.
> 
> 
> Q.703 says:
> 
> | The receiving level 2 also withholds acknowledgement of the message signal
> | unit, which triggered off the congestion detection, and of message signal
> | units received during the congestion situation; that is fill-in signal units
> | or message signal units are sent as usual, but with the backward sequence
> | number and backwards indicator bit assigned the values which are contained
> | in the last transmitted signal unit before the congestion is recognized.
> 
> It is rather important that the busy end not acknowledge messages while it is
> in receive congestion.  This is because if the receive congestion results in
> link failure, acknowledged messages caught at the congested receiver will be
> unavailable for retransmission on an alternate link at the transmitter.
> 
> Also, delivering these messages to MTP3 late will aggravate missequencing of
> these delayed messages with subsequent messages transmitted on the alternate
> link after changeover.
> 
> Change M2PA to say:
> 
> < M2PA SHALL continue transmitting message while it is in receive congestion
> < but MUST NOT acknowledge the message that triggered sending of the Link
> < Status Busy message, nor any messages received before the sending of Link
> < Status Busy Ended.
> 
> M2PA-09 says:
> 
> 
>>When the peer M2PA receives the first Link Status Busy message, it
>>SHALL start the Remote Congestion timer T6. Additional Link Status
>>Busy messages received while T6 is running do not cause T6 to be
>>reset.
> 
> 
> Q.703 says:
> 
> | At the remote end of the signalling link, every reception of a link status
> | singal unit containing indication "B" causes the excessive delay of
> | acknolwedgement timer T7 to be restarted, if the timer is already running.
> | In addition first reception of the link status signal unit containing a
> | status indication "B" starts a longer supervision timer T7 (see 12.3) only
> | if there are message signal units in the retransmission buffer.  Should
> | timer T6 expire, link failure indication is generated.
> 
> Change M2PA to say:
> 
> < When the peer M2PA receives the first Link Status Busy message, it SHALL
> < start the Remote Congestion timer T6 if there are messages in the
> < retransmission buffer awaiting acknowledgement (i.e, T7 is running).
> < In addition, M2PA will stop the T7 timer if it is running.  Additional Link
> < Status Busy messages received while T6 is running do not cause T6 to be
> < reset and do not cause T7 to be started.  Also, while T6 is running, T7
> < SHALL NOT be started.
> 
> < When the peer M2PA receives the Link Status Busy Ended message and T6
> < has not expired, it SHALL stop T6 if T6 is running and start T7 if there are
> < messages awaiting acknowledgement in the retransmission buffer.
> 
> Now, I know that we said that Link Status messages are not used to acknowledge
> DATA messages, but it makes sense to permit the Link Status Busy Ended message
> to acknowledge outstanding User Data messages, so that upon its receipt, T7
> might not need to be started if all messages in the retransmission buffer are
> acknolwedged by the Link Status Busy Ended message.
> 
> The above also kind of changes the idea of T7 metering the oldest message in
> the retransmission buffer.  It would now measure the oldest message in the
> retransmission buffer since the receipt of the last Link Status Busy Ended.
> 
> M2PA-09 says:
> 
> 
>>The peer M2PA SHOULD continue transmitting and acknowledging messages
>>while its T6 timer is running, i.e., while the other end is Busy.
> 
> 
> But T6 is not necessarily running during the busy condition.
> 
> Change M2PA to say:
> 
> < The peer M2PA SHOULD continue transmitting and acknowledging messages
> < while the other end is Busy: i.e, between receiving Link Status Busy
> < and Link Status Busy Ended.
> 
> --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.  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 exim@www1.ietf.org  Tue Jul 29 18:07:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA12110
	for <sigtran-archive@odin.ietf.org>; Tue, 29 Jul 2003 18:07:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hccc-0000qF-Kh
	for sigtran-archive@odin.ietf.org; Tue, 29 Jul 2003 18:07:06 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6TM76SN003223
	for sigtran-archive@odin.ietf.org; Tue, 29 Jul 2003 18:07:06 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hccX-0000n1-GN; Tue, 29 Jul 2003 18:07:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hcbc-0000iB-6t
	for sigtran@optimus.ietf.org; Tue, 29 Jul 2003 18:06:04 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA11920
	for <sigtran@ietf.org>; Tue, 29 Jul 2003 18:05:57 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hcbZ-0002Ns-00
	for sigtran@ietf.org; Tue, 29 Jul 2003 18:06:01 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[vdrBNByxTiJWQkGcgvnjrNTuj6xT4HjB])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hcbY-0002Np-00
	for sigtran@ietf.org; Tue, 29 Jul 2003 18:06:00 -0400
Received: from ns.pigworks.openss7.net (IDENT:HgRhIv0fVjddgS8tHLLHLqLBHWovjqGb@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6TM5sE30193;
	Tue, 29 Jul 2003 16:05:54 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6TM5ra05733;
	Tue, 29 Jul 2003 16:05:53 -0600
Date: Tue, 29 Jul 2003 16:05:53 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tom George <Tom.George@alcatel.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - T6
Message-ID: <20030729160553.A5718@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tom George <Tom.George@alcatel.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030712002921.A21136@openss7.org> <3F26EEFA.7040203@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: <3F26EEFA.7040203@alcatel.com>; from Tom.George@alcatel.com on Tue, Jul 29, 2003 at 05:02:34PM -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>

Tom,

Right.  That was my thinking too.

--brian

Tom George wrote:                                                                      (Tue, 29 Jul 2003 17:02:34)
> Brian,
> 
> The old text about ignoring the BSN in the LS messages has been taken 
> out in version 9. We explicitly mention that the BSN in the LS PO 
> message can be used, but there is no reason to restrict the use of the 
> BSN in other LS messages.
> 
> Right?
> 
> tomG.
> 
> Brian F. G. Bidulock wrote:
>    [...]
> > Now, I know that we said that Link Status messages are not used to acknowledge
> > DATA messages, but it makes sense to permit the Link Status Busy Ended message
> > to acknowledge outstanding User Data messages, so that upon its receipt, T7
> > might not need to be started if all messages in the retransmission buffer are
> > acknolwedged by the Link Status Busy Ended message.
> > 
> > The above also kind of changes the idea of T7 metering the oldest message in
> > the retransmission buffer.  It would now measure the oldest message in the
> > retransmission buffer since the receipt of the last Link Status Busy Ended.
>    [...]
> 
> -- 
> * 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.  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 exim@www1.ietf.org  Tue Jul 29 22:10:15 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA19819
	for <sigtran-archive@odin.ietf.org>; Tue, 29 Jul 2003 22:10:15 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hgPX-0003Qe-1H
	for sigtran-archive@odin.ietf.org; Tue, 29 Jul 2003 22:09:51 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6U29pSc013159
	for sigtran-archive@odin.ietf.org; Tue, 29 Jul 2003 22:09:51 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hgOi-0003OV-UD; Tue, 29 Jul 2003 22:09:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hgO9-0003Mi-2U
	for sigtran@optimus.ietf.org; Tue, 29 Jul 2003 22:08:25 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA19801
	for <sigtran@ietf.org>; Tue, 29 Jul 2003 22:08:18 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hgO6-0004BD-00
	for sigtran@ietf.org; Tue, 29 Jul 2003 22:08:22 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19hgO5-0004BA-00
	for sigtran@ietf.org; Tue, 29 Jul 2003 22:08:21 -0400
Received: from [172.16.1.40] (helo=ccpu00251)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19hgNw-00076C-Qf
	for <sigtran@ietf.org>; Tue, 29 Jul 2003 19:08:12 -0700
Reply-To: <umesh@ccpu.com>
From: "Umesh Vats" <umesh@ccpu.com>
To: <sigtran@ietf.org>
Date: Tue, 29 Jul 2003 19:14:36 -0700
Message-ID: <ACEMIJONPBBPDLOKCABHOEIHCBAA.umesh@ccpu.com>
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 IMO, Build 9.0.2416 (9.0.2911.0)
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Importance: Normal
Content-Transfer-Encoding: 7bit
Subject: [Sigtran] M3UA : IPSP DE behaviour
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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

Suppose we have two Double Ended IPSPs A and B. AS1 is application server
on IPSP-A and AS2 is application server on IPSP-B

     AS1 A----------B AS2

Now, IPSP-A sends ASPAC for for AS1 to B. IPSP-B doesn't activate AS2.
After this, can IPSP-B send DATA messages for AS1 to IPSP-A in any scenario
?

Thanks
Umesh Vats
Sr Software Engg.
extn : 8986


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



From exim@www1.ietf.org  Wed Jul 30 00:25:09 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA22154
	for <sigtran-archive@odin.ietf.org>; Wed, 30 Jul 2003 00:25:09 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hiW5-000759-4z
	for sigtran-archive@odin.ietf.org; Wed, 30 Jul 2003 00:24:45 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6U4Ojxu027202
	for sigtran-archive@odin.ietf.org; Wed, 30 Jul 2003 00:24:45 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hiVN-00072T-Mn; Wed, 30 Jul 2003 00:24:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hiUp-00071t-Rx
	for sigtran@optimus.ietf.org; Wed, 30 Jul 2003 00:23:28 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA22038
	for <sigtran@ietf.org>; Wed, 30 Jul 2003 00:23:21 -0400 (EDT)
From: kakaul@hss.hns.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hiUn-0004dk-00
	for sigtran@ietf.org; Wed, 30 Jul 2003 00:23:25 -0400
Received: from [202.54.26.202] (helo=hindon.hss.co.in)
	by ietf-mx with esmtp (Exim 4.12)
	id 19hiUl-0004dR-00
	for sigtran@ietf.org; Wed, 30 Jul 2003 00:23:24 -0400
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h6U4J1s16426;
	Wed, 30 Jul 2003 09:49:01 +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 ESMTP id h6U4Mou25127;
	Wed, 30 Jul 2003 09:52:51 +0530 (IST)
Subject: Re: [Sigtran] M3UA : IPSP DE behaviour
To: <umesh@ccpu.com>
Cc: <sigtran@ietf.org>
X-Mailer: Lotus Notes Release 5.0.5  September 22, 2000
Message-ID: <OF79E83B5A.4B07B003-ON65256D73.00169177@hss.hns.com>
Date: Wed, 30 Jul 2003 09:47:21 +0530
X-MIMETrack: Serialize by Router on Sandesh/HSS(Release 6.0|September 26, 2002) at 30/07/2003
 09:46:01 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>





Umesh

  IPSP B can send DATA messages for AS1 to IPSP A only if on receiving
ASPAC message from IPSP A , IPSP B sends ASPAC ack for AS1 to IPSP A.Not
activating AS2 means IPSP B will not be able to receive any message for
AS2.So in a way only data for AS1 will flow from IPSP B to IPSP A.

Regards
Kamesh Kaul
www.hssworld.com
Hughes Software Systems




"Umesh Vats" <umesh@ccpu.com>@ietf.org on 07/30/2003 07:44:36 AM

Please respond to <umesh@ccpu.com>

Sent by:    sigtran-admin@ietf.org


To:    <sigtran@ietf.org>
cc:

Subject:    [Sigtran] M3UA : IPSP DE behaviour


Suppose we have two Double Ended IPSPs A and B. AS1 is application server
on IPSP-A and AS2 is application server on IPSP-B

     AS1 A----------B AS2

Now, IPSP-A sends ASPAC for for AS1 to B. IPSP-B doesn't activate AS2.
After this, can IPSP-B send DATA messages for AS1 to IPSP-A in any scenario
?

Thanks
Umesh Vats
Sr Software Engg.
extn : 8986


_______________________________________________
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 exim@www1.ietf.org  Wed Jul 30 10:04:14 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA01444
	for <sigtran-archive@odin.ietf.org>; Wed, 30 Jul 2003 10:04:14 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hrYT-0001Xl-MN
	for sigtran-archive@odin.ietf.org; Wed, 30 Jul 2003 10:03:49 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6UE3nJm005917
	for sigtran-archive@odin.ietf.org; Wed, 30 Jul 2003 10:03:49 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hrXh-0001VP-HI; Wed, 30 Jul 2003 10:03:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hrXU-0001VE-6s
	for sigtran@optimus.ietf.org; Wed, 30 Jul 2003 10:02:48 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA01356
	for <sigtran@ietf.org>; Wed, 30 Jul 2003 10:02:40 -0400 (EDT)
From: kakaul@hss.hns.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hrXQ-0000M6-00
	for sigtran@ietf.org; Wed, 30 Jul 2003 10:02:44 -0400
Received: from [202.54.26.202] (helo=hindon.hss.co.in)
	by ietf-mx with esmtp (Exim 4.12)
	id 19hrXO-0000Kr-00
	for sigtran@ietf.org; Wed, 30 Jul 2003 10:02:43 -0400
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h6UDwHs17944
	for <sigtran@ietf.org>; Wed, 30 Jul 2003 19:28:17 +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 ESMTP id h6UE29b25436
	for <sigtran@ietf.org>; Wed, 30 Jul 2003 19:32:09 +0530 (IST)
Subject: [Sigtran] M3UA : Network appearance in data
To: sigtran@ietf.org
X-Mailer: Lotus Notes Release 5.0.5  September 22, 2000
Message-ID: <OFF36AC4A0.CA56134E-ON65256D73.004C5E8B@hss.hns.com>
Date: Wed, 30 Jul 2003 19:26:40 +0530
X-MIMETrack: Serialize by Router on Sandesh/HSS(Release 6.0|September 26, 2002) at 30/07/2003
 07:25:19 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
      If RC is semi-optional parameter in M3UA Payload Data message what is
the need of network appearance parameter     in Payload Data message even
though it is an optional parameter in RFC 3332. From RC we can directly get
the network appearance associated with the data message.Please comment as i
have missed the earlier discussions on this.

Kamesh Kaul
www.hssworld.com
Hughes Software Systems

"DISCLAIMER: This message is proprietary to Hughes Software Systems Limited
(HSS) and is intended solely for the use of the individual to whom it is
addressed. It may contain  privileged or confidential information and
should not be circulated or used for any purpose other than for what it is
intended. If you have received this message in error, please notify the
originator immediately. If you are not the intended recipient, you are
notified that you are strictly prohibited from using, copying, altering, or
disclosing the contents of this message. HSS accepts no responsibility for
loss or damage arising from the use of the information transmitted by this
email including damage from virus."


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



From exim@www1.ietf.org  Wed Jul 30 11:31:07 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA05011
	for <sigtran-archive@odin.ietf.org>; Wed, 30 Jul 2003 11:31:07 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hsuW-0004iJ-Lh
	for sigtran-archive@odin.ietf.org; Wed, 30 Jul 2003 11:30:42 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6UFUetg018113
	for sigtran-archive@odin.ietf.org; Wed, 30 Jul 2003 11:30:40 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hstt-0004e3-VS; Wed, 30 Jul 2003 11:30:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hst3-0004dS-Ev
	for sigtran@optimus.ietf.org; Wed, 30 Jul 2003 11:29:09 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA04968
	for <sigtran@ietf.org>; Wed, 30 Jul 2003 11:29:04 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hst2-00016q-00
	for sigtran@ietf.org; Wed, 30 Jul 2003 11:29:08 -0400
Received: from auds953.usa.alcatel.com ([143.209.238.6])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hst1-00016B-00
	for sigtran@ietf.org; Wed, 30 Jul 2003 11:29:07 -0400
Received: from alcatel.com (localhost [127.0.0.1])
	by auds953.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6UFSaLX004977;
	Wed, 30 Jul 2003 10:28:36 -0500 (CDT)
Message-ID: <3F27E424.5090007@alcatel.com>
Date: Wed, 30 Jul 2003 10:28:36 -0500
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
User-Agent: Mozilla/5.0 (X11; U; SunOS sun4u; en-US; rv:1.0.1) Gecko/20020920 Netscape/7.0
X-Accept-Language: en-us, en
MIME-Version: 1.0
CC: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - LPO
References: <99522F0FD7EFD611A47D0002A58EDAB7C84A8F@mchh2a8e.mchh.siemens.de> <20030725174358.B29483@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

Brian, Andrew, Franz, et. al.,

I changed the 2nd paragraph in 4.1.4 to this:

=====
When the local processor outage condition ends, M2PA SHALL send a Link
Status Processor Recovered message to its peer. This message is used
to signal the end of the processor outage condition, instead of an MSU
or FISU as in MTP2. The BSN in the Link Status Processor Recovered
message is set to the FSN of the last User Data message received from
the peer M2PA. The BSN value in the Link Status Processor Recovered
message is used by the peer to resynchronize its sequence numbers, if
this is required by MTP2.
=====

Here is the old paragraph for reference:
=====
When the local processor outage condition ends, M2PA SHALL send a Link
Status Processor Recovered message to its peer. This message is used
to signal in the end of the processor outage condition, instead of an
MSU or FISU as in MTP2. The BSN value in this message is used by the
peer to resynchronize its sequence numbers, if this is required by
MTP2.
=====

Is this the only change that needs to be made from this discussion thread?

tomG.

