From sigtran-admin@ietf.org  Mon Sep  2 00:03:44 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA03540
	for <sigtran-archive@lists.ietf.org>; Mon, 2 Sep 2002 00:03:44 -0400 (EDT)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id g8244So29215;
	Mon, 2 Sep 2002 00:04:29 -0400
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id g8242so29181
	for <sigtran@optimus.ietf.org>; Mon, 2 Sep 2002 00:02:54 -0400
Received: from mta0 ([61.144.161.2])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA03495
	for <sigtran@ietf.org>; Mon, 2 Sep 2002 00:01:10 -0400 (EDT)
Received: from l19463 (mta0 [172.17.1.62])
 by mta0.huawei.com (iPlanet Messaging Server 5.2 HotFix 0.8 (built Jul 12
 2002)) with ESMTPA id <0H1S00LOYLSMJQ@mta0.huawei.com> for sigtran@ietf.org;
 Mon, 02 Sep 2002 12:00:23 +0800 (CST)
Date: Sun, 01 Sep 2002 12:08:13 +0800
From: Ming Lin <mlin@huawei.com>
Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-m2pa-06.txt
To: Tom George <Tom.George@alcatel.com>
Cc: sigtran@ietf.org
Message-id: <000701c2516d$319c30c0$2a180b0a@huawei.com>
MIME-version: 1.0
X-MIMEOLE: Produced By Microsoft MimeOLE V5.50.4522.1200
X-Mailer: Microsoft Outlook Express 5.50.4522.1200
Content-type: text/plain; charset=iso-8859-1
Content-transfer-encoding: 7BIT
X-Priority: 3
X-MSMail-priority: Normal
References: <200208301150.HAA12342@ietf.org> <3D6F8F82.61E1FF94@alcatel.com>
Content-Transfer-Encoding: 7BIT
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7BIT

Dear Tom,

I think we should remove RETRIEVAL state, this state will result in many problem but no use,
for example the No. 18 issue.

"18. 08/09/02 Howard May "[Sigtran] M2PA - Minor corrections?" 
     In Figure 7, RETRIEVAL receives MTP Start. Go to AIP, not OOS.
 A: Yes.  [Done.]"

----- Original Message ----- 
From: "Tom George" <Tom.George@alcatel.com>
Cc: <sigtran@ietf.org>
Sent: Friday, August 30, 2002 11:30 PM
Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-m2pa-06.txt


> M2PA Changes for version 6
> 08/28/2002
> 
> Here are the changes in draft-ietf-sigtran-m2pa-06.txt (M2PAv6).
> 
> If anyone wants to study the changes in detail, I have a "diff" file I
> can send.
> 
> This is not a complete list of changes. I did not include minor
> rewording.
> 
> 
> This is the list of comments received on M2PAv5 on the SIGTRAN list.
> 
> ==================================================================
> 
> Here is my list of comments. For the most part, they are arranged by
> the thread from the SIGTRAN list.
> 
> Everyone please make comments. I think the discussion would go more
> smoothly if you only talked about one item per e-mail.
> 
> 
> 1. Section 2.2. FSN/BSN should be 24 bits, not 23.
> 
> A: This was a typo. It will be 24 in M2PAv6.  [Done.]
> 
> 2. 05/06/02 Jeff Craig "[Sigtran] (M2PA) What actions should be taken
>    if LSA or User Data Received in PROVING?" There is no guidance
>    about anomalous events during alignment/proving.
> 
> A: I agree that some text should be added. I have not worked out the
> details and would welcome suggestions.
> 
>    My ideas:
> 
>    User Data received during PROVING should be discarded.
> 
>    Link Status Alignment received during PROVING should be ignored.
> 
>    [Done.]
> 
> 3. 05/10/02 Jeff Craig, "[Sigtran] (M2PA) Please standardize M2PA
>    protocol timer identifiers" and Jeff Craig 06/21/02 "m2pa comment:
>    timers". Line up timer names in Q.703 and M2PA. Add the T7 timer.
> 
> A: This is what I got out of the discussion:
> 
>    Keep same names for T1, T3, T6. [Done.]
> 
>    Rename M2PA T2 to T4. [Done.]
> 
>    Don't use "timer" names for the rate parameters.  [Done.]
> 
>    T7 - see #11 below.  [Done.]
> 
> 4. 05/17/02 Jeff Craig, "[Sigtran] (M2PA) Please add a Network
>    Appearance parameter to the M2PA protocol". A suggestion that a NA
>    parameter be added to M2PA.
> 
> A: I prefer not to add this.  [Done.]
> 
> 5. 05/22/02 anssharma, "[Sigtran] M2PA Clarification". In level 2 flow
>    control, M2PA stops transmitting. What does it do with the
>    messages?
> 
> A: I think is an implementation matter.  [Done.]
> 
> 6. 05/24/02 Lisa Gorgone, "[Sigtran] M2PA:Questions on sequence
>    number". Some questions about bad values of FSNR and BSNR.
> 
> A: M2PAv5 (4.2.1) says to fail the link if one FSNR is out of order. I
> think that is OK. There is nothing about bad BSNR values. I would be
> inclined to fail the link in this case as well, since one end is
> acking messages that the other does not have in its transmit buffer.
> 
>    I have a question about the BSNR values on the list. 2 out of 3
>    consecutive messages with bad BSNRs?
> 
>    Eliminate acks on the Status stream. All acks go over the Data
>    stream. If there is no User Data message to be sent, send one with
>    no data in it. Get rid of Link Status In Service, since its only
>    purpose was to carry an ack. [Done.]
> 
> 7. 05/24/02 Brian Bidulock "[Sigtran] Re: M2PA Clarification". Does
>    M2PA need to withhold messages from SCTP during busy?
> 
> A: This is how Q.703 works, so we should copy it. Before there were
> M2PA sequence numbers, messages were withheld while the peer was Busy
> to prevent message loss in case the link later failed. Now that M2PA
> has its own sequence numbers, this is no longer a danger. So M2PA can
> continue transmitting while its peer is Busy. 
> 
>    I updated secion 4.1.5 Level 2 Flow Control.  [Done.]
> 
> 8. 06/18/02 amish.chana "[Sigtran] M2PA Comment: Multi-homing". Should
>    M2PA support multi-homing?
> 
> A: It is allowed. The only change needed is in 4.1.2 so that it says
> "the IP address (or IP addresses if multi-homing is used)".  [Done.]
> 
> 9. 06/19/02 Lisa Gorgone "Re: [Sigtran] M2PA Comment:Retrieval
>    State". Does M2PA need a RETRIEVAL state? Does RETRIEVAL state
>    cause harm?
> 
> A: I would like to have more discussion about this. 
> 
>    Since there has not been more discussion, I will leave this as is
>    for M2PA version 6.  [Done.]
> 
> 10. 06/21/02 Javier Pastor-Balbas "[Sigtran] M2PA Comment:
>     Multi-homing". Also 06/21/02 "[Sigtran] Multiplexed M2PA links on
>     an association".  Should M2PA allow multiple links per
>     association?
> 
> A: The disadvantage of this is that if the association dies, you lose
> all the links on the association. This is one of the main reasons I
> did not allow for multiple links in an association. 
> 
>    No change to the draft is needed.  [Done.]
> 
> 11. 06/21/02 Andrew Booth "[Sigtran] M2PA Comment: T7 Timer" and
>     several of the earlier threads. Should there be an equivalent of
>     MTP2 T7?
> 
> A: OK, you finally convinced me. 8-) I added the T7 timer in 4.2.1
> Sending/receiving messages and 4.1.5 Level 2 Flow Control.  [Done.]
> 
> 12. 06/25/02 Lyndon Ong "RE: [Sigtran] M2PA Queries - summary". Just a
>     summary of the discussion.
> 
> A: No answer needed. [Done.]
> 
> 13. 06/26/02 Karen King "[Sigtran] A couple of comments on M2PA
>     processor outage". Recommend adding text to 4.1.4 Processor Outage
>     about flushing buffers as in Q.703.
> 
> A: Even though this is in Q.703, the action is taken by MTP3. Section
> 4.1.4 (next to last paragraph) does allow for the flush. Flush is
> described in 4.2.4. So I think this is taken care of (although there
> may be problems in 4.2.4 - that's another thread).  [Done.]
> 
> 14. 06/28/02 Ray Singh "[Sigtran] Comment on M2PA - Performance
>     Monitoring". Add back the monitoring of SCTP association that was
>     in draft version 3.
> 
> A: The reason I took this out is that I never had any solid ideas
> about what SCTP parameters should be monitored and what the acceptable
> values would be. If we could come up with something helpful, I am not
> opposed to adding the monitoring back in. But I don't see the point of
> adding it if the criteria will be some implementation-dependent
> parameters having implementation-dependent values. 8-( [Done.]
> 
> 15. 07/12/02 anssharma "[Sigtran] M2PA - Flush Buffers". Questions
>     about Flush.
> 
> A: These are good questions but I have not had time to think about
> them.
> 
> 
> 16. 07/19/02 Ray Singh "[Sigtran] Comment on M2PA - Slow Start". What
>     kind of warning is needed about slow start?
> 
> A: I proposed a general warning about slow start to replace specific
> warnings throughout the document.
> 
> alignment, changeover, path failure (multihoming), reduced volume. See
> Tolga, 07/26/02. [Done.]
> 
> 
> 17. 07/22/02 Tolga Asveren "[Sigtran] M2PA : LS messages on stream
>     0". Are there problems with sending Status and Data on different
>     streams? What mimics MTP2 best?
> 
> A: The problem with putting LS and Data on the same stream is that a
> LS message can get stuck behind a Data retransmission. This is
> contrary to the message priority of MTP2. (Brian Bidulock 08/13/2002
> 21:00 pointed out that this is a problem that MTP2 does not have.)
> 
> The problem with using SCTP unordered transmission for LS messages is
> that they can get out of order with each other. For example, a LS Proc
> Outage could get out of order with a LS Proc Outage Ended. 
> 
> MTP does not require that LSSUs and MTUs be associated with each other
> in any particular way. It has MTP2 put LSSUs ahead of MTUs in the
> transmit buffer. So it should not be required that M2PA transmit LS in
> order with Data.
> 
> Having M2PA put LS at the front of the M2PA transmit queue helps some,
> but it does not do anything about the messages that are in the SCTP
> send queue. To truly imitate MTP2 at this point, we would need a way
> to give a message to SCTP and tell it to put the message at the front
> of its send queue.
> 
> (Or, we could restrict SCTP to just 1 message in its send buffer. This
> would allow M2PA to prioritize like MTP2. But this would create severe
> problems of its own. No one seemed to like this option.)
> 
> Sending LS and Data on different streams allows for a partial
> imitation of the MTP2 transmission priority. If the receiving M2PA
> gives priority to reading incoming LS messages (on stream 0) over Data
> messages (on stream 1), then LS messages can be received sooner than
> if they waited behind the Data messages. Also, it is possible for
> SCTP, by some implementation-dependent method or some future SCTP
> extension, to give a higher priority to sending messages on stream 0.
> 
> My conclusion is to leave the draft alone. Having separate Status and
> Data streams does not exactly imitate MTP2, but it is as close as we
> can get using SCTP.  [Done.]
> 
> Some related matters:
> 
> I will add to 4.1.7 so that it says that M2PA transmits LS messages
> before Data messages. This is in addition to the attempt to prioritize
> SCTP stream 0. [That was already there. Done.]
> 
> The FSN should be ignored in the LS messages. FSN should be
> incremented only for Data messages. [That was already there. Done.]
> 
> I will add a warning that one has to watch out for a wraparound of the
> FSN/BSN numbers. This is unlikely with 24 bits, but should probably be
> mentioned. [Done.]
> 
> 
> 18. 08/09/02 Howard May "[Sigtran] M2PA - Minor corrections?" 
> 
>     In Figure 7, RETRIEVAL receives MTP Start. Go to AIP, not OOS.
> 
> A: Yes.  [Done.]
> 
> 
> ==================================================================
> 
> More detailed list of changes:
> 
> 2.2  M2PA Header: Changed BSN and FSN to 24 bits in diagram.
> 
> 2.3.2 Link Status: Removed In Service (10) from list of values.
> 
> 2.3.2.1 Link Status Proving, 1st para: Removed statement about SCTP
> slow start.
> 
> 3.  M2PA Link State Control, Figure 7: M2PA Link State Transition
> Diagram:
> 
>     T2 is renamed T4.
> 
>     Added T7 expiry.
> 
>     In RETRIEVAL, MTP3 Start goes to AIP instead of OOS.
> 
> 4.1.2 MTP and SCTP Entities, 2nd para: Added "(or IP addresses, if
> multi-homing is used)".
> 
> 4.1.3 Link Alignment, 1st para: Removed list item 3 about SCTP slow
> start.
> 
> 4.1.3 Link Alignment, several places: Renamed T2 to T4.
> 
> 4.1.3 Link Alignment, para "Anomalous...": Added this paragraph.
> 
> 4.1.5 Level 2 Flow Control, para "The peer M2PA should...": Changed
> "shall cease transmitting messages" to "should continue transmitting
> messages".
> 
> 4.1.5 Level 2 Flow Control, para "The peer M2PA shall not...": Added
> this paragraph about T7.
> 
> 4.2.1 Sending and receiving messages, para "When M2PA receives a User
> Data message...": Specified that this is a _User Data_ message.
> 
> 4.2.1 Sending and receiving messages: Removed paragraph stating that
> ack may be sent in User Data or Link Status.
> 
> 4.2.1 Sending and receiving messages, para "M2PA should follow the
> criterion...": Added this paragraph about 2-out-of-3 criterion.
> 
> 4.2.1 Sending and receiving messages, para "M2PA should ignore...":
> Added this paragraph.
> 
> 4.2.1 Sending and receiving messages, para "Note: In all
> calculations...": Added this paragraph.
> 
> 4.2.1 Sending and receiving messages, para "If there is no other User
> Data...": Changed this to say M2PA sends User Data with no payload
> instead of LS In Service.
> 
> 4.2.1 Sending and receiving messages, last 2 paragraphs: Added these
> paragraphs about T7.
> 
> 4.3 SCTP Considerations: Added this section to discuss slow start.
> 
> 5.1  Link Initialization, several places: Renamed T2 to T4.
> 
> 7.1 SCTP Payload Protocol Identifier: Registered User Port Number is
> 3565. Removed UDP.
> 
> 9.  References: Updated some of the references.
> 
> 10.  Author's Addresses: New address for Brian Bidulock.
> 
> (end)
> 
> 
> -- 
> * Tom George          972-519-3168           Tom.George@alcatel.com   *
> * PB2-1323       Intranet:  http://www.ssd.usa.alcatel.com/~tgeorge/  *
> * Alcatel USA, Inc.; 1000 Coit Rd, MS PB2-STPA; Plano TX 75075-5813   *
> * I actively disclaim speaking for Alcatel USA, Inc. I do not purport *
> * to advertise or represent its position,  operations, or activities. *
> _______________________________________________
> Sigtran mailing list
> Sigtran@ietf.org
> https://www1.ietf.org/mailman/listinfo/sigtran
_______________________________________________
Sigtran mailing list
Sigtran@ietf.org
https://www1.ietf.org/mailman/listinfo/sigtran


From sigtran-admin@ietf.org  Mon Sep  2 04:47:59 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA16962
	for <sigtran-archive@lists.ietf.org>; Mon, 2 Sep 2002 04:47:59 -0400 (EDT)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id g828mpo21049;
	Mon, 2 Sep 2002 04:48:52 -0400
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id g828lZo20983
	for <sigtran@optimus.ietf.org>; Mon, 2 Sep 2002 04:47:35 -0400
Received: from odin.isw.intel.com (swfdns01.isw.intel.com [192.55.37.143])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA16899
	for <sigtran@ietf.org>; Mon, 2 Sep 2002 04:45:50 -0400 (EDT)
Received: from swsmsxvs01.isw.intel.com (swsmsxvs01.isw.intel.com [172.28.130.22])
	by odin.isw.intel.com (8.11.6/8.11.6/d: solo.mc,v 1.43 2002/08/30 20:06:11 dmccart Exp $) with SMTP id g828krY14544
	for <sigtran@ietf.org>; Mon, 2 Sep 2002 08:46:53 GMT
Received: from swsmsx17.isw.intel.com ([172.28.130.21])
 by swsmsxvs01.isw.intel.com (NAVGW 2.5.2.11) with SMTP id M2002090209452900344
 ; Mon, 02 Sep 2002 09:45:29 +0100
Received: by swsmsx17.isw.intel.com with Internet Mail Service (5.5.2653.19)
	id <QT1M3G17>; Mon, 2 Sep 2002 09:51:07 +0100
Message-ID: <4A3192E7220D9241A1B774B074C6E1210171F226@swsmsx103.isw.intel.com>
From: "May, Howard" <howard.may@intel.com>
To: "'Ming Lin'" <mlin@huawei.com>
Cc: sigtran@ietf.org
Date: Mon, 2 Sep 2002 09:46:20 +0100 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Sigtran] M2PA - Retrieval State
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Dear Ming

I think Issue 18, included in your email below, is resolved by Toms
correction (well done with the ascii art work Tom!).  Ming could you post a
summary of the other problems you see with the Retrieval state.

