
From philip.eardley@bt.com  Wed Aug  5 05:55:55 2009
Return-Path: <philip.eardley@bt.com>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 3419E3A6BE4 for <tsvwg@core3.amsl.com>; Wed,  5 Aug 2009 05:55:55 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.991
X-Spam-Level: 
X-Spam-Status: No, score=-2.991 tagged_above=-999 required=5 tests=[AWL=0.608,  BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id lxTXacWBPMf1 for <tsvwg@core3.amsl.com>; Wed,  5 Aug 2009 05:55:54 -0700 (PDT)
Received: from smtp1.smtp.bt.com (smtp1.smtp.bt.com [217.32.164.137]) by core3.amsl.com (Postfix) with ESMTP id 2C20C3A687B for <tsvwg@ietf.org>; Wed,  5 Aug 2009 05:55:53 -0700 (PDT)
Received: from E03MVB1-UKBR.domain1.systemhost.net ([193.113.197.107]) by smtp1.smtp.bt.com with Microsoft SMTPSVC(6.0.3790.3959);  Wed, 5 Aug 2009 13:55:55 +0100
X-MimeOLE: Produced By Microsoft Exchange V6.5
Content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Date: Wed, 5 Aug 2009 13:55:55 +0100
Message-ID: <4A916DBC72536E419A0BD955EDECEDEC05DF88FE@E03MVB1-UKBR.domain1.systemhost.net>
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Thread-Topic: comment on draft-ietf-tsvwg-ecn-tunnel
Thread-Index: AcoVzBGqkXuqNWa1TVC5JRT+0cgOUA==
From: <philip.eardley@bt.com>
To: <tsvwg@ietf.org>
X-OriginalArrivalTime: 05 Aug 2009 12:55:55.0874 (UTC) FILETIME=[1257DC20:01CA15CC]
Subject: [tsvwg] comment on draft-ietf-tsvwg-ecn-tunnel
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 05 Aug 2009 12:55:55 -0000

hi
i wanted to follow up on this from stockholm. i've tried to catch up =
with bob & david's discussion, sorry for where my reading is incomplete.
=20
- tunnel egress behaviour for 3168 & 4301 is inconsistent, amongst other =
motivations for change to a single consistent tunnel egress behaviour
=20
- one set of changes is about
inner: not-ect
outer: isn't not-ect
which should neve be seen.
i think bob & david have converged (?). i have no desire to upset their =
convergence!
=20
- another cahnge is about:
inner: ect(0)
outer: ect(1)
decaps header: ect(1)
this is a change from the current specs, although it's somethng that =
should never be seen.=20
making this change enables 2 levels of congestion severity.
i think bob & david both agree with this change.
=20
the point where there is disagreement:
David: in environements without 2 levels of congestion severity, then =
this is an error, so the tunnel egress MAY raise an alarm.
Bob: don't raise an alarm.=20
=20
David, my argument against your view is: this 'MAY raise an alarm' is =
actually different from all the other combinations of codepoints where =
the egress 'MAY raise an alarm'. in those other cases it's always the =
right thing for the egress to do. in this case however, if it's a domain =
that uses 2 levels of congestion severity, then it MUST NOT raise an =
alarm (becuase it's the expected behaviour), whereas if it's a domain =
that uses 1 level of congestion severity, then it MAY raise an alarm =
(becuase something's wrong).
to me the simplest thing would be not to raise an alarm, so there doesnt =
have to be different configuration depending on the domain. if the std =
does mention the alarm then somehow it has to be very clear that this is =
different from the other alarms.
=20
best wishes
phil

From nishida@sfc.wide.ad.jp  Mon Aug 10 01:14:22 2009
Return-Path: <nishida@sfc.wide.ad.jp>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 16F2E3A6C14 for <tsvwg@core3.amsl.com>; Mon, 10 Aug 2009 01:14:22 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 4.521
X-Spam-Level: ****
X-Spam-Status: No, score=4.521 tagged_above=-999 required=5 tests=[AWL=2.128,  BAYES_05=-1.11, HELO_EQ_JP=1.244, HOST_EQ_JP=1.265, RELAY_IS_203=0.994]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 4CS2G+cLNQ1U for <tsvwg@core3.amsl.com>; Mon, 10 Aug 2009 01:14:21 -0700 (PDT)
Received: from mail.sfc.wide.ad.jp (mail.sfc.wide.ad.jp [203.178.142.146]) by core3.amsl.com (Postfix) with ESMTP id 5C4E63A6873 for <tsvwg@ietf.org>; Mon, 10 Aug 2009 01:14:21 -0700 (PDT)
Received: from localhost (cpu.sfc.wide.ad.jp [IPv6:2001:200:0:8803:203:178:142:143]) by mail.sfc.wide.ad.jp (Postfix) with ESMTP id A67104C540 for <tsvwg@ietf.org>; Mon, 10 Aug 2009 17:14:20 +0900 (JST)
Date: Mon, 10 Aug 2009 17:14:20 +0900 (JST)
Message-Id: <20090810.171420.104076446.nishida@sfc.wide.ad.jp>
To: tsvwg@ietf.org
From: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>
X-Mailer: Mew version 6.2 on Emacs 22.1 / Mule 5.0 (SAKAKI)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Subject: [tsvwg] Q about path failure detection in SCTP
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 10 Aug 2009 08:14:22 -0000

Hello people,

I sometimes write sctp applications for multihome environments and notice that
there is always significant long delay (over 30 secs) to switch over to secondary 
path when the primary path becomes unavailable. 