Brian F. G. Bidulock wrote:
> Franz,
> 
> (2) is already stated.  I have no problem with adding (1) to the description
> of handling of buffers at the side experiencing processor outage.
> 
> --brian
> 
> Gaiser Franz wrote:                                                                 (Fri, 25 Jul 2003 13:33:55)
> 
>>Hi All,
>>
>>From a developer point of view I would prefer that the statements about
>>resync are described more precisely and not in a way that the developer has
>>to interpret it especially when he was not involved in the development
>>process of M2PA.
>> 
>>Why don't complement it with the additional remarks i.e. made by Andrew that 
>>(1) the BSN in the LS PR is set to the last message received from the M2PA. 
>>(2) the sender always increments the FSN by 1 when sending MSUs. 
>>
>>best regards, Franz
>>
>>
>>
>>
>>
>>-----Ursprungliche Nachricht-----
>>Von: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
>>Gesendet: Donnerstag, 24. Juli 2003 18:11
>>An: Andrew Booth
>>Cc: 'sigtran@ietf.org'
>>Betreff: Re: [Sigtran] M2PA Last Call Comment - LPO
>>
>>
>>Andrew,
>>
>>On the first point, performance related issues are outside the scope
>>of the protocol.
>>
>>On the second point, nowhere in M2PA-09 does it say to stop incrementing
>>BSN.  Just to withhold acknowledgements and buffer during PO.  The
>>peer is welcome, as it says in M2PA-09, to resync off of the BSN in the
>>LPR.  The fact that it doesn't change the FSN for ANSI does not mean that
>>it is not performing the resynchronization.
>>
>>Look at the resync wording in M2PA-09 again.
>>
>>I don't think it need changing.
>>
>>--brian
>>
>>Andrew Booth wrote:                                                                 (Thu, 24 Jul 2003 11:01:02)
>>
>>>Hi Brian,
>>>
>>>I don't think synchronization of sequence numbers after PO is necessary 
>>>or desirable in M2PA.  Can you see any advantage to it?
>>>
>>>I'd like to highlight this point by bringing up your comment after 
>>>example 2 below: in order to avoid potential race conditions the side 
>>>sending LS Processor Recovered should set the BSN to the last FSN 
>>>received by M2PA.  By setting the BSN in this way we expect is that the 
>>>peer will continue sending FSNs "in sequence" i.e. that the 
>>>resynchronization will not actually change the next FSN the peer will 
>>>send.  If that's the behaviour we want, and I think it is, then I 
>>>propose that we specify it directly rather than carry over an 
>>>unnecessary procedure from MTP2.
>>>
>>>More comments below.
>>>
>>>Brian F. G. Bidulock wrote:
>>>
>>>
>>>>Andrew,
>>>>
>>>>Please see comments below...
>>>>
>>>>Andrew Booth wrote:                                                                 (Wed, 23 Jul 2003 21:58:07)
>>>> 
>>>>
>>>>
>>>>>Hi Franz,
>>>>>
>>>>>Before answering your question I'll ask another: do we really need to 
>>>>>resynchronize sequence numbers in M2PA, or can we just mandate that the 
>>>>>sender continues to increment the FSN on every transmitted MSU?
>>>>>- This seems simpler than resynchronizing.
>>>>>- Since M2PA does not retransmit, the motivation for resynchronization 
>>>>>doesn't seem to be present as it does in MTP2.
>>>>>- Some of the examples below give examples where resequencing can go 
>>>>>wrong.  Although these situations are unlikely, the consequences are 
>>>>>unpleasant.
>>>>>
>>>>>Now for my best answer to your question: Option (2), but if there are no 
>>>>>unacknowledged messages then the BSN should be set to the FSN of the 
>>>>>last acknowledged message.
>>>>>Note also that if the receiver of the LS PR uses the BSN to free 
>>>>>messages in the RTB (as you indicated below) then they have to be extra 
>>>>>careful when validating BSNs in ensuing Data messages that may have been 
>>>>>sent before the LS PR. I don't think this is a good idea.
>>>>>
>>>>>Some potential issues with resynchronizing sequence numbers:
>>>>>
>>>>>Note that the sender of LS PR cannot know if the receiver will use the 
>>>>>BSN to resynchronize sequence numbers or not, since it is the MTP in RPO 
>>>>>that decides when RPO is "long duration" via the MTP3 T1 timer.
>>>>>This leads to race conditions in which one side flushes and the other 
>>>>>does not, though they are unlikely.  For example, the following could 
>>>>>occur if an MSU was delayed longer than MTP3 T1 on the IP network:
>>>>>
>>>>>
>>>>>    MTP3                  M2PA                        M2PA
>>>>>      |                     |                          |
>>>>>      |                     |<--Data (FSN 0)-----------|
>>>>>      |                     |                          |
>>>>>      |                     |                     +----|
>>>>>      |----LPO------------->|                    /     |
>>>>>      |                     |----------LS PO----/----->|
>>>>>      |                     |                  /       |
>>>>>      |                     |                 /        |<--Flush
>>>>>      |----LPR + Continue-->|                /         |
>>>>>      |                     |----LS PR------/--------->| resync FSN
>>>>>      |                     |    (BSN 0)   /           |
>>>>>      |                     |             /            |
>>>>>      |                     |<--Data-----+             |
>>>>>      |                     |   (FSN 1)                |
>>>>>      |                     |                          |
>>>>>      |             Discard |<--Data-------------------|
>>>>>      |           (bad FSN) |   (FSN 0)                |
>>>>>   
>>>>>
>>>>
>>>>This is not really probable.  This would mean that the MTP RPO timer
>>>>expires when the LPO timer hasn't (a very narrow window) and that the
>>>>Data (FSN 1) was delayed longer than the RPO timer (which shouldn't
>>>>really happen).
>>>>
>>>
>>>Agreed, it is unlikely.  If SCTP parameters are set correctly it might 
>>>even be impossible.
>>>
>>>
>>>>Set your SCTP parameters (or lifetimes) so that messages not
>>>>acknowledged within the RPO timeout will drop the SCTP association and
>>>>you won't have this problem.
>>>>
>>>>Do you see?  The "long duration" of the procedure (T1 or about 1.0 to
>>>>2.0 seconds) is used so that old messages stored in buffers are not sent
>>>>later.  RTB clear is performed immediately before changeover (after T1
>>>>expiry).  One doesn't want a recovering processor to transfer old
>>>>messages after changeover has occured or a lot of missequencing is going
>>>>to happen in the SS7 network.
>>>>
>>>
>>>Yes I understand, though I think MTP3 T1 is 0.5 to 1.2 seconds.
>>>
>>>
>>>>Now, if you have SCTP parameters set to allow messages to be buffered IN
>>>>THE NETWORK for longer than T1, you have a real big problem: both the
>>>>time-controlled changeover and controlled rerouting procedures will
>>>>result in missequenced messages, even without processor outage.
>>>>
>>>
>>>Agreed.  Should this be stated explicitly in M2PA?
>>>
>>>
>>>>So, if you have a problem above, becuase your SCTP will not drop the
>>>>link before delaying a data message longer than T1, you have the greater
>>>>problem that every time-controlled changeover or controlled rerouting
>>>>procedure for those M2PA links are going to missequence messagses.
>>>>
>>>>Therefore, you need to tighten your SCTP parameters.  If you cannot
>>>>(e.g. the outher end is a really bad implemetnation), you really need to
>>>>bring the link down when the duplicate FSN arrives.
>>>>
>>>>I think you have an error here too: BSN 0 acknowledges the receipt of
>>>>data with FSN 0.  After the resync, the next data sent should have BSN
>>>>1.
>>>>
>>>
>>>Agreed, the last FSN should be 1 (still invalid).
>>>
>>>
>>>>
>>>> 
>>>>
>>>>
>>>>>The following example could occur if the side in LPO flushes while the 
>>>>>side in RPO does not (perhaps they have different T1 timer settings), 
>>>>>and if the side in LPO sends a BSN different from the last received by M2PA:
>>>>>
>>>>>    MTP3                  M2PA                        M2PA
>>>>>      |                     |                          |
>>>>>      |                     |<--Data (FSN 0)-----------|
>>>>>      |                     |                          |
>>>>>      |                     |                     +----|
>>>>>      |----LPO------------->|                    /     |
>>>>>      |                     |----------LS PO----/----->|
>>>>>      |                     |                  /       |
>>>>>      |                     |                 /        |
>>>>>      |               To RB |<--Data---------+         |
>>>>>      |                     |   (FSN 1)                |
>>>>>      |                     |                          |
>>>>>      |----LPR + Flush----->|                          |
>>>>>      |                     |----LS PR---------------->|
>>>>>      |                     |    (BSN 0)               |<- Continue,
>>>>>      |                     |                          |   don't resync FSN
>>>>>      |                     |                          |
>>>>>      |                     |                          |
>>>>>      |             Discard |<--Data (FSN 2)-----------|
>>>>>      |    (expected FSN 1) |                          |
>>>>>      |                     |                          |
>>>>>      |             Discard |<--Data (FSN 3)-----------|
>>>>>      |    (expected FSN 1) |                          |
>>>>>   
>>>>>
>>>>
>>>>Again, having one side flush without the other is a small time window,
>>>>however, why are you sending LS PR (BSN 0)?  I think you should send
>>>>LS PR (BSN 1) in this instance.  Then there is no problem.
>>>>
>>>
>>>I agree this is unlikely, but I don't think the time window is 
>>>necessarily that small: if one side sets T1 to 0.5s and the other to 
>>>1.2s you have 0.7s to hit this.
>>>
>>>
>>>>MTP2 would send BSN 0 here, but it is also discarding messages with the
>>>>expectation of retransmission.  M2PA buffers but does not acknowledge
>>>>messages.  The purpose of delaying acknowledgement is so that the peer
>>>>M2PA keeps these messages in the retransmission buffer for changeover.
>>>>(Remember than some variants collect from the RTB instead of clearing
>>>>it, unlike ANSI).  But, for M2PA, the BSN should be incremented when
>>>>the data is buffered to accomplish what happens here.
>>>>
>>>>So this one is not a problem either.
>>>>
>>>
>>>Agreed, it's not a problem, but it does highlight the point I made 
>>>above: we should set the BSN so that when resynchronizing sequence 
>>>numbers the peer doesn't actually have to change anything.  It seems 
>>>strange to have an exchange of BSN values for this purpose.
>>>
>>>Thoughts?
>>>Andrew
>>>
>>>
>>>>--brian
>>>>
>>>> 
>>>>
>>>>
>>>>>I think this possible condition provides reasonable incentive to always 
>>>>>set the BSN in the LS Processor Recovered to the last message received 
>>>>
>>>>>from the M2PA peer.  In that case, there doesn't seem to be any point in 
>>>>
>>>>>resynching sequence numbers, just let the peer increment the FSN on 
>>>>>every MSU.
>>>>>
>>>>>In summary, these examples are unlikely but why complicate things? 
>>>>>Unless there is a good reason to keep the MTP2 resynching procedures 
>>>>>lets get rid of them in favour of the simpler approach: the sender 
>>>>>always increments the FSN by 1 when sending MSUs.
>>>>>
>>>>>Thoughts?
>>>>>Andrew
>>>>>
>>>>>Gaiser Franz wrote:
>>>>>
>>>>>   
>>>>>
>>>>>
>>>>>>Hi All,
>>>>>>
>>>>>>implementing Processor Outage Handling of M2PA draft 09 the following questions came up:
>>>>>>
>>>>>>The Processor Outage chapter 4.1.4 says
>>>>>>
>>>>>>When the local processor outage conditions ends, M2PA shall send a Link
>>>>>>Status Processor Recovered message to its peer. This message is used
>>>>>>to signal in the end of the processor outage condition, instead of an
>>>>>>MSU or FISU as in MTP2. The BSN value in this message is used by the 
>>>>>>peer to resynchronize its sequence numbers, if this is required by
>>>>>>MTP2.
>>>>>>
>>>>>>What does M2PA draft 09 with BSN in this context mean, is it (assume, MTP3 has send Continue Command)
>>>>>>
>>>>>>(1) the FSN of the last received message before LPO or 
>>>>>>(2) the FSN of the last message in the incoming messages that were queued and unacknowledged
>>>>>>during the processor outage (with that the remote side can release the messages in its RTB) ? 
>>>>>>
>>>>>>
>>>>>>
>>>>>>thanks
>>>>>>best regards, Franz 
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>_______________________________________________
>>>>>>Sigtran mailing list
>>>>>>Sigtran@ietf.org
>>>>>>https://www1.ietf.org/mailman/listinfo/sigtran
>>>>>>
>>>>>>
>>>>>>
>>>>>>     
>>>>>>
>>>>>
>>>>>-- 
>>>>>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.
>>>>>
>>>>>   
>>>>>
>>>>
>>>> 
>>>>
>>>
>>>-- 
>>>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.
>>>
>>
>>-- 
>>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
> 
> 


-- 
* 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.  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 exim@www1.ietf.org  Wed Jul 30 18:42:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA22269
	for <sigtran-archive@odin.ietf.org>; Wed, 30 Jul 2003 18:42:33 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hze5-0000Xm-Ol
	for sigtran-archive@odin.ietf.org; Wed, 30 Jul 2003 18:42:09 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6UMg9NV002090
	for sigtran-archive@odin.ietf.org; Wed, 30 Jul 2003 18:42:09 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hzdx-0000XH-Co; Wed, 30 Jul 2003 18:42:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19hzdH-0000Wr-OE
	for sigtran@optimus.ietf.org; Wed, 30 Jul 2003 18:41:19 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA22251
	for <sigtran@ietf.org>; Wed, 30 Jul 2003 18:41:12 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hzdE-0004hV-00
	for sigtran@ietf.org; Wed, 30 Jul 2003 18:41:16 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[sWSyyN1QaChOqd93CNM3SWdjf6LJsKDn])
	by ietf-mx with esmtp (Exim 4.12)
	id 19hzdD-0004hQ-00
	for sigtran@ietf.org; Wed, 30 Jul 2003 18:41:16 -0400
Received: from ns.pigworks.openss7.net (IDENT:7w2dLetCdjzkS/oe+Fnhur7zR74eTMER@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6UMfFE22896;
	Wed, 30 Jul 2003 16:41:15 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6UMfEq14541;
	Wed, 30 Jul 2003 16:41:14 -0600
Date: Wed, 30 Jul 2003 16:41:14 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: kakaul@hss.hns.com
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA : Network appearance in data
Message-ID: <20030730164114.A14486@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: kakaul@hss.hns.com, sigtran@ietf.org
References: <OFF36AC4A0.CA56134E-ON65256D73.004C5E8B@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: <OFF36AC4A0.CA56134E-ON65256D73.004C5E8B@hss.hns.com>; from kakaul@hss.hns.com on Wed, Jul 30, 2003 at 07:26:40PM +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>

kakaul,

You are correct.  It is not necessary but we agreed that it could be
there.

The reason the RFC is that way is because that was what we finally
agreed to after very long debate.  See the ladders in the mail archive
with subject lines like: "RC in SSNM and DATA," and "RC and NA".

--brian

kakaul@hss.hns.com wrote:                                                           (Wed, 30 Jul 2003 19:26:40)
> 
> 
> 
> 
> Hi
>       If RC is semi-optional parameter in M3UA Payload Data message what is
> the need of network appearance parameter     in Payload Data message even
> though it is an optional parameter in RFC 3332. From RC we can directly get
> the network appearance associated with the data message.Please comment as i
> have missed the earlier discussions on this.
> 
> Kamesh Kaul
> www.hssworld.com
> Hughes Software Systems
> 
> "DISCLAIMER: This message is proprietary to Hughes Software Systems Limited
> (HSS) and is intended solely for the use of the individual to whom it is
> addressed. It may contain  privileged or confidential information and
> should not be circulated or used for any purpose other than for what it is
> intended. If you have received this message in error, please notify the
> originator immediately. If you are not the intended recipient, you are
> notified that you are strictly prohibited from using, copying, altering, or
> disclosing the contents of this message. HSS accepts no responsibility for
> loss or damage arising from the use of the information transmitted by this
> email including damage from virus."
> 
> 
> _______________________________________________
> 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 exim@www1.ietf.org  Thu Jul 31 00:37:36 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA29101
	for <sigtran-archive@odin.ietf.org>; Thu, 31 Jul 2003 00:37:36 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19i5Bf-0002a4-T7
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 00:37:12 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6V4bBfX009918
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 00:37:11 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19i5BV-0002WC-Sd; Thu, 31 Jul 2003 00:37:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19i5Ai-0002UU-0X
	for sigtran@optimus.ietf.org; Thu, 31 Jul 2003 00:36:12 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA29076
	for <sigtran@ietf.org>; Thu, 31 Jul 2003 00:36:05 -0400 (EDT)
From: kakaul@hss.hns.com
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19i5Af-00070h-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 00:36:09 -0400
Received: from [202.54.26.202] (helo=hindon.hss.co.in)
	by ietf-mx with esmtp (Exim 4.12)
	id 19i5Ad-00070C-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 00:36:08 -0400
Received: from ultra.hss.co.in (ultra [192.168.100.5])
	by hindon.hss.co.in (8.10.0/8.10.0) with ESMTP id h6V4Ugs12367;
	Thu, 31 Jul 2003 10:00:42 +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 ESMTP id h6V4YLU28791;
	Thu, 31 Jul 2003 10:04:21 +0530 (IST)
Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
To: "BAND,RICHARD (HP-France,ex1)" <richard.band@hp.com>
Cc: sigtran@ietf.org, umesh@ccpu.com
X-Mailer: Lotus Notes Release 5.0.5  September 22, 2000
Message-ID: <OF57248D73.A3A928EE-ON65256D74.0013FA63@hss.hns.com>
Date: Thu, 31 Jul 2003 09:58:48 +0530
X-MIMETrack: Serialize by Router on Sandesh/HSS(Release 6.0|September 26, 2002) at 31/07/2003
 09:57:41 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>





Hi Richard,
  You are right that there is no point in activating only AS1in IPSP DE
mode in the given example. Unless both the AS 's AS1 and AS2 are not active
user to user traffic (bidirectional) cannot be started in IPSP model.Any
way IPSP DE mode for that reason is kept optional.Bidirectional traffic
between IPSP A and IPSP B will only start when both the AS's are active. In
my openion let IPSP B indicate availability of remote AS1as and when it
sends ASPAC ack for AS1 to IPSP A.If IPSP B is interested in sending data
for remote AS1 it can send and IPSP A will receive the data, if some one
wants single direction traffic.But yes i do argee that there is no
practical scenario in which single direction traffic is desired in IPSP
model.

Kamesh Kaul
www.hssworld.com
Hughes Software Systems




"BAND,RICHARD (HP-France,ex1)" <richard.band@hp.com> on 07/30/2003 05:33:53
PM

To:    Kamesh Kaul/HSS@HSS, umesh@ccpu.com
cc:    sigtran@ietf.org

Subject:    RE: [Sigtran] M3UA : IPSP DE behaviour


Hi Kamesh, Umesh,

I agree that the protocol allows traffic from IPSP B to A without having
traffic from IPSP A to B.

However, I do think that there is a problem with the notifications for the
AS on each side:
In this situation, how should the IPSPs A and B indicate the availability
of
the remote AS to the local AS?
Should IPSP A indicate that AS 2 is available, even though it knows AS 2
cannot reply?
Should IPSP B indicate that AS 1 in unavailable, as AS 2 cannot send any
messages to AS 1?
If so, what should IPSP B do when AS 1 sends traffic? Discard the messages,
or transfer the data to AS 2, which will then receive traffic from an
unavailable destination?

In my opinion, both IPSPs should indicate that the remote AS is
unavailable.
I would say that the DE IPSP exchange mode allows each side to indicate its
readiness to receive traffic, but traffic should not be started unless both
sides are active.
It seems to me that this is the only to avoid the problems described above.

Please let me know your opinion on this point.

Regards,

Richard Band

-----Original Message-----
From: kakaul@hss.hns.com [mailto:kakaul@hss.hns.com]
Sent: Wednesday, July 30, 2003 06:17
To: umesh@ccpu.com
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M3UA : IPSP DE behaviour




Umesh

  IPSP B can send DATA messages for AS1 to IPSP A only if on receiving
ASPAC message from IPSP A , IPSP B sends ASPAC ack for AS1 to IPSP A.Not
activating AS2 means IPSP B will not be able to receive any message for
AS2.So in a way only data for AS1 will flow from IPSP B to IPSP A.

Regards
Kamesh Kaul
www.hssworld.com
Hughes Software Systems




"Umesh Vats" <umesh@ccpu.com>@ietf.org on 07/30/2003 07:44:36 AM

Please respond to <umesh@ccpu.com>

Sent by:    sigtran-admin@ietf.org


To:    <sigtran@ietf.org>
cc:

Subject:    [Sigtran] M3UA : IPSP DE behaviour


Suppose we have two Double Ended IPSPs A and B. AS1 is application server
on IPSP-A and AS2 is application server on IPSP-B

     AS1 A----------B AS2

Now, IPSP-A sends ASPAC for for AS1 to B. IPSP-B doesn't activate AS2.
After this, can IPSP-B send DATA messages for AS1 to IPSP-A in any scenario
?