I currently believe the Retrieval state to be worthwhile.  As the link can
only leave the Retrieval state after an event from MTP3 it guarantees that
MTP3 will be able to perform and complete retrieval procedures before the
link comes back into service.   If the Retrieval state is not used then I
think we would be left in a situation where the link could come back into
service before Retrieval had been completed.  There is then an issue of
either allowing MTP3 and M2PA to fight over messages in the TX buffer (MTP3
trying to Retrieve them and M2PA trying to transmit them - clearly
undesirable) or perhaps when the link fails moving messages left in the Tx
buffer to another buffer until MTP3 decides whether it wants them or not (My
gut feeling is that this is where more problems lie rather than the use of
the Retrieval state)

Cheers

> -----Original Message-----
> From: Ming Lin [mailto:mlin@huawei.com]
> Sent: 01 September 2002 05:08
> To: Tom George
> Cc: sigtran@ietf.org
> Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-m2pa-06.txt
> 
> 
> Dear Tom,
> 
> I think we should remove RETRIEVAL state, this state will 
> result in many problem but no use,
> for example the No. 18 issue.
> 
> "18. 08/09/02 Howard May "[Sigtran] M2PA - Minor corrections?" 
>      In Figure 7, RETRIEVAL receives MTP Start. Go to AIP, not OOS.
>  A: Yes.  [Done.]"
> 
> ----- Original Message ----- 
> From: "Tom George" <Tom.George@alcatel.com>
> Cc: <sigtran@ietf.org>
> Sent: Friday, August 30, 2002 11:30 PM
> Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-m2pa-06.txt
> 
> 
> > M2PA Changes for version 6
> > 08/28/2002
> > 
> > Here are the changes in draft-ietf-sigtran-m2pa-06.txt (M2PAv6).
> > 
> > If anyone wants to study the changes in detail, I have a 
> "diff" file I
> > can send.
> > 
> > This is not a complete list of changes. I did not include minor
> > rewording.
> > 
> > 
> > This is the list of comments received on M2PAv5 on the SIGTRAN list.
> > 
> > ==================================================================
> > 
> > Here is my list of comments. For the most part, they are arranged by
> > the thread from the SIGTRAN list.
> > 
> > Everyone please make comments. I think the discussion would go more
> > smoothly if you only talked about one item per e-mail.
> > 
> > 
> > 1. Section 2.2. FSN/BSN should be 24 bits, not 23.
> > 
> > A: This was a typo. It will be 24 in M2PAv6.  [Done.]
> > 
> > 2. 05/06/02 Jeff Craig "[Sigtran] (M2PA) What actions 
> should be taken
> >    if LSA or User Data Received in PROVING?" There is no guidance
> >    about anomalous events during alignment/proving.
> > 
> > A: I agree that some text should be added. I have not worked out the
> > details and would welcome suggestions.
> > 
> >    My ideas:
> > 
> >    User Data received during PROVING should be discarded.
> > 
> >    Link Status Alignment received during PROVING should be ignored.
> > 
> >    [Done.]
> > 
> > 3. 05/10/02 Jeff Craig, "[Sigtran] (M2PA) Please standardize M2PA
> >    protocol timer identifiers" and Jeff Craig 06/21/02 
> "m2pa comment:
> >    timers". Line up timer names in Q.703 and M2PA. Add the T7 timer.
> > 
> > A: This is what I got out of the discussion:
> > 
> >    Keep same names for T1, T3, T6. [Done.]
> > 
> >    Rename M2PA T2 to T4. [Done.]
> > 
> >    Don't use "timer" names for the rate parameters.  [Done.]
> > 
> >    T7 - see #11 below.  [Done.]
> > 
> > 4. 05/17/02 Jeff Craig, "[Sigtran] (M2PA) Please add a Network
> >    Appearance parameter to the M2PA protocol". A suggestion 
> that a NA
> >    parameter be added to M2PA.
> > 
> > A: I prefer not to add this.  [Done.]
> > 
> > 5. 05/22/02 anssharma, "[Sigtran] M2PA Clarification". In 
> level 2 flow
> >    control, M2PA stops transmitting. What does it do with the
> >    messages?
> > 
> > A: I think is an implementation matter.  [Done.]
> > 
> > 6. 05/24/02 Lisa Gorgone, "[Sigtran] M2PA:Questions on sequence
> >    number". Some questions about bad values of FSNR and BSNR.
> > 
> > A: M2PAv5 (4.2.1) says to fail the link if one FSNR is out 
> of order. I
> > think that is OK. There is nothing about bad BSNR values. I would be
> > inclined to fail the link in this case as well, since one end is
> > acking messages that the other does not have in its transmit buffer.
> > 
> >    I have a question about the BSNR values on the list. 2 out of 3
> >    consecutive messages with bad BSNRs?
> > 
> >    Eliminate acks on the Status stream. All acks go over the Data
> >    stream. If there is no User Data message to be sent, 
> send one with
> >    no data in it. Get rid of Link Status In Service, since its only
> >    purpose was to carry an ack. [Done.]
> > 
> > 7. 05/24/02 Brian Bidulock "[Sigtran] Re: M2PA Clarification". Does
> >    M2PA need to withhold messages from SCTP during busy?
> > 
> > A: This is how Q.703 works, so we should copy it. Before there were
> > M2PA sequence numbers, messages were withheld while the 
> peer was Busy
> > to prevent message loss in case the link later failed. Now that M2PA
> > has its own sequence numbers, this is no longer a danger. 
> So M2PA can
> > continue transmitting while its peer is Busy. 
> > 
> >    I updated secion 4.1.5 Level 2 Flow Control.  [Done.]
> > 
> > 8. 06/18/02 amish.chana "[Sigtran] M2PA Comment: 
> Multi-homing". Should
> >    M2PA support multi-homing?
> > 
> > A: It is allowed. The only change needed is in 4.1.2 so that it says
> > "the IP address (or IP addresses if multi-homing is used)".  [Done.]
> > 
> > 9. 06/19/02 Lisa Gorgone "Re: [Sigtran] M2PA Comment:Retrieval
> >    State". Does M2PA need a RETRIEVAL state? Does RETRIEVAL state
> >    cause harm?
> > 
> > A: I would like to have more discussion about this. 
> > 
> >    Since there has not been more discussion, I will leave this as is
> >    for M2PA version 6.  [Done.]
> > 
> > 10. 06/21/02 Javier Pastor-Balbas "[Sigtran] M2PA Comment:
> >     Multi-homing". Also 06/21/02 "[Sigtran] Multiplexed 
> M2PA links on
> >     an association".  Should M2PA allow multiple links per
> >     association?
> > 
> > A: The disadvantage of this is that if the association 
> dies, you lose
> > all the links on the association. This is one of the main reasons I
> > did not allow for multiple links in an association. 
> > 
> >    No change to the draft is needed.  [Done.]
> > 
> > 11. 06/21/02 Andrew Booth "[Sigtran] M2PA Comment: T7 Timer" and
> >     several of the earlier threads. Should there be an equivalent of
> >     MTP2 T7?
> > 
> > A: OK, you finally convinced me. 8-) I added the T7 timer in 4.2.1
> > Sending/receiving messages and 4.1.5 Level 2 Flow Control.  [Done.]
> > 
> > 12. 06/25/02 Lyndon Ong "RE: [Sigtran] M2PA Queries - 
> summary". Just a
> >     summary of the discussion.
> > 
> > A: No answer needed. [Done.]
> > 
> > 13. 06/26/02 Karen King "[Sigtran] A couple of comments on M2PA
> >     processor outage". Recommend adding text to 4.1.4 
> Processor Outage
> >     about flushing buffers as in Q.703.
> > 
> > A: Even though this is in Q.703, the action is taken by 
> MTP3. Section
> > 4.1.4 (next to last paragraph) does allow for the flush. Flush is
> > described in 4.2.4. So I think this is taken care of (although there
> > may be problems in 4.2.4 - that's another thread).  [Done.]
> > 
> > 14. 06/28/02 Ray Singh "[Sigtran] Comment on M2PA - Performance
> >     Monitoring". Add back the monitoring of SCTP 
> association that was
> >     in draft version 3.
> > 
> > A: The reason I took this out is that I never had any solid ideas
> > about what SCTP parameters should be monitored and what the 
> acceptable
> > values would be. If we could come up with something 
> helpful, I am not
> > opposed to adding the monitoring back in. But I don't see 
> the point of
> > adding it if the criteria will be some implementation-dependent
> > parameters having implementation-dependent values. 8-( [Done.]
> > 
> > 15. 07/12/02 anssharma "[Sigtran] M2PA - Flush Buffers". Questions
> >     about Flush.
> > 
> > A: These are good questions but I have not had time to think about
> > them.
> > 
> > 
> > 16. 07/19/02 Ray Singh "[Sigtran] Comment on M2PA - Slow 
> Start". What
> >     kind of warning is needed about slow start?
> > 
> > A: I proposed a general warning about slow start to replace specific
> > warnings throughout the document.
> > 
> > alignment, changeover, path failure (multihoming), reduced 
> volume. See
> > Tolga, 07/26/02. [Done.]
> > 
> > 
> > 17. 07/22/02 Tolga Asveren "[Sigtran] M2PA : LS messages on stream
> >     0". Are there problems with sending Status and Data on different
> >     streams? What mimics MTP2 best?
> > 
> > A: The problem with putting LS and Data on the same stream is that a
> > LS message can get stuck behind a Data retransmission. This is
> > contrary to the message priority of MTP2. (Brian Bidulock 08/13/2002
> > 21:00 pointed out that this is a problem that MTP2 does not have.)
> > 
> > The problem with using SCTP unordered transmission for LS 
> messages is
> > that they can get out of order with each other. For 
> example, a LS Proc
> > Outage could get out of order with a LS Proc Outage Ended. 
> > 
> > MTP does not require that LSSUs and MTUs be associated with 
> each other
> > in any particular way. It has MTP2 put LSSUs ahead of MTUs in the
> > transmit buffer. So it should not be required that M2PA 
> transmit LS in
> > order with Data.
> > 
> > Having M2PA put LS at the front of the M2PA transmit queue 
> helps some,
> > but it does not do anything about the messages that are in the SCTP
> > send queue. To truly imitate MTP2 at this point, we would need a way
> > to give a message to SCTP and tell it to put the message at 
> the front
> > of its send queue.
> > 
> > (Or, we could restrict SCTP to just 1 message in its send 
> buffer. This
> > would allow M2PA to prioritize like MTP2. But this would 
> create severe
> > problems of its own. No one seemed to like this option.)
> > 
> > Sending LS and Data on different streams allows for a partial
> > imitation of the MTP2 transmission priority. If the receiving M2PA
> > gives priority to reading incoming LS messages (on stream 
> 0) over Data
> > messages (on stream 1), then LS messages can be received sooner than
> > if they waited behind the Data messages. Also, it is possible for
> > SCTP, by some implementation-dependent method or some future SCTP
> > extension, to give a higher priority to sending messages on 
> stream 0.
> > 
> > My conclusion is to leave the draft alone. Having separate 
> Status and
> > Data streams does not exactly imitate MTP2, but it is as close as we
> > can get using SCTP.  [Done.]
> > 
> > Some related matters:
> > 
> > I will add to 4.1.7 so that it says that M2PA transmits LS messages
> > before Data messages. This is in addition to the attempt to 
> prioritize
> > SCTP stream 0. [That was already there. Done.]
> > 
> > The FSN should be ignored in the LS messages. FSN should be
> > incremented only for Data messages. [That was already there. Done.]
> > 
> > I will add a warning that one has to watch out for a 
> wraparound of the
> > FSN/BSN numbers. This is unlikely with 24 bits, but should 
> probably be
> > mentioned. [Done.]
> > 
> > 
> > 18. 08/09/02 Howard May "[Sigtran] M2PA - Minor corrections?" 
> > 
> >     In Figure 7, RETRIEVAL receives MTP Start. Go to AIP, not OOS.
> > 
> > A: Yes.  [Done.]
> > 
> > 
> > ==================================================================
> > 
> > More detailed list of changes:
> > 
> > 2.2  M2PA Header: Changed BSN and FSN to 24 bits in diagram.
> > 
> > 2.3.2 Link Status: Removed In Service (10) from list of values.
> > 
> > 2.3.2.1 Link Status Proving, 1st para: Removed statement about SCTP
> > slow start.
> > 
> > 3.  M2PA Link State Control, Figure 7: M2PA Link State Transition
> > Diagram:
> > 
> >     T2 is renamed T4.
> > 
> >     Added T7 expiry.
> > 
> >     In RETRIEVAL, MTP3 Start goes to AIP instead of OOS.
> > 
> > 4.1.2 MTP and SCTP Entities, 2nd para: Added "(or IP addresses, if
> > multi-homing is used)".
> > 
> > 4.1.3 Link Alignment, 1st para: Removed list item 3 about SCTP slow
> > start.
> > 
> > 4.1.3 Link Alignment, several places: Renamed T2 to T4.
> > 
> > 4.1.3 Link Alignment, para "Anomalous...": Added this paragraph.
> > 
> > 4.1.5 Level 2 Flow Control, para "The peer M2PA should...": Changed
> > "shall cease transmitting messages" to "should continue transmitting
> > messages".
> > 
> > 4.1.5 Level 2 Flow Control, para "The peer M2PA shall not...": Added
> > this paragraph about T7.
> > 
> > 4.2.1 Sending and receiving messages, para "When M2PA 
> receives a User
> > Data message...": Specified that this is a _User Data_ message.
> > 
> > 4.2.1 Sending and receiving messages: Removed paragraph stating that
> > ack may be sent in User Data or Link Status.
> > 
> > 4.2.1 Sending and receiving messages, para "M2PA should follow the
> > criterion...": Added this paragraph about 2-out-of-3 criterion.
> > 
> > 4.2.1 Sending and receiving messages, para "M2PA should ignore...":
> > Added this paragraph.
> > 
> > 4.2.1 Sending and receiving messages, para "Note: In all
> > calculations...": Added this paragraph.
> > 
> > 4.2.1 Sending and receiving messages, para "If there is no 
> other User
> > Data...": Changed this to say M2PA sends User Data with no payload
> > instead of LS In Service.
> > 
> > 4.2.1 Sending and receiving messages, last 2 paragraphs: Added these
> > paragraphs about T7.
> > 
> > 4.3 SCTP Considerations: Added this section to discuss slow start.
> > 
> > 5.1  Link Initialization, several places: Renamed T2 to T4.
> > 
> > 7.1 SCTP Payload Protocol Identifier: Registered User Port Number is
> > 3565. Removed UDP.
> > 
> > 9.  References: Updated some of the references.
> > 
> > 10.  Author's Addresses: New address for Brian Bidulock.
> > 
> > (end)
> > 
> > 
> > -- 
> > * Tom George          972-519-3168           
> Tom.George@alcatel.com   *
> > * PB2-1323       Intranet:  
> http://www.ssd.usa.alcatel.com/~tgeorge/  *
> > * Alcatel USA, Inc.; 1000 Coit Rd, MS PB2-STPA; Plano TX 
> 75075-5813   *
> > * I actively disclaim speaking for Alcatel USA, Inc. I do 
> not purport *
> > * to advertise or represent its position,  operations, or 
> activities. *
> > _______________________________________________
> > Sigtran mailing list
> > Sigtran@ietf.org
> > https://www1.ietf.org/mailman/listinfo/sigtran
> _______________________________________________
> Sigtran mailing 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 sigtran-admin@ietf.org  Mon Sep  2 06:01:10 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA17951
	for <sigtran-archive@lists.ietf.org>; Mon, 2 Sep 2002 06:01:10 -0400 (EDT)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id g82A1lo25186;
	Mon, 2 Sep 2002 06:01:48 -0400
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id g82A0do25118
	for <sigtran@optimus.ietf.org>; Mon, 2 Sep 2002 06:00:39 -0400
Received: from mta0 ([61.144.161.2])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA17918
	for <sigtran@ietf.org>; Mon, 2 Sep 2002 05:59:03 -0400 (EDT)
Received: from l19463 (mta0 [172.17.1.62])
 by mta0.huawei.com (iPlanet Messaging Server 5.2 HotFix 0.8 (built Jul 12
 2002)) with ESMTPA id <0H1T000HE2D8ML@mta0.huawei.com> for sigtran@ietf.org;
 Mon, 02 Sep 2002 17:58:21 +0800 (CST)
Date: Sun, 01 Sep 2002 18:06:06 +0800
From: Ming Lin <mlin@huawei.com>
Subject: Re: [Sigtran] M2PA - Retrieval State
To: "May, Howard" <howard.may@intel.com>
Cc: sigtran@ietf.org
Message-id: <003301c2519f$308ea5a0$2a180b0a@huawei.com>
MIME-version: 1.0
X-MIMEOLE: Produced By Microsoft MimeOLE V5.50.4522.1200
X-Mailer: Microsoft Outlook Express 5.50.4522.1200
Content-type: text/plain; charset=iso-8859-1
Content-transfer-encoding: 7BIT
X-Priority: 3
X-MSMail-priority: Normal
References: <4A3192E7220D9241A1B774B074C6E1210171F226@swsmsx103.isw.intel.com>
Content-Transfer-Encoding: 7BIT
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7BIT

Dear May,

First, there is not retrieval state in mtp2, but it works well.
Second, in mtp2, the link comes into service only after MTP3 in both side START link,
so the link will not come into service during retrieving. 
After retrieving, MTP3 will issue start command.
Third, most of event happening in OOS can happen in RETRIEVAL state.