This is because sctp sender tries to retransmit data packets (usualy) 5 times to the
primary path before SCTP considers it is unavailable. In my understanding, this is 
a normal behavior of SCTP since RFC4960 specifies that an endpoint should always 
transmit to the primary path.

However, it seems to me that this behavior impairs the usability of SCTP. 
For example, if we can utilize the retransmission scheme proposed in the 
CMT paper [1] which use "potentially-failed" state, we might be able to 
switch over to the other path more smoothly.

I just would like to know if there is any reason that we don't adopt this kind of 
approach in the standard. It seems to me there is no major drawback in this approach.
Or am I misunderstanding something?

[1]
P. Natarajan, N, Ekiz, P. Amer, R. Stewart, 
"Concurrent Multipath Transfer using SCTP Multihoming: Transmission Policies using
Potentially-failed Destination State", LNCS 4982

Thanks,
--
Yoshifumi Nishida
nishida@sfc.wide.ad.jp

From randall@lakerest.net  Mon Aug 10 13:23:28 2009
Return-Path: <randall@lakerest.net>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 71C803A6F01 for <tsvwg@core3.amsl.com>; Mon, 10 Aug 2009 13:23:28 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.299
X-Spam-Level: 
X-Spam-Status: No, score=-2.299 tagged_above=-999 required=5 tests=[AWL=0.300,  BAYES_00=-2.599]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id TZ440uu+Zt+4 for <tsvwg@core3.amsl.com>; Mon, 10 Aug 2009 13:23:18 -0700 (PDT)
Received: from lakerest.net (unknown [IPv6:2001:240:585:2:213:d4ff:fef3:2d8d]) by core3.amsl.com (Postfix) with ESMTP id 5BADD3A67E5 for <tsvwg@ietf.org>; Mon, 10 Aug 2009 13:23:17 -0700 (PDT)
Received: from sc-cihuang-01.corp.extremenetworks.com (mail3.extremenetworks.com [207.179.9.4]) (authenticated bits=0) by lakerest.net (8.14.3/8.14.3) with ESMTP id n7AKNDCE043123 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Mon, 10 Aug 2009 16:23:15 -0400 (EDT) (envelope-from randall@lakerest.net)
DKIM-Signature: a=rsa-sha1; c=simple/simple; d=lakerest.net; s=mail; t=1249935796; h=Cc:Message-Id:From:To:In-Reply-To:Content-Type: Content-Transfer-Encoding:Mime-Version:Subject:Date:References: X-Mailer; b=sx6V5bnZcqzU6dQGgdLj2gm8ZkRfCh3gjFqrUvVxwWsXAGavgoc8noE FfacL53W6AGbqrz+T6EMUkUDTF/w9JA==
Message-Id: <195857BC-4901-4E1E-A39E-139B262CE4B7@lakerest.net>
From: Randy Stewart <randall@lakerest.net>
To: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>
In-Reply-To: <20090810.171420.104076446.nishida@sfc.wide.ad.jp>
Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes
Content-Transfer-Encoding: 7bit
Mime-Version: 1.0 (Apple Message framework v936)
Date: Mon, 10 Aug 2009 13:23:08 -0700
References: <20090810.171420.104076446.nishida@sfc.wide.ad.jp>
X-Mailer: Apple Mail (2.936)
Cc: tsvwg@ietf.org
Subject: Re: [tsvwg] Q about path failure detection in SCTP
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 10 Aug 2009 20:23:28 -0000

Nishida-san:

I think you are probably correct. You can also find
some similar ideas put out by Armando Caro in some
of his research... don't have the citations at my finger
tips..

His idea on multiple triggers for failure were I think what
Jana pulled in to the CMT work...

I am not sure about putting it in the standard I would have
to think about if its allowed (it might well be in the current
standard)...