Thanks
Umesh Vats
Sr Software Engg.
extn : 8986


_______________________________________________
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 exim@www1.ietf.org  Thu Jul 31 03:49:39 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA15097
	for <sigtran-archive@odin.ietf.org>; Thu, 31 Jul 2003 03:49:39 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19i8BU-0003G3-Bg
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 03:49:12 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6V7nCLA012515
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 03:49:12 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19i8BJ-0003Ci-NG; Thu, 31 Jul 2003 03:49:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19i8Af-0003Bc-U2
	for sigtran@optimus.ietf.org; Thu, 31 Jul 2003 03:48:22 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA15086
	for <sigtran@ietf.org>; Thu, 31 Jul 2003 03:48:17 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19i8Ac-0000TI-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 03:48:18 -0400
Received: from thoth.sbs.de ([192.35.17.2])
	by ietf-mx with esmtp (Exim 4.12)
	id 19i8Ab-0000TF-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 03:48:17 -0400
Received: from mail2.siemens.de (mail2.siemens.de [139.25.208.11])
	by thoth.sbs.de (8.11.7/8.11.7) with ESMTP id h6V7mFQ07704;
	Thu, 31 Jul 2003 09:48:15 +0200 (MEST)
Received: from moody.mchh.siemens.de (moody.mchh.siemens.de [139.21.205.85])
	by mail2.siemens.de (8.11.7/8.11.7) with ESMTP id h6V7mDJ01418;
	Thu, 31 Jul 2003 09:48:13 +0200 (MEST)
Received: from mchh273e.demchh201e.icn.siemens.de (mchh273e.mchh.siemens.de [139.21.200.83])
	by moody.mchh.siemens.de (8.9.3/8.9.1) with ESMTP id JAA05367;
	Thu, 31 Jul 2003 09:48:13 +0200 (MET DST)
Received: by mchh273e.mchh.siemens.de with Internet Mail Service (5.5.2656.59)
	id <PLMKSVKL>; Thu, 31 Jul 2003 09:47:36 +0200
Message-ID: <99522F0FD7EFD611A47D0002A58EDAB7C84A98@mchh2a8e.mchh.siemens.de>
From: Gaiser Franz <franz.gaiser@siemens.com>
To: "'Tom George'" <Tom.George@alcatel.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: AW: [Sigtran] M2PA Last Call Comment - LPO
Date: Thu, 31 Jul 2003 09:47:34 +0200
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2656.59)
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>

Tom,

I think we need an additional remark to support the following situation mentioned by Andrew (see figure below):

To avoid that after CONTINUE at the right side the restarted T7 timer will expire (Data with fsn=1 has not been acknowledged) we should permit the Link Status Busy message to acknowledge outstanding User data messages, so that upon its receipt, T7 might not need to be started if all messages in the RTB are acknowledged by the LS PR message.


The following example could occur if the side in LPO flushes while the 
side in RPO does not (perhaps they have different T1 timer settings)

     MTP3                  M2PA                        M2PA
       |                     |                          |
       |                     |<--Data (FSN 0)-----------|
       |                     |                          |
       |                     |                     +----|
       |----LPO------------->|                    /     |
       |                     |----------LS PO----/----->|
       |                     |                  /       |
       |                     |                 /        |
       |               To RB |<--Data---------+         |
       |                     |   (FSN 1)                |
       |                     |                          |
       |----LPR + Flush----->|                          |
       |                     |----LS PR---------------->|
       |                     |    (BSN 1)               |<- Continue,
       |                     |                          |   don't resync FSN
       |                     |                          |
       |                     |                          |
       |               To RB |<--Data (FSN 2)-----------|


Best regards, Franz


-----Ursprungliche Nachricht-----
Von: Tom George [mailto:Tom.George@alcatel.com]
Gesendet: Mittwoch, 30. Juli 2003 17:29
Cc: 'sigtran@ietf.org'
Betreff: Re: [Sigtran] M2PA Last Call Comment - LPO


Brian, Andrew, Franz, et. al.,

I changed the 2nd paragraph in 4.1.4 to this:

=====
When the local processor outage condition ends, M2PA SHALL send a Link
Status Processor Recovered message to its peer. This message is used
to signal the end of the processor outage condition, instead of an MSU
or FISU as in MTP2. The BSN in the Link Status Processor Recovered
message is set to the FSN of the last User Data message received from
the peer M2PA. The BSN value in the Link Status Processor Recovered
message is used by the peer to resynchronize its sequence numbers, if
this is required by MTP2.
=====

Here is the old paragraph for reference:
=====
When the local processor outage condition ends, M2PA SHALL send a Link
Status Processor Recovered message to its peer. This message is used
to signal in the end of the processor outage condition, instead of an
MSU or FISU as in MTP2. The BSN value in this message is used by the
peer to resynchronize its sequence numbers, if this is required by
MTP2.
=====

Is this the only change that needs to be made from this discussion thread?

tomG.

Brian F. G. Bidulock wrote:
> Franz,
> 
> (2) is already stated.  I have no problem with adding (1) to the description
> of handling of buffers at the side experiencing processor outage.
> 
> --brian
> 
> Gaiser Franz wrote:                                                                 (Fri, 25 Jul 2003 13:33:55)
> 
>>Hi All,
>>
>>From a developer point of view I would prefer that the statements about
>>resync are described more precisely and not in a way that the developer has
>>to interpret it especially when he was not involved in the development
>>process of M2PA.
>> 
>>Why don't complement it with the additional remarks i.e. made by Andrew that 
>>(1) the BSN in the LS PR is set to the last message received from the M2PA. 
>>(2) the sender always increments the FSN by 1 when sending MSUs. 
>>
>>best regards, Franz
>>
>>
>>
>>
>>
>>-----Ursprungliche Nachricht-----
>>Von: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
>>Gesendet: Donnerstag, 24. Juli 2003 18:11
>>An: Andrew Booth
>>Cc: 'sigtran@ietf.org'
>>Betreff: Re: [Sigtran] M2PA Last Call Comment - LPO
>>
>>
>>Andrew,
>>
>>On the first point, performance related issues are outside the scope
>>of the protocol.
>>
>>On the second point, nowhere in M2PA-09 does it say to stop incrementing
>>BSN.  Just to withhold acknowledgements and buffer during PO.  The
>>peer is welcome, as it says in M2PA-09, to resync off of the BSN in the
>>LPR.  The fact that it doesn't change the FSN for ANSI does not mean that
>>it is not performing the resynchronization.
>>
>>Look at the resync wording in M2PA-09 again.
>>
>>I don't think it need changing.
>>
>>--brian
>>
>>Andrew Booth wrote:                                                                 (Thu, 24 Jul 2003 11:01:02)
>>
>>>Hi Brian,
>>>
>>>I don't think synchronization of sequence numbers after PO is necessary 
>>>or desirable in M2PA.  Can you see any advantage to it?
>>>
>>>I'd like to highlight this point by bringing up your comment after 
>>>example 2 below: in order to avoid potential race conditions the side 
>>>sending LS Processor Recovered should set the BSN to the last FSN 
>>>received by M2PA.  By setting the BSN in this way we expect is that the 
>>>peer will continue sending FSNs "in sequence" i.e. that the 
>>>resynchronization will not actually change the next FSN the peer will 
>>>send.  If that's the behaviour we want, and I think it is, then I 
>>>propose that we specify it directly rather than carry over an 
>>>unnecessary procedure from MTP2.
>>>
>>>More comments below.
>>>
>>>Brian F. G. Bidulock wrote:
>>>
>>>
>>>>Andrew,
>>>>
>>>>Please see comments below...
>>>>
>>>>Andrew Booth wrote:                                                                 (Wed, 23 Jul 2003 21:58:07)
>>>> 
>>>>
>>>>
>>>>>Hi Franz,
>>>>>
>>>>>Before answering your question I'll ask another: do we really need to 
>>>>>resynchronize sequence numbers in M2PA, or can we just mandate that the 
>>>>>sender continues to increment the FSN on every transmitted MSU?
>>>>>- This seems simpler than resynchronizing.
>>>>>- Since M2PA does not retransmit, the motivation for resynchronization 
>>>>>doesn't seem to be present as it does in MTP2.
>>>>>- Some of the examples below give examples where resequencing can go 
>>>>>wrong.  Although these situations are unlikely, the consequences are 
>>>>>unpleasant.
>>>>>
>>>>>Now for my best answer to your question: Option (2), but if there are no 
>>>>>unacknowledged messages then the BSN should be set to the FSN of the 
>>>>>last acknowledged message.
>>>>>Note also that if the receiver of the LS PR uses the BSN to free 
>>>>>messages in the RTB (as you indicated below) then they have to be extra 
>>>>>careful when validating BSNs in ensuing Data messages that may have been 
>>>>>sent before the LS PR. I don't think this is a good idea.
>>>>>
>>>>>Some potential issues with resynchronizing sequence numbers:
>>>>>
>>>>>Note that the sender of LS PR cannot know if the receiver will use the 
>>>>>BSN to resynchronize sequence numbers or not, since it is the MTP in RPO 
>>>>>that decides when RPO is "long duration" via the MTP3 T1 timer.
>>>>>This leads to race conditions in which one side flushes and the other 
>>>>>does not, though they are unlikely.  For example, the following could 
>>>>>occur if an MSU was delayed longer than MTP3 T1 on the IP network:
>>>>>
>>>>>
>>>>>    MTP3                  M2PA                        M2PA
>>>>>      |                     |                          |
>>>>>      |                     |<--Data (FSN 0)-----------|
>>>>>      |                     |                          |
>>>>>      |                     |                     +----|
>>>>>      |----LPO------------->|                    /     |
>>>>>      |                     |----------LS PO----/----->|
>>>>>      |                     |                  /       |
>>>>>      |                     |                 /        |<--Flush
>>>>>      |----LPR + Continue-->|                /         |
>>>>>      |                     |----LS PR------/--------->| resync FSN
>>>>>      |                     |    (BSN 0)   /           |
>>>>>      |                     |             /            |
>>>>>      |                     |<--Data-----+             |
>>>>>      |                     |   (FSN 1)                |
>>>>>      |                     |                          |
>>>>>      |             Discard |<--Data-------------------|
>>>>>      |           (bad FSN) |   (FSN 0)                |
>>>>>   
>>>>>
>>>>
>>>>This is not really probable.  This would mean that the MTP RPO timer
>>>>expires when the LPO timer hasn't (a very narrow window) and that the
>>>>Data (FSN 1) was delayed longer than the RPO timer (which shouldn't
>>>>really happen).
>>>>
>>>
>>>Agreed, it is unlikely.  If SCTP parameters are set correctly it might 
>>>even be impossible.
>>>
>>>
>>>>Set your SCTP parameters (or lifetimes) so that messages not
>>>>acknowledged within the RPO timeout will drop the SCTP association and
>>>>you won't have this problem.
>>>>
>>>>Do you see?  The "long duration" of the procedure (T1 or about 1.0 to
>>>>2.0 seconds) is used so that old messages stored in buffers are not sent
>>>>later.  RTB clear is performed immediately before changeover (after T1
>>>>expiry).  One doesn't want a recovering processor to transfer old
>>>>messages after changeover has occured or a lot of missequencing is going
>>>>to happen in the SS7 network.
>>>>
>>>
>>>Yes I understand, though I think MTP3 T1 is 0.5 to 1.2 seconds.
>>>
>>>
>>>>Now, if you have SCTP parameters set to allow messages to be buffered IN
>>>>THE NETWORK for longer than T1, you have a real big problem: both the
>>>>time-controlled changeover and controlled rerouting procedures will
>>>>result in missequenced messages, even without processor outage.
>>>>
>>>
>>>Agreed.  Should this be stated explicitly in M2PA?
>>>
>>>
>>>>So, if you have a problem above, becuase your SCTP will not drop the
>>>>link before delaying a data message longer than T1, you have the greater
>>>>problem that every time-controlled changeover or controlled rerouting
>>>>procedure for those M2PA links are going to missequence messagses.
>>>>
>>>>Therefore, you need to tighten your SCTP parameters.  If you cannot
>>>>(e.g. the outher end is a really bad implemetnation), you really need to
>>>>bring the link down when the duplicate FSN arrives.
>>>>
>>>>I think you have an error here too: BSN 0 acknowledges the receipt of
>>>>data with FSN 0.  After the resync, the next data sent should have BSN
>>>>1.
>>>>
>>>
>>>Agreed, the last FSN should be 1 (still invalid).
>>>
>>>
>>>>
>>>> 
>>>>
>>>>
>>>>>The following example could occur if the side in LPO flushes while the 
>>>>>side in RPO does not (perhaps they have different T1 timer settings), 
>>>>>and if the side in LPO sends a BSN different from the last received by M2PA:
>>>>>
>>>>>    MTP3                  M2PA                        M2PA
>>>>>      |                     |                          |
>>>>>      |                     |<--Data (FSN 0)-----------|
>>>>>      |                     |                          |
>>>>>      |                     |                     +----|
>>>>>      |----LPO------------->|                    /     |
>>>>>      |                     |----------LS PO----/----->|
>>>>>      |                     |                  /       |
>>>>>      |                     |                 /        |
>>>>>      |               To RB |<--Data---------+         |
>>>>>      |                     |   (FSN 1)                |
>>>>>      |                     |                          |
>>>>>      |----LPR + Flush----->|                          |
>>>>>      |                     |----LS PR---------------->|
>>>>>      |                     |    (BSN 0)               |<- Continue,
>>>>>      |                     |                          |   don't resync FSN
>>>>>      |                     |                          |
>>>>>      |                     |                          |
>>>>>      |             Discard |<--Data (FSN 2)-----------|
>>>>>      |    (expected FSN 1) |                          |
>>>>>      |                     |                          |
>>>>>      |             Discard |<--Data (FSN 3)-----------|
>>>>>      |    (expected FSN 1) |                          |
>>>>>   
>>>>>
>>>>
>>>>Again, having one side flush without the other is a small time window,
>>>>however, why are you sending LS PR (BSN 0)?  I think you should send
>>>>LS PR (BSN 1) in this instance.  Then there is no problem.
>>>>
>>>
>>>I agree this is unlikely, but I don't think the time window is 
>>>necessarily that small: if one side sets T1 to 0.5s and the other to 
>>>1.2s you have 0.7s to hit this.
>>>
>>>
>>>>MTP2 would send BSN 0 here, but it is also discarding messages with the
>>>>expectation of retransmission.  M2PA buffers but does not acknowledge
>>>>messages.  The purpose of delaying acknowledgement is so that the peer
>>>>M2PA keeps these messages in the retransmission buffer for changeover.
>>>>(Remember than some variants collect from the RTB instead of clearing
>>>>it, unlike ANSI).  But, for M2PA, the BSN should be incremented when
>>>>the data is buffered to accomplish what happens here.
>>>>
>>>>So this one is not a problem either.
>>>>
>>>
>>>Agreed, it's not a problem, but it does highlight the point I made 
>>>above: we should set the BSN so that when resynchronizing sequence 
>>>numbers the peer doesn't actually have to change anything.  It seems 
>>>strange to have an exchange of BSN values for this purpose.
>>>
>>>Thoughts?
>>>Andrew
>>>
>>>
>>>>--brian
>>>>
>>>> 
>>>>
>>>>
>>>>>I think this possible condition provides reasonable incentive to always 
>>>>>set the BSN in the LS Processor Recovered to the last message received 
>>>>
>>>>>from the M2PA peer.  In that case, there doesn't seem to be any point in 
>>>>
>>>>>resynching sequence numbers, just let the peer increment the FSN on 
>>>>>every MSU.
>>>>>
>>>>>In summary, these examples are unlikely but why complicate things? 
>>>>>Unless there is a good reason to keep the MTP2 resynching procedures 
>>>>>lets get rid of them in favour of the simpler approach: the sender 
>>>>>always increments the FSN by 1 when sending MSUs.
>>>>>
>>>>>Thoughts?
>>>>>Andrew
>>>>>
>>>>>Gaiser Franz wrote:
>>>>>
>>>>>   
>>>>>
>>>>>
>>>>>>Hi All,
>>>>>>
>>>>>>implementing Processor Outage Handling of M2PA draft 09 the following questions came up:
>>>>>>
>>>>>>The Processor Outage chapter 4.1.4 says
>>>>>>
>>>>>>When the local processor outage conditions ends, M2PA shall send a Link
>>>>>>Status Processor Recovered message to its peer. This message is used
>>>>>>to signal in the end of the processor outage condition, instead of an
>>>>>>MSU or FISU as in MTP2. The BSN value in this message is used by the 
>>>>>>peer to resynchronize its sequence numbers, if this is required by
>>>>>>MTP2.
>>>>>>
>>>>>>What does M2PA draft 09 with BSN in this context mean, is it (assume, MTP3 has send Continue Command)
>>>>>>
>>>>>>(1) the FSN of the last received message before LPO or 
>>>>>>(2) the FSN of the last message in the incoming messages that were queued and unacknowledged
>>>>>>during the processor outage (with that the remote side can release the messages in its RTB) ? 
>>>>>>
>>>>>>
>>>>>>
>>>>>>thanks
>>>>>>best regards, Franz 
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>_______________________________________________
>>>>>>Sigtran mailing list
>>>>>>Sigtran@ietf.org
>>>>>>https://www1.ietf.org/mailman/listinfo/sigtran
>>>>>>
>>>>>>
>>>>>>
>>>>>>     
>>>>>>
>>>>>
>>>>>-- 
>>>>>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.
>>>>>
>>>>>   
>>>>>
>>>>
>>>> 
>>>>
>>>
>>>-- 
>>>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.
>>>
>>
>>-- 
>>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
> 
> 


-- 
* 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.  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 exim@www1.ietf.org  Thu Jul 31 03:49:42 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA15115
	for <sigtran-archive@odin.ietf.org>; Thu, 31 Jul 2003 03:49:42 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19i8BU-0003G2-BX
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 03:49:15 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6V7nChi012514
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 03:49:12 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19i8BK-0003Cy-GH; Thu, 31 Jul 2003 03:49:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19i8B4-0003CB-L7
	for sigtran@optimus.ietf.org; Thu, 31 Jul 2003 03:48:46 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA15089
	for <sigtran@ietf.org>; Thu, 31 Jul 2003 03:48:43 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19i8B2-0000TT-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 03:48:44 -0400
Received: from zdemail03.zdem.compaq.com ([161.114.112.27])
	by ietf-mx with esmtp (Exim 4.12)
	id 19i8B1-0000TE-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 03:48:43 -0400
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail03.zdem.compaq.com (Postfix) with ESMTP
	id 432481279; Thu, 31 Jul 2003 09:48:08 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Thu, 31 Jul 2003 09:48:08 +0200
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
X-MimeOLE: Produced By Microsoft Exchange V6.0.6375.0
Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
Date: Thu, 31 Jul 2003 09:48:07 +0200
Message-ID: <06CAA4D3753C53408D2CE6340B68444214C478@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA : IPSP DE behaviour
Thread-Index: AcNXHcDhKHymXSYIQpWnGzVhfDRdiwAGAM/w
From: "Zannin, Francois" <francois.zannin@hp.com>
To: <kakaul@hss.hns.com>, "Band, Richard" <richard.band@hp.com>
Cc: <sigtran@ietf.org>, <umesh@ccpu.com>
X-OriginalArrivalTime: 31 Jul 2003 07:48:08.0263 (UTC) FILETIME=[153D4D70:01C35738]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

If we consider that the IPSP DE is a replicated SG/ASP model (as already
mentionned
Before), then SG/ASP protocol indicate that the SG cannot send AND
receive traffic=20
from an ASP if the concerned AS(RC) is not activated.=20

When two IPSP are hosting several AS, it seem pertinent to authorize the
traffic only=20
between "ACTIVE AS" in the IPSP.