Here is the mail that I sent before and the reply:
============================================
Hi all,

I lose tracking of m2pa for a long time, so I am not clear that why we need 'RETRIEVAL ' state 
in M2PA Link State Transition Diagram.
I think 'RETRIEVAL ' is same as 'OOS' state, 
the evens happening in RETRIEVAL state also could happen in OOS state.

And there is not this state in MTP2(Q.703).

Could somebody tell me why we need RETRIEVAL state in M2PA?

Thanks&Regards
Lin Ming

===================================================
We agree with you: we couldn't understand the reason why a retrieval
state is needed.
Furthermore, we think that passing through this state can lead to a
delay in the subsequent alignment procedure. Following the flow chart on

pages 20-21 of the draft, one can see that if MTP3 sends a "Start"
rather
than triggering the retrieval procedure, M2PA moves to OOS state whit no

further actions. From there it needs again an MTP3 "Start" to begin the
alignment procedure.
We think that this behaviour should be avoided. When MTP3 sends a
"Start" the
alignment has to begin immediately.

Regards
Lisa & Aldo
========================================

Cheers
Ming Lin
----- Original Message ----- 
From: "May, Howard" <howard.may@intel.com>
To: "'Ming Lin'" <mlin@huawei.com>
Cc: <sigtran@ietf.org>
Sent: Monday, September 02, 2002 4:46 PM
Subject: [Sigtran] M2PA - Retrieval State


> Dear Ming
> 
> I think Issue 18, included in your email below, is resolved by Toms
> correction (well done with the ascii art work Tom!).  Ming could you post a
> summary of the other problems you see with the Retrieval state.
> 
> I currently believe the Retrieval state to be worthwhile.  As the link can
> only leave the Retrieval state after an event from MTP3 it guarantees that
> MTP3 will be able to perform and complete retrieval procedures before the
> link comes back into service.   If the Retrieval state is not used then I
> think we would be left in a situation where the link could come back into
> service before Retrieval had been completed.  There is then an issue of
> either allowing MTP3 and M2PA to fight over messages in the TX buffer (MTP3
> trying to Retrieve them and M2PA trying to transmit them - clearly
> undesirable) or perhaps when the link fails moving messages left in the Tx
> buffer to another buffer until MTP3 decides whether it wants them or not (My
> gut feeling is that this is where more problems lie rather than the use of
> the Retrieval state)
> 
> Cheers
> 
> > -----Original Message-----
> > From: Ming Lin [mailto:mlin@huawei.com]
> > Sent: 01 September 2002 05:08
> > To: Tom George
> > Cc: sigtran@ietf.org
> > Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-m2pa-06.txt
> > 
> > 
> > Dear Tom,
> > 
> > I think we should remove RETRIEVAL state, this state will 
> > result in many problem but no use,
> > for example the No. 18 issue.
> > 
> > "18. 08/09/02 Howard May "[Sigtran] M2PA - Minor corrections?" 
> >      In Figure 7, RETRIEVAL receives MTP Start. Go to AIP, not OOS.
> >  A: Yes.  [Done.]"
> > 
> > ----- Original Message ----- 
> > From: "Tom George" <Tom.George@alcatel.com>
> > Cc: <sigtran@ietf.org>
> > Sent: Friday, August 30, 2002 11:30 PM
> > Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-m2pa-06.txt
> > 
> > 
> > > M2PA Changes for version 6
> > > 08/28/2002
> > > 
> > > Here are the changes in draft-ietf-sigtran-m2pa-06.txt (M2PAv6).
> > > 
> > > If anyone wants to study the changes in detail, I have a 
> > "diff" file I
> > > can send.
> > > 
> > > This is not a complete list of changes. I did not include minor
> > > rewording.
> > > 
> > > 
> > > This is the list of comments received on M2PAv5 on the SIGTRAN list.
> > > 
> > > ==================================================================
> > > 
> > > Here is my list of comments. For the most part, they are arranged by
> > > the thread from the SIGTRAN list.
> > > 
> > > Everyone please make comments. I think the discussion would go more
> > > smoothly if you only talked about one item per e-mail.
> > > 
> > > 
> > > 1. Section 2.2. FSN/BSN should be 24 bits, not 23.
> > > 
> > > A: This was a typo. It will be 24 in M2PAv6.  [Done.]
> > > 
> > > 2. 05/06/02 Jeff Craig "[Sigtran] (M2PA) What actions 
> > should be taken
> > >    if LSA or User Data Received in PROVING?" There is no guidance
> > >    about anomalous events during alignment/proving.
> > > 
> > > A: I agree that some text should be added. I have not worked out the
> > > details and would welcome suggestions.
> > > 
> > >    My ideas:
> > > 
> > >    User Data received during PROVING should be discarded.
> > > 
> > >    Link Status Alignment received during PROVING should be ignored.
> > > 
> > >    [Done.]
> > > 
> > > 3. 05/10/02 Jeff Craig, "[Sigtran] (M2PA) Please standardize M2PA
> > >    protocol timer identifiers" and Jeff Craig 06/21/02 
> > "m2pa comment:
> > >    timers". Line up timer names in Q.703 and M2PA. Add the T7 timer.
> > > 
> > > A: This is what I got out of the discussion:
> > > 
> > >    Keep same names for T1, T3, T6. [Done.]
> > > 
> > >    Rename M2PA T2 to T4. [Done.]
> > > 
> > >    Don't use "timer" names for the rate parameters.  [Done.]
> > > 
> > >    T7 - see #11 below.  [Done.]
> > > 
> > > 4. 05/17/02 Jeff Craig, "[Sigtran] (M2PA) Please add a Network
> > >    Appearance parameter to the M2PA protocol". A suggestion 
> > that a NA
> > >    parameter be added to M2PA.
> > > 
> > > A: I prefer not to add this.  [Done.]
> > > 
> > > 5. 05/22/02 anssharma, "[Sigtran] M2PA Clarification". In 
> > level 2 flow
> > >    control, M2PA stops transmitting. What does it do with the
> > >    messages?
> > > 
> > > A: I think is an implementation matter.  [Done.]
> > > 
> > > 6. 05/24/02 Lisa Gorgone, "[Sigtran] M2PA:Questions on sequence
> > >    number". Some questions about bad values of FSNR and BSNR.
> > > 
> > > A: M2PAv5 (4.2.1) says to fail the link if one FSNR is out 
> > of order. I
> > > think that is OK. There is nothing about bad BSNR values. I would be
> > > inclined to fail the link in this case as well, since one end is
> > > acking messages that the other does not have in its transmit buffer.
> > > 
> > >    I have a question about the BSNR values on the list. 2 out of 3
> > >    consecutive messages with bad BSNRs?
> > > 
> > >    Eliminate acks on the Status stream. All acks go over the Data
> > >    stream. If there is no User Data message to be sent, 
> > send one with
> > >    no data in it. Get rid of Link Status In Service, since its only
> > >    purpose was to carry an ack. [Done.]
> > > 
> > > 7. 05/24/02 Brian Bidulock "[Sigtran] Re: M2PA Clarification". Does
> > >    M2PA need to withhold messages from SCTP during busy?
> > > 
> > > A: This is how Q.703 works, so we should copy it. Before there were
> > > M2PA sequence numbers, messages were withheld while the 
> > peer was Busy
> > > to prevent message loss in case the link later failed. Now that M2PA
> > > has its own sequence numbers, this is no longer a danger. 
> > So M2PA can
> > > continue transmitting while its peer is Busy. 
> > > 
> > >    I updated secion 4.1.5 Level 2 Flow Control.  [Done.]
> > > 
> > > 8. 06/18/02 amish.chana "[Sigtran] M2PA Comment: 
> > Multi-homing". Should
> > >    M2PA support multi-homing?
> > > 
> > > A: It is allowed. The only change needed is in 4.1.2 so that it says
> > > "the IP address (or IP addresses if multi-homing is used)".  [Done.]
> > > 
> > > 9. 06/19/02 Lisa Gorgone "Re: [Sigtran] M2PA Comment:Retrieval
> > >    State". Does M2PA need a RETRIEVAL state? Does RETRIEVAL state
> > >    cause harm?
> > > 
> > > A: I would like to have more discussion about this. 
> > > 
> > >    Since there has not been more discussion, I will leave this as is
> > >    for M2PA version 6.  [Done.]
> > > 
> > > 10. 06/21/02 Javier Pastor-Balbas "[Sigtran] M2PA Comment:
> > >     Multi-homing". Also 06/21/02 "[Sigtran] Multiplexed 
> > M2PA links on
> > >     an association".  Should M2PA allow multiple links per
> > >     association?
> > > 
> > > A: The disadvantage of this is that if the association 
> > dies, you lose
> > > all the links on the association. This is one of the main reasons I
> > > did not allow for multiple links in an association. 
> > > 
> > >    No change to the draft is needed.  [Done.]
> > > 
> > > 11. 06/21/02 Andrew Booth "[Sigtran] M2PA Comment: T7 Timer" and
> > >     several of the earlier threads. Should there be an equivalent of
> > >     MTP2 T7?
> > > 
> > > A: OK, you finally convinced me. 8-) I added the T7 timer in 4.2.1
> > > Sending/receiving messages and 4.1.5 Level 2 Flow Control.  [Done.]
> > > 
> > > 12. 06/25/02 Lyndon Ong "RE: [Sigtran] M2PA Queries - 
> > summary". Just a
> > >     summary of the discussion.
> > > 
> > > A: No answer needed. [Done.]
> > > 
> > > 13. 06/26/02 Karen King "[Sigtran] A couple of comments on M2PA
> > >     processor outage". Recommend adding text to 4.1.4 
> > Processor Outage
> > >     about flushing buffers as in Q.703.
> > > 
> > > A: Even though this is in Q.703, the action is taken by 
> > MTP3. Section
> > > 4.1.4 (next to last paragraph) does allow for the flush. Flush is
> > > described in 4.2.4. So I think this is taken care of (although there
> > > may be problems in 4.2.4 - that's another thread).  [Done.]
> > > 
> > > 14. 06/28/02 Ray Singh "[Sigtran] Comment on M2PA - Performance
> > >     Monitoring". Add back the monitoring of SCTP 
> > association that was
> > >     in draft version 3.
> > > 
> > > A: The reason I took this out is that I never had any solid ideas
> > > about what SCTP parameters should be monitored and what the 
> > acceptable
> > > values would be. If we could come up with something 
> > helpful, I am not
> > > opposed to adding the monitoring back in. But I don't see 
> > the point of
> > > adding it if the criteria will be some implementation-dependent
> > > parameters having implementation-dependent values. 8-( [Done.]
> > > 
> > > 15. 07/12/02 anssharma "[Sigtran] M2PA - Flush Buffers". Questions
> > >     about Flush.
> > > 
> > > A: These are good questions but I have not had time to think about
> > > them.
> > > 
> > > 
> > > 16. 07/19/02 Ray Singh "[Sigtran] Comment on M2PA - Slow 
> > Start". What
> > >     kind of warning is needed about slow start?
> > > 
> > > A: I proposed a general warning about slow start to replace specific
> > > warnings throughout the document.
> > > 
> > > alignment, changeover, path failure (multihoming), reduced 
> > volume. See
> > > Tolga, 07/26/02. [Done.]
> > > 
> > > 
> > > 17. 07/22/02 Tolga Asveren "[Sigtran] M2PA : LS messages on stream
> > >     0". Are there problems with sending Status and Data on different
> > >     streams? What mimics MTP2 best?
> > > 
> > > A: The problem with putting LS and Data on the same stream is that a
> > > LS message can get stuck behind a Data retransmission. This is
> > > contrary to the message priority of MTP2. (Brian Bidulock 08/13/2002
> > > 21:00 pointed out that this is a problem that MTP2 does not have.)
> > > 
> > > The problem with using SCTP unordered transmission for LS 
> > messages is
> > > that they can get out of order with each other. For 
> > example, a LS Proc
> > > Outage could get out of order with a LS Proc Outage Ended. 
> > > 
> > > MTP does not require that LSSUs and MTUs be associated with 
> > each other
> > > in any particular way. It has MTP2 put LSSUs ahead of MTUs in the
> > > transmit buffer. So it should not be required that M2PA 
> > transmit LS in
> > > order with Data.
> > > 
> > > Having M2PA put LS at the front of the M2PA transmit queue 
> > helps some,
> > > but it does not do anything about the messages that are in the SCTP
> > > send queue. To truly imitate MTP2 at this point, we would need a way
> > > to give a message to SCTP and tell it to put the message at 
> > the front
> > > of its send queue.
> > > 
> > > (Or, we could restrict SCTP to just 1 message in its send 
> > buffer. This
> > > would allow M2PA to prioritize like MTP2. But this would 
> > create severe
> > > problems of its own. No one seemed to like this option.)
> > > 
> > > Sending LS and Data on different streams allows for a partial
> > > imitation of the MTP2 transmission priority. If the receiving M2PA
> > > gives priority to reading incoming LS messages (on stream 
> > 0) over Data
> > > messages (on stream 1), then LS messages can be received sooner than
> > > if they waited behind the Data messages. Also, it is possible for
> > > SCTP, by some implementation-dependent method or some future SCTP
> > > extension, to give a higher priority to sending messages on 
> > stream 0.
> > > 
> > > My conclusion is to leave the draft alone. Having separate 
> > Status and
> > > Data streams does not exactly imitate MTP2, but it is as close as we
> > > can get using SCTP.  [Done.]
> > > 
> > > Some related matters:
> > > 
> > > I will add to 4.1.7 so that it says that M2PA transmits LS messages
> > > before Data messages. This is in addition to the attempt to 
> > prioritize
> > > SCTP stream 0. [That was already there. Done.]
> > > 
> > > The FSN should be ignored in the LS messages. FSN should be
> > > incremented only for Data messages. [That was already there. Done.]
> > > 
> > > I will add a warning that one has to watch out for a 
> > wraparound of the
> > > FSN/BSN numbers. This is unlikely with 24 bits, but should 
> > probably be
> > > mentioned. [Done.]
> > > 
> > > 
> > > 18. 08/09/02 Howard May "[Sigtran] M2PA - Minor corrections?" 
> > > 
> > >     In Figure 7, RETRIEVAL receives MTP Start. Go to AIP, not OOS.
> > > 
> > > A: Yes.  [Done.]
> > > 
> > > 
> > > ==================================================================
> > > 
> > > More detailed list of changes:
> > > 
> > > 2.2  M2PA Header: Changed BSN and FSN to 24 bits in diagram.
> > > 
> > > 2.3.2 Link Status: Removed In Service (10) from list of values.
> > > 
> > > 2.3.2.1 Link Status Proving, 1st para: Removed statement about SCTP
> > > slow start.
> > > 
> > > 3.  M2PA Link State Control, Figure 7: M2PA Link State Transition
> > > Diagram:
> > > 
> > >     T2 is renamed T4.
> > > 
> > >     Added T7 expiry.
> > > 
> > >     In RETRIEVAL, MTP3 Start goes to AIP instead of OOS.
> > > 
> > > 4.1.2 MTP and SCTP Entities, 2nd para: Added "(or IP addresses, if
> > > multi-homing is used)".
> > > 
> > > 4.1.3 Link Alignment, 1st para: Removed list item 3 about SCTP slow
> > > start.
> > > 
> > > 4.1.3 Link Alignment, several places: Renamed T2 to T4.
> > > 
> > > 4.1.3 Link Alignment, para "Anomalous...": Added this paragraph.
> > > 
> > > 4.1.5 Level 2 Flow Control, para "The peer M2PA should...": Changed
> > > "shall cease transmitting messages" to "should continue transmitting
> > > messages".
> > > 
> > > 4.1.5 Level 2 Flow Control, para "The peer M2PA shall not...": Added
> > > this paragraph about T7.
> > > 
> > > 4.2.1 Sending and receiving messages, para "When M2PA 
> > receives a User
> > > Data message...": Specified that this is a _User Data_ message.
> > > 
> > > 4.2.1 Sending and receiving messages: Removed paragraph stating that
> > > ack may be sent in User Data or Link Status.
> > > 
> > > 4.2.1 Sending and receiving messages, para "M2PA should follow the
> > > criterion...": Added this paragraph about 2-out-of-3 criterion.
> > > 
> > > 4.2.1 Sending and receiving messages, para "M2PA should ignore...":
> > > Added this paragraph.
> > > 
> > > 4.2.1 Sending and receiving messages, para "Note: In all
> > > calculations...": Added this paragraph.
> > > 
> > > 4.2.1 Sending and receiving messages, para "If there is no 
> > other User
> > > Data...": Changed this to say M2PA sends User Data with no payload
> > > instead of LS In Service.
> > > 
> > > 4.2.1 Sending and receiving messages, last 2 paragraphs: Added these
> > > paragraphs about T7.
> > > 
> > > 4.3 SCTP Considerations: Added this section to discuss slow start.
> > > 
> > > 5.1  Link Initialization, several places: Renamed T2 to T4.
> > > 
> > > 7.1 SCTP Payload Protocol Identifier: Registered User Port Number is
> > > 3565. Removed UDP.
> > > 
> > > 9.  References: Updated some of the references.
> > > 
> > > 10.  Author's Addresses: New address for Brian Bidulock.
> > > 
> > > (end)
> > > 
> > > 
> > > -- 
> > > * Tom George          972-519-3168           
> > Tom.George@alcatel.com   *
> > > * PB2-1323       Intranet:  
> > http://www.ssd.usa.alcatel.com/~tgeorge/  *
> > > * Alcatel USA, Inc.; 1000 Coit Rd, MS PB2-STPA; Plano TX 
> > 75075-5813   *
> > > * I actively disclaim speaking for Alcatel USA, Inc. I do 
> > not purport *
> > > * to advertise or represent its position,  operations, or 
> > activities. *
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > _______________________________________________
> > Sigtran mailing 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 sigtran-admin@ietf.org  Mon Sep  2 06:03:19 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA18027
	for <sigtran-archive@lists.ietf.org>; Mon, 2 Sep 2002 06:03:19 -0400 (EDT)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id g82A45o25321;
	Mon, 2 Sep 2002 06:04:05 -0400
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id g82A3jo25285
	for <sigtran@optimus.ietf.org>; Mon, 2 Sep 2002 06:03:45 -0400
Received: from gw.openss7.com (IDENT:6tDVzK0/actx2G+i66oeUw+LQ1egapwn@gw.openss7.com [142.179.199.224])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA18002
	for <sigtran@ietf.org>; Mon, 2 Sep 2002 06:02:10 -0400 (EDT)
Received: from ns.pigworks.openss7.net (ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id g82A3hN01932;
	Mon, 2 Sep 2002 04:03:43 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id g82A3g210023;
	Mon, 2 Sep 2002 04:03:42 -0600
Date: Mon, 2 Sep 2002 04:03:42 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "May, Howard" <howard.may@intel.com>
Cc: "'Ming Lin'" <mlin@huawei.com>, sigtran@ietf.org
Subject: Re: [Sigtran] M2PA - Retrieval State
Message-ID: <20020902040342.A8050@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "May, Howard" <howard.may@intel.com>,
	'Ming Lin' <mlin@huawei.com>, sigtran@ietf.org
References: <4A3192E7220D9241A1B774B074C6E1210171F226@swsmsx103.isw.intel.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: <4A3192E7220D9241A1B774B074C6E1210171F226@swsmsx103.isw.intel.com>; from howard.may@intel.com on Mon, Sep 02, 2002 at 09:46:20AM +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>