R
On Aug 10, 2009, at 1:14 AM, Yoshifumi Nishida wrote:

>
> Hello people,
>
> I sometimes write sctp applications for multihome environments and  
> notice that
> there is always significant long delay (over 30 secs) to switch over  
> to secondary
> path when the primary path becomes unavailable.
>
> This is because sctp sender tries to retransmit data packets  
> (usualy) 5 times to the
> primary path before SCTP considers it is unavailable. In my  
> understanding, this is
> a normal behavior of SCTP since RFC4960 specifies that an endpoint  
> should always
> transmit to the primary path.
>
> However, it seems to me that this behavior impairs the usability of  
> SCTP.
> For example, if we can utilize the retransmission scheme proposed in  
> the
> CMT paper [1] which use "potentially-failed" state, we might be able  
> to
> switch over to the other path more smoothly.
>
> I just would like to know if there is any reason that we don't adopt  
> this kind of
> approach in the standard. It seems to me there is no major drawback  
> in this approach.
> Or am I misunderstanding something?
>
> [1]
> P. Natarajan, N, Ekiz, P. Amer, R. Stewart,
> "Concurrent Multipath Transfer using SCTP Multihoming: Transmission  
> Policies using
> Potentially-failed Destination State", LNCS 4982
>
> Thanks,
> --
> Yoshifumi Nishida
> nishida@sfc.wide.ad.jp
>

-----
Randall Stewart
randall@lakerest.net





From nishida@sfc.wide.ad.jp  Mon Aug 10 16:59:42 2009
Return-Path: <nishida@sfc.wide.ad.jp>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 0E3743A6822 for <tsvwg@core3.amsl.com>; Mon, 10 Aug 2009 16:59:42 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 3.276
X-Spam-Level: ***
X-Spam-Status: No, score=3.276 tagged_above=-999 required=5 tests=[AWL=2.372,  BAYES_00=-2.599, HELO_EQ_JP=1.244, HOST_EQ_JP=1.265, RELAY_IS_203=0.994]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id bQPrhzqG0lYS for <tsvwg@core3.amsl.com>; Mon, 10 Aug 2009 16:59:41 -0700 (PDT)
Received: from mail.sfc.wide.ad.jp (mail.sfc.wide.ad.jp [203.178.142.146]) by core3.amsl.com (Postfix) with ESMTP id 7117628C1D7 for <tsvwg@ietf.org>; Mon, 10 Aug 2009 16:59:37 -0700 (PDT)
Received: from localhost (cpu.sfc.wide.ad.jp [IPv6:2001:200:0:8803:203:178:142:143]) by mail.sfc.wide.ad.jp (Postfix) with ESMTP id 88A494C536; Tue, 11 Aug 2009 08:59:36 +0900 (JST)
Date: Tue, 11 Aug 2009 08:59:36 +0900 (JST)
Message-Id: <20090811.085936.183042386.nishida@sfc.wide.ad.jp>
To: randall@lakerest.net
From: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>
In-Reply-To: <195857BC-4901-4E1E-A39E-139B262CE4B7@lakerest.net>
References: <20090810.171420.104076446.nishida@sfc.wide.ad.jp> <195857BC-4901-4E1E-A39E-139B262CE4B7@lakerest.net>
X-Mailer: Mew version 6.2 on Emacs 22.1 / Mule 5.0 (SAKAKI)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Cc: tsvwg@ietf.org
Subject: Re: [tsvwg] Q about path failure detection in SCTP
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 10 Aug 2009 23:59:42 -0000

Dear Randy,

Thank you so much for your feedback. 
In my understanding, there is no good way to solve this in the current standard
without some cooperations from applications.
The section 6.4 in RFC4960 says:

  By default, an endpoint SHOULD always transmit to the primary path,
  unless the SCTP user explicitly specifies the destination transport
  address (and possibly source transport address) to use.

If we follow this, we have to transmit data to the primary path until it has been changed.
However, the threshold of 'Path.Max.Retrans' prevents to change primary path quickly.

Thanks,
--
Yoshifumi Nishida
nishida@sfc.wide.ad.jp


From: Randy Stewart <randall@lakerest.net>
Subject: Re: [tsvwg] Q about path failure detection in SCTP
Date: Mon, 10 Aug 2009 13:23:08 -0700
Message-ID: <195857BC-4901-4E1E-A39E-139B262CE4B7@lakerest.net>