An active status is reported to the application ONLY if the local AS
(the local application)
And the remote AS are activated. If the remote OR the local AS is
deactivated, then
A pause is reported to the application.

My opinion is that only ACTIVE AS (between IPSP) can use the association
to exchange messages.

Francois Z.

> -----Original Message-----
> From: kakaul@hss.hns.com [mailto:kakaul@hss.hns.com]=20
> Sent: Thursday, July 31, 2003 6:29 AM
> To: Band, Richard
> Cc: sigtran@ietf.org; umesh@ccpu.com
> Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
>=20
>=20
>=20
>=20
>=20
>=20
> Hi Richard,
>   You are right that there is no point in activating only=20
> AS1 in IPSP DE mode in the given example. Unless both the AS=20
> 's AS1 and AS2 are not active user to user traffic=20
> (bidirectional) cannot be started in IPSP model.Any way IPSP=20
> DE mode for that reason is kept optional.Bidirectional=20
> traffic between IPSP A and IPSP B will only start when both=20
> the AS's are active. In my openion let IPSP B indicate=20
> availability of remote AS1as and when it sends ASPAC ack for=20
> AS1 to IPSP A.If IPSP B is interested in sending data for=20
> remote AS1 it can send and IPSP A will receive the data, if=20
> some one wants single direction traffic.But yes i do argee=20
> that there is no practical scenario in which single direction=20
> traffic is desired in IPSP model.
>=20
> Kamesh Kaul
> www.hssworld.com
> Hughes Software Systems
>=20
>=20
>=20
>=20
> "BAND,RICHARD (HP-France,ex1)" <richard.band@hp.com> on=20
> 07/30/2003 05:33:53 PM
>=20
> To:    Kamesh Kaul/HSS@HSS, umesh@ccpu.com
> cc:    sigtran@ietf.org
>=20
> Subject:    RE: [Sigtran] M3UA : IPSP DE behaviour
>=20
>=20
> Hi Kamesh, Umesh,
>=20
> I agree that the protocol allows traffic from IPSP B to A=20
> without having traffic from IPSP A to B.
>=20
> However, I do think that there is a problem with the=20
> notifications for the AS on each side: In this situation, how=20
> should the IPSPs A and B indicate the availability of the=20
> remote AS to the local AS? Should IPSP A indicate that AS 2=20
> is available, even though it knows AS 2 cannot reply? Should=20
> IPSP B indicate that AS 1 in unavailable, as AS 2 cannot send=20
> any messages to AS 1? If so, what should IPSP B do when AS 1=20
> sends traffic? Discard the messages, or transfer the data to=20
> AS 2, which will then receive traffic from an unavailable destination?
>=20
> In my opinion, both IPSPs should indicate that the remote AS=20
> is unavailable. I would say that the DE IPSP exchange mode=20
> allows each side to indicate its readiness to receive=20
> traffic, but traffic should not be started unless both sides=20
> are active. It seems to me that this is the only to avoid the=20
> problems described above.
>=20
> Please let me know your opinion on this point.
>=20
> Regards,
>=20
> Richard Band
>=20
> -----Original Message-----
> From: kakaul@hss.hns.com [mailto:kakaul@hss.hns.com]
> Sent: Wednesday, July 30, 2003 06:17
> To: umesh@ccpu.com
> Cc: sigtran@ietf.org
> Subject: Re: [Sigtran] M3UA : IPSP DE behaviour
>=20
>=20
>=20
>=20
> Umesh
>=20
>   IPSP B can send DATA messages for AS1 to IPSP A only if on=20
> receiving ASPAC message from IPSP A , IPSP B sends ASPAC ack=20
> for AS1 to IPSP A.Not activating AS2 means IPSP B will not be=20
> able to receive any message for AS2.So in a way only data for=20
> AS1 will flow from IPSP B to IPSP A.
>=20
> Regards
> Kamesh Kaul
> www.hssworld.com
> Hughes Software Systems
>=20
>=20
>=20
>=20
> "Umesh Vats" <umesh@ccpu.com>@ietf.org on 07/30/2003 07:44:36 AM
>=20
> Please respond to <umesh@ccpu.com>
>=20
> Sent by:    sigtran-admin@ietf.org
>=20
>=20
> To:    <sigtran@ietf.org>
> cc:
>=20
> Subject:    [Sigtran] M3UA : IPSP DE behaviour
>=20
>=20
> Suppose we have two Double Ended IPSPs A and B. AS1 is=20
> application server on IPSP-A and AS2 is application server on IPSP-B
>=20
>      AS1 A----------B AS2
>=20
> Now, IPSP-A sends ASPAC for for AS1 to B. IPSP-B doesn't=20
> activate AS2. After this, can IPSP-B send DATA messages for=20
> AS1 to IPSP-A in any scenario ?
>=20
> Thanks
> Umesh Vats
> Sr Software Engg.
> extn : 8986
>=20
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
>  https://www1.ietf.org/mailman/listinfo/sigtran
>=20
>=20
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
>  https://www1.ietf.org/mailman/listinfo/sigtran
>=20
>=20
>=20
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>=20

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



From exim@www1.ietf.org  Thu Jul 31 03:55:36 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA15257
	for <sigtran-archive@odin.ietf.org>; Thu, 31 Jul 2003 03:55:36 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19i8HE-0003Xm-UF
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 03:55:09 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6V7t8cu013623
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 03:55:08 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19i8H8-0003Vt-T7; Thu, 31 Jul 2003 03:55:02 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19i8Gk-0003V3-Oz
	for sigtran@optimus.ietf.org; Thu, 31 Jul 2003 03:54:38 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA15227
	for <sigtran@ietf.org>; Thu, 31 Jul 2003 03:54:35 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19i8Gi-0000WA-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 03:54:36 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[dLcL/tK3q+mqeVcQgbXhQebNZ7ty9a7r])
	by ietf-mx with esmtp (Exim 4.12)
	id 19i8Gg-0000W6-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 03:54:35 -0400
Received: from ns.pigworks.openss7.net (IDENT:6nlJNEyCRcwWnJ5pbkfNRbiBquMwSFFO@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6V7sYE24833;
	Thu, 31 Jul 2003 01:54:34 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6V7sX920111;
	Thu, 31 Jul 2003 01:54:33 -0600
Date: Thu, 31 Jul 2003 01:54:33 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Gaiser Franz <franz.gaiser@siemens.com>
Cc: "'Tom George'" <Tom.George@alcatel.com>,
        "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - LPO
Message-ID: <20030731015433.A20078@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Gaiser Franz <franz.gaiser@siemens.com>,
	'Tom George' <Tom.George@alcatel.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <99522F0FD7EFD611A47D0002A58EDAB7C84A98@mchh2a8e.mchh.siemens.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: <99522F0FD7EFD611A47D0002A58EDAB7C84A98@mchh2a8e.mchh.siemens.de>; from franz.gaiser@siemens.com on Thu, Jul 31, 2003 at 09:47:34AM +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>

Gaiser,

I think that is already acheived by Tom's wording (underlined below):

> =====
> When the local processor outage condition ends, M2PA SHALL send a Link
> Status Processor Recovered message to its peer. This message is used
> to signal the end of the processor outage condition, instead of an MSU
> or FISU as in MTP2. The BSN in the Link Status Processor Recovered
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> message is set to the FSN of the last User Data message received from
  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> the peer M2PA. The BSN value in the Link Status Processor Recovered
  ^^^^^^^^^^^^^^
> message is used by the peer to resynchronize its sequence numbers, if
> this is required by MTP2.
> =====

--brian

Gaiser Franz wrote:                                                                 (Thu, 31 Jul 2003 09:47:34)
> Tom,
> 
> I think we need an additional remark to support the following situation
> mentioned by Andrew (see figure below):
> 
> To avoid that after CONTINUE at the right side the restarted T7 timer will
> expire (Data with fsn=1 has not been acknowledged) we should permit the Link
> Status Busy message to acknowledge outstanding User data messages, so that
> upon its receipt, T7 might not need to be started if all messages in the RTB
> are acknowledged by the LS PR message.
> 
> 
> The following example could occur if the side in LPO flushes while the 
> side in RPO does not (perhaps they have different T1 timer settings)
> 
>      MTP3                  M2PA                        M2PA
>        |                     |                          |
>        |                     |<--Data (FSN 0)-----------|
>        |                     |                          |
>        |                     |                     +----|
>        |----LPO------------->|                    /     |
>        |                     |----------LS PO----/----->|
>        |                     |                  /       |
>        |                     |                 /        |
>        |               To RB |<--Data---------+         |
>        |                     |   (FSN 1)                |
>        |                     |                          |
>        |----LPR + Flush----->|                          |
>        |                     |----LS PR---------------->|
>        |                     |    (BSN 1)               |<- Continue,
>        |                     |                          |   don't resync FSN
>        |                     |                          |
>        |                     |                          |
>        |               To RB |<--Data (FSN 2)-----------|
> 
> 
> Best regards, Franz
> 
> 
> -----Ursprungliche Nachricht-----
> Von: Tom George [mailto:Tom.George@alcatel.com]
> Gesendet: Mittwoch, 30. Juli 2003 17:29
> Cc: 'sigtran@ietf.org'
> Betreff: Re: [Sigtran] M2PA Last Call Comment - LPO
> 
> 
> Brian, Andrew, Franz, et. al.,
> 
> I changed the 2nd paragraph in 4.1.4 to this:
> 
> =====
> When the local processor outage condition ends, M2PA SHALL send a Link
> Status Processor Recovered message to its peer. This message is used
> to signal the end of the processor outage condition, instead of an MSU
> or FISU as in MTP2. The BSN in the Link Status Processor Recovered
> message is set to the FSN of the last User Data message received from
> the peer M2PA. The BSN value in the Link Status Processor Recovered
> message is used by the peer to resynchronize its sequence numbers, if
> this is required by MTP2.
> =====
> 
> Here is the old paragraph for reference:
> =====
> When the local processor outage condition ends, M2PA SHALL send a Link
> Status Processor Recovered message to its peer. This message is used
> to signal in the end of the processor outage condition, instead of an
> MSU or FISU as in MTP2. The BSN value in this message is used by the
> peer to resynchronize its sequence numbers, if this is required by
> MTP2.
> =====
> 
> Is this the only change that needs to be made from this discussion thread?
> 
> tomG.
> 
> Brian F. G. Bidulock wrote:
> > Franz,
> > 
> > (2) is already stated.  I have no problem with adding (1) to the description
> > of handling of buffers at the side experiencing processor outage.
> > 
> > --brian
> > 
> > Gaiser Franz wrote:                                                                 (Fri, 25 Jul 2003 13:33:55)
> > 
> >>Hi All,
> >>
> >>From a developer point of view I would prefer that the statements about
> >>resync are described more precisely and not in a way that the developer has
> >>to interpret it especially when he was not involved in the development
> >>process of M2PA.
> >> 
> >>Why don't complement it with the additional remarks i.e. made by Andrew that 
> >>(1) the BSN in the LS PR is set to the last message received from the M2PA. 
> >>(2) the sender always increments the FSN by 1 when sending MSUs. 
> >>
> >>best regards, Franz
> >>
> >>
> >>
> >>
> >>
> >>-----Ursprungliche Nachricht-----
> >>Von: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> >>Gesendet: Donnerstag, 24. Juli 2003 18:11
> >>An: Andrew Booth
> >>Cc: 'sigtran@ietf.org'
> >>Betreff: Re: [Sigtran] M2PA Last Call Comment - LPO
> >>
> >>
> >>Andrew,
> >>
> >>On the first point, performance related issues are outside the scope
> >>of the protocol.
> >>
> >>On the second point, nowhere in M2PA-09 does it say to stop incrementing
> >>BSN.  Just to withhold acknowledgements and buffer during PO.  The
> >>peer is welcome, as it says in M2PA-09, to resync off of the BSN in the
> >>LPR.  The fact that it doesn't change the FSN for ANSI does not mean that
> >>it is not performing the resynchronization.
> >>
> >>Look at the resync wording in M2PA-09 again.
> >>
> >>I don't think it need changing.
> >>
> >>--brian
> >>
> >>Andrew Booth wrote:                                                                 (Thu, 24 Jul 2003 11:01:02)
> >>
> >>>Hi Brian,
> >>>
> >>>I don't think synchronization of sequence numbers after PO is necessary 
> >>>or desirable in M2PA.  Can you see any advantage to it?
> >>>
> >>>I'd like to highlight this point by bringing up your comment after 
> >>>example 2 below: in order to avoid potential race conditions the side 
> >>>sending LS Processor Recovered should set the BSN to the last FSN 
> >>>received by M2PA.  By setting the BSN in this way we expect is that the 
> >>>peer will continue sending FSNs "in sequence" i.e. that the 
> >>>resynchronization will not actually change the next FSN the peer will 
> >>>send.  If that's the behaviour we want, and I think it is, then I 
> >>>propose that we specify it directly rather than carry over an 
> >>>unnecessary procedure from MTP2.
> >>>
> >>>More comments below.
> >>>
> >>>Brian F. G. Bidulock wrote:
> >>>
> >>>
> >>>>Andrew,
> >>>>
> >>>>Please see comments below...
> >>>>
> >>>>Andrew Booth wrote:                                                                 (Wed, 23 Jul 2003 21:58:07)
> >>>> 
> >>>>
> >>>>
> >>>>>Hi Franz,
> >>>>>
> >>>>>Before answering your question I'll ask another: do we really need to 
> >>>>>resynchronize sequence numbers in M2PA, or can we just mandate that the 
> >>>>>sender continues to increment the FSN on every transmitted MSU?
> >>>>>- This seems simpler than resynchronizing.
> >>>>>- Since M2PA does not retransmit, the motivation for resynchronization 
> >>>>>doesn't seem to be present as it does in MTP2.
> >>>>>- Some of the examples below give examples where resequencing can go 
> >>>>>wrong.  Although these situations are unlikely, the consequences are 
> >>>>>unpleasant.
> >>>>>
> >>>>>Now for my best answer to your question: Option (2), but if there are no 
> >>>>>unacknowledged messages then the BSN should be set to the FSN of the 
> >>>>>last acknowledged message.
> >>>>>Note also that if the receiver of the LS PR uses the BSN to free 
> >>>>>messages in the RTB (as you indicated below) then they have to be extra 
> >>>>>careful when validating BSNs in ensuing Data messages that may have been 
> >>>>>sent before the LS PR. I don't think this is a good idea.
> >>>>>
> >>>>>Some potential issues with resynchronizing sequence numbers:
> >>>>>
> >>>>>Note that the sender of LS PR cannot know if the receiver will use the 
> >>>>>BSN to resynchronize sequence numbers or not, since it is the MTP in RPO 
> >>>>>that decides when RPO is "long duration" via the MTP3 T1 timer.
> >>>>>This leads to race conditions in which one side flushes and the other 
> >>>>>does not, though they are unlikely.  For example, the following could 
> >>>>>occur if an MSU was delayed longer than MTP3 T1 on the IP network:
> >>>>>
> >>>>>
> >>>>>    MTP3                  M2PA                        M2PA
> >>>>>      |                     |                          |
> >>>>>      |                     |<--Data (FSN 0)-----------|
> >>>>>      |                     |                          |
> >>>>>      |                     |                     +----|
> >>>>>      |----LPO------------->|                    /     |
> >>>>>      |                     |----------LS PO----/----->|
> >>>>>      |                     |                  /       |
> >>>>>      |                     |                 /        |<--Flush
> >>>>>      |----LPR + Continue-->|                /         |
> >>>>>      |                     |----LS PR------/--------->| resync FSN
> >>>>>      |                     |    (BSN 0)   /           |
> >>>>>      |                     |             /            |
> >>>>>      |                     |<--Data-----+             |
> >>>>>      |                     |   (FSN 1)                |
> >>>>>      |                     |                          |
> >>>>>      |             Discard |<--Data-------------------|
> >>>>>      |           (bad FSN) |   (FSN 0)                |
> >>>>>   
> >>>>>
> >>>>
> >>>>This is not really probable.  This would mean that the MTP RPO timer
> >>>>expires when the LPO timer hasn't (a very narrow window) and that the
> >>>>Data (FSN 1) was delayed longer than the RPO timer (which shouldn't
> >>>>really happen).
> >>>>
> >>>
> >>>Agreed, it is unlikely.  If SCTP parameters are set correctly it might 
> >>>even be impossible.
> >>>
> >>>
> >>>>Set your SCTP parameters (or lifetimes) so that messages not
> >>>>acknowledged within the RPO timeout will drop the SCTP association and
> >>>>you won't have this problem.
> >>>>
> >>>>Do you see?  The "long duration" of the procedure (T1 or about 1.0 to
> >>>>2.0 seconds) is used so that old messages stored in buffers are not sent
> >>>>later.  RTB clear is performed immediately before changeover (after T1
> >>>>expiry).  One doesn't want a recovering processor to transfer old
> >>>>messages after changeover has occured or a lot of missequencing is going
> >>>>to happen in the SS7 network.
> >>>>
> >>>
> >>>Yes I understand, though I think MTP3 T1 is 0.5 to 1.2 seconds.
> >>>
> >>>
> >>>>Now, if you have SCTP parameters set to allow messages to be buffered IN
> >>>>THE NETWORK for longer than T1, you have a real big problem: both the
> >>>>time-controlled changeover and controlled rerouting procedures will
> >>>>result in missequenced messages, even without processor outage.
> >>>>
> >>>
> >>>Agreed.  Should this be stated explicitly in M2PA?
> >>>
> >>>
> >>>>So, if you have a problem above, becuase your SCTP will not drop the
> >>>>link before delaying a data message longer than T1, you have the greater
> >>>>problem that every time-controlled changeover or controlled rerouting
> >>>>procedure for those M2PA links are going to missequence messagses.
> >>>>
> >>>>Therefore, you need to tighten your SCTP parameters.  If you cannot
> >>>>(e.g. the outher end is a really bad implemetnation), you really need to
> >>>>bring the link down when the duplicate FSN arrives.
> >>>>
> >>>>I think you have an error here too: BSN 0 acknowledges the receipt of
> >>>>data with FSN 0.  After the resync, the next data sent should have BSN
> >>>>1.
> >>>>
> >>>
> >>>Agreed, the last FSN should be 1 (still invalid).
> >>>
> >>>
> >>>>
> >>>> 
> >>>>
> >>>>
> >>>>>The following example could occur if the side in LPO flushes while the 
> >>>>>side in RPO does not (perhaps they have different T1 timer settings), 
> >>>>>and if the side in LPO sends a BSN different from the last received by M2PA:
> >>>>>
> >>>>>    MTP3                  M2PA                        M2PA
> >>>>>      |                     |                          |
> >>>>>      |                     |<--Data (FSN 0)-----------|
> >>>>>      |                     |                          |
> >>>>>      |                     |                     +----|
> >>>>>      |----LPO------------->|                    /     |
> >>>>>      |                     |----------LS PO----/----->|
> >>>>>      |                     |                  /       |
> >>>>>      |                     |                 /        |
> >>>>>      |               To RB |<--Data---------+         |
> >>>>>      |                     |   (FSN 1)                |
> >>>>>      |                     |                          |
> >>>>>      |----LPR + Flush----->|                          |
> >>>>>      |                     |----LS PR---------------->|
> >>>>>      |                     |    (BSN 0)               |<- Continue,
> >>>>>      |                     |                          |   don't resync FSN
> >>>>>      |                     |                          |
> >>>>>      |                     |                          |
> >>>>>      |             Discard |<--Data (FSN 2)-----------|
> >>>>>      |    (expected FSN 1) |                          |
> >>>>>      |                     |                          |
> >>>>>      |             Discard |<--Data (FSN 3)-----------|
> >>>>>      |    (expected FSN 1) |                          |
> >>>>>   
> >>>>>
> >>>>
> >>>>Again, having one side flush without the other is a small time window,
> >>>>however, why are you sending LS PR (BSN 0)?  I think you should send
> >>>>LS PR (BSN 1) in this instance.  Then there is no problem.
> >>>>
> >>>
> >>>I agree this is unlikely, but I don't think the time window is 
> >>>necessarily that small: if one side sets T1 to 0.5s and the other to 
> >>>1.2s you have 0.7s to hit this.
> >>>
> >>>
> >>>>MTP2 would send BSN 0 here, but it is also discarding messages with the
> >>>>expectation of retransmission.  M2PA buffers but does not acknowledge
> >>>>messages.  The purpose of delaying acknowledgement is so that the peer
> >>>>M2PA keeps these messages in the retransmission buffer for changeover.
> >>>>(Remember than some variants collect from the RTB instead of clearing
> >>>>it, unlike ANSI).  But, for M2PA, the BSN should be incremented when
> >>>>the data is buffered to accomplish what happens here.
> >>>>
> >>>>So this one is not a problem either.
> >>>>
> >>>
> >>>Agreed, it's not a problem, but it does highlight the point I made 
> >>>above: we should set the BSN so that when resynchronizing sequence 
> >>>numbers the peer doesn't actually have to change anything.  It seems 
> >>>strange to have an exchange of BSN values for this purpose.
> >>>
> >>>Thoughts?
> >>>Andrew
> >>>
> >>>
> >>>>--brian
> >>>>
> >>>> 
> >>>>
> >>>>
> >>>>>I think this possible condition provides reasonable incentive to always 
> >>>>>set the BSN in the LS Processor Recovered to the last message received 
> >>>>
> >>>>>from the M2PA peer.  In that case, there doesn't seem to be any point in 
> >>>>
> >>>>>resynching sequence numbers, just let the peer increment the FSN on 
> >>>>>every MSU.
> >>>>>
> >>>>>In summary, these examples are unlikely but why complicate things? 
> >>>>>Unless there is a good reason to keep the MTP2 resynching procedures 
> >>>>>lets get rid of them in favour of the simpler approach: the sender 
> >>>>>always increments the FSN by 1 when sending MSUs.
> >>>>>
> >>>>>Thoughts?
> >>>>>Andrew
> >>>>>
> >>>>>Gaiser Franz wrote:
> >>>>>
> >>>>>   
> >>>>>
> >>>>>
> >>>>>>Hi All,
> >>>>>>
> >>>>>>implementing Processor Outage Handling of M2PA draft 09 the following questions came up:
> >>>>>>
> >>>>>>The Processor Outage chapter 4.1.4 says
> >>>>>>
> >>>>>>When the local processor outage conditions ends, M2PA shall send a Link
> >>>>>>Status Processor Recovered message to its peer. This message is used
> >>>>>>to signal in the end of the processor outage condition, instead of an
> >>>>>>MSU or FISU as in MTP2. The BSN value in this message is used by the 
> >>>>>>peer to resynchronize its sequence numbers, if this is required by
> >>>>>>MTP2.
> >>>>>>
> >>>>>>What does M2PA draft 09 with BSN in this context mean, is it (assume, MTP3 has send Continue Command)
> >>>>>>
> >>>>>>(1) the FSN of the last received message before LPO or 
> >>>>>>(2) the FSN of the last message in the incoming messages that were queued and unacknowledged
> >>>>>>during the processor outage (with that the remote side can release the messages in its RTB) ? 
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>>thanks
> >>>>>>best regards, Franz 
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>>_______________________________________________
> >>>>>>Sigtran mailing list
> >>>>>>Sigtran@ietf.org
> >>>>>>https://www1.ietf.org/mailman/listinfo/sigtran
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>>     
> >>>>>>
> >>>>>
> >>>>>-- 
> >>>>>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.
> >>>>>
> >>>>>   
> >>>>>
> >>>>
> >>>> 
> >>>>
> >>>
> >>>-- 
> >>>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.
> >>>
> >>
> >>-- 
> >>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
> > 
> > 
> 
> 
> -- 
> * 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.  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 exim@www1.ietf.org  Thu Jul 31 08:04:34 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA21808
	for <sigtran-archive@odin.ietf.org>; Thu, 31 Jul 2003 08:04:34 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iCAB-0004fC-8u
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 08:04:07 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6VC47iW017925
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 08:04:07 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iCA4-0004es-K7; Thu, 31 Jul 2003 08:04:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iC9C-0004eU-0i
	for sigtran@optimus.ietf.org; Thu, 31 Jul 2003 08:03:06 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA21787
	for <sigtran@ietf.org>; Thu, 31 Jul 2003 08:03:02 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19iC9A-0002mk-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 08:03:04 -0400
Received: from web12806.mail.yahoo.com ([216.136.174.41])
	by ietf-mx with smtp (Exim 4.12)
	id 19iC9A-0002mb-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 08:03:04 -0400
Message-ID: <20030731120303.11329.qmail@web12806.mail.yahoo.com>
Received: from [208.255.120.2] by web12806.mail.yahoo.com via HTTP; Thu, 31 Jul 2003 05:03:03 PDT
Date: Thu, 31 Jul 2003 05:03:03 -0700 (PDT)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
To: sigtran@ietf.org
In-Reply-To: <06CAA4D3753C53408D2CE6340B68444214C478@vbeexc02.emea.cpqcorp.net>
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>