Howard,

I agree with Ming.  There is no difference between RETRIEVAL state and
OOS state.  Q.703 gets (and even MTP3b) along quite nicely without an
LSC RETRIEVAL state: both only use an Out of Service state.  For consistency
with MTP2/MTP3b the RETRIEVAL state needs to be removed.

M2PA/Figure 7 should be made to be consistent with Figure 2/Q.703. Figure
2/Q.703 has no RETRIEVAL state.  The only states listed in Q.703 overview
are:

        Figure 2/Q.703               M2PA/Figure 7
        ------------------           ------------------
        Power Off                    IDLE
        Out of Service               OOS
        Initial Alignment            AIP, PROVING
        Aligned Ready                ALIGNED READY
        In Service                   INS
        Processor Outage             --
        Aligned Not Ready            --
        --                           RETRIEVAL

It would be good if M2PA could at least conform to the Q.703 state machine
overview diagram (even if it falls short of the detailed SDLs).  Several
things should be done to M2PA/Figure 7 to accomplish this:  RETRIEVAL state
should be removed (MTP3 does not distinguish it from OOS state).  PROVING
state should be removed--well collapsed with AIP--(MTP3 does not distinguish
Proving from Initial Alignment), and two states need to be added to the
diagram which MTP3 is concerned about: PROCESSOR OUTAGE and ALIGNED NOT READY.
PROCESSOR OUTAGE state results from local processor outage condition
experienced in the INS state.  ALIGNED NOT READY state results from local
processor outage condition experienced in the ALIGNED READY state.  Thus, the
diagram should be updated as follows:

                                                                        
                    +-----------+                                       
                    |  POWER    |                                       
                    |   OFF     |                                       
                    +-----------+                                       
                          |                                             
                          | Power On                                    
                          |                                             
        Flush Buffers     |  +-------------------------+                
        OR Retreival      |  |                         |                
               +-------+  |  |                         |                
               |       |  V  V                         |                
               |    +-----------+                      |                
               +--->|  OUT OF   |                      |                
  +---------------->|  SERVICE  |<--+                  |                
  |                 +-----------+   | Link Configured  |                
  |                       |   |     | (Associate)      |                
  |                       |   +-----+                  |                
  |                       |                            |                
  |                       | MTP3 Start                 |                
  |                       V                            |                
  |                 +-----------+                      |                
  |                 |  INITIAL  |                      |                
  +<----------------| ALIGNMENT |--------------------->+                
  |      MTP3 Stop  +-----------+     SCTP Comm Error  |                
  |   OR T1 Expiry        |  |     OR SCTP Comm Lost   |                
  |   OR T4 Expiry        |  |                         |                
  |   OR Receive LS OOS   |  +-------------------------|---------+      
  |                       | Send and                   |         |      
  |                       |      Receive LS Alignment  |         |      
  |                       V                            |         |      
  |                 +-----------+        LPO/LPR       |   +-----------+
  |                 |  ALIGNED  |<---------------------|-->|  ALIGNED  |
  +<----------------|   READY   |--------------------->+   | NOT READY |
  |      MTP3 Stop  +-----------+     SCTP Comm Error      +-----------+
  |   OR T3 Expiry        |        OR SCTP Comm Lost             |      
  |   OR Receive LS OOS   |                                      |      
  |                       | Receive LS Ready                     |      
  |                       | OR Receive User Data                 |      
  |                       |                                      |      
  |                       V                                      V      
  |                 +-----------+        LPO/LPR           +-----------+
  |                 |    IN     |<------------------------>| PROCESSOR |
  |                 |  SERVICE  |                          |   OUTAGE  |
  |                 +-----------+                          +-----------+
  |                       |                                      |      
  |                       |    MTP3 Stop                         |      
  |                       | OR Receive LS OOS                    |      
  |                       | OR SCTP Comm Error                   |      
  |                       | OR SCTP Comm Lost                    |      
  |                       | OR T6 Expiry                         |      
  |                       | OR T7 Expiry                         |      
  |                       | OR M2PA Link Failure                 |      
  |                       |                                      |      
  +<----------------------+<-------------------------------------+      
                                                                        
                                                                        
             Figure 7: M2PA Link State Transition Diagram               
                                                                        

Text should be updated to be consistent with this diagram.