 > Nishida-san:
 > 
 > I think you are probably correct. You can also find
 > some similar ideas put out by Armando Caro in some
 > of his research... don't have the citations at my finger
 > tips..
 > 
 > His idea on multiple triggers for failure were I think what
 > Jana pulled in to the CMT work...
 > 
 > I am not sure about putting it in the standard I would have
 > to think about if its allowed (it might well be in the current
 > standard)...
 > 
 > 
 > R
 > On Aug 10, 2009, at 1:14 AM, Yoshifumi Nishida wrote:
 > 
 > >
 > > Hello people,
 > >
 > > I sometimes write sctp applications for multihome environments and
 > > notice that
 > > there is always significant long delay (over 30 secs) to switch over
 > > to secondary
 > > path when the primary path becomes unavailable.
 > >
 > > This is because sctp sender tries to retransmit data packets (usualy)
 > > 5 times to the
 > > primary path before SCTP considers it is unavailable. In my
 > > understanding, this is
 > > a normal behavior of SCTP since RFC4960 specifies that an endpoint
 > > should always
 > > transmit to the primary path.
 > >
 > > However, it seems to me that this behavior impairs the usability of
 > > SCTP.
 > > For example, if we can utilize the retransmission scheme proposed in
 > > the
 > > CMT paper [1] which use "potentially-failed" state, we might be able
 > > to
 > > switch over to the other path more smoothly.
 > >
 > > I just would like to know if there is any reason that we don't adopt
 > > this kind of
 > > approach in the standard. It seems to me there is no major drawback in
 > > this approach.
 > > Or am I misunderstanding something?
 > >
 > > [1]
 > > P. Natarajan, N, Ekiz, P. Amer, R. Stewart,
 > > "Concurrent Multipath Transfer using SCTP Multihoming: Transmission
 > > Policies using
 > > Potentially-failed Destination State", LNCS 4982
 > >
 > > Thanks,
 > > --
 > > Yoshifumi Nishida
 > > nishida@sfc.wide.ad.jp
 > >
 > 
 > -----
 > Randall Stewart
 > randall@lakerest.net
 > 
 > 
 > 
 > 

From randall@lakerest.net  Mon Aug 10 17:17:40 2009
Return-Path: <randall@lakerest.net>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id C19CC28C205 for <tsvwg@core3.amsl.com>; Mon, 10 Aug 2009 17:17:40 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.449
X-Spam-Level: 
X-Spam-Status: No, score=-2.449 tagged_above=-999 required=5 tests=[AWL=0.150,  BAYES_00=-2.599]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id hizKwuHHASHN for <tsvwg@core3.amsl.com>; Mon, 10 Aug 2009 17:17:39 -0700 (PDT)
Received: from lakerest.net (unknown [IPv6:2001:240:585:2:213:d4ff:fef3:2d8d]) by core3.amsl.com (Postfix) with ESMTP id 052A828C1FF for <tsvwg@ietf.org>; Mon, 10 Aug 2009 17:17:38 -0700 (PDT)
Received: from rrs.extremenetworks.com (extrgate1.extremenetworks.com [207.179.9.4]) (authenticated bits=0) by lakerest.net (8.14.3/8.14.3) with ESMTP id n7B0HaPL052634 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Mon, 10 Aug 2009 20:17:38 -0400 (EDT) (envelope-from randall@lakerest.net)
DKIM-Signature: a=rsa-sha1; c=simple/simple; d=lakerest.net; s=mail; t=1249949859; h=Cc:Message-Id:From:To:In-Reply-To:Content-Type: Content-Transfer-Encoding:Mime-Version:Subject:Date:References: X-Mailer; b=FMMqP3pr2/yFFNbsNk7BWA5lUvVFHFhu1b1X1Wpt+4kJmoJnALkmMFC bW6LljuFPNjOt4LrmbemtQReVfmUpig==
Message-Id: <FAD184A2-08B1-4C56-8728-0E3F91DB78D2@lakerest.net>
From: Randy Stewart <randall@lakerest.net>
To: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>
In-Reply-To: <20090811.085936.183042386.nishida@sfc.wide.ad.jp>
Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes
Content-Transfer-Encoding: 7bit
Mime-Version: 1.0 (Apple Message framework v936)
Date: Mon, 10 Aug 2009 17:17:30 -0700
References: <20090810.171420.104076446.nishida@sfc.wide.ad.jp> <195857BC-4901-4E1E-A39E-139B262CE4B7@lakerest.net> <20090811.085936.183042386.nishida@sfc.wide.ad.jp>
X-Mailer: Apple Mail (2.936)
Cc: tsvwg@ietf.org
Subject: Re: [tsvwg] Q about path failure detection in SCTP
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 11 Aug 2009 00:17:40 -0000

On Aug 10, 2009, at 4:59 PM, Yoshifumi Nishida wrote:

>
> Dear Randy,
>
> Thank you so much for your feedback.
> In my understanding, there is no good way to solve this in the  
> current standard
> without some cooperations from applications.
> The section 6.4 in RFC4960 says:
>
>  By default, an endpoint SHOULD always transmit to the primary path,
>  unless the SCTP user explicitly specifies the destination transport
>  address (and possibly source transport address) to use.
>
> If we follow this, we have to transmit data to the primary path  
> until it has been changed.
> However, the threshold of 'Path.Max.Retrans' prevents to change  
> primary path quickly.
>


True, but I was more thinking of the mechanisms that have to
do with primary path... i.e. when does one failover..... this is
what some of Armando's work talked about. I need to go reread
that sections... and only if its a MUST do you violate the
spec by the way... even though SHOULD means you really really really
should do it unless you have a good reason not to. And if it causes
no harm to switch path's early (as some of Armando's work indicated) I
think we might be able to do it..