Unidirectional traffic is a general problem for
DE-IPSP model and I think will stay to be like this,
because it is not a real perr-to-peer model but a 2
client-server relationships coupled together.
IMO, using ASPAC-ACK to indicate availability of local
AS won't do good. What will happen for the scenarios,
where one IPSP serves multiple AS and when some of
them are available but some of them are not? Should
IPSP reply with ASPAC-ACK for a received ASPAC? SE
model was solving this problem by corraleting two AS
to each other on RK level. 
--- "Zannin, Francois" <francois.zannin@hp.com> wrote:
> If we consider that the IPSP DE is a replicated
> SG/ASP model (as already
> mentionned
> Before), then SG/ASP protocol indicate that the SG
> cannot send AND
> receive traffic 
> from an ASP if the concerned AS(RC) is not
> activated. 
> 
> When two IPSP are hosting several AS, it seem
> pertinent to authorize the
> traffic only 
> between "ACTIVE AS" in the IPSP.
> 
> An active status is reported to the application ONLY
> if the local AS
> (the local application)
> And the remote AS are activated. If the remote OR
> the local AS is
> deactivated, then
> A pause is reported to the application.
> 
> My opinion is that only ACTIVE AS (between IPSP) can
> use the association
> to exchange messages.
> 
> Francois Z.
> 
> > -----Original Message-----
> > From: kakaul@hss.hns.com
> [mailto:kakaul@hss.hns.com] 
> > Sent: Thursday, July 31, 2003 6:29 AM
> > To: Band, Richard
> > Cc: sigtran@ietf.org; umesh@ccpu.com
> > Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
> > 
> > 
> > 
> > 
> > 
> > 
> > Hi Richard,
> >   You are right that there is no point in
> activating only 
> > AS1 in IPSP DE mode in the given example. Unless
> both the AS 
> > 's AS1 and AS2 are not active user to user traffic
> 
> > (bidirectional) cannot be started in IPSP
> model.Any way IPSP 
> > DE mode for that reason is kept
> optional.Bidirectional 
> > traffic between IPSP A and IPSP B will only start
> when both 
> > the AS's are active. In my openion let IPSP B
> indicate 
> > availability of remote AS1as and when it sends
> ASPAC ack for 
> > AS1 to IPSP A.If IPSP B is interested in sending
> data for 
> > remote AS1 it can send and IPSP A will receive the
> data, if 
> > some one wants single direction traffic.But yes i
> do argee 
> > that there is no practical scenario in which
> single direction 
> > traffic is desired in IPSP model.
> > 
> > Kamesh Kaul
> > www.hssworld.com
> > Hughes Software Systems
> > 
> > 
> > 
> > 
> > "BAND,RICHARD (HP-France,ex1)"
> <richard.band@hp.com> on 
> > 07/30/2003 05:33:53 PM
> > 
> > To:    Kamesh Kaul/HSS@HSS, umesh@ccpu.com
> > cc:    sigtran@ietf.org
> > 
> > Subject:    RE: [Sigtran] M3UA : IPSP DE behaviour
> > 
> > 
> > Hi Kamesh, Umesh,
> > 
> > I agree that the protocol allows traffic from IPSP
> B to A 
> > without having traffic from IPSP A to B.
> > 
> > However, I do think that there is a problem with
> the 
> > notifications for the AS on each side: In this
> situation, how 
> > should the IPSPs A and B indicate the availability
> of the 
> > remote AS to the local AS? Should IPSP A indicate
> that AS 2 
> > is available, even though it knows AS 2 cannot
> reply? Should 
> > IPSP B indicate that AS 1 in unavailable, as AS 2
> cannot send 
> > any messages to AS 1? If so, what should IPSP B do
> when AS 1 
> > sends traffic? Discard the messages, or transfer
> the data to 
> > AS 2, which will then receive traffic from an
> unavailable destination?
> > 
> > In my opinion, both IPSPs should indicate that the
> remote AS 
> > is unavailable. I would say that the DE IPSP
> exchange mode 
> > allows each side to indicate its readiness to
> receive 
> > traffic, but traffic should not be started unless
> both sides 
> > are active. It seems to me that this is the only
> to avoid the 
> > problems described above.
> > 
> > Please let me know your opinion on this point.
> > 
> > Regards,
> > 
> > Richard Band
> > 
> > -----Original Message-----
> > From: kakaul@hss.hns.com
> [mailto:kakaul@hss.hns.com]
> > Sent: Wednesday, July 30, 2003 06:17
> > To: umesh@ccpu.com
> > Cc: sigtran@ietf.org
> > Subject: Re: [Sigtran] M3UA : IPSP DE behaviour
> > 
> > 
> > 
> > 
> > Umesh
> > 
> >   IPSP B can send DATA messages for AS1 to IPSP A
> only if on 
> > receiving ASPAC message from IPSP A , IPSP B sends
> ASPAC ack 
> > for AS1 to IPSP A.Not activating AS2 means IPSP B
> will not be 
> > able to receive any message for AS2.So in a way
> only data for 
> > AS1 will flow from IPSP B to IPSP A.
> > 
> > Regards
> > Kamesh Kaul
> > www.hssworld.com
> > Hughes Software Systems
> > 
> > 
> > 
> > 
> > "Umesh Vats" <umesh@ccpu.com>@ietf.org on
> 07/30/2003 07:44:36 AM
> > 
> > Please respond to <umesh@ccpu.com>
> > 
> > Sent by:    sigtran-admin@ietf.org
> > 
> > 
> > To:    <sigtran@ietf.org>
> > cc:
> > 
> > Subject:    [Sigtran] M3UA : IPSP DE behaviour
> > 
> > 
> > Suppose we have two Double Ended IPSPs A and B.
> AS1 is 
> > application server on IPSP-A and AS2 is
> application server on IPSP-B
> > 
> >      AS1 A----------B AS2
> > 
> > Now, IPSP-A sends ASPAC for for AS1 to B. IPSP-B
> doesn't 
> > activate AS2. After this, can IPSP-B send DATA
> messages for 
> > AS1 to IPSP-A in any scenario ?
> > 
> > Thanks
> > Umesh Vats
> > Sr Software Engg.
> > extn : 8986
> > 
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> >  https://www1.ietf.org/mailman/listinfo/sigtran
> > 
> > 
> > 
> > _______________________________________________
> > Sigtran mailing list
> 
=== message truncated ===


__________________________________
Do you Yahoo!?
Yahoo! SiteBuilder - Free, easy-to-use web site design software
http://sitebuilder.yahoo.com

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



From exim@www1.ietf.org  Thu Jul 31 08:13:34 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA22038
	for <sigtran-archive@odin.ietf.org>; Thu, 31 Jul 2003 08:13:34 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iCIt-0005BB-03
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 08:13:07 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6VCD6rT019904
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 08:13:06 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iCIn-00059Q-8Y; Thu, 31 Jul 2003 08:13:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iCIK-00058o-WA
	for sigtran@optimus.ietf.org; Thu, 31 Jul 2003 08:12:33 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA22005
	for <sigtran@ietf.org>; Thu, 31 Jul 2003 08:12:29 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19iCIJ-0002rf-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 08:12:31 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19iCIJ-0002rc-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 08:12:31 -0400
Received: from httpd by smtp.ccpu.com with local (Exim 4.20 #1 (Debian))
	id 19iCGt-0005TH-1L; Thu, 31 Jul 2003 05:11:03 -0700
From: "Vivek Toky" <vivektoky@ccpu.com>
To: "Zannin, Francois" <francois.zannin@hp.com>, <kakaul@hss.hns.com>,
        "Band, Richard" <richard.band@hp.com>, <sigtran@ietf.org>,
        <umesh@ccpu.com>
Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
X-Mailer: NeoMail 1.25
X-IPAddress: 172.16.2.11
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Message-Id: <E19iCGt-0005TH-1L@smtp.ccpu.com>
Date: Thu, 31 Jul 2003 05:11:03 -0700
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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 should be done in the following case:


      |IPSP_A                             |
  AS1 |      ---------------------- IPSP_C| AS2
      |IPSP_B                             |

IPSP_A and IPSP_B are in loadshare mode.

  when AS1 has activated itself by sending
ASP Active from both the IPSPs (A & B) but 
AS2 has sent ASP Active just to IPSP_A. 

What should IPSP_C do on receiving traffic for AS1?

a) Loadshare the traffic towards IPSP_A and IPSP_B

or 

b) It should send all the traffic to IPSP_A and 
   wait untill it sends ASP Active IPSP_B as well.
   In the above case IPSP_B cannot send the traffic 
   towards IPSP_C but IPSP_C can.

thanks,
Vivek.

> If we consider that the IPSP DE is a replicated SG/ASP model (as 
already
> mentionned
> Before), then SG/ASP protocol indicate that the SG cannot send AND
> receive traffic 
> from an ASP if the concerned AS(RC) is not activated. 
> 
> When two IPSP are hosting several AS, it seem pertinent to authorize 
the
> traffic only 
> between "ACTIVE AS" in the IPSP.
> 
> An active status is reported to the application ONLY if the local AS
> (the local application)
> And the remote AS are activated. If the remote OR the local AS is
> deactivated, then
> A pause is reported to the application.
> 
> My opinion is that only ACTIVE AS (between IPSP) can use the 
association
> to exchange messages.
> 
> Francois Z.
> 
> > -----Original Message-----
> > From: kakaul@hss.hns.com [mailto:kakaul@hss.hns.com] 
> > Sent: Thursday, July 31, 2003 6:29 AM
> > To: Band, Richard
> > Cc: sigtran@ietf.org; umesh@ccpu.com
> > Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
> > 
> > 
> > 
> > 
> > 
> > 
> > Hi Richard,
> >   You are right that there is no point in activating only 
> > AS1 in IPSP DE mode in the given example. Unless both the AS 
> > 's AS1 and AS2 are not active user to user traffic 
> > (bidirectional) cannot be started in IPSP model.Any way IPSP 
> > DE mode for that reason is kept optional.Bidirectional 
> > traffic between IPSP A and IPSP B will only start when both 
> > the AS's are active. In my openion let IPSP B indicate 
> > availability of remote AS1as and when it sends ASPAC ack for 
> > AS1 to IPSP A.If IPSP B is interested in sending data for 
> > remote AS1 it can send and IPSP A will receive the data, if 
> > some one wants single direction traffic.But yes i do argee 
> > that there is no practical scenario in which single direction 
> > traffic is desired in IPSP model.
> > 
> > Kamesh Kaul
> > www.hssworld.com
> > Hughes Software Systems
> > 
> > 
> > 
> > 
> > "BAND,RICHARD (HP-France,ex1)" <richard.band@hp.com> on 
> > 07/30/2003 05:33:53 PM
> > 
> > To:    Kamesh Kaul/HSS@HSS, umesh@ccpu.com
> > cc:    sigtran@ietf.org
> > 
> > Subject:    RE: [Sigtran] M3UA : IPSP DE behaviour
> > 
> > 
> > Hi Kamesh, Umesh,
> > 
> > I agree that the protocol allows traffic from IPSP B to A 
> > without having traffic from IPSP A to B.
> > 
> > However, I do think that there is a problem with the 
> > notifications for the AS on each side: In this situation, how 
> > should the IPSPs A and B indicate the availability of the 
> > remote AS to the local AS? Should IPSP A indicate that AS 2 
> > is available, even though it knows AS 2 cannot reply? Should 
> > IPSP B indicate that AS 1 in unavailable, as AS 2 cannot send 
> > any messages to AS 1? If so, what should IPSP B do when AS 1 
> > sends traffic? Discard the messages, or transfer the data to 
> > AS 2, which will then receive traffic from an unavailable 
destination?
> > 
> > In my opinion, both IPSPs should indicate that the remote AS 
> > is unavailable. I would say that the DE IPSP exchange mode 
> > allows each side to indicate its readiness to receive 
> > traffic, but traffic should not be started unless both sides 
> > are active. It seems to me that this is the only to avoid the 
> > problems described above.
> > 
> > Please let me know your opinion on this point.
> > 
> > Regards,
> > 
> > Richard Band
> > 
> > -----Original Message-----
> > From: kakaul@hss.hns.com [mailto:kakaul@hss.hns.com]
> > Sent: Wednesday, July 30, 2003 06:17
> > To: umesh@ccpu.com
> > Cc: sigtran@ietf.org
> > Subject: Re: [Sigtran] M3UA : IPSP DE behaviour
> > 
> > 
> > 
> > 
> > Umesh
> > 
> >   IPSP B can send DATA messages for AS1 to IPSP A only if on 
> > receiving ASPAC message from IPSP A , IPSP B sends ASPAC ack 
> > for AS1 to IPSP A.Not activating AS2 means IPSP B will not be 
> > able to receive any message for AS2.So in a way only data for 
> > AS1 will flow from IPSP B to IPSP A.
> > 
> > Regards
> > Kamesh Kaul
> > www.hssworld.com
> > Hughes Software Systems
> > 
> > 
> > 
> > 
> > "Umesh Vats" <umesh@ccpu.com>@ietf.org on 07/30/2003 07:44:36 AM
> > 
> > Please respond to <umesh@ccpu.com>
> > 
> > Sent by:    sigtran-admin@ietf.org
> > 
> > 
> > To:    <sigtran@ietf.org>
> > cc:
> > 
> > Subject:    [Sigtran] M3UA : IPSP DE behaviour
> > 
> > 
> > Suppose we have two Double Ended IPSPs A and B. AS1 is 
> > application server on IPSP-A and AS2 is application server on IPSP-
B
> > 
> >      AS1 A----------B AS2
> > 
> > Now, IPSP-A sends ASPAC for for AS1 to B. IPSP-B doesn't 
> > activate AS2. After this, can IPSP-B send DATA messages for 
> > AS1 to IPSP-A in any scenario ?
> > 
> > Thanks
> > Umesh Vats
> > Sr Software Engg.
> > extn : 8986
> > 
> > 
> > _______________________________________________
> > 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
> 
> 