--brian

May, Howard wrote:                                                                    (Mon, 02 Sep 2002 09:46:20)
> Dear Ming
> 
> I think Issue 18, included in your email below, is resolved by Toms
> correction (well done with the ascii art work Tom!).  Ming could you post a
> summary of the other problems you see with the Retrieval state.
> 
> I currently believe the Retrieval state to be worthwhile.  As the link can
> only leave the Retrieval state after an event from MTP3 it guarantees that
> MTP3 will be able to perform and complete retrieval procedures before the
> link comes back into service.   If the Retrieval state is not used then I
> think we would be left in a situation where the link could come back into
> service before Retrieval had been completed.  There is then an issue of
> either allowing MTP3 and M2PA to fight over messages in the TX buffer (MTP3
> trying to Retrieve them and M2PA trying to transmit them - clearly
> undesirable) or perhaps when the link fails moving messages left in the Tx
> buffer to another buffer until MTP3 decides whether it wants them or not (My
> gut feeling is that this is where more problems lie rather than the use of
> the Retrieval state)
> 
> Cheers
> 
> > -----Original Message-----
> > From: Ming Lin [mailto:mlin@huawei.com]
> > Sent: 01 September 2002 05:08
> > To: Tom George
> > Cc: sigtran@ietf.org
> > Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-m2pa-06.txt
> > 
> > 
> > Dear Tom,
> > 
> > I think we should remove RETRIEVAL state, this state will 
> > result in many problem but no use,
> > for example the No. 18 issue.
> > 
> > "18. 08/09/02 Howard May "[Sigtran] M2PA - Minor corrections?" 
> >      In Figure 7, RETRIEVAL receives MTP Start. Go to AIP, not OOS.
> >  A: Yes.  [Done.]"
> > 
> > ----- Original Message ----- 
> > From: "Tom George" <Tom.George@alcatel.com>
> > Cc: <sigtran@ietf.org>
> > Sent: Friday, August 30, 2002 11:30 PM
> > Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-m2pa-06.txt
> > 
> > 
> > > M2PA Changes for version 6
> > > 08/28/2002
> > > 
> > > Here are the changes in draft-ietf-sigtran-m2pa-06.txt (M2PAv6).
> > > 
> > > If anyone wants to study the changes in detail, I have a 
> > "diff" file I
> > > can send.
> > > 
> > > This is not a complete list of changes. I did not include minor
> > > rewording.
> > > 
> > > 
> > > This is the list of comments received on M2PAv5 on the SIGTRAN list.
> > > 
> > > ==================================================================
> > > 
> > > Here is my list of comments. For the most part, they are arranged by
> > > the thread from the SIGTRAN list.
> > > 
> > > Everyone please make comments. I think the discussion would go more
> > > smoothly if you only talked about one item per e-mail.
> > > 
> > > 
> > > 1. Section 2.2. FSN/BSN should be 24 bits, not 23.
> > > 
> > > A: This was a typo. It will be 24 in M2PAv6.  [Done.]
> > > 
> > > 2. 05/06/02 Jeff Craig "[Sigtran] (M2PA) What actions 
> > should be taken
> > >    if LSA or User Data Received in PROVING?" There is no guidance
> > >    about anomalous events during alignment/proving.
> > > 
> > > A: I agree that some text should be added. I have not worked out the
> > > details and would welcome suggestions.
> > > 
> > >    My ideas:
> > > 
> > >    User Data received during PROVING should be discarded.
> > > 
> > >    Link Status Alignment received during PROVING should be ignored.
> > > 
> > >    [Done.]
> > > 
> > > 3. 05/10/02 Jeff Craig, "[Sigtran] (M2PA) Please standardize M2PA
> > >    protocol timer identifiers" and Jeff Craig 06/21/02 
> > "m2pa comment:
> > >    timers". Line up timer names in Q.703 and M2PA. Add the T7 timer.
> > > 
> > > A: This is what I got out of the discussion:
> > > 
> > >    Keep same names for T1, T3, T6. [Done.]
> > > 
> > >    Rename M2PA T2 to T4. [Done.]
> > > 
> > >    Don't use "timer" names for the rate parameters.  [Done.]
> > > 
> > >    T7 - see #11 below.  [Done.]
> > > 
> > > 4. 05/17/02 Jeff Craig, "[Sigtran] (M2PA) Please add a Network
> > >    Appearance parameter to the M2PA protocol". A suggestion 
> > that a NA
> > >    parameter be added to M2PA.
> > > 
> > > A: I prefer not to add this.  [Done.]
> > > 
> > > 5. 05/22/02 anssharma, "[Sigtran] M2PA Clarification". In 
> > level 2 flow
> > >    control, M2PA stops transmitting. What does it do with the
> > >    messages?
> > > 
> > > A: I think is an implementation matter.  [Done.]
> > > 
> > > 6. 05/24/02 Lisa Gorgone, "[Sigtran] M2PA:Questions on sequence
> > >    number". Some questions about bad values of FSNR and BSNR.
> > > 
> > > A: M2PAv5 (4.2.1) says to fail the link if one FSNR is out 
> > of order. I
> > > think that is OK. There is nothing about bad BSNR values. I would be
> > > inclined to fail the link in this case as well, since one end is
> > > acking messages that the other does not have in its transmit buffer.
> > > 
> > >    I have a question about the BSNR values on the list. 2 out of 3
> > >    consecutive messages with bad BSNRs?
> > > 
> > >    Eliminate acks on the Status stream. All acks go over the Data
> > >    stream. If there is no User Data message to be sent, 
> > send one with
> > >    no data in it. Get rid of Link Status In Service, since its only
> > >    purpose was to carry an ack. [Done.]
> > > 
> > > 7. 05/24/02 Brian Bidulock "[Sigtran] Re: M2PA Clarification". Does
> > >    M2PA need to withhold messages from SCTP during busy?
> > > 
> > > A: This is how Q.703 works, so we should copy it. Before there were
> > > M2PA sequence numbers, messages were withheld while the 
> > peer was Busy
> > > to prevent message loss in case the link later failed. Now that M2PA
> > > has its own sequence numbers, this is no longer a danger. 
> > So M2PA can
> > > continue transmitting while its peer is Busy. 
> > > 
> > >    I updated secion 4.1.5 Level 2 Flow Control.  [Done.]
> > > 
> > > 8. 06/18/02 amish.chana "[Sigtran] M2PA Comment: 
> > Multi-homing". Should
> > >    M2PA support multi-homing?
> > > 
> > > A: It is allowed. The only change needed is in 4.1.2 so that it says
> > > "the IP address (or IP addresses if multi-homing is used)".  [Done.]
> > > 
> > > 9. 06/19/02 Lisa Gorgone "Re: [Sigtran] M2PA Comment:Retrieval
> > >    State". Does M2PA need a RETRIEVAL state? Does RETRIEVAL state
> > >    cause harm?
> > > 
> > > A: I would like to have more discussion about this. 
> > > 
> > >    Since there has not been more discussion, I will leave this as is
> > >    for M2PA version 6.  [Done.]
> > > 
> > > 10. 06/21/02 Javier Pastor-Balbas "[Sigtran] M2PA Comment:
> > >     Multi-homing". Also 06/21/02 "[Sigtran] Multiplexed 
> > M2PA links on
> > >     an association".  Should M2PA allow multiple links per
> > >     association?
> > > 
> > > A: The disadvantage of this is that if the association 
> > dies, you lose
> > > all the links on the association. This is one of the main reasons I
> > > did not allow for multiple links in an association. 
> > > 
> > >    No change to the draft is needed.  [Done.]
> > > 
> > > 11. 06/21/02 Andrew Booth "[Sigtran] M2PA Comment: T7 Timer" and
> > >     several of the earlier threads. Should there be an equivalent of
> > >     MTP2 T7?
> > > 
> > > A: OK, you finally convinced me. 8-) I added the T7 timer in 4.2.1
> > > Sending/receiving messages and 4.1.5 Level 2 Flow Control.  [Done.]
> > > 
> > > 12. 06/25/02 Lyndon Ong "RE: [Sigtran] M2PA Queries - 
> > summary". Just a
> > >     summary of the discussion.
> > > 
> > > A: No answer needed. [Done.]
> > > 
> > > 13. 06/26/02 Karen King "[Sigtran] A couple of comments on M2PA
> > >     processor outage". Recommend adding text to 4.1.4 
> > Processor Outage
> > >     about flushing buffers as in Q.703.
> > > 
> > > A: Even though this is in Q.703, the action is taken by 
> > MTP3. Section
> > > 4.1.4 (next to last paragraph) does allow for the flush. Flush is
> > > described in 4.2.4. So I think this is taken care of (although there
> > > may be problems in 4.2.4 - that's another thread).  [Done.]
> > > 
> > > 14. 06/28/02 Ray Singh "[Sigtran] Comment on M2PA - Performance
> > >     Monitoring". Add back the monitoring of SCTP 
> > association that was
> > >     in draft version 3.
> > > 
> > > A: The reason I took this out is that I never had any solid ideas
> > > about what SCTP parameters should be monitored and what the 
> > acceptable
> > > values would be. If we could come up with something 
> > helpful, I am not
> > > opposed to adding the monitoring back in. But I don't see 
> > the point of
> > > adding it if the criteria will be some implementation-dependent
> > > parameters having implementation-dependent values. 8-( [Done.]
> > > 
> > > 15. 07/12/02 anssharma "[Sigtran] M2PA - Flush Buffers". Questions
> > >     about Flush.
> > > 
> > > A: These are good questions but I have not had time to think about
> > > them.
> > > 
> > > 
> > > 16. 07/19/02 Ray Singh "[Sigtran] Comment on M2PA - Slow 
> > Start". What
> > >     kind of warning is needed about slow start?
> > > 
> > > A: I proposed a general warning about slow start to replace specific
> > > warnings throughout the document.
> > > 
> > > alignment, changeover, path failure (multihoming), reduced 
> > volume. See
> > > Tolga, 07/26/02. [Done.]
> > > 
> > > 
> > > 17. 07/22/02 Tolga Asveren "[Sigtran] M2PA : LS messages on stream
> > >     0". Are there problems with sending Status and Data on different
> > >     streams? What mimics MTP2 best?
> > > 
> > > A: The problem with putting LS and Data on the same stream is that a
> > > LS message can get stuck behind a Data retransmission. This is
> > > contrary to the message priority of MTP2. (Brian Bidulock 08/13/2002
> > > 21:00 pointed out that this is a problem that MTP2 does not have.)
> > > 
> > > The problem with using SCTP unordered transmission for LS 
> > messages is
> > > that they can get out of order with each other. For 
> > example, a LS Proc
> > > Outage could get out of order with a LS Proc Outage Ended. 
> > > 
> > > MTP does not require that LSSUs and MTUs be associated with 
> > each other
> > > in any particular way. It has MTP2 put LSSUs ahead of MTUs in the
> > > transmit buffer. So it should not be required that M2PA 
> > transmit LS in
> > > order with Data.
> > > 
> > > Having M2PA put LS at the front of the M2PA transmit queue 
> > helps some,
> > > but it does not do anything about the messages that are in the SCTP
> > > send queue. To truly imitate MTP2 at this point, we would need a way
> > > to give a message to SCTP and tell it to put the message at 
> > the front
> > > of its send queue.
> > > 
> > > (Or, we could restrict SCTP to just 1 message in its send 
> > buffer. This
> > > would allow M2PA to prioritize like MTP2. But this would 
> > create severe
> > > problems of its own. No one seemed to like this option.)
> > > 
> > > Sending LS and Data on different streams allows for a partial
> > > imitation of the MTP2 transmission priority. If the receiving M2PA
> > > gives priority to reading incoming LS messages (on stream 
> > 0) over Data
> > > messages (on stream 1), then LS messages can be received sooner than
> > > if they waited behind the Data messages. Also, it is possible for
> > > SCTP, by some implementation-dependent method or some future SCTP
> > > extension, to give a higher priority to sending messages on 
> > stream 0.
> > > 
> > > My conclusion is to leave the draft alone. Having separate 
> > Status and
> > > Data streams does not exactly imitate MTP2, but it is as close as we
> > > can get using SCTP.  [Done.]
> > > 
> > > Some related matters:
> > > 
> > > I will add to 4.1.7 so that it says that M2PA transmits LS messages
> > > before Data messages. This is in addition to the attempt to 
> > prioritize
> > > SCTP stream 0. [That was already there. Done.]
> > > 
> > > The FSN should be ignored in the LS messages. FSN should be
> > > incremented only for Data messages. [That was already there. Done.]
> > > 
> > > I will add a warning that one has to watch out for a 
> > wraparound of the
> > > FSN/BSN numbers. This is unlikely with 24 bits, but should 
> > probably be
> > > mentioned. [Done.]
> > > 
> > > 
> > > 18. 08/09/02 Howard May "[Sigtran] M2PA - Minor corrections?" 
> > > 
> > >     In Figure 7, RETRIEVAL receives MTP Start. Go to AIP, not OOS.
> > > 
> > > A: Yes.  [Done.]
> > > 
> > > 
> > > ==================================================================
> > > 
> > > More detailed list of changes:
> > > 
> > > 2.2  M2PA Header: Changed BSN and FSN to 24 bits in diagram.
> > > 
> > > 2.3.2 Link Status: Removed In Service (10) from list of values.
> > > 
> > > 2.3.2.1 Link Status Proving, 1st para: Removed statement about SCTP
> > > slow start.
> > > 
> > > 3.  M2PA Link State Control, Figure 7: M2PA Link State Transition
> > > Diagram:
> > > 
> > >     T2 is renamed T4.
> > > 
> > >     Added T7 expiry.
> > > 
> > >     In RETRIEVAL, MTP3 Start goes to AIP instead of OOS.
> > > 
> > > 4.1.2 MTP and SCTP Entities, 2nd para: Added "(or IP addresses, if
> > > multi-homing is used)".
> > > 
> > > 4.1.3 Link Alignment, 1st para: Removed list item 3 about SCTP slow
> > > start.
> > > 
> > > 4.1.3 Link Alignment, several places: Renamed T2 to T4.
> > > 
> > > 4.1.3 Link Alignment, para "Anomalous...": Added this paragraph.
> > > 
> > > 4.1.5 Level 2 Flow Control, para "The peer M2PA should...": Changed
> > > "shall cease transmitting messages" to "should continue transmitting
> > > messages".
> > > 
> > > 4.1.5 Level 2 Flow Control, para "The peer M2PA shall not...": Added
> > > this paragraph about T7.
> > > 
> > > 4.2.1 Sending and receiving messages, para "When M2PA 
> > receives a User
> > > Data message...": Specified that this is a _User Data_ message.
> > > 
> > > 4.2.1 Sending and receiving messages: Removed paragraph stating that
> > > ack may be sent in User Data or Link Status.
> > > 
> > > 4.2.1 Sending and receiving messages, para "M2PA should follow the
> > > criterion...": Added this paragraph about 2-out-of-3 criterion.
> > > 
> > > 4.2.1 Sending and receiving messages, para "M2PA should ignore...":
> > > Added this paragraph.
> > > 
> > > 4.2.1 Sending and receiving messages, para "Note: In all
> > > calculations...": Added this paragraph.
> > > 
> > > 4.2.1 Sending and receiving messages, para "If there is no 
> > other User
> > > Data...": Changed this to say M2PA sends User Data with no payload
> > > instead of LS In Service.
> > > 
> > > 4.2.1 Sending and receiving messages, last 2 paragraphs: Added these
> > > paragraphs about T7.
> > > 
> > > 4.3 SCTP Considerations: Added this section to discuss slow start.
> > > 
> > > 5.1  Link Initialization, several places: Renamed T2 to T4.
> > > 
> > > 7.1 SCTP Payload Protocol Identifier: Registered User Port Number is
> > > 3565. Removed UDP.
> > > 
> > > 9.  References: Updated some of the references.
> > > 
> > > 10.  Author's Addresses: New address for Brian Bidulock.
> > > 
> > > (end)
> > > 
> > > 
> > > -- 
> > > * Tom George          972-519-3168           
> > Tom.George@alcatel.com   *
> > > * PB2-1323       Intranet:  
> > http://www.ssd.usa.alcatel.com/~tgeorge/  *
> > > * Alcatel USA, Inc.; 1000 Coit Rd, MS PB2-STPA; Plano TX 
> > 75075-5813   *
> > > * I actively disclaim speaking for Alcatel USA, Inc. I do 
> > not purport *
> > > * to advertise or represent its position,  operations, or 
> > activities. *
> > > _______________________________________________
> > > Sigtran mailing list
> > > Sigtran@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/sigtran
> > _______________________________________________
> > Sigtran mailing 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 sigtran-admin@ietf.org  Mon Sep  2 08:48:22 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA20828
	for <sigtran-archive@lists.ietf.org>; Mon, 2 Sep 2002 08:48:22 -0400 (EDT)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id g82Cmlo02325;
	Mon, 2 Sep 2002 08:48:49 -0400
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id g82Clfo02264
	for <sigtran@optimus.ietf.org>; Mon, 2 Sep 2002 08:47:41 -0400
Received: from odin.isw.intel.com (swfdns01.isw.intel.com [192.55.37.143])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA20769
	for <sigtran@ietf.org>; Mon, 2 Sep 2002 08:46:03 -0400 (EDT)
Received: from swsmsxvs01.isw.intel.com (swsmsxvs01.isw.intel.com [172.28.130.22])
	by odin.isw.intel.com (8.11.6/8.11.6/d: solo.mc,v 1.43 2002/08/30 20:06:11 dmccart Exp $) with SMTP id g82Cl7u18620
	for <sigtran@ietf.org>; Mon, 2 Sep 2002 12:47:07 GMT
Received: from swsmsx17.isw.intel.com ([172.28.130.21])
 by swsmsxvs01.isw.intel.com (NAVGW 2.5.2.11) with SMTP id M2002090213454205813
 ; Mon, 02 Sep 2002 13:45:42 +0100
Received: by swsmsx17.isw.intel.com with Internet Mail Service (5.5.2653.19)
	id <QT1M3MGF>; Mon, 2 Sep 2002 13:51:21 +0100
Message-ID: <4A3192E7220D9241A1B774B074C6E1210171F231@swsmsx103.isw.intel.com>
From: "May, Howard" <howard.may@intel.com>
To: "'bidulock@openss7.org'" <bidulock@openss7.org>
Cc: sigtran@ietf.org
Subject: RE: [Sigtran] M2PA - Retrieval State
Date: Mon, 2 Sep 2002 13:46:33 +0100 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: sigtran-admin@ietf.org
Errors-To: sigtran-admin@ietf.org
X-BeenThere: sigtran@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=unsubscribe>
List-Id: Signaling Transport <sigtran.ietf.org>
List-Post: <mailto:sigtran@ietf.org>
List-Help: <mailto:sigtran-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/sigtran>,
	<mailto:sigtran-request@ietf.org?subject=subscribe>

Brian, Ming,

I concur with your points, I do not believe that the Retrieval State adds
anything and now would support it's removal.  

With regards aligning more closely with Figure 2/Q.703.  Assuming you are
correct when you say 'MTP3 is concerned about: PROCESSOR OUTAGE and ALIGNED
NOT READY' then shouldn't their addition be mandatory?