But let me look at it tonight when I get back to my hotel room and
have some time to think on it.

R


> Thanks,
> --
> Yoshifumi Nishida
> nishida@sfc.wide.ad.jp
>
>
> From: Randy Stewart <randall@lakerest.net>
> Subject: Re: [tsvwg] Q about path failure detection in SCTP
> Date: Mon, 10 Aug 2009 13:23:08 -0700
> Message-ID: <195857BC-4901-4E1E-A39E-139B262CE4B7@lakerest.net>
>
>> Nishida-san:
>>
>> I think you are probably correct. You can also find
>> some similar ideas put out by Armando Caro in some
>> of his research... don't have the citations at my finger
>> tips..
>>
>> His idea on multiple triggers for failure were I think what
>> Jana pulled in to the CMT work...
>>
>> I am not sure about putting it in the standard I would have
>> to think about if its allowed (it might well be in the current
>> standard)...
>>
>>
>> R
>> On Aug 10, 2009, at 1:14 AM, Yoshifumi Nishida wrote:
>>
>>>
>>> Hello people,
>>>
>>> I sometimes write sctp applications for multihome environments and
>>> notice that
>>> there is always significant long delay (over 30 secs) to switch over
>>> to secondary
>>> path when the primary path becomes unavailable.
>>>
>>> This is because sctp sender tries to retransmit data packets  
>>> (usualy)
>>> 5 times to the
>>> primary path before SCTP considers it is unavailable. In my
>>> understanding, this is
>>> a normal behavior of SCTP since RFC4960 specifies that an endpoint
>>> should always
>>> transmit to the primary path.
>>>
>>> However, it seems to me that this behavior impairs the usability of
>>> SCTP.
>>> For example, if we can utilize the retransmission scheme proposed in
>>> the
>>> CMT paper [1] which use "potentially-failed" state, we might be able
>>> to
>>> switch over to the other path more smoothly.
>>>
>>> I just would like to know if there is any reason that we don't adopt
>>> this kind of
>>> approach in the standard. It seems to me there is no major  
>>> drawback in
>>> this approach.
>>> Or am I misunderstanding something?
>>>
>>> [1]
>>> P. Natarajan, N, Ekiz, P. Amer, R. Stewart,
>>> "Concurrent Multipath Transfer using SCTP Multihoming: Transmission
>>> Policies using
>>> Potentially-failed Destination State", LNCS 4982
>>>
>>> Thanks,
>>> --
>>> Yoshifumi Nishida
>>> nishida@sfc.wide.ad.jp
>>>
>>
>> -----
>> Randall Stewart
>> randall@lakerest.net
>>
>>
>>
>>
>

-----
Randall Stewart
randall@lakerest.net





From Michael.Tuexen@lurchi.franken.de  Tue Aug 11 00:43:23 2009
Return-Path: <Michael.Tuexen@lurchi.franken.de>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id BDBC93A6FA6 for <tsvwg@core3.amsl.com>; Tue, 11 Aug 2009 00:43:23 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 2.223
X-Spam-Level: **
X-Spam-Status: No, score=2.223 tagged_above=-999 required=5 tests=[AWL=0.185,  BAYES_00=-2.599, HOST_EQ_DIP_TDIAL=2.144, HOST_MISMATCH_NET=0.311,  MIME_8BIT_HEADER=0.3, RCVD_IN_PBL=0.905, RCVD_IN_SORBS_DUL=0.877, RDNS_DYNAMIC=0.1]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id AcJJ2ncS9h2h for <tsvwg@core3.amsl.com>; Tue, 11 Aug 2009 00:43:23 -0700 (PDT)
Received: from mail-n.franken.de (drew.ipv6.franken.de [IPv6:2001:638:a02:a001:20e:cff:fe4a:feaa]) by core3.amsl.com (Postfix) with ESMTP id 90C823A6FA4 for <tsvwg@ietf.org>; Tue, 11 Aug 2009 00:43:22 -0700 (PDT)
Received: from [192.168.1.100] (p508FEBBD.dip.t-dialin.net [80.143.235.189]) by mail-n.franken.de (Postfix) with ESMTP id 3A9AA1C0C0BCC; Tue, 11 Aug 2009 09:43:23 +0200 (CEST)
Message-Id: <A996D8C5-D745-4D14-BD6A-FF2C2EA5D67F@lurchi.franken.de>
From: =?ISO-8859-1?Q?Michael_T=FCxen?= <Michael.Tuexen@lurchi.franken.de>
To: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>
In-Reply-To: <20090810.171420.104076446.nishida@sfc.wide.ad.jp>
Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes
Content-Transfer-Encoding: 7bit
Mime-Version: 1.0 (Apple Message framework v936)
Date: Tue, 11 Aug 2009 09:43:22 +0200
References: <20090810.171420.104076446.nishida@sfc.wide.ad.jp>
X-Mailer: Apple Mail (2.936)
Cc: tsvwg@ietf.org
Subject: Re: [tsvwg] Q about path failure detection in SCTP
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 11 Aug 2009 07:43:23 -0000