-- 
Continuous Computing
http://www.ccpu.com

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



From exim@www1.ietf.org  Thu Jul 31 08:22:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA22256
	for <sigtran-archive@odin.ietf.org>; Thu, 31 Jul 2003 08:22:33 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iCRa-0005Tj-Hh
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 08:22:06 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6VCM64C021043
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 08:22:06 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iCRU-0005TB-7U; Thu, 31 Jul 2003 08:22:00 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iCQo-0005SV-Un
	for sigtran@optimus.ietf.org; Thu, 31 Jul 2003 08:21:18 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA22230
	for <sigtran@ietf.org>; Thu, 31 Jul 2003 08:21:15 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19iCQn-0002xM-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 08:21:17 -0400
Received: from web12801.mail.yahoo.com ([216.136.174.36])
	by ietf-mx with smtp (Exim 4.12)
	id 19iCQm-0002xJ-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 08:21:16 -0400
Message-ID: <20030731122116.20884.qmail@web12801.mail.yahoo.com>
Received: from [208.255.120.2] by web12801.mail.yahoo.com via HTTP; Thu, 31 Jul 2003 05:21:16 PDT
Date: Thu, 31 Jul 2003 05:21:16 -0700 (PDT)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
To: sigtran@ietf.org
In-Reply-To: <E19iCGt-0005TH-1L@smtp.ccpu.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>

Vivek,
--- Vivek Toky <vivektoky@ccpu.com> wrote:
> Hi All,
>  What should be done in the following case:
> 
> 
>       |IPSP_A                             |
>   AS1 |      ---------------------- IPSP_C| AS2
>       |IPSP_B                             |
> 
> IPSP_A and IPSP_B are in loadshare mode.
> 
>   when AS1 has activated itself by sending
> ASP Active from both the IPSPs (A & B) but 
> AS2 has sent ASP Active just to IPSP_A. 
> 
> What should IPSP_C do on receiving traffic for AS1?
> 
> a) Loadshare the traffic towards IPSP_A and IPSP_B
> 
> or 
> 
> b) It should send all the traffic to IPSP_A and 
>    wait untill it sends ASP Active IPSP_B as well.
>    In the above case IPSP_B cannot send the traffic 
>    towards IPSP_C but IPSP_C can.
[TOLGA]It should send all traffic to IPSPS_A. Which
case do you refer as "the above case", a) or b) ?
> 
> thanks,
> Vivek.
> 


__________________________________
Do you Yahoo!?
Yahoo! SiteBuilder - Free, easy-to-use web site design software
http://sitebuilder.yahoo.com

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



From exim@www1.ietf.org  Thu Jul 31 08:43:34 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA23035
	for <sigtran-archive@odin.ietf.org>; Thu, 31 Jul 2003 08:43:34 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iClv-0006fp-Hc
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 08:43:08 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6VCh7aE025635
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 08:43:07 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iClp-0006a3-LE; Thu, 31 Jul 2003 08:43:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iClL-0006Yu-Mf
	for sigtran@optimus.ietf.org; Thu, 31 Jul 2003 08:42:31 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA22989
	for <sigtran@ietf.org>; Thu, 31 Jul 2003 08:42:27 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19iClK-0003BM-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 08:42:30 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19iClJ-0003BG-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 08:42:29 -0400
Received: from httpd by smtp.ccpu.com with local (Exim 4.20 #1 (Debian))
	id 19iClD-0006gS-04; Thu, 31 Jul 2003 05:42:23 -0700
From: "Vivek Toky" <vivektoky@ccpu.com>
To: Tolga Asveren <tolga_asveren@yahoo.com>, sigtran@ietf.org
Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
X-Mailer: NeoMail 1.25
X-IPAddress: 172.16.2.11
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Message-Id: <E19iClD-0006gS-04@smtp.ccpu.com>
Date: Thu, 31 Jul 2003 05:42:23 -0700
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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 Tolga,
  "Above case" means 
"when AS1 has activated itself by sending
 ASP Active from both the IPSPs (A & B) but 
 AS2 has sent ASP Active just to IPSP_A."

In my opinion IPSP_C should loadshare the traffic
as it has received ASP active from both the IPSPs
so AS1 is accessible through both of these.
However, it is the responsibility of AS1 to route
all the traffic towards AS2 through IPSP_A.

In either case I think it is worth mentioning in 
Implementor's guide whether Unidirectional traffic
is possible between associations(in case of 
IPSP DE) or not.It will avoid any further
confusion or possible interoperatibility issues.

thanks,

-Vivek.

> Vivek,
> --- Vivek Toky <vivektoky@ccpu.com> wrote:
> > Hi All,
> >  What should be done in the following case:
> > 
> > 
> >       |IPSP_A                             |
> >   AS1 |      ---------------------- IPSP_C| AS2
> >       |IPSP_B                             |
> > 
> > IPSP_A and IPSP_B are in loadshare mode.
> > 
> >   when AS1 has activated itself by sending
> > ASP Active from both the IPSPs (A & B) but 
> > AS2 has sent ASP Active just to IPSP_A. 
> > 
> > What should IPSP_C do on receiving traffic for AS1?
> > 
> > a) Loadshare the traffic towards IPSP_A and IPSP_B
> > 
> > or 
> > 
> > b) It should send all the traffic to IPSP_A and 
> >    wait untill it sends ASP Active IPSP_B as well.
> >    In the above case IPSP_B cannot send the traffic 
> >    towards IPSP_C but IPSP_C can.
> [TOLGA]It should send all traffic to IPSPS_A. Which
> case do you refer as "the above case", a) or b) ?
> > 
> > thanks,
> > Vivek.
> > 
> 
> 
> __________________________________
> Do you Yahoo!?
> Yahoo! SiteBuilder - Free, easy-to-use web site design software
> http://sitebuilder.yahoo.com
> 
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> 
> 

-- 
Continuous Computing
http://www.ccpu.com

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



From exim@www1.ietf.org  Thu Jul 31 12:27:05 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA03292
	for <sigtran-archive@odin.ietf.org>; Thu, 31 Jul 2003 12:27:05 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iGGF-0001GE-PC
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 12:26:39 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6VGQdrK004842
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 12:26:39 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iGFd-0001BL-TP; Thu, 31 Jul 2003 12:26:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iGEs-0001AJ-QU
	for sigtran@optimus.ietf.org; Thu, 31 Jul 2003 12:25:14 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA03260
	for <sigtran@ietf.org>; Thu, 31 Jul 2003 12:25:09 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19iGEr-0005VA-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 12:25:13 -0400
Received: from zdemail04.zdem.compaq.com ([161.114.112.28])
	by ietf-mx with esmtp (Exim 4.12)
	id 19iGEp-0005Uy-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 12:25:12 -0400
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail04.zdem.compaq.com (Postfix) with ESMTP
	id 81FA7C7F; Thu, 31 Jul 2003 18:24:34 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Thu, 31 Jul 2003 18:24:34 +0200
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
X-MimeOLE: Produced By Microsoft Exchange V6.0.6375.0
Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
Date: Thu, 31 Jul 2003 18:24:33 +0200
Message-ID: <06CAA4D3753C53408D2CE6340B68444214C479@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA : IPSP DE behaviour
Thread-Index: AcNXXASR0pg83jq9RZO3hsRKVkrn+QAIuapg
From: "Zannin, Francois" <francois.zannin@hp.com>
To: "Tolga Asveren" <tolga_asveren@yahoo.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 31 Jul 2003 16:24:34.0438 (UTC) FILETIME=[3A706260:01C35780]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

My response inline.

Francois Z.

> -----Original Message-----
> From: Tolga Asveren [mailto:tolga_asveren@yahoo.com]=20
> Sent: Thursday, July 31, 2003 2:03 PM
> To: sigtran@ietf.org
> Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
>=20
>=20
> Unidirectional traffic is a general problem for
> DE-IPSP model and I think will stay to be like this,
> because it is not a real perr-to-peer model but a 2=20
> client-server relationships coupled together. IMO, using=20
> ASPAC-ACK to indicate availability of local AS won't do good.=20

In DE, ASPAC-ACK only aknowledge the activation of a remote AS, means
LOCAL IPSP can send=20
and receive data between for this remote AS and it's own Local AS that
are ACTIVE.
If no local AS is active, no local AS will be able to send or receive
traffic from a remote AS.

> What will happen for the scenarios, where one IPSP serves=20
> multiple AS and when some of them are available but some of=20
> them are not? Should IPSP reply with ASPAC-ACK for a received=20
> ASPAC? SE model was solving this problem by corraleting two=20
> AS to each other on RK level.=20

We are in DOUBLE echange, an ACTIVE/ACTIE_ACK is only valid for an AS
located on the IPSP that=20
Initiates this ASPTM exchange.

If An IPSP serve multiple AS, it just sends ASP_ACTIVE for its local AS
ready to operate.
Then a remote IPSP will be allowed to send traffic only for AS
activated.

> --- "Zannin, Francois" <francois.zannin@hp.com> wrote:
> > If we consider that the IPSP DE is a replicated
> > SG/ASP model (as already
> > mentionned
> > Before), then SG/ASP protocol indicate that the SG cannot send AND
> > receive traffic from an ASP if the concerned AS(RC) is not
activated.=20
> >=20
> > When two IPSP are hosting several AS, it seem pertinent to authorize
the traffic only
> > between "ACTIVE AS" in the IPSP.
> >=20
> > An active status is reported to the application ONLY if the local AS
> > (the local application) And the remote AS are activated. If the
remote OR
> > the local AS is deactivated, then A pause is reported to the
application.
> >=20
> > My opinion is that only ACTIVE AS (between IPSP) can use the
association
> > to exchange messages.
> >=20
> > Francois Z.
> >=20
> > > -----Original Message-----
> > > From: kakaul@hss.hns.com
> > [mailto:kakaul@hss.hns.com]
> > > Sent: Thursday, July 31, 2003 6:29 AM
> > > To: Band, Richard
> > > Cc: sigtran@ietf.org; umesh@ccpu.com
> > > Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
> > >=20
> > >=20
> > >=20
> > >=20
> > >=20
> > >=20
> > > Hi Richard,
> > >   You are right that there is no point in
> > activating only
> > > AS1 in IPSP DE mode in the given example. Unless
> > both the AS
> > > 's AS1 and AS2 are not active user to user traffic
> >=20
> > > (bidirectional) cannot be started in IPSP
> > model.Any way IPSP
> > > DE mode for that reason is kept
> > optional.Bidirectional
> > > traffic between IPSP A and IPSP B will only start
> > when both
> > > the AS's are active. In my openion let IPSP B
> > indicate
> > > availability of remote AS1as and when it sends
> > ASPAC ack for
> > > AS1 to IPSP A.If IPSP B is interested in sending
> > data for
> > > remote AS1 it can send and IPSP A will receive the
> > data, if
> > > some one wants single direction traffic.But yes i
> > do argee
> > > that there is no practical scenario in which
> > single direction
> > > traffic is desired in IPSP model.
> > >=20
> > > Kamesh Kaul
> > > www.hssworld.com
> > > Hughes Software Systems
> > >=20
> > >=20
> > >=20
> > >=20
> > > "BAND,RICHARD (HP-France,ex1)"
> > <richard.band@hp.com> on
> > > 07/30/2003 05:33:53 PM
> > >=20
> > > To:    Kamesh Kaul/HSS@HSS, umesh@ccpu.com
> > > cc:    sigtran@ietf.org
> > >=20
> > > Subject:    RE: [Sigtran] M3UA : IPSP DE behaviour
> > >=20
> > >=20
> > > Hi Kamesh, Umesh,
> > >=20
> > > I agree that the protocol allows traffic from IPSP
> > B to A
> > > without having traffic from IPSP A to B.
> > >=20
> > > However, I do think that there is a problem with
> > the
> > > notifications for the AS on each side: In this
> > situation, how
> > > should the IPSPs A and B indicate the availability
> > of the
> > > remote AS to the local AS? Should IPSP A indicate
> > that AS 2
> > > is available, even though it knows AS 2 cannot
> > reply? Should
> > > IPSP B indicate that AS 1 in unavailable, as AS 2
> > cannot send
> > > any messages to AS 1? If so, what should IPSP B do
> > when AS 1
> > > sends traffic? Discard the messages, or transfer
> > the data to
> > > AS 2, which will then receive traffic from an
> > unavailable destination?
> > >=20
> > > In my opinion, both IPSPs should indicate that the
> > remote AS
> > > is unavailable. I would say that the DE IPSP
> > exchange mode
> > > allows each side to indicate its readiness to
> > receive
> > > traffic, but traffic should not be started unless
> > both sides
> > > are active. It seems to me that this is the only
> > to avoid the
> > > problems described above.
> > >=20
> > > Please let me know your opinion on this point.
> > >=20
> > > Regards,
> > >=20
> > > Richard Band
> > >=20
> > > -----Original Message-----
> > > From: kakaul@hss.hns.com
> > [mailto:kakaul@hss.hns.com]
> > > Sent: Wednesday, July 30, 2003 06:17
> > > To: umesh@ccpu.com
> > > Cc: sigtran@ietf.org
> > > Subject: Re: [Sigtran] M3UA : IPSP DE behaviour
> > >=20
> > >=20
> > >=20
> > >=20
> > > Umesh
> > >=20
> > >   IPSP B can send DATA messages for AS1 to IPSP A
> > only if on
> > > receiving ASPAC message from IPSP A , IPSP B sends
> > ASPAC ack
> > > for AS1 to IPSP A.Not activating AS2 means IPSP B
> > will not be
> > > able to receive any message for AS2.So in a way
> > only data for
> > > AS1 will flow from IPSP B to IPSP A.
> > >=20
> > > Regards
> > > Kamesh Kaul
> > > www.hssworld.com
> > > Hughes Software Systems
> > >=20
> > >=20
> > >=20
> > >=20
> > > "Umesh Vats" <umesh@ccpu.com>@ietf.org on
> > 07/30/2003 07:44:36 AM
> > >=20
> > > Please respond to <umesh@ccpu.com>
> > >=20
> > > Sent by:    sigtran-admin@ietf.org
> > >=20
> > >=20
> > > To:    <sigtran@ietf.org>
> > > cc:
> > >=20
> > > Subject:    [Sigtran] M3UA : IPSP DE behaviour
> > >=20
> > >=20
> > > Suppose we have two Double Ended IPSPs A and B.
> > AS1 is
> > > application server on IPSP-A and AS2 is
> > application server on IPSP-B
> > >=20
> > >      AS1 A----------B AS2
> > >=20
> > > Now, IPSP-A sends ASPAC for for AS1 to B. IPSP-B
> > doesn't
> > > activate AS2. After this, can IPSP-B send DATA
> > messages for
> > > AS1 to IPSP-A in any scenario ?
> > >=20
> > > Thanks
> > > Umesh Vats
> > > Sr Software Engg.
> > > extn : 8986
> > >=20
> > >=20
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org  https://www1.ietf.org/mailman/listinfo/sigtran
> > >=20
> > >=20
> > >=20
> > > _______________________________________________
> > > Sigtran mailing list
> >=20
> =3D=3D=3D message truncated =3D=3D=3D
>=20
>=20
> __________________________________
> Do you Yahoo!?
> Yahoo! SiteBuilder - Free, easy-to-use web site design=20
> software http://sitebuilder.yahoo.com
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>=20

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



From exim@www1.ietf.org  Thu Jul 31 12:33:06 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA03486
	for <sigtran-archive@odin.ietf.org>; Thu, 31 Jul 2003 12:33:05 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iGM1-0001cI-ED
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 12:32:37 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6VGWbrb006214
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 12:32:37 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iGLR-0001Xt-Lb; Thu, 31 Jul 2003 12:32:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iGL1-0001Xd-5u
	for sigtran@optimus.ietf.org; Thu, 31 Jul 2003 12:31:35 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA03459
	for <sigtran@ietf.org>; Thu, 31 Jul 2003 12:31:30 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19iGKz-0005XX-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 12:31:33 -0400
Received: from zdemail03.zdem.compaq.com ([161.114.112.27])
	by ietf-mx with esmtp (Exim 4.12)
	id 19iGKy-0005XT-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 12:31:33 -0400
Received: from demexg11.emea.cpqcorp.net (demexg11.emea.cpqcorp.net [16.41.86.63])
	by zdemail03.zdem.compaq.com (Postfix) with ESMTP
	id EC7FA10B7; Thu, 31 Jul 2003 18:31:01 +0200 (CEST)
Received: from vbeexc02.emea.cpqcorp.net ([16.188.146.229]) by demexg11.emea.cpqcorp.net with Microsoft SMTPSVC(5.0.2195.6673);
	 Thu, 31 Jul 2003 18:31:01 +0200
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
X-MimeOLE: Produced By Microsoft Exchange V6.0.6375.0
Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
Date: Thu, 31 Jul 2003 18:30:55 +0200
Message-ID: <06CAA4D3753C53408D2CE6340B68444214C47A@vbeexc02.emea.cpqcorp.net>
Thread-Topic: [Sigtran] M3UA : IPSP DE behaviour
Thread-Index: AcNXYW4CFXYsgaEeRtqDRDGtP2QQbwAHumnA
From: "Zannin, Francois" <francois.zannin@hp.com>
To: "Vivek Toky" <vivektoky@ccpu.com>,
        "Tolga Asveren" <tolga_asveren@yahoo.com>, <sigtran@ietf.org>
X-OriginalArrivalTime: 31 Jul 2003 16:31:01.0847 (UTC) FILETIME=[215A4E70:01C35781]
Content-Transfer-Encoding: quoted-printable
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-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: quoted-printable

See my opigion online

> -----Original Message-----
> From: Vivek Toky [mailto:vivektoky@ccpu.com]=20
> Sent: Thursday, July 31, 2003 2:42 PM
> To: Tolga Asveren; sigtran@ietf.org
> Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
>=20
>=20
> Hi Tolga,
>   "Above case" means=20
> "when AS1 has activated itself by sending
>  ASP Active from both the IPSPs (A & B) but=20
>  AS2 has sent ASP Active just to IPSP_A."
>=20
> In my opinion IPSP_C should loadshare the traffic
> as it has received ASP active from both the IPSPs
> so AS1 is accessible through both of these.
> However, it is the responsibility of AS1 to route
> all the traffic towards AS2 through IPSP_A.