Howard

> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> Sent: 02 September 2002 11:04
> To: May, Howard
> Cc: 'Ming Lin'; sigtran@ietf.org
> Subject: Re: [Sigtran] M2PA - Retrieval State
> 
> 
> Howard,
> 
> I agree with Ming.  There is no difference between RETRIEVAL state and
> OOS state.  Q.703 gets (and even MTP3b) along quite nicely without an
> LSC RETRIEVAL state: both only use an Out of Service state.  
> For consistency
> with MTP2/MTP3b the RETRIEVAL state needs to be removed.
> 
> M2PA/Figure 7 should be made to be consistent with Figure 
> 2/Q.703. Figure
> 2/Q.703 has no RETRIEVAL state.  The only states listed in 
> Q.703 overview
> are:
> 
>         Figure 2/Q.703               M2PA/Figure 7
>         ------------------           ------------------
>         Power Off                    IDLE
>         Out of Service               OOS
>         Initial Alignment            AIP, PROVING
>         Aligned Ready                ALIGNED READY
>         In Service                   INS
>         Processor Outage             --
>         Aligned Not Ready            --
>         --                           RETRIEVAL
> 
> It would be good if M2PA could at least conform to the Q.703 
> state machine
> overview diagram (even if it falls short of the detailed 
> SDLs).  Several
> things should be done to M2PA/Figure 7 to accomplish this:  
> RETRIEVAL state
> should be removed (MTP3 does not distinguish it from OOS 
> state).  PROVING
> state should be removed--well collapsed with AIP--(MTP3 does 
> not distinguish
> Proving from Initial Alignment), and two states need to be 
> added to the
> diagram which MTP3 is concerned about: PROCESSOR OUTAGE and 
> ALIGNED NOT READY.
> PROCESSOR OUTAGE state results from local processor outage condition
> experienced in the INS state.  ALIGNED NOT READY state 
> results from local
> processor outage condition experienced in the ALIGNED READY 
> state.  Thus, the
> diagram should be updated as follows:
> 
>                                                               
>           
>                     +-----------+                             
>           
>                     |  POWER    |                             
>           
>                     |   OFF     |                             
>           
>                     +-----------+                             
>           
>                           |                                   
>           
>                           | Power On                          
>           
>                           |                                   
>           
>         Flush Buffers     |  +-------------------------+      
>           
>         OR Retreival      |  |                         |      
>           
>                +-------+  |  |                         |      
>           
>                |       |  V  V                         |      
>           
>                |    +-----------+                      |      
>           
>                +--->|  OUT OF   |                      |      
>           
>   +---------------->|  SERVICE  |<--+                  |      
>           
>   |                 +-----------+   | Link Configured  |      
>           
>   |                       |   |     | (Associate)      |      
>           
>   |                       |   +-----+                  |      
>           
>   |                       |                            |      
>           
>   |                       | MTP3 Start                 |      
>           
>   |                       V                            |      
>           
>   |                 +-----------+                      |      
>           
>   |                 |  INITIAL  |                      |      
>           
>   +<----------------| ALIGNMENT |--------------------->+      
>           
>   |      MTP3 Stop  +-----------+     SCTP Comm Error  |      
>           
>   |   OR T1 Expiry        |  |     OR SCTP Comm Lost   |      
>           
>   |   OR T4 Expiry        |  |                         |      
>           
>   |   OR Receive LS OOS   |  
> +-------------------------|---------+      
>   |                       | Send and                   |      
>    |      
>   |                       |      Receive LS Alignment  |      
>    |      
>   |                       V                            |      
>    |      
>   |                 +-----------+        LPO/LPR       |   
> +-----------+
>   |                 |  ALIGNED  |<---------------------|-->|  
> ALIGNED  |
>   +<----------------|   READY   |--------------------->+   | 
> NOT READY |
>   |      MTP3 Stop  +-----------+     SCTP Comm Error      
> +-----------+
>   |   OR T3 Expiry        |        OR SCTP Comm Lost          
>    |      
>   |   OR Receive LS OOS   |                                   
>    |      
>   |                       | Receive LS Ready                  
>    |      
>   |                       | OR Receive User Data              
>    |      
>   |                       |                                   
>    |      
>   |                       V                                   
>    V      
>   |                 +-----------+        LPO/LPR           
> +-----------+
>   |                 |    IN     |<------------------------>| 
> PROCESSOR |
>   |                 |  SERVICE  |                          |  
>  OUTAGE  |
>   |                 +-----------+                          
> +-----------+
>   |                       |                                   
>    |      
>   |                       |    MTP3 Stop                      
>    |      
>   |                       | OR Receive LS OOS                 
>    |      
>   |                       | OR SCTP Comm Error                
>    |      
>   |                       | OR SCTP Comm Lost                 
>    |      
>   |                       | OR T6 Expiry                      
>    |      
>   |                       | OR T7 Expiry                      
>    |      
>   |                       | OR M2PA Link Failure              
>    |      
>   |                       |                                   
>    |      
>   
> +<----------------------+<-------------------------------------+      
>                                                               
>           
>                                                               
>           
>              Figure 7: M2PA Link State Transition Diagram     
>           
>                                                               
>           
> 
> Text should be updated to be consistent with this diagram.
> 
> --brian
> 
> May, Howard wrote:                                            
>                         (Mon, 02 Sep 2002 09:46:20)
> > Dear Ming
> > 
> > I think Issue 18, included in your email below, is resolved by Toms
> > correction (well done with the ascii art work Tom!).  Ming 
> could you post a
> > summary of the other problems you see with the Retrieval state.
> > 
> > I currently believe the Retrieval state to be worthwhile.  
> As the link can
> > only leave the Retrieval state after an event from MTP3 it 
> guarantees that
> > MTP3 will be able to perform and complete retrieval 
> procedures before the
> > link comes back into service.   If the Retrieval state is 
> not used then I
> > think we would be left in a situation where the link could 
> come back into
> > service before Retrieval had been completed.  There is then 
> an issue of
> > either allowing MTP3 and M2PA to fight over messages in the 
> TX buffer (MTP3
> > trying to Retrieve them and M2PA trying to transmit them - clearly
> > undesirable) or perhaps when the link fails moving messages 
> left in the Tx
> > buffer to another buffer until MTP3 decides whether it 
> wants them or not (My
> > gut feeling is that this is where more problems lie rather 
> than the use of
> > the Retrieval state)
> > 
> > Cheers
> > 
> > > -----Original Message-----
> > > From: Ming Lin [mailto:mlin@huawei.com]
> > > Sent: 01 September 2002 05:08
> > > To: Tom George
> > > Cc: sigtran@ietf.org
> > > Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-m2pa-06.txt
> > > 
> > > 
> > > Dear Tom,
> > > 
> > > I think we should remove RETRIEVAL state, this state will 
> > > result in many problem but no use,
> > > for example the No. 18 issue.
> > > 
> > > "18. 08/09/02 Howard May "[Sigtran] M2PA - Minor corrections?" 
> > >      In Figure 7, RETRIEVAL receives MTP Start. Go to 
> AIP, not OOS.
> > >  A: Yes.  [Done.]"
> > > 
> > > ----- Original Message ----- 
> > > From: "Tom George" <Tom.George@alcatel.com>
> > > Cc: <sigtran@ietf.org>
> > > Sent: Friday, August 30, 2002 11:30 PM
> > > Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-m2pa-06.txt
> > > 
> > > 
> > > > M2PA Changes for version 6
> > > > 08/28/2002
> > > > 
> > > > Here are the changes in draft-ietf-sigtran-m2pa-06.txt (M2PAv6).
> > > > 
> > > > If anyone wants to study the changes in detail, I have a 
> > > "diff" file I
> > > > can send.
> > > > 
> > > > This is not a complete list of changes. I did not include minor
> > > > rewording.
> > > > 
> > > > 
> > > > This is the list of comments received on M2PAv5 on the 
> SIGTRAN list.
> > > > 
> > > > 
> ==================================================================
> > > > 
> > > > Here is my list of comments. For the most part, they 
> are arranged by
> > > > the thread from the SIGTRAN list.
> > > > 
> > > > Everyone please make comments. I think the discussion 
> would go more
> > > > smoothly if you only talked about one item per e-mail.
> > > > 
> > > > 
> > > > 1. Section 2.2. FSN/BSN should be 24 bits, not 23.
> > > > 
> > > > A: This was a typo. It will be 24 in M2PAv6.  [Done.]
> > > > 
> > > > 2. 05/06/02 Jeff Craig "[Sigtran] (M2PA) What actions 
> > > should be taken
> > > >    if LSA or User Data Received in PROVING?" There is 
> no guidance
> > > >    about anomalous events during alignment/proving.
> > > > 
> > > > A: I agree that some text should be added. I have not 
> worked out the
> > > > details and would welcome suggestions.
> > > > 
> > > >    My ideas:
> > > > 
> > > >    User Data received during PROVING should be discarded.
> > > > 
> > > >    Link Status Alignment received during PROVING should 
> be ignored.
> > > > 
> > > >    [Done.]
> > > > 
> > > > 3. 05/10/02 Jeff Craig, "[Sigtran] (M2PA) Please 
> standardize M2PA
> > > >    protocol timer identifiers" and Jeff Craig 06/21/02 
> > > "m2pa comment:
> > > >    timers". Line up timer names in Q.703 and M2PA. Add 
> the T7 timer.
> > > > 
> > > > A: This is what I got out of the discussion:
> > > > 
> > > >    Keep same names for T1, T3, T6. [Done.]
> > > > 
> > > >    Rename M2PA T2 to T4. [Done.]
> > > > 
> > > >    Don't use "timer" names for the rate parameters.  [Done.]
> > > > 
> > > >    T7 - see #11 below.  [Done.]
> > > > 
> > > > 4. 05/17/02 Jeff Craig, "[Sigtran] (M2PA) Please add a Network
> > > >    Appearance parameter to the M2PA protocol". A suggestion 
> > > that a NA
> > > >    parameter be added to M2PA.
> > > > 
> > > > A: I prefer not to add this.  [Done.]
> > > > 
> > > > 5. 05/22/02 anssharma, "[Sigtran] M2PA Clarification". In 
> > > level 2 flow
> > > >    control, M2PA stops transmitting. What does it do with the
> > > >    messages?
> > > > 
> > > > A: I think is an implementation matter.  [Done.]
> > > > 
> > > > 6. 05/24/02 Lisa Gorgone, "[Sigtran] M2PA:Questions on sequence
> > > >    number". Some questions about bad values of FSNR and BSNR.
> > > > 
> > > > A: M2PAv5 (4.2.1) says to fail the link if one FSNR is out 
> > > of order. I
> > > > think that is OK. There is nothing about bad BSNR 
> values. I would be
> > > > inclined to fail the link in this case as well, since one end is
> > > > acking messages that the other does not have in its 
> transmit buffer.
> > > > 
> > > >    I have a question about the BSNR values on the list. 
> 2 out of 3
> > > >    consecutive messages with bad BSNRs?
> > > > 
> > > >    Eliminate acks on the Status stream. All acks go 
> over the Data
> > > >    stream. If there is no User Data message to be sent, 
> > > send one with
> > > >    no data in it. Get rid of Link Status In Service, 
> since its only
> > > >    purpose was to carry an ack. [Done.]
> > > > 
> > > > 7. 05/24/02 Brian Bidulock "[Sigtran] Re: M2PA 
> Clarification". Does
> > > >    M2PA need to withhold messages from SCTP during busy?
> > > > 
> > > > A: This is how Q.703 works, so we should copy it. 
> Before there were
> > > > M2PA sequence numbers, messages were withheld while the 
> > > peer was Busy
> > > > to prevent message loss in case the link later failed. 
> Now that M2PA
> > > > has its own sequence numbers, this is no longer a danger. 
> > > So M2PA can
> > > > continue transmitting while its peer is Busy. 
> > > > 
> > > >    I updated secion 4.1.5 Level 2 Flow Control.  [Done.]
> > > > 
> > > > 8. 06/18/02 amish.chana "[Sigtran] M2PA Comment: 
> > > Multi-homing". Should
> > > >    M2PA support multi-homing?
> > > > 
> > > > A: It is allowed. The only change needed is in 4.1.2 so 
> that it says
> > > > "the IP address (or IP addresses if multi-homing is 
> used)".  [Done.]
> > > > 
> > > > 9. 06/19/02 Lisa Gorgone "Re: [Sigtran] M2PA Comment:Retrieval
> > > >    State". Does M2PA need a RETRIEVAL state? Does 
> RETRIEVAL state
> > > >    cause harm?
> > > > 
> > > > A: I would like to have more discussion about this. 
> > > > 
> > > >    Since there has not been more discussion, I will 
> leave this as is
> > > >    for M2PA version 6.  [Done.]
> > > > 
> > > > 10. 06/21/02 Javier Pastor-Balbas "[Sigtran] M2PA Comment:
> > > >     Multi-homing". Also 06/21/02 "[Sigtran] Multiplexed 
> > > M2PA links on
> > > >     an association".  Should M2PA allow multiple links per
> > > >     association?
> > > > 
> > > > A: The disadvantage of this is that if the association 
> > > dies, you lose
> > > > all the links on the association. This is one of the 
> main reasons I
> > > > did not allow for multiple links in an association. 
> > > > 
> > > >    No change to the draft is needed.  [Done.]
> > > > 
> > > > 11. 06/21/02 Andrew Booth "[Sigtran] M2PA Comment: T7 Timer" and
> > > >     several of the earlier threads. Should there be an 
> equivalent of
> > > >     MTP2 T7?
> > > > 
> > > > A: OK, you finally convinced me. 8-) I added the T7 
> timer in 4.2.1
> > > > Sending/receiving messages and 4.1.5 Level 2 Flow 
> Control.  [Done.]
> > > > 
> > > > 12. 06/25/02 Lyndon Ong "RE: [Sigtran] M2PA Queries - 
> > > summary". Just a
> > > >     summary of the discussion.
> > > > 
> > > > A: No answer needed. [Done.]
> > > > 
> > > > 13. 06/26/02 Karen King "[Sigtran] A couple of comments on M2PA
> > > >     processor outage". Recommend adding text to 4.1.4 
> > > Processor Outage
> > > >     about flushing buffers as in Q.703.
> > > > 
> > > > A: Even though this is in Q.703, the action is taken by 
> > > MTP3. Section
> > > > 4.1.4 (next to last paragraph) does allow for the 
> flush. Flush is
> > > > described in 4.2.4. So I think this is taken care of 
> (although there
> > > > may be problems in 4.2.4 - that's another thread).  [Done.]
> > > > 
> > > > 14. 06/28/02 Ray Singh "[Sigtran] Comment on M2PA - Performance
> > > >     Monitoring". Add back the monitoring of SCTP 
> > > association that was
> > > >     in draft version 3.
> > > > 
> > > > A: The reason I took this out is that I never had any 
> solid ideas
> > > > about what SCTP parameters should be monitored and what the 
> > > acceptable
> > > > values would be. If we could come up with something 
> > > helpful, I am not
> > > > opposed to adding the monitoring back in. But I don't see 
> > > the point of
> > > > adding it if the criteria will be some implementation-dependent
> > > > parameters having implementation-dependent values. 8-( [Done.]
> > > > 
> > > > 15. 07/12/02 anssharma "[Sigtran] M2PA - Flush 
> Buffers". Questions
> > > >     about Flush.
> > > > 
> > > > A: These are good questions but I have not had time to 
> think about
> > > > them.
> > > > 
> > > > 
> > > > 16. 07/19/02 Ray Singh "[Sigtran] Comment on M2PA - Slow 
> > > Start". What
> > > >     kind of warning is needed about slow start?
> > > > 
> > > > A: I proposed a general warning about slow start to 
> replace specific
> > > > warnings throughout the document.
> > > > 
> > > > alignment, changeover, path failure (multihoming), reduced 
> > > volume. See
> > > > Tolga, 07/26/02. [Done.]
> > > > 
> > > > 
> > > > 17. 07/22/02 Tolga Asveren "[Sigtran] M2PA : LS 
> messages on stream
> > > >     0". Are there problems with sending Status and Data 
> on different
> > > >     streams? What mimics MTP2 best?
> > > > 
> > > > A: The problem with putting LS and Data on the same 
> stream is that a
> > > > LS message can get stuck behind a Data retransmission. This is
> > > > contrary to the message priority of MTP2. (Brian 
> Bidulock 08/13/2002
> > > > 21:00 pointed out that this is a problem that MTP2 does 
> not have.)
> > > > 
> > > > The problem with using SCTP unordered transmission for LS 
> > > messages is
> > > > that they can get out of order with each other. For 
> > > example, a LS Proc
> > > > Outage could get out of order with a LS Proc Outage Ended. 
> > > > 
> > > > MTP does not require that LSSUs and MTUs be associated with 
> > > each other
> > > > in any particular way. It has MTP2 put LSSUs ahead of 
> MTUs in the
> > > > transmit buffer. So it should not be required that M2PA 
> > > transmit LS in
> > > > order with Data.
> > > > 
> > > > Having M2PA put LS at the front of the M2PA transmit queue 
> > > helps some,
> > > > but it does not do anything about the messages that are 
> in the SCTP
> > > > send queue. To truly imitate MTP2 at this point, we 
> would need a way
> > > > to give a message to SCTP and tell it to put the message at 
> > > the front
> > > > of its send queue.
> > > > 
> > > > (Or, we could restrict SCTP to just 1 message in its send 
> > > buffer. This
> > > > would allow M2PA to prioritize like MTP2. But this would 
> > > create severe
> > > > problems of its own. No one seemed to like this option.)
> > > > 
> > > > Sending LS and Data on different streams allows for a partial
> > > > imitation of the MTP2 transmission priority. If the 
> receiving M2PA
> > > > gives priority to reading incoming LS messages (on stream 
> > > 0) over Data
> > > > messages (on stream 1), then LS messages can be 
> received sooner than
> > > > if they waited behind the Data messages. Also, it is 
> possible for
> > > > SCTP, by some implementation-dependent method or some 
> future SCTP
> > > > extension, to give a higher priority to sending messages on 
> > > stream 0.
> > > > 
> > > > My conclusion is to leave the draft alone. Having separate 
> > > Status and
> > > > Data streams does not exactly imitate MTP2, but it is 
> as close as we
> > > > can get using SCTP.  [Done.]
> > > > 
> > > > Some related matters:
> > > > 
> > > > I will add to 4.1.7 so that it says that M2PA transmits 
> LS messages
> > > > before Data messages. This is in addition to the attempt to 
> > > prioritize
> > > > SCTP stream 0. [That was already there. Done.]
> > > > 
> > > > The FSN should be ignored in the LS messages. FSN should be
> > > > incremented only for Data messages. [That was already 
> there. Done.]
> > > > 
> > > > I will add a warning that one has to watch out for a 
> > > wraparound of the
> > > > FSN/BSN numbers. This is unlikely with 24 bits, but should 
> > > probably be
> > > > mentioned. [Done.]
> > > > 
> > > > 
> > > > 18. 08/09/02 Howard May "[Sigtran] M2PA - Minor corrections?" 
> > > > 
> > > >     In Figure 7, RETRIEVAL receives MTP Start. Go to 
> AIP, not OOS.
> > > > 
> > > > A: Yes.  [Done.]
> > > > 
> > > > 
> > > > 
> ==================================================================
> > > > 
> > > > More detailed list of changes:
> > > > 
> > > > 2.2  M2PA Header: Changed BSN and FSN to 24 bits in diagram.
> > > > 
> > > > 2.3.2 Link Status: Removed In Service (10) from list of values.
> > > > 
> > > > 2.3.2.1 Link Status Proving, 1st para: Removed 
> statement about SCTP
> > > > slow start.
> > > > 
> > > > 3.  M2PA Link State Control, Figure 7: M2PA Link State 
> Transition
> > > > Diagram:
> > > > 
> > > >     T2 is renamed T4.
> > > > 
> > > >     Added T7 expiry.
> > > > 
> > > >     In RETRIEVAL, MTP3 Start goes to AIP instead of OOS.
> > > > 
> > > > 4.1.2 MTP and SCTP Entities, 2nd para: Added "(or IP 
> addresses, if
> > > > multi-homing is used)".
> > > > 
> > > > 4.1.3 Link Alignment, 1st para: Removed list item 3 
> about SCTP slow
> > > > start.
> > > > 
> > > > 4.1.3 Link Alignment, several places: Renamed T2 to T4.
> > > > 
> > > > 4.1.3 Link Alignment, para "Anomalous...": Added this paragraph.
> > > > 
> > > > 4.1.5 Level 2 Flow Control, para "The peer M2PA 
> should...": Changed
> > > > "shall cease transmitting messages" to "should continue 
> transmitting
> > > > messages".
> > > > 
> > > > 4.1.5 Level 2 Flow Control, para "The peer M2PA shall 
> not...": Added
> > > > this paragraph about T7.
> > > > 
> > > > 4.2.1 Sending and receiving messages, para "When M2PA 
> > > receives a User
> > > > Data message...": Specified that this is a _User Data_ message.
> > > > 
> > > > 4.2.1 Sending and receiving messages: Removed paragraph 
> stating that
> > > > ack may be sent in User Data or Link Status.
> > > > 
> > > > 4.2.1 Sending and receiving messages, para "M2PA should 
> follow the
> > > > criterion...": Added this paragraph about 2-out-of-3 criterion.
> > > > 
> > > > 4.2.1 Sending and receiving messages, para "M2PA should 
> ignore...":
> > > > Added this paragraph.
> > > > 
> > > > 4.2.1 Sending and receiving messages, para "Note: In all
> > > > calculations...": Added this paragraph.
> > > > 
> > > > 4.2.1 Sending and receiving messages, para "If there is no 
> > > other User
> > > > Data...": Changed this to say M2PA sends User Data with 
> no payload
> > > > instead of LS In Service.
> > > > 
> > > > 4.2.1 Sending and receiving messages, last 2 
> paragraphs: Added these
> > > > paragraphs about T7.
> > > > 
> > > > 4.3 SCTP Considerations: Added this section to discuss 
> slow start.
> > > > 
> > > > 5.1  Link Initialization, several places: Renamed T2 to T4.
> > > > 
> > > > 7.1 SCTP Payload Protocol Identifier: Registered User 
> Port Number is
> > > > 3565. Removed UDP.
> > > > 
> > > > 9.  References: Updated some of the references.
> > > > 
> > > > 10.  Author's Addresses: New address for Brian Bidulock.
> > > > 
> > > > (end)
> > > > 
> > > > 
> > > > -- 
> > > > * Tom George          972-519-3168           
> > > Tom.George@alcatel.com   *
> > > > * PB2-1323       Intranet:  
> > > http://www.ssd.usa.alcatel.com/~tgeorge/  *
> > > > * Alcatel USA, Inc.; 1000 Coit Rd, MS PB2-STPA; Plano TX 
> > > 75075-5813   *
> > > > * I actively disclaim speaking for Alcatel USA, Inc. I do 
> > > not purport *
> > > > * to advertise or represent its position,  operations, or 
> > > activities. *
> > > > _______________________________________________
> > > > Sigtran mailing list
> > > > Sigtran@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > _______________________________________________
> > > Sigtran mailing 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 sigtran-admin@ietf.org  Mon Sep  2 16:26:00 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA00766
	for <sigtran-archive@lists.ietf.org>; Mon, 2 Sep 2002 16:26:00 -0400 (EDT)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id g82KQmo28143;
	Mon, 2 Sep 2002 16:26:49 -0400
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id g82KP7o28073
	for <sigtran@optimus.ietf.org>; Mon, 2 Sep 2002 16:25:07 -0400
Received: from gw.openss7.com (IDENT:5BVOdisakRcG6qzSwNw4wKN7c2BfzyCA@fw.openss7.com [142.179.197.31] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA00748
	for <sigtran@ietf.org>; Mon, 2 Sep 2002 16:23:25 -0400 (EDT)
Received: from ns.pigworks.openss7.net (ns1.evil.openss7.net [192.168.9.1])
	by gw.openss7.com (8.11.6/8.11.6) with ESMTP id g82KOsN11134;
	Mon, 2 Sep 2002 14:24:54 -0600
Received: (from brian@localhost)
	by ns.pigworks.openss7.net (8.11.6/8.11.6) id g82KOs122636;
	Mon, 2 Sep 2002 14:24:54 -0600
Date: Mon, 2 Sep 2002 14:24:54 -0600
From: "Brian F. G. Bidulock" <bidulock@openss7.org>
To: "May, Howard" <howard.may@intel.com>
Cc: sigtran@ietf.org
Subject: Re: [Sigtran] M2PA - Retrieval State
Message-ID: <20020902142454.B22260@openss7.org>
Reply-To: bidulock@openss7.org
Mail-Followup-To: "May, Howard" <howard.may@intel.com>, sigtran@ietf.org
References: <4A3192E7220D9241A1B774B074C6E1210171F231@swsmsx103.isw.intel.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: <4A3192E7220D9241A1B774B074C6E1210171F231@swsmsx103.isw.intel.com>; from howard.may@intel.com on Mon, Sep 02, 2002 at 01:46:33PM +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>