On Aug 10, 2009, at 10:14 AM, Yoshifumi Nishida wrote:

>
> Hello people,
>
> I sometimes write sctp applications for multihome environments and  
> notice that
> there is always significant long delay (over 30 secs) to switch over  
> to secondary
> path when the primary path becomes unavailable.
>
> This is because sctp sender tries to retransmit data packets  
> (usualy) 5 times to the
> primary path before SCTP considers it is unavailable. In my  
> understanding, this is
> a normal behavior of SCTP since RFC4960 specifies that an endpoint  
> should always
> transmit to the primary path.
Have you you tried to reduce the number of retransmissions before  
considering
a path as failed? This setting does have an impact on the failover time.
>
> However, it seems to me that this behavior impairs the usability of  
> SCTP.
> For example, if we can utilize the retransmission scheme proposed in  
> the
> CMT paper [1] which use "potentially-failed" state, we might be able  
> to
> switch over to the other path more smoothly.
>
> I just would like to know if there is any reason that we don't adopt  
> this kind of
> approach in the standard. It seems to me there is no major drawback  
> in this approach.
> Or am I misunderstanding something?
Getting this adopted needs an analysis if it does not hurt the network,
as the CMT stuff is not standardized due due to congestion control  
issues.
However, this might change in the near future, since there seems to be
a way to resolve this issue.

However, if these technical point is resolved it needs people willing
to write the ID and the TSVWG to adopt it as a WG item.
>
> [1]
> P. Natarajan, N, Ekiz, P. Amer, R. Stewart,
> "Concurrent Multipath Transfer using SCTP Multihoming: Transmission  
> Policies using
> Potentially-failed Destination State", LNCS 4982
>
> Thanks,
> --
> Yoshifumi Nishida
> nishida@sfc.wide.ad.jp
>


From root@core3.amsl.com  Tue Aug 11 01:15:01 2009
Return-Path: <root@core3.amsl.com>
X-Original-To: tsvwg@ietf.org
Delivered-To: tsvwg@core3.amsl.com
Received: by core3.amsl.com (Postfix, from userid 0) id A2F2C3A68F7; Tue, 11 Aug 2009 01:15:01 -0700 (PDT)
From: Internet-Drafts@ietf.org
To: i-d-announce@ietf.org
Content-Type: Multipart/Mixed; Boundary="NextPart"
Mime-Version: 1.0
Message-Id: <20090811081501.A2F2C3A68F7@core3.amsl.com>
Date: Tue, 11 Aug 2009 01:15:01 -0700 (PDT)
Cc: tsvwg@ietf.org
Subject: [tsvwg] I-D Action:draft-ietf-tsvwg-iana-ports-02.txt
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 11 Aug 2009 08:15:01 -0000

--NextPart

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


	Title           : Internet Assigned Numbers Authority (IANA) Procedures for the Management of the Transport Protocol Port Number and Service Name Registry
	Author(s)       : M. Cotton, et al.
	Filename        : draft-ietf-tsvwg-iana-ports-02.txt
	Pages           : 24
	Date            : 2009-08-11

This document defines the procedures that the Internet Assigned
Numbers Authority (IANA) uses when handling registration and other
requests related to the transport protocol port number and service
name registry.  It also discusses the rationale and principles behind
these procedures and how they facilitate the long-term sustainability
of the registry.

This document updates RFC2780 by obsoleting Sections 8 and 9.1 of
that RFC, and it updates the IANA allocation procedures for DCCP as
defined in RFC4340.

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-tsvwg-iana-ports-02.txt

Internet-Drafts are also available by anonymous FTP at:
ftp://ftp.ietf.org/internet-drafts/

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: Message/External-body;
	name="draft-ietf-tsvwg-iana-ports-02.txt";
	site="ftp.ietf.org";
	access-type="anon-ftp";
	directory="internet-drafts"