If IPSP_B receive some traffic from IPSP_C BUT no AS is ACTIVE for it on
IPSP_C can it route this traffic???
CAN a SG send to a ss7 network a message receive from an INACTIVE ASP !
I don't think so!
IPSP DE must be considered as two ASP/SG head to tail.

>=20
> In either case I think it is worth mentioning in=20
> Implementor's guide whether Unidirectional traffic
> is possible between associations(in case of=20
> IPSP DE) or not.It will avoid any further
> confusion or possible interoperatibility issues.
>=20
> thanks,
>=20
> -Vivek.
>=20
> > Vivek,
> > --- Vivek Toky <vivektoky@ccpu.com> wrote:
> > > Hi All,
> > >  What should be done in the following case:
> > >=20
> > >=20
> > >       |IPSP_A                             |
> > >   AS1 |      ---------------------- IPSP_C| AS2
> > >       |IPSP_B                             |
> > >=20
> > > IPSP_A and IPSP_B are in loadshare mode.
> > >=20
> > >   when AS1 has activated itself by sending
> > > ASP Active from both the IPSPs (A & B) but
> > > AS2 has sent ASP Active just to IPSP_A.=20
> > >=20
> > > What should IPSP_C do on receiving traffic for AS1?
> > >=20
> > > a) Loadshare the traffic towards IPSP_A and IPSP_B
> > >=20
> > > or
> > >=20
> > > b) It should send all the traffic to IPSP_A and=20
> > >    wait untill it sends ASP Active IPSP_B as well.
> > >    In the above case IPSP_B cannot send the traffic=20
> > >    towards IPSP_C but IPSP_C can.
> > [TOLGA]It should send all traffic to IPSPS_A. Which
> > case do you refer as "the above case", a) or b) ?
> > >=20
> > > thanks,
> > > Vivek.
> > >=20
> >=20
> >=20
> > __________________________________
> > Do you Yahoo!?
> > Yahoo! SiteBuilder - Free, easy-to-use web site design software=20
> > http://sitebuilder.yahoo.com
> >=20
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> >=20
> >=20
>=20
> --=20
> Continuous Computing
> http://www.ccpu.com
>=20
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
>=20

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



From exim@www1.ietf.org  Thu Jul 31 12:55:01 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA04144
	for <sigtran-archive@odin.ietf.org>; Thu, 31 Jul 2003 12:55:01 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iGhI-0002f2-MZ
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 12:54:36 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6VGsaZ7010227
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 12:54:36 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iGgj-0002aH-1R; Thu, 31 Jul 2003 12:54:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iGgD-0002Yl-1E
	for sigtran@optimus.ietf.org; Thu, 31 Jul 2003 12:53:29 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA04050
	for <sigtran@ietf.org>; Thu, 31 Jul 2003 12:53:23 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19iGgB-0005ff-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 12:53:27 -0400
Received: from web12812.mail.yahoo.com ([216.136.174.127])
	by ietf-mx with smtp (Exim 4.12)
	id 19iGgA-0005fc-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 12:53:26 -0400
Message-ID: <20030731165324.6413.qmail@web12812.mail.yahoo.com>
Received: from [208.255.120.2] by web12812.mail.yahoo.com via HTTP; Thu, 31 Jul 2003 09:53:24 PDT
Date: Thu, 31 Jul 2003 09:53:24 -0700 (PDT)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
To: sigtran@ietf.org
In-Reply-To: <06CAA4D3753C53408D2CE6340B68444214C479@vbeexc02.emea.cpqcorp.net>
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>

Francois,
--- "Zannin, Francois" <francois.zannin@hp.com> wrote:
> My response inline.
> 
> Francois Z.
> 
> > -----Original Message-----
> > From: Tolga Asveren
> [mailto:tolga_asveren@yahoo.com] 
> > Sent: Thursday, July 31, 2003 2:03 PM
> > To: sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
> > 
> > 
> > Unidirectional traffic is a general problem for
> > DE-IPSP model and I think will stay to be like
> this,
> > because it is not a real perr-to-peer model but a
> 2 
> > client-server relationships coupled together. IMO,
> using 
> > ASPAC-ACK to indicate availability of local AS
> won't do good. 
> 
> In DE, ASPAC-ACK only aknowledge the activation of a
> remote AS, means
> LOCAL IPSP can send 
> and receive data between for this remote AS and it's
> own Local AS that
> are ACTIVE.
> If no local AS is active, no local AS will be able
> to send or receive
> traffic from a remote AS.
[TOLGA]What I said was to show that trying to send
ASPAC-ACK only when local AS is ACTIVE to prevent
unidirectional flow for DE-IPSP (to my understanding
something which is proposed in this thread) is not
usefull.
> 
> > What will happen for the scenarios, where one IPSP
> serves 
> > multiple AS and when some of them are available
> but some of 
> > them are not? Should IPSP reply with ASPAC-ACK for
> a received 
> > ASPAC? SE model was solving this problem by
> corraleting two 
> > AS to each other on RK level. 
> 
> We are in DOUBLE echange, an ACTIVE/ACTIE_ACK is
> only valid for an AS
> located on the IPSP that 
> Initiates this ASPTM exchange.
[TOLGA]Yes, this is the point. DE-IPSP has the
potential problem of unidirectional traffic.
> 
> If An IPSP serve multiple AS, it just sends
> ASP_ACTIVE for its local AS
> ready to operate.
> Then a remote IPSP will be allowed to send traffic
> only for AS
> activated.
> 
> > --- "Zannin, Francois" <francois.zannin@hp.com>
> wrote:
> > > If we consider that the IPSP DE is a replicated
> > > SG/ASP model (as already
> > > mentionned
> > > Before), then SG/ASP protocol indicate that the
> SG cannot send AND
> > > receive traffic from an ASP if the concerned
> AS(RC) is not
> activated. 
> > > 
> > > When two IPSP are hosting several AS, it seem
> pertinent to authorize
> the traffic only
> > > between "ACTIVE AS" in the IPSP.
> > > 
> > > An active status is reported to the application
> ONLY if the local AS
> > > (the local application) And the remote AS are
> activated. If the
> remote OR
> > > the local AS is deactivated, then A pause is
> reported to the
> application.
> > > 
> > > My opinion is that only ACTIVE AS (between IPSP)
> can use the
> association
> > > to exchange messages.
> > > 
> > > Francois Z.
> > > 
> > > > -----Original Message-----
> > > > From: kakaul@hss.hns.com
> > > [mailto:kakaul@hss.hns.com]
> > > > Sent: Thursday, July 31, 2003 6:29 AM
> > > > To: Band, Richard
> > > > Cc: sigtran@ietf.org; umesh@ccpu.com
> > > > Subject: RE: [Sigtran] M3UA : IPSP DE
> behaviour
> > > > 
> > > > 
> > > > 
> > > > 
> > > > 
> > > > 
> > > > Hi Richard,
> > > >   You are right that there is no point in
> > > activating only
> > > > AS1 in IPSP DE mode in the given example.
> Unless
> > > both the AS
> > > > 's AS1 and AS2 are not active user to user
> traffic
> > > 
> > > > (bidirectional) cannot be started in IPSP
> > > model.Any way IPSP
> > > > DE mode for that reason is kept
> > > optional.Bidirectional
> > > > traffic between IPSP A and IPSP B will only
> start
> > > when both
> > > > the AS's are active. In my openion let IPSP B
> > > indicate
> > > > availability of remote AS1as and when it sends
> > > ASPAC ack for
> > > > AS1 to IPSP A.If IPSP B is interested in
> sending
> > > data for
> > > > remote AS1 it can send and IPSP A will receive
> the
> > > data, if
> > > > some one wants single direction traffic.But
> yes i
> > > do argee
> > > > that there is no practical scenario in which
> > > single direction
> > > > traffic is desired in IPSP model.
> > > > 
> > > > Kamesh Kaul
> > > > www.hssworld.com
> > > > Hughes Software Systems
> > > > 
> > > > 
> > > > 
> > > > 
> > > > "BAND,RICHARD (HP-France,ex1)"
> > > <richard.band@hp.com> on
> > > > 07/30/2003 05:33:53 PM
> > > > 
> > > > To:    Kamesh Kaul/HSS@HSS, umesh@ccpu.com
> > > > cc:    sigtran@ietf.org
> > > > 
> > > > Subject:    RE: [Sigtran] M3UA : IPSP DE
> behaviour
> > > > 
> > > > 
> > > > Hi Kamesh, Umesh,
> > > > 
> > > > I agree that the protocol allows traffic from
> IPSP
> > > B to A
> > > > without having traffic from IPSP A to B.
> > > > 
> > > > However, I do think that there is a problem
> with
> > > the
> > > > notifications for the AS on each side: In this
> > > situation, how
> > > > should the IPSPs A and B indicate the
> availability
> > > of the
> > > > remote AS to the local AS? Should IPSP A
> indicate
> > > that AS 2
> > > > is available, even though it knows AS 2 cannot
> > > reply? Should
> > > > IPSP B indicate that AS 1 in unavailable, as
> AS 2
> > > cannot send
> > > > any messages to AS 1? If so, what should IPSP
> B do
> > > when AS 1
> > > > sends traffic? Discard the messages, or
> transfer
> > > the data to
> > > > AS 2, which will then receive traffic from an
> > > unavailable destination?
> > > > 
> > > > In my opinion, both IPSPs should indicate that
> the
> > > remote AS
> > > > is unavailable. I would say that the DE IPSP
> > > exchange mode
> > > > allows each side to indicate its readiness to
> > > receive
> > > > traffic, but traffic should not be started
> unless
> > > both sides
> > > > are active. It seems to me that this is the
> only
> > > to avoid the
> > > > problems described above.
> > > > 
> > > > Please let me know your opinion on this point.
> > > > 
> > > > Regards,
> 
=== message truncated ===


__________________________________
Do you Yahoo!?
Yahoo! SiteBuilder - Free, easy-to-use web site design software
http://sitebuilder.yahoo.com

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



From exim@www1.ietf.org  Thu Jul 31 13:06:05 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA04458
	for <sigtran-archive@odin.ietf.org>; Thu, 31 Jul 2003 13:06:05 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iGrz-00035T-Ng
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 13:05:40 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6VH5dFm011863
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 13:05:39 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iGrN-000318-JV; Thu, 31 Jul 2003 13:05:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iGr6-00030j-5n
	for sigtran@optimus.ietf.org; Thu, 31 Jul 2003 13:04:44 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA04439
	for <sigtran@ietf.org>; Thu, 31 Jul 2003 13:04:38 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19iGr4-0005jO-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 13:04:42 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19iGr3-0005jL-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 13:04:41 -0400
Received: from httpd by smtp.ccpu.com with local (Exim 4.20 #1 (Debian))
	id 19iGqu-0001jg-Nx; Thu, 31 Jul 2003 10:04:32 -0700
From: "Vivek Toky" <vivektoky@ccpu.com>
To: "Zannin, Francois" <francois.zannin@hp.com>,
        "Vivek Toky" <vivektoky@ccpu.com>,
        "Tolga Asveren" <tolga_asveren@yahoo.com>, <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
X-Mailer: NeoMail 1.25
X-IPAddress: 66.75.130.161
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Message-Id: <E19iGqu-0001jg-Nx@smtp.ccpu.com>
Date: Thu, 31 Jul 2003 10:04:32 -0700
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Please see my comments inline

> See my opigion online
> 
> > -----Original Message-----
> > From: Vivek Toky [mailto:vivektoky@ccpu.com] 
> > Sent: Thursday, July 31, 2003 2:42 PM
> > To: Tolga Asveren; sigtran@ietf.org
> > Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
> > 
> > 
> > Hi Tolga,
> >   "Above case" means 
> > "when AS1 has activated itself by sending
> >  ASP Active from both the IPSPs (A & B) but 
> >  AS2 has sent ASP Active just to IPSP_A."
> > 
> > In my opinion IPSP_C should loadshare the traffic
> > as it has received ASP active from both the IPSPs
> > so AS1 is accessible through both of these.
> > However, it is the responsibility of AS1 to route
> > all the traffic towards AS2 through IPSP_A.
> 
> If IPSP_B receive some traffic from IPSP_C BUT no AS is ACTIVE for 
it on
> IPSP_C can it route this traffic???
> CAN a SG send to a ss7 network a message receive from an INACTIVE 
ASP !
> I don't think so!
> IPSP DE must be considered as two ASP/SG head to tail.
> 
[Vivek]
AS is ACTIVE on IPSP_B (remember it sent ASP Active message)
As far as the SGP - ASP example see the following..

1. It is the AS which is Active or Inactive not ASP
2. ASPTM ptocedures are between SGPs and ASPs not SG and ASPs

The local AS on the ASP may be active through some other
SGP (say SGPy)to which an ACTIVE message has been sent.
So SGP (say SGPx) should route the traffic to SS7 coming from an 
ASP which is inactive for an AS, because that AS is 
Active through some other SGP and SS7 destination may still 
be able to send the traffic towards that AS through some 
other SGP

In M3UA outgoing and Incoming Traffic flows are independent,
so why restrict the traffic ?

> > 
> > In either case I think it is worth mentioning in 
> > Implementor's guide whether Unidirectional traffic
> > is possible between associations(in case of 
> > IPSP DE) or not.It will avoid any further
> > confusion or possible interoperatibility issues.
> > 
> > thanks,
> > 
> > -Vivek.
> > 
> > > Vivek,
> > > --- Vivek Toky <vivektoky@ccpu.com> wrote:
> > > > Hi All,
> > > >  What should be done in the following case:
> > > > 
> > > > 
> > > >       |IPSP_A                             |
> > > >   AS1 |      ---------------------- IPSP_C| AS2
> > > >       |IPSP_B                             |
> > > > 
> > > > IPSP_A and IPSP_B are in loadshare mode.
> > > > 
> > > >   when AS1 has activated itself by sending
> > > > ASP Active from both the IPSPs (A & B) but
> > > > AS2 has sent ASP Active just to IPSP_A. 
> > > > 
> > > > What should IPSP_C do on receiving traffic for AS1?
> > > > 
> > > > a) Loadshare the traffic towards IPSP_A and IPSP_B
> > > > 
> > > > or
> > > > 
> > > > b) It should send all the traffic to IPSP_A and 
> > > >    wait untill it sends ASP Active IPSP_B as well.
> > > >    In the above case IPSP_B cannot send the traffic 
> > > >    towards IPSP_C but IPSP_C can.
> > > [TOLGA]It should send all traffic to IPSPS_A. Which
> > > case do you refer as "the above case", a) or b) ?
> > > > 
> > > > thanks,
> > > > Vivek.
> > > > 
> > > 
> > > 
> > > __________________________________
> > > Do you Yahoo!?
> > > Yahoo! SiteBuilder - Free, easy-to-use web site design software 
> > > http://sitebuilder.yahoo.com
> > > 
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> > > 
> > 
> > -- 
> > Continuous Computing
> > http://www.ccpu.com
> > 
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> > 
> 
> 

-- 
Continuous Computing
http://www.ccpu.com

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



From exim@www1.ietf.org  Thu Jul 31 13:21:03 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA04847
	for <sigtran-archive@odin.ietf.org>; Thu, 31 Jul 2003 13:21:03 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iH6U-0003vN-E1
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 13:20:38 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6VHKc9F015040
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 13:20:38 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iH5t-0003ph-T1; Thu, 31 Jul 2003 13:20:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iH5Q-0003ot-Ta
	for sigtran@optimus.ietf.org; Thu, 31 Jul 2003 13:19:32 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA04823
	for <sigtran@ietf.org>; Thu, 31 Jul 2003 13:19:27 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19iH5P-0005nn-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 13:19:31 -0400
Received: from web12803.mail.yahoo.com ([216.136.174.38])
	by ietf-mx with smtp (Exim 4.12)
	id 19iH5O-0005nk-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 13:19:30 -0400
Message-ID: <20030731171929.82038.qmail@web12803.mail.yahoo.com>
Received: from [208.255.120.2] by web12803.mail.yahoo.com via HTTP; Thu, 31 Jul 2003 10:19:29 PDT
Date: Thu, 31 Jul 2003 10:19:29 -0700 (PDT)
From: Tolga Asveren <tolga_asveren@yahoo.com>
Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
To: sigtran@ietf.org
In-Reply-To: <E19iGqu-0001jg-Nx@smtp.ccpu.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>