Howard,

Another approach would be to ignore processor outage states altogther
ala Q.2210; however, M2PA already closely parallels the processor outage
procedures of Q.703: the states are just missing from the state diagram.

--brian

May, Howard wrote:                                (Mon, 02 Sep 2002 13:46:33)
> Brian, Ming,
> 
> I concur with your points, I do not believe that the Retrieval State adds
> anything and now would support it's removal.  
> 
> With regards aligning more closely with Figure 2/Q.703.  Assuming you are
> correct when you say 'MTP3 is concerned about: PROCESSOR OUTAGE and ALIGNED
> NOT READY' then shouldn't their addition be mandatory?
> 
> Howard
> 
> > -----Original Message-----
> > From: Brian F. G. Bidulock [mailto:bidulock@openss7.org]
> > Sent: 02 September 2002 11:04
> > To: May, Howard
> > Cc: 'Ming Lin'; sigtran@ietf.org
> > Subject: Re: [Sigtran] M2PA - Retrieval State
> > 
> > 
> > Howard,
> > 
> > I agree with Ming.  There is no difference between RETRIEVAL state and
> > OOS state.  Q.703 gets (and even MTP3b) along quite nicely without an
> > LSC RETRIEVAL state: both only use an Out of Service state.  
> > For consistency
> > with MTP2/MTP3b the RETRIEVAL state needs to be removed.
> > 
> > M2PA/Figure 7 should be made to be consistent with Figure 
> > 2/Q.703. Figure
> > 2/Q.703 has no RETRIEVAL state.  The only states listed in 
> > Q.703 overview
> > are:
> > 
> >         Figure 2/Q.703               M2PA/Figure 7
> >         ------------------           ------------------
> >         Power Off                    IDLE
> >         Out of Service               OOS
> >         Initial Alignment            AIP, PROVING
> >         Aligned Ready                ALIGNED READY
> >         In Service                   INS
> >         Processor Outage             --
> >         Aligned Not Ready            --
> >         --                           RETRIEVAL
> > 
> > It would be good if M2PA could at least conform to the Q.703 
> > state machine
> > overview diagram (even if it falls short of the detailed 
> > SDLs).  Several
> > things should be done to M2PA/Figure 7 to accomplish this:  
> > RETRIEVAL state
> > should be removed (MTP3 does not distinguish it from OOS 
> > state).  PROVING
> > state should be removed--well collapsed with AIP--(MTP3 does 
> > not distinguish
> > Proving from Initial Alignment), and two states need to be 
> > added to the
> > diagram which MTP3 is concerned about: PROCESSOR OUTAGE and 
> > ALIGNED NOT READY.
> > PROCESSOR OUTAGE state results from local processor outage condition
> > experienced in the INS state.  ALIGNED NOT READY state 
> > results from local
> > processor outage condition experienced in the ALIGNED READY 
> > state.  Thus, the
> > diagram should be updated as follows:
> > 
> >                                                               
> >           
> >                     +-----------+                             
> >           
> >                     |  POWER    |                             
> >           
> >                     |   OFF     |                             
> >           
> >                     +-----------+                             
> >           
> >                           |                                   
> >           
> >                           | Power On                          
> >           
> >                           |                                   
> >           
> >         Flush Buffers     |  +-------------------------+      
> >           
> >         OR Retreival      |  |                         |      
> >           
> >                +-------+  |  |                         |      
> >           
> >                |       |  V  V                         |      
> >           
> >                |    +-----------+                      |      
> >           
> >                +--->|  OUT OF   |                      |      
> >           
> >   +---------------->|  SERVICE  |<--+                  |      
> >           
> >   |                 +-----------+   | Link Configured  |      
> >           
> >   |                       |   |     | (Associate)      |      
> >           
> >   |                       |   +-----+                  |      
> >           
> >   |                       |                            |      
> >           
> >   |                       | MTP3 Start                 |      
> >           
> >   |                       V                            |      
> >           
> >   |                 +-----------+                      |      
> >           
> >   |                 |  INITIAL  |                      |      
> >           
> >   +<----------------| ALIGNMENT |--------------------->+      
> >           
> >   |      MTP3 Stop  +-----------+     SCTP Comm Error  |      
> >           
> >   |   OR T1 Expiry        |  |     OR SCTP Comm Lost   |      
> >           
> >   |   OR T4 Expiry        |  |                         |      
> >           
> >   |   OR Receive LS OOS   |  
> > +-------------------------|---------+      
> >   |                       | Send and                   |      
> >    |      
> >   |                       |      Receive LS Alignment  |      
> >    |      
> >   |                       V                            |      
> >    |      
> >   |                 +-----------+        LPO/LPR       |   
> > +-----------+
> >   |                 |  ALIGNED  |<---------------------|-->|  
> > ALIGNED  |
> >   +<----------------|   READY   |--------------------->+   | 
> > NOT READY |
> >   |      MTP3 Stop  +-----------+     SCTP Comm Error      
> > +-----------+
> >   |   OR T3 Expiry        |        OR SCTP Comm Lost          
> >    |      
> >   |   OR Receive LS OOS   |                                   
> >    |      
> >   |                       | Receive LS Ready                  
> >    |      
> >   |                       | OR Receive User Data              
> >    |      
> >   |                       |                                   
> >    |      
> >   |                       V                                   
> >    V      
> >   |                 +-----------+        LPO/LPR           
> > +-----------+
> >   |                 |    IN     |<------------------------>| 
> > PROCESSOR |
> >   |                 |  SERVICE  |                          |  
> >  OUTAGE  |
> >   |                 +-----------+                          
> > +-----------+
> >   |                       |                                   
> >    |      
> >   |                       |    MTP3 Stop                      
> >    |      
> >   |                       | OR Receive LS OOS                 
> >    |      
> >   |                       | OR SCTP Comm Error                
> >    |      
> >   |                       | OR SCTP Comm Lost                 
> >    |      
> >   |                       | OR T6 Expiry                      
> >    |      
> >   |                       | OR T7 Expiry                      
> >    |      
> >   |                       | OR M2PA Link Failure              
> >    |      
> >   |                       |                                   
> >    |      
> >   
> > +<----------------------+<-------------------------------------+      
> >                                                               
> >           
> >                                                               
> >           
> >              Figure 7: M2PA Link State Transition Diagram     
> >           
> >                                                               
> >           
> > 
> > Text should be updated to be consistent with this diagram.
> > 
> > --brian
> > 
> > May, Howard wrote:                                            
> >                         (Mon, 02 Sep 2002 09:46:20)
> > > Dear Ming
> > > 
> > > I think Issue 18, included in your email below, is resolved by Toms
> > > correction (well done with the ascii art work Tom!).  Ming 
> > could you post a
> > > summary of the other problems you see with the Retrieval state.
> > > 
> > > I currently believe the Retrieval state to be worthwhile.  
> > As the link can
> > > only leave the Retrieval state after an event from MTP3 it 
> > guarantees that
> > > MTP3 will be able to perform and complete retrieval 
> > procedures before the
> > > link comes back into service.   If the Retrieval state is 
> > not used then I
> > > think we would be left in a situation where the link could 
> > come back into
> > > service before Retrieval had been completed.  There is then 
> > an issue of
> > > either allowing MTP3 and M2PA to fight over messages in the 
> > TX buffer (MTP3
> > > trying to Retrieve them and M2PA trying to transmit them - clearly
> > > undesirable) or perhaps when the link fails moving messages 
> > left in the Tx
> > > buffer to another buffer until MTP3 decides whether it 
> > wants them or not (My
> > > gut feeling is that this is where more problems lie rather 
> > than the use of
> > > the Retrieval state)
> > > 
> > > Cheers
> > > 
> > > > -----Original Message-----
> > > > From: Ming Lin [mailto:mlin@huawei.com]
> > > > Sent: 01 September 2002 05:08
> > > > To: Tom George
> > > > Cc: sigtran@ietf.org
> > > > Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-m2pa-06.txt
> > > > 
> > > > 
> > > > Dear Tom,
> > > > 
> > > > I think we should remove RETRIEVAL state, this state will 
> > > > result in many problem but no use,
> > > > for example the No. 18 issue.
> > > > 
> > > > "18. 08/09/02 Howard May "[Sigtran] M2PA - Minor corrections?" 
> > > >      In Figure 7, RETRIEVAL receives MTP Start. Go to 
> > AIP, not OOS.
> > > >  A: Yes.  [Done.]"
> > > > 
> > > > ----- Original Message ----- 
> > > > From: "Tom George" <Tom.George@alcatel.com>
> > > > Cc: <sigtran@ietf.org>
> > > > Sent: Friday, August 30, 2002 11:30 PM
> > > > Subject: Re: [Sigtran] I-D ACTION:draft-ietf-sigtran-m2pa-06.txt
> > > > 
> > > > 
> > > > > M2PA Changes for version 6
> > > > > 08/28/2002
> > > > > 
> > > > > Here are the changes in draft-ietf-sigtran-m2pa-06.txt (M2PAv6).
> > > > > 
> > > > > If anyone wants to study the changes in detail, I have a 
> > > > "diff" file I
> > > > > can send.
> > > > > 
> > > > > This is not a complete list of changes. I did not include minor
> > > > > rewording.
> > > > > 
> > > > > 
> > > > > This is the list of comments received on M2PAv5 on the 
> > SIGTRAN list.
> > > > > 
> > > > > 
> > ==================================================================
> > > > > 
> > > > > Here is my list of comments. For the most part, they 
> > are arranged by
> > > > > the thread from the SIGTRAN list.
> > > > > 
> > > > > Everyone please make comments. I think the discussion 
> > would go more
> > > > > smoothly if you only talked about one item per e-mail.
> > > > > 
> > > > > 
> > > > > 1. Section 2.2. FSN/BSN should be 24 bits, not 23.
> > > > > 
> > > > > A: This was a typo. It will be 24 in M2PAv6.  [Done.]
> > > > > 
> > > > > 2. 05/06/02 Jeff Craig "[Sigtran] (M2PA) What actions 
> > > > should be taken
> > > > >    if LSA or User Data Received in PROVING?" There is 
> > no guidance
> > > > >    about anomalous events during alignment/proving.
> > > > > 
> > > > > A: I agree that some text should be added. I have not 
> > worked out the
> > > > > details and would welcome suggestions.
> > > > > 
> > > > >    My ideas:
> > > > > 
> > > > >    User Data received during PROVING should be discarded.
> > > > > 
> > > > >    Link Status Alignment received during PROVING should 
> > be ignored.
> > > > > 
> > > > >    [Done.]
> > > > > 
> > > > > 3. 05/10/02 Jeff Craig, "[Sigtran] (M2PA) Please 
> > standardize M2PA
> > > > >    protocol timer identifiers" and Jeff Craig 06/21/02 
> > > > "m2pa comment:
> > > > >    timers". Line up timer names in Q.703 and M2PA. Add 
> > the T7 timer.
> > > > > 
> > > > > A: This is what I got out of the discussion:
> > > > > 
> > > > >    Keep same names for T1, T3, T6. [Done.]
> > > > > 
> > > > >    Rename M2PA T2 to T4. [Done.]
> > > > > 
> > > > >    Don't use "timer" names for the rate parameters.  [Done.]
> > > > > 
> > > > >    T7 - see #11 below.  [Done.]
> > > > > 
> > > > > 4. 05/17/02 Jeff Craig, "[Sigtran] (M2PA) Please add a Network
> > > > >    Appearance parameter to the M2PA protocol". A suggestion 
> > > > that a NA
> > > > >    parameter be added to M2PA.
> > > > > 
> > > > > A: I prefer not to add this.  [Done.]
> > > > > 
> > > > > 5. 05/22/02 anssharma, "[Sigtran] M2PA Clarification". In 
> > > > level 2 flow
> > > > >    control, M2PA stops transmitting. What does it do with the
> > > > >    messages?
> > > > > 
> > > > > A: I think is an implementation matter.  [Done.]
> > > > > 
> > > > > 6. 05/24/02 Lisa Gorgone, "[Sigtran] M2PA:Questions on sequence
> > > > >    number". Some questions about bad values of FSNR and BSNR.
> > > > > 
> > > > > A: M2PAv5 (4.2.1) says to fail the link if one FSNR is out 
> > > > of order. I
> > > > > think that is OK. There is nothing about bad BSNR 
> > values. I would be
> > > > > inclined to fail the link in this case as well, since one end is
> > > > > acking messages that the other does not have in its 
> > transmit buffer.
> > > > > 
> > > > >    I have a question about the BSNR values on the list. 
> > 2 out of 3
> > > > >    consecutive messages with bad BSNRs?
> > > > > 
> > > > >    Eliminate acks on the Status stream. All acks go 
> > over the Data
> > > > >    stream. If there is no User Data message to be sent, 
> > > > send one with
> > > > >    no data in it. Get rid of Link Status In Service, 