Content-Type: text/plain
Content-ID: <2009-08-11011113.I-D@ietf.org>


--NextPart--

From nishida@sfc.wide.ad.jp  Tue Aug 11 01:29:45 2009
Return-Path: <nishida@sfc.wide.ad.jp>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 990E53A6E81 for <tsvwg@core3.amsl.com>; Tue, 11 Aug 2009 01:29:45 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 3.06
X-Spam-Level: ***
X-Spam-Status: No, score=3.06 tagged_above=-999 required=5 tests=[AWL=2.156, BAYES_00=-2.599, HELO_EQ_JP=1.244, HOST_EQ_JP=1.265, RELAY_IS_203=0.994]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 4EbxMcBX-dlo for <tsvwg@core3.amsl.com>; Tue, 11 Aug 2009 01:29:44 -0700 (PDT)
Received: from mail.sfc.wide.ad.jp (mail.sfc.wide.ad.jp [203.178.142.146]) by core3.amsl.com (Postfix) with ESMTP id BD5603A6F93 for <tsvwg@ietf.org>; Tue, 11 Aug 2009 01:29:44 -0700 (PDT)
Received: from localhost (cpu.sfc.wide.ad.jp [IPv6:2001:200:0:8803:203:178:142:143]) by mail.sfc.wide.ad.jp (Postfix) with ESMTP id A5B6A4C173; Tue, 11 Aug 2009 17:29:43 +0900 (JST)
Date: Tue, 11 Aug 2009 17:29:43 +0900 (JST)
Message-Id: <20090811.172943.209950324.nishida@sfc.wide.ad.jp>
To: Michael.Tuexen@lurchi.franken.de
From: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>
In-Reply-To: <A996D8C5-D745-4D14-BD6A-FF2C2EA5D67F@lurchi.franken.de>
References: <20090810.171420.104076446.nishida@sfc.wide.ad.jp> <A996D8C5-D745-4D14-BD6A-FF2C2EA5D67F@lurchi.franken.de>
X-Mailer: Mew version 6.2 on Emacs 22.1 / Mule 5.0 (SAKAKI)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable
Cc: tsvwg@ietf.org
Subject: Re: [tsvwg] Q about path failure detection in SCTP
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 11 Aug 2009 08:29:45 -0000

From: Michael T=FCxen <Michael.Tuexen@lurchi.franken.de>
Subject: Re: [tsvwg] Q about path failure detection in SCTP
Date: Tue, 11 Aug 2009 09:43:22 +0200
Message-ID: <A996D8C5-D745-4D14-BD6A-FF2C2EA5D67F@lurchi.franken.de>

 > > This is because sctp sender tries to retransmit data packets (usua=
ly)
 > > 5 times to the
 > > primary path before SCTP considers it is unavailable. In my
 > > understanding, this is
 > > a normal behavior of SCTP since RFC4960 specifies that an endpoint=

 > > should always
 > > transmit to the primary path.
 > Have you you tried to reduce the number of retransmissions before
 > considering
 > a path as failed? This setting does have an impact on the failover
 > time.

Do you mean reducing Path.Max.Retrans? If so, yes. that's an option.
However, it might cause path flapping. also the recommended value in RF=
C4960 is 5.
I think it would be better to have more sophisticated way for this.

 > > I just would like to know if there is any reason that we don't ado=
pt
 > > this kind of
 > > approach in the standard. It seems to me there is no major drawbac=
k in
 > > this approach.
 > > Or am I misunderstanding something?
 > Getting this adopted needs an analysis if it does not hurt the
 > network,
 > as the CMT stuff is not standardized due due to congestion control
 > issues.
 > However, this might change in the near future, since there seems to =
be
 > a way to resolve this issue.
 > However, if these technical point is resolved it needs people willin=
g
 > to write the ID and the TSVWG to adopt it as a WG item.

My rough idea is something like this.

1) If timeout happens on the primary path, set the status of this path =
to "suspicious"
   =

2) If the status of the primary path is "suspicious", send data to the =
secondary,
   while send keep-alive to the primary.

3) If keep-alive ack is received from the primary, set the status of th=
e primay to
   "normal" and back to normal data transmission.
   If keep-alive cannot be received, increase error counter on each tim=
eout.
   If the counter exceeds Path.Max.Retrans, change primary

As you suggest, we need to analyze the impact of 2), but I tend to thin=
k =

it's mostly OK in many cases..