Vivek,
--- Vivek Toky <vivektoky@ccpu.com> wrote:
> Please see my comments inline
> 
> > See my opigion online
> > 
> > > -----Original Message-----
> > > From: Vivek Toky [mailto:vivektoky@ccpu.com] 
> > > Sent: Thursday, July 31, 2003 2:42 PM
> > > To: Tolga Asveren; sigtran@ietf.org
> > > Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
> > > 
> > > 
> > > Hi Tolga,
> > >   "Above case" means 
> > > "when AS1 has activated itself by sending
> > >  ASP Active from both the IPSPs (A & B) but 
> > >  AS2 has sent ASP Active just to IPSP_A."
> > > 
> > > In my opinion IPSP_C should loadshare the
> traffic
> > > as it has received ASP active from both the
> IPSPs
> > > so AS1 is accessible through both of these.
> > > However, it is the responsibility of AS1 to
> route
> > > all the traffic towards AS2 through IPSP_A.
> > 
> > If IPSP_B receive some traffic from IPSP_C BUT no
> AS is ACTIVE for 
> it on
> > IPSP_C can it route this traffic???
> > CAN a SG send to a ss7 network a message receive
> from an INACTIVE 
> ASP !
> > I don't think so!
> > IPSP DE must be considered as two ASP/SG head to
> tail.
> > 
> [Vivek]
> AS is ACTIVE on IPSP_B (remember it sent ASP Active
> message)
[TOLGA]State of an AS is in the context of an ASP/SGP
pair. That ASP (IPSP_B) sends ACTIVE does not mean
that AS is ACTIVE for the SGP, unless SGP sends
ASPAC-ACK (IPSP_C in your example)
> As far as the SGP - ASP example see the following..
> 
> 1. It is the AS which is Active or Inactive not ASP
> 2. ASPTM ptocedures are between SGPs and ASPs not SG
> and ASPs
> 
> The local AS on the ASP may be active through some
> other
> SGP (say SGPy)to which an ACTIVE message has been
> sent.
> So SGP (say SGPx) should route the traffic to SS7
> coming from an 
> ASP which is inactive for an AS, because that AS is 
> Active through some other SGP and SS7 destination
> may still 
> be able to send the traffic towards that AS through
> some 
> other SGP
> 
> In M3UA outgoing and Incoming Traffic flows are
> independent,
> so why restrict the traffic ?
> 
> > > 
> > > In either case I think it is worth mentioning in
> 
> > > Implementor's guide whether Unidirectional
> traffic
> > > is possible between associations(in case of 
> > > IPSP DE) or not.It will avoid any further
> > > confusion or possible interoperatibility issues.
> > > 
> > > thanks,
> > > 
> > > -Vivek.
> > > 
> > > > Vivek,
> > > > --- Vivek Toky <vivektoky@ccpu.com> wrote:
> > > > > Hi All,
> > > > >  What should be done in the following case:
> > > > > 
> > > > > 
> > > > >       |IPSP_A                             |
> > > > >   AS1 |      ---------------------- IPSP_C|
> AS2
> > > > >       |IPSP_B                             |
> > > > > 
> > > > > IPSP_A and IPSP_B are in loadshare mode.
> > > > > 
> > > > >   when AS1 has activated itself by sending
> > > > > ASP Active from both the IPSPs (A & B) but
> > > > > AS2 has sent ASP Active just to IPSP_A. 
> > > > > 
> > > > > What should IPSP_C do on receiving traffic
> for AS1?
> > > > > 
> > > > > a) Loadshare the traffic towards IPSP_A and
> IPSP_B
> > > > > 
> > > > > or
> > > > > 
> > > > > b) It should send all the traffic to IPSP_A
> and 
> > > > >    wait untill it sends ASP Active IPSP_B as
> well.
> > > > >    In the above case IPSP_B cannot send the
> traffic 
> > > > >    towards IPSP_C but IPSP_C can.
> > > > [TOLGA]It should send all traffic to IPSPS_A.
> Which
> > > > case do you refer as "the above case", a) or
> b) ?
> > > > > 
> > > > > thanks,
> > > > > Vivek.
> > > > > 
> > > > 
> > > > 
> > > > __________________________________
> > > > Do you Yahoo!?
> > > > Yahoo! SiteBuilder - Free, easy-to-use web
> site design software 
> > > > http://sitebuilder.yahoo.com
> > > > 
> > > >
> _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> > > > 
> > > > 
> > > 
> > > -- 
> > > Continuous Computing
> > > http://www.ccpu.com
> > > 
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> > 
> > 
> 
> -- 
> Continuous Computing
> http://www.ccpu.com


__________________________________
Do you Yahoo!?
Yahoo! SiteBuilder - Free, easy-to-use web site design software
http://sitebuilder.yahoo.com

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



From exim@www1.ietf.org  Thu Jul 31 13:46:05 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA05775
	for <sigtran-archive@odin.ietf.org>; Thu, 31 Jul 2003 13:46:05 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iHUg-0005Fi-GD
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 13:45:38 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6VHjcuB020183
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 13:45:38 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iHU5-00058Y-OC; Thu, 31 Jul 2003 13:45:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iHTz-00057i-Ix
	for sigtran@optimus.ietf.org; Thu, 31 Jul 2003 13:44:55 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA05697
	for <sigtran@ietf.org>; Thu, 31 Jul 2003 13:44:51 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19iHTx-00063M-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 13:44:53 -0400
Received: from defender.ccpu.com ([216.54.134.34] helo=smtp.ccpu.com)
	by ietf-mx with esmtp (Exim 4.12)
	id 19iHTw-00063H-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 13:44:52 -0400
Received: from [172.16.1.203] (helo=CCPU00186)
	by smtp.ccpu.com with smtp (Exim 4.20 #1 (Debian))
	id 19iHTp-0004DH-6U; Thu, 31 Jul 2003 10:44:45 -0700
From: "VIvek Toky" <vivektoky@ccpu.com>
To: "Tolga Asveren" <tolga_asveren@yahoo.com>, <sigtran@ietf.org>
Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
Date: Thu, 31 Jul 2003 10:49:39 -0700
Message-ID: <HKEBKFBNPIDALDOKLGKDEELBCDAA.vivektoky@ccpu.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="US-ASCII"
Content-Transfer-Encoding: 7bit
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.6604 (9.0.2911.0)
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
In-Reply-To: <20030731171929.82038.qmail@web12803.mail.yahoo.com>
Importance: Normal
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,

-----Original Message-----
From: sigtran-admin@ietf.org [mailto:sigtran-admin@ietf.org]On Behalf Of
Tolga Asveren
Sent: Thursday, July 31, 2003 10:19 AM
To: sigtran@ietf.org
Subject: RE: [Sigtran] M3UA : IPSP DE behaviour


Vivek,
--- Vivek Toky <vivektoky@ccpu.com> wrote:
> Please see my comments inline
> 
> > See my opigion online
> > 
> > > -----Original Message-----
> > > From: Vivek Toky [mailto:vivektoky@ccpu.com] 
> > > Sent: Thursday, July 31, 2003 2:42 PM
> > > To: Tolga Asveren; sigtran@ietf.org
> > > Subject: RE: [Sigtran] M3UA : IPSP DE behaviour
> > > 
> > > 
> > > Hi Tolga,
> > >   "Above case" means 
> > > "when AS1 has activated itself by sending
> > >  ASP Active from both the IPSPs (A & B) but 
> > >  AS2 has sent ASP Active just to IPSP_A."
> > > 
> > > In my opinion IPSP_C should loadshare the
> traffic
> > > as it has received ASP active from both the
> IPSPs
> > > so AS1 is accessible through both of these.
> > > However, it is the responsibility of AS1 to
> route
> > > all the traffic towards AS2 through IPSP_A.
> > 
> > If IPSP_B receive some traffic from IPSP_C BUT no
> AS is ACTIVE for 
> it on
> > IPSP_C can it route this traffic???
> > CAN a SG send to a ss7 network a message receive
> from an INACTIVE 
> ASP !
> > I don't think so!
> > IPSP DE must be considered as two ASP/SG head to
> tail.
> > 
> [Vivek]
> AS is ACTIVE on IPSP_B (remember it sent ASP Active
> message)
[TOLGA]State of an AS is in the context of an ASP/SGP
pair. That ASP (IPSP_B) sends ACTIVE does not mean
that AS is ACTIVE for the SGP, unless SGP sends
ASPAC-ACK (IPSP_C in your example)

You are right, state of AS is in context of ASP, this was
what I wanted to say. I am assuming that SGP did send
ACTIVE ack back and ASP( for that AS) is active through
the SGP from which it got the ACK back. Is it fine now...


> As far as the SGP - ASP example see the following..
> 
> 1. It is the AS which is Active or Inactive not ASP
> 2. ASPTM ptocedures are between SGPs and ASPs not SG
> and ASPs
> 
> The local AS on the ASP may be active through some
> other
> SGP (say SGPy)to which an ACTIVE message has been
> sent.
> So SGP (say SGPx) should route the traffic to SS7
> coming from an 
> ASP which is inactive for an AS, because that AS is 
> Active through some other SGP and SS7 destination
> may still 
> be able to send the traffic towards that AS through
> some 
> other SGP
> 
> In M3UA outgoing and Incoming Traffic flows are
> independent,
> so why restrict the traffic ?
> 
> > > 
> > > In either case I think it is worth mentioning in
> 
> > > Implementor's guide whether Unidirectional
> traffic
> > > is possible between associations(in case of 
> > > IPSP DE) or not.It will avoid any further
> > > confusion or possible interoperatibility issues.
> > > 
> > > thanks,
> > > 
> > > -Vivek.
> > > 
> > > > Vivek,
> > > > --- Vivek Toky <vivektoky@ccpu.com> wrote:
> > > > > Hi All,
> > > > >  What should be done in the following case:
> > > > > 
> > > > > 
> > > > >       |IPSP_A                             |
> > > > >   AS1 |      ---------------------- IPSP_C|
> AS2
> > > > >       |IPSP_B                             |
> > > > > 
> > > > > IPSP_A and IPSP_B are in loadshare mode.
> > > > > 
> > > > >   when AS1 has activated itself by sending
> > > > > ASP Active from both the IPSPs (A & B) but
> > > > > AS2 has sent ASP Active just to IPSP_A. 
> > > > > 
> > > > > What should IPSP_C do on receiving traffic
> for AS1?
> > > > > 
> > > > > a) Loadshare the traffic towards IPSP_A and
> IPSP_B
> > > > > 
> > > > > or
> > > > > 
> > > > > b) It should send all the traffic to IPSP_A
> and 
> > > > >    wait untill it sends ASP Active IPSP_B as
> well.
> > > > >    In the above case IPSP_B cannot send the
> traffic 
> > > > >    towards IPSP_C but IPSP_C can.
> > > > [TOLGA]It should send all traffic to IPSPS_A.
> Which
> > > > case do you refer as "the above case", a) or
> b) ?
> > > > > 
> > > > > thanks,
> > > > > Vivek.
> > > > > 
> > > > 
> > > > 
> > > > __________________________________
> > > > Do you Yahoo!?
> > > > Yahoo! SiteBuilder - Free, easy-to-use web
> site design software 
> > > > http://sitebuilder.yahoo.com
> > > > 
> > > >
> _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
> > > > 
> > > > 
> > > 
> > > -- 
> > > Continuous Computing
> > > http://www.ccpu.com
> > > 
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > 
> > 
> > 
> 
> -- 
> Continuous Computing
> http://www.ccpu.com


__________________________________
Do you Yahoo!?
Yahoo! SiteBuilder - Free, easy-to-use web site design software
http://sitebuilder.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 exim@www1.ietf.org  Thu Jul 31 18:30:33 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA17470
	for <sigtran-archive@odin.ietf.org>; Thu, 31 Jul 2003 18:30:33 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iLw0-0008UM-LH
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 18:30:08 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6VMU8A5032616
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 18:30:08 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iLvt-0008Tq-G1; Thu, 31 Jul 2003 18:30:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iLvh-0008TT-IZ
	for sigtran@optimus.ietf.org; Thu, 31 Jul 2003 18:29:49 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA17455
	for <sigtran@ietf.org>; Thu, 31 Jul 2003 18:29:43 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19iLve-0000Oy-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 18:29:46 -0400
Received: from auds952.usa.alcatel.com ([143.209.238.7])
	by ietf-mx with esmtp (Exim 4.12)
	id 19iLvd-0000Ol-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 18:29:45 -0400
Received: from alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6VMTFB3023522;
	Thu, 31 Jul 2003 17:29:15 -0500 (CDT)
Message-ID: <3F29983B.2040505@alcatel.com>
Date: Thu, 31 Jul 2003 17:29:15 -0500
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
User-Agent: Mozilla/5.0 (X11; U; SunOS sun4u; en-US; rv:1.0.1) Gecko/20020920 Netscape/7.0
X-Accept-Language: en-us, en
MIME-Version: 1.0
CC: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - RPO
References: <829F074A10F98943A218DC363D09C92A25022F@w2ksjexg06.oni.com> <20030712170744.A11077@openss7.org> <3F15C7F2.6070704@pt.com> <20030716191143.F6725@openss7.org> <3F18328D.1080706@pt.com> <20030718120053.C14760@openss7.org> <3F184959.2040103@pt.com> <20030718133957.A16698@openss7.org> <3F1867CB.6060108@pt.com> <20030724134845.D11627@openss7.org> <3F26BAA2.1080908@pt.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

Brian, Andrew, et. al.,

So where do we stand on this issue? If it is true that some MTP3 
implementations would accept incoming messages during RPO and some would 
not, then we should probably cover both cases.

tomG.

Andrew Booth wrote:
> Hi Brian,
> 
> Sorry for the slow reply.
> 
> I was thinking mostly about letting the RTB drain and forgot about the 
> TB.  So if you want to let the TB drain, use the BSN in the LS PR is 
> used to resynchronize sequence numbers as in the original text.  Is the 
> rest of the proposed text ok?
   [...]

-- 
* 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.  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 exim@www1.ietf.org  Thu Jul 31 18:41:30 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA17716
	for <sigtran-archive@odin.ietf.org>; Thu, 31 Jul 2003 18:41:30 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iM6c-0000Y1-Jq
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 18:41:06 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6VMf6Z1002094
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 18:41:06 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iM6X-0000WK-4c; Thu, 31 Jul 2003 18:41:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iM5n-0000Ub-D1
	for sigtran@optimus.ietf.org; Thu, 31 Jul 2003 18:40:15 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA17663
	for <sigtran@ietf.org>; Thu, 31 Jul 2003 18:40:09 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19iM5k-0000S4-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 18:40:12 -0400
Received: from auds952.usa.alcatel.com ([143.209.238.7])
	by ietf-mx with esmtp (Exim 4.12)
	id 19iM5j-0000Rw-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 18:40:11 -0400
Received: from alcatel.com (localhost [127.0.0.1])
	by auds952.usa.alcatel.com (8.12.8p1/8.12.8) with ESMTP id h6VMdfB3024388;
	Thu, 31 Jul 2003 17:39:41 -0500 (CDT)
Message-ID: <3F299AAD.7060307@alcatel.com>
Date: Thu, 31 Jul 2003 17:39:41 -0500
From: Tom George <Tom.George@alcatel.com>
Organization: Alcatel
User-Agent: Mozilla/5.0 (X11; U; SunOS sun4u; en-US; rv:1.0.1) Gecko/20020920 Netscape/7.0
X-Accept-Language: en-us, en
MIME-Version: 1.0
CC: IETF sigtran <sigtran@ietf.org>
Subject: Re: FW: [Sigtran] M2PA Last Call Comment
References: <D73559B38D51D4119CB800508B9527B505AC19EB@mchh266e.mchh.siemens.de>
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

Hanns,

The reference to "the applicable MTP2 standard" comes from section 4. 
The links you labelled Q.703 and Q.2140 are irrelevant to deciding which 
protocol is "applicable" to the M2PA links. But both ends have to use 
the same one.

I suspect that I am not explaining what I mean very well....

tomG.

Schwarzbauer Hanns Juergen wrote:
   [...]
 > [HJS] Let's think about this issue the other way round: both STP's
 > only have Q.703 to the rest of the SS7 network but use Q.2140 between
 > then. I don't remember any discussion in ITU-T about any dependence
 > for proving on the SAAL links on the other Q.703 links on both sides.
 > It was simply "another level 2 protocol". So there was no problem at
 > all, especially as all data links are decoupled from each other by
 > MTP L3!
 >
   [...]
 > [HJS] I think the text is misleading: the words "if this is allowed
 > by the applicable MTP2 standard" couple the M2PA procedure to the TDM
 > side. My proposal reads liks:
 >
 > section 4.1.3 The proving period MAY be omitted upon bilateral
 > agreement between the concerned peers.
   [...diagram snipped because my "rewrap" mangled it...]
 > In that scenario I even
 > doubt that your text allows STP A and STP B to omit the proving. The
 > text could be interpreted to allow omission of proving only if this
 > is applicable on both ends. I consider this correlation as not
 > necessary at all, as all MTP L2 have to be considered independent
 > from each other (due to MTP3). M2PA from the beginning was considered
 > a "new Level 2" and therefore should not be designed a way to have
 > built in dependencies from existing level 2 protocols.
   [...]

-- 
* 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.  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 exim@www1.ietf.org  Thu Jul 31 18:51:35 2003
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA17870
	for <sigtran-archive@odin.ietf.org>; Thu, 31 Jul 2003 18:51:35 -0400 (EDT)
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iMGN-0000rb-EA
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 18:51:11 -0400
Received: (from exim@localhost)
	by www1.ietf.org (8.12.8/8.12.8/Submit) id h6VMpBxJ003298
	for sigtran-archive@odin.ietf.org; Thu, 31 Jul 2003 18:51:11 -0400
Received: from localhost.localdomain ([127.0.0.1] helo=www1.ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iMGD-0000pg-Qr; Thu, 31 Jul 2003 18:51:01 -0400
Received: from odin.ietf.org ([132.151.1.176] helo=ietf.org)
	by optimus.ietf.org with esmtp (Exim 4.20)
	id 19iMFs-0000pQ-Fx
	for sigtran@optimus.ietf.org; Thu, 31 Jul 2003 18:50:40 -0400
Received: from ietf-mx (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA17852
	for <sigtran@ietf.org>; Thu, 31 Jul 2003 18:50:34 -0400 (EDT)
Received: from ietf-mx ([132.151.6.1])
	by ietf-mx with esmtp (Exim 4.12)
	id 19iMFp-0000Vf-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 18:50:37 -0400
Received: from gw.openss7.com ([142.179.199.224] ident=[8lmt5xruMk3ED0+pNWSWT7dQFvGqJ+Zx])
	by ietf-mx with esmtp (Exim 4.12)
	id 19iMFo-0000Va-00
	for sigtran@ietf.org; Thu, 31 Jul 2003 18:50:36 -0400
Received: from ns.pigworks.openss7.net (IDENT:C3awC1H2YHjkYHXfpeso3rzL98uxtlhb@ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id h6VMoWE05167;
	Thu, 31 Jul 2003 16:50:32 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id h6VMoVj06763;
	Thu, 31 Jul 2003 16:50:31 -0600
Date: Thu, 31 Jul 2003 16:50:31 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: Tom George <Tom.George@alcatel.com>
Cc: "'sigtran@ietf.org'" <sigtran@ietf.org>
Subject: Re: [Sigtran] M2PA Last Call Comment - RPO
Message-ID: <20030731165031.A6734@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: Tom George <Tom.George@alcatel.com>,
	"'sigtran@ietf.org'" <sigtran@ietf.org>
References: <3F15C7F2.6070704@pt.com> <20030716191143.F6725@openss7.org> <3F18328D.1080706@pt.com> <20030718120053.C14760@openss7.org> <3F184959.2040103@pt.com> <20030718133957.A16698@openss7.org> <3F1867CB.6060108@pt.com> <20030724134845.D11627@openss7.org> <3F26BAA2.1080908@pt.com> <3F29983B.2040505@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: <3F29983B.2040505@alcatel.com>; from Tom.George@alcatel.com on Thu, Jul 31, 2003 at 05:29:15PM -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>

Tom,

See the original note on this thread for suggested changes.

--brian

Tom George wrote:                                                                      (Thu, 31 Jul 2003 17:29:15)
> Brian, Andrew, et. al.,
> 
> So where do we stand on this issue? If it is true that some MTP3 
> implementations would accept incoming messages during RPO and some would 
> not, then we should probably cover both cases.
> 
> tomG.
> 
> Andrew Booth wrote:
> > Hi Brian,
> > 
> > Sorry for the slow reply.
> > 
> > I was thinking mostly about letting the RTB drain and forgot about the 
> > TB.  So if you want to let the TB drain, use the BSN in the LS PR is 
> > used to resynchronize sequence numbers as in the original text.  Is the 
> > rest of the proposed text ok?
>    [...]
> 
> -- 
> * 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.  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