> > since its only
> > > > >    purpose was to carry an ack. [Done.]
> > > > > 
> > > > > 7. 05/24/02 Brian Bidulock "[Sigtran] Re: M2PA 
> > Clarification". Does
> > > > >    M2PA need to withhold messages from SCTP during busy?
> > > > > 
> > > > > A: This is how Q.703 works, so we should copy it. 
> > Before there were
> > > > > M2PA sequence numbers, messages were withheld while the 
> > > > peer was Busy
> > > > > to prevent message loss in case the link later failed. 
> > Now that M2PA
> > > > > has its own sequence numbers, this is no longer a danger. 
> > > > So M2PA can
> > > > > continue transmitting while its peer is Busy. 
> > > > > 
> > > > >    I updated secion 4.1.5 Level 2 Flow Control.  [Done.]
> > > > > 
> > > > > 8. 06/18/02 amish.chana "[Sigtran] M2PA Comment: 
> > > > Multi-homing". Should
> > > > >    M2PA support multi-homing?
> > > > > 
> > > > > A: It is allowed. The only change needed is in 4.1.2 so 
> > that it says
> > > > > "the IP address (or IP addresses if multi-homing is 
> > used)".  [Done.]
> > > > > 
> > > > > 9. 06/19/02 Lisa Gorgone "Re: [Sigtran] M2PA Comment:Retrieval
> > > > >    State". Does M2PA need a RETRIEVAL state? Does 
> > RETRIEVAL state
> > > > >    cause harm?
> > > > > 
> > > > > A: I would like to have more discussion about this. 
> > > > > 
> > > > >    Since there has not been more discussion, I will 
> > leave this as is
> > > > >    for M2PA version 6.  [Done.]
> > > > > 
> > > > > 10. 06/21/02 Javier Pastor-Balbas "[Sigtran] M2PA Comment:
> > > > >     Multi-homing". Also 06/21/02 "[Sigtran] Multiplexed 
> > > > M2PA links on
> > > > >     an association".  Should M2PA allow multiple links per
> > > > >     association?
> > > > > 
> > > > > A: The disadvantage of this is that if the association 
> > > > dies, you lose
> > > > > all the links on the association. This is one of the 
> > main reasons I
> > > > > did not allow for multiple links in an association. 
> > > > > 
> > > > >    No change to the draft is needed.  [Done.]
> > > > > 
> > > > > 11. 06/21/02 Andrew Booth "[Sigtran] M2PA Comment: T7 Timer" and
> > > > >     several of the earlier threads. Should there be an 
> > equivalent of
> > > > >     MTP2 T7?
> > > > > 
> > > > > A: OK, you finally convinced me. 8-) I added the T7 
> > timer in 4.2.1
> > > > > Sending/receiving messages and 4.1.5 Level 2 Flow 
> > Control.  [Done.]
> > > > > 
> > > > > 12. 06/25/02 Lyndon Ong "RE: [Sigtran] M2PA Queries - 
> > > > summary". Just a
> > > > >     summary of the discussion.
> > > > > 
> > > > > A: No answer needed. [Done.]
> > > > > 
> > > > > 13. 06/26/02 Karen King "[Sigtran] A couple of comments on M2PA
> > > > >     processor outage". Recommend adding text to 4.1.4 
> > > > Processor Outage
> > > > >     about flushing buffers as in Q.703.
> > > > > 
> > > > > A: Even though this is in Q.703, the action is taken by 
> > > > MTP3. Section
> > > > > 4.1.4 (next to last paragraph) does allow for the 
> > flush. Flush is
> > > > > described in 4.2.4. So I think this is taken care of 
> > (although there
> > > > > may be problems in 4.2.4 - that's another thread).  [Done.]
> > > > > 
> > > > > 14. 06/28/02 Ray Singh "[Sigtran] Comment on M2PA - Performance
> > > > >     Monitoring". Add back the monitoring of SCTP 
> > > > association that was
> > > > >     in draft version 3.
> > > > > 
> > > > > A: The reason I took this out is that I never had any 
> > solid ideas
> > > > > about what SCTP parameters should be monitored and what the 
> > > > acceptable
> > > > > values would be. If we could come up with something 
> > > > helpful, I am not
> > > > > opposed to adding the monitoring back in. But I don't see 
> > > > the point of
> > > > > adding it if the criteria will be some implementation-dependent
> > > > > parameters having implementation-dependent values. 8-( [Done.]
> > > > > 
> > > > > 15. 07/12/02 anssharma "[Sigtran] M2PA - Flush 
> > Buffers". Questions
> > > > >     about Flush.
> > > > > 
> > > > > A: These are good questions but I have not had time to 
> > think about
> > > > > them.
> > > > > 
> > > > > 
> > > > > 16. 07/19/02 Ray Singh "[Sigtran] Comment on M2PA - Slow 
> > > > Start". What
> > > > >     kind of warning is needed about slow start?
> > > > > 
> > > > > A: I proposed a general warning about slow start to 
> > replace specific
> > > > > warnings throughout the document.
> > > > > 
> > > > > alignment, changeover, path failure (multihoming), reduced 
> > > > volume. See
> > > > > Tolga, 07/26/02. [Done.]
> > > > > 
> > > > > 
> > > > > 17. 07/22/02 Tolga Asveren "[Sigtran] M2PA : LS 
> > messages on stream
> > > > >     0". Are there problems with sending Status and Data 
> > on different
> > > > >     streams? What mimics MTP2 best?
> > > > > 
> > > > > A: The problem with putting LS and Data on the same 
> > stream is that a
> > > > > LS message can get stuck behind a Data retransmission. This is
> > > > > contrary to the message priority of MTP2. (Brian 
> > Bidulock 08/13/2002
> > > > > 21:00 pointed out that this is a problem that MTP2 does 
> > not have.)
> > > > > 
> > > > > The problem with using SCTP unordered transmission for LS 
> > > > messages is
> > > > > that they can get out of order with each other. For 
> > > > example, a LS Proc
> > > > > Outage could get out of order with a LS Proc Outage Ended. 
> > > > > 
> > > > > MTP does not require that LSSUs and MTUs be associated with 
> > > > each other
> > > > > in any particular way. It has MTP2 put LSSUs ahead of 
> > MTUs in the
> > > > > transmit buffer. So it should not be required that M2PA 
> > > > transmit LS in
> > > > > order with Data.
> > > > > 
> > > > > Having M2PA put LS at the front of the M2PA transmit queue 
> > > > helps some,
> > > > > but it does not do anything about the messages that are 
> > in the SCTP
> > > > > send queue. To truly imitate MTP2 at this point, we 
> > would need a way
> > > > > to give a message to SCTP and tell it to put the message at 
> > > > the front
> > > > > of its send queue.
> > > > > 
> > > > > (Or, we could restrict SCTP to just 1 message in its send 
> > > > buffer. This
> > > > > would allow M2PA to prioritize like MTP2. But this would 
> > > > create severe
> > > > > problems of its own. No one seemed to like this option.)
> > > > > 
> > > > > Sending LS and Data on different streams allows for a partial
> > > > > imitation of the MTP2 transmission priority. If the 
> > receiving M2PA
> > > > > gives priority to reading incoming LS messages (on stream 
> > > > 0) over Data
> > > > > messages (on stream 1), then LS messages can be 
> > received sooner than
> > > > > if they waited behind the Data messages. Also, it is 
> > possible for
> > > > > SCTP, by some implementation-dependent method or some 
> > future SCTP
> > > > > extension, to give a higher priority to sending messages on 
> > > > stream 0.
> > > > > 
> > > > > My conclusion is to leave the draft alone. Having separate 
> > > > Status and
> > > > > Data streams does not exactly imitate MTP2, but it is 
> > as close as we
> > > > > can get using SCTP.  [Done.]
> > > > > 
> > > > > Some related matters:
> > > > > 
> > > > > I will add to 4.1.7 so that it says that M2PA transmits 
> > LS messages
> > > > > before Data messages. This is in addition to the attempt to 
> > > > prioritize
> > > > > SCTP stream 0. [That was already there. Done.]
> > > > > 
> > > > > The FSN should be ignored in the LS messages. FSN should be
> > > > > incremented only for Data messages. [That was already 
> > there. Done.]
> > > > > 
> > > > > I will add a warning that one has to watch out for a 
> > > > wraparound of the
> > > > > FSN/BSN numbers. This is unlikely with 24 bits, but should 
> > > > probably be
> > > > > mentioned. [Done.]
> > > > > 
> > > > > 
> > > > > 18. 08/09/02 Howard May "[Sigtran] M2PA - Minor corrections?" 
> > > > > 
> > > > >     In Figure 7, RETRIEVAL receives MTP Start. Go to 
> > AIP, not OOS.
> > > > > 
> > > > > A: Yes.  [Done.]
> > > > > 
> > > > > 
> > > > > 
> > ==================================================================
> > > > > 
> > > > > More detailed list of changes:
> > > > > 
> > > > > 2.2  M2PA Header: Changed BSN and FSN to 24 bits in diagram.
> > > > > 
> > > > > 2.3.2 Link Status: Removed In Service (10) from list of values.
> > > > > 
> > > > > 2.3.2.1 Link Status Proving, 1st para: Removed 
> > statement about SCTP
> > > > > slow start.
> > > > > 
> > > > > 3.  M2PA Link State Control, Figure 7: M2PA Link State 
> > Transition
> > > > > Diagram:
> > > > > 
> > > > >     T2 is renamed T4.
> > > > > 
> > > > >     Added T7 expiry.
> > > > > 
> > > > >     In RETRIEVAL, MTP3 Start goes to AIP instead of OOS.
> > > > > 
> > > > > 4.1.2 MTP and SCTP Entities, 2nd para: Added "(or IP 
> > addresses, if
> > > > > multi-homing is used)".
> > > > > 
> > > > > 4.1.3 Link Alignment, 1st para: Removed list item 3 
> > about SCTP slow
> > > > > start.
> > > > > 
> > > > > 4.1.3 Link Alignment, several places: Renamed T2 to T4.
> > > > > 
> > > > > 4.1.3 Link Alignment, para "Anomalous...": Added this paragraph.
> > > > > 
> > > > > 4.1.5 Level 2 Flow Control, para "The peer M2PA 
> > should...": Changed
> > > > > "shall cease transmitting messages" to "should continue 
> > transmitting
> > > > > messages".
> > > > > 
> > > > > 4.1.5 Level 2 Flow Control, para "The peer M2PA shall 
> > not...": Added
> > > > > this paragraph about T7.
> > > > > 
> > > > > 4.2.1 Sending and receiving messages, para "When M2PA 
> > > > receives a User
> > > > > Data message...": Specified that this is a _User Data_ message.
> > > > > 
> > > > > 4.2.1 Sending and receiving messages: Removed paragraph 
> > stating that
> > > > > ack may be sent in User Data or Link Status.
> > > > > 
> > > > > 4.2.1 Sending and receiving messages, para "M2PA should 
> > follow the
> > > > > criterion...": Added this paragraph about 2-out-of-3 criterion.
> > > > > 
> > > > > 4.2.1 Sending and receiving messages, para "M2PA should 
> > ignore...":
> > > > > Added this paragraph.
> > > > > 
> > > > > 4.2.1 Sending and receiving messages, para "Note: In all
> > > > > calculations...": Added this paragraph.
> > > > > 
> > > > > 4.2.1 Sending and receiving messages, para "If there is no 
> > > > other User
> > > > > Data...": Changed this to say M2PA sends User Data with 
> > no payload
> > > > > instead of LS In Service.
> > > > > 
> > > > > 4.2.1 Sending and receiving messages, last 2 
> > paragraphs: Added these
> > > > > paragraphs about T7.
> > > > > 
> > > > > 4.3 SCTP Considerations: Added this section to discuss 
> > slow start.
> > > > > 
> > > > > 5.1  Link Initialization, several places: Renamed T2 to T4.
> > > > > 
> > > > > 7.1 SCTP Payload Protocol Identifier: Registered User 
> > Port Number is
> > > > > 3565. Removed UDP.
> > > > > 
> > > > > 9.  References: Updated some of the references.
> > > > > 
> > > > > 10.  Author's Addresses: New address for Brian Bidulock.
> > > > > 
> > > > > (end)
> > > > > 
> > > > > 
> > > > > -- 
> > > > > * Tom George          972-519-3168           
> > > > Tom.George@alcatel.com   *
> > > > > * PB2-1323       Intranet:  
> > > > http://www.ssd.usa.alcatel.com/~tgeorge/  *
> > > > > * Alcatel USA, Inc.; 1000 Coit Rd, MS PB2-STPA; Plano TX 
> > > > 75075-5813   *
> > > > > * I actively disclaim speaking for Alcatel USA, Inc. I do 
> > > > not purport *
> > > > > * to advertise or represent its position,  operations, or 
> > > > activities. *
> > > > > _______________________________________________
> > > > > Sigtran mailing list
> > > > > Sigtran@ietf.org
> > > > > https://www1.ietf.org/mailman/listinfo/sigtran
> > > > _______________________________________________
> > > > Sigtran mailing 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 sigtran-admin@ietf.org  Tue Sep  3 07:42:24 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA22410
	for <sigtran-archive@lists.ietf.org>; Tue, 3 Sep 2002 07:42:24 -0400 (EDT)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id g83Bh8o18227;
	Tue, 3 Sep 2002 07:43:09 -0400
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id g83BaMo17368
	for <sigtran@optimus.ietf.org>; Tue, 3 Sep 2002 07:36:22 -0400
Received: from CNRI.Reston.VA.US (localhost [127.0.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA22105;
	Tue, 3 Sep 2002 07:34:45 -0400 (EDT)
Message-Id: <200209031134.HAA22105@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: Tue, 03 Sep 2002 07:34:45 -0400
Subject: [Sigtran] I-D ACTION:draft-ietf-sigtran-m2pa-06.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-06.txt
	Pages		: 50
	Date		: 2002-8-29
	
This Internet Draft defines a protocol supporting the transport of
Signaling System Number 7 (SS7) Message Transfer Part (MTP) Layer 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 employing the MTP Level 3
protocol. The SS7 Signaling Points may also employ standard SS7 links
using the SS7 MTP Layer 2 to provide transport of MTP Layer 3
signaling messages.

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-sigtran-m2pa-06.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-06.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-06.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:	<2002-8-29142404.I-D@ietf.org>

ENCODING mime
FILE /internet-drafts/draft-ietf-sigtran-m2pa-06.txt

--OtherAccess
Content-Type: Message/External-body;
	name="draft-ietf-sigtran-m2pa-06.txt";
	site="ftp.ietf.org";
	access-type="anon-ftp";
	directory="internet-drafts"

Content-Type: text/plain
Content-ID:	<2002-8-29142404.I-D@ietf.org>

--OtherAccess--

--NextPart--


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