Thanks,
--
Yoshifumi Nishida
nishida@sfc.wide.ad.jp

From Michael.Tuexen@lurchi.franken.de  Tue Aug 11 02:15:55 2009
Return-Path: <Michael.Tuexen@lurchi.franken.de>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id C90E93A6FAF for <tsvwg@core3.amsl.com>; Tue, 11 Aug 2009 02:15:55 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 2.195
X-Spam-Level: **
X-Spam-Status: No, score=2.195 tagged_above=-999 required=5 tests=[AWL=0.157,  BAYES_00=-2.599, HOST_EQ_DIP_TDIAL=2.144, HOST_MISMATCH_NET=0.311,  MIME_8BIT_HEADER=0.3, RCVD_IN_PBL=0.905, RCVD_IN_SORBS_DUL=0.877, RDNS_DYNAMIC=0.1]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 1dDT+pId15Zi for <tsvwg@core3.amsl.com>; Tue, 11 Aug 2009 02:15:55 -0700 (PDT)
Received: from mail-n.franken.de (drew.ipv6.franken.de [IPv6:2001:638:a02:a001:20e:cff:fe4a:feaa]) by core3.amsl.com (Postfix) with ESMTP id 855EF3A6F13 for <tsvwg@ietf.org>; Tue, 11 Aug 2009 02:15:54 -0700 (PDT)
Received: from [192.168.1.100] (p508FE5BB.dip.t-dialin.net [80.143.229.187]) by mail-n.franken.de (Postfix) with ESMTP id 98A211C0C0BCC; Tue, 11 Aug 2009 11:15:56 +0200 (CEST)
Message-Id: <2E52235B-E074-49A2-BAE0-113D4668BECE@lurchi.franken.de>
From: =?ISO-8859-1?Q?Michael_T=FCxen?= <Michael.Tuexen@lurchi.franken.de>
To: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>
In-Reply-To: <20090811.172943.209950324.nishida@sfc.wide.ad.jp>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed; delsp=yes
Content-Transfer-Encoding: quoted-printable
Mime-Version: 1.0 (Apple Message framework v936)
Date: Tue, 11 Aug 2009 11:15:55 +0200
References: <20090810.171420.104076446.nishida@sfc.wide.ad.jp> <A996D8C5-D745-4D14-BD6A-FF2C2EA5D67F@lurchi.franken.de> <20090811.172943.209950324.nishida@sfc.wide.ad.jp>
X-Mailer: Apple Mail (2.936)
Cc: tsvwg@ietf.org
Subject: Re: [tsvwg] Q about path failure detection in SCTP
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 11 Aug 2009 09:15:55 -0000

On Aug 11, 2009, at 10:29 AM, Yoshifumi Nishida wrote:

>
> From: Michael T=FCxen <Michael.Tuexen@lurchi.franken.de>
> Subject: Re: [tsvwg] Q about path failure detection in SCTP
> Date: Tue, 11 Aug 2009 09:43:22 +0200
> Message-ID: <A996D8C5-D745-4D14-BD6A-FF2C2EA5D67F@lurchi.franken.de>
>
>>> This is because sctp sender tries to retransmit data packets =20
>>> (usualy)
>>> 5 times to the
>>> primary path before SCTP considers it is unavailable. In my
>>> understanding, this is
>>> a normal behavior of SCTP since RFC4960 specifies that an endpoint
>>> should always
>>> transmit to the primary path.
>> Have you you tried to reduce the number of retransmissions before
>> considering
>> a path as failed? This setting does have an impact on the failover
>> time.
>
> Do you mean reducing Path.Max.Retrans? If so, yes. that's an option.
> However, it might cause path flapping. also the recommended value in =20=

> RFC4960 is 5.
> I think it would be better to have more sophisticated way for this.
>
>>> I just would like to know if there is any reason that we don't adopt
>>> this kind of
>>> approach in the standard. It seems to me there is no major =20
>>> drawback in
>>> this approach.
>>> Or am I misunderstanding something?
>> Getting this adopted needs an analysis if it does not hurt the
>> network,
>> as the CMT stuff is not standardized due due to congestion control
>> issues.
>> However, this might change in the near future, since there seems to =20=

>> be
>> a way to resolve this issue.
>> However, if these technical point is resolved it needs people willing
>> to write the ID and the TSVWG to adopt it as a WG item.
>
> My rough idea is something like this.
>
> 1) If timeout happens on the primary path, set the status of this =20
> path to "suspicious"
So a path having experienced a congestion indication (timeout) is =20
suspicious?
>
> 2) If the status of the primary path is "suspicious", send data to =20
> the secondary,
>   while send keep-alive to the primary.
>
> 3) If keep-alive ack is received from the primary, set the status of =20=

> the primay to
>   "normal" and back to normal data transmission.
>   If keep-alive cannot be received, increase error counter on each =20
> timeout.
>   If the counter exceeds Path.Max.Retrans, change primary
>
> As you suggest, we need to analyze the impact of 2), but I tend to =20
> think
> it's mostly OK in many cases..
>
> Thanks,
> --
> Yoshifumi Nishida
> nishida@sfc.wide.ad.jp
>


From randall@lakerest.net  Tue Aug 11 06:28:27 2009
Return-Path: <randall@lakerest.net>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id E25523A70B6 for <tsvwg@core3.amsl.com>; Tue, 11 Aug 2009 06:28:27 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.299
X-Spam-Level: 
X-Spam-Status: No, score=-2.299 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, MIME_8BIT_HEADER=0.3]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id T3bJPOh37ZRf for <tsvwg@core3.amsl.com>; Tue, 11 Aug 2009 06:28:27 -0700 (PDT)
Received: from lakerest.net (unknown [IPv6:2001:240:585:2:213:d4ff:fef3:2d8d]) by core3.amsl.com (Postfix) with ESMTP id 834143A70B1 for <tsvwg@ietf.org>; Tue, 11 Aug 2009 06:28:26 -0700 (PDT)
Received: from [12.236.17.84] ([12.236.17.84]) (authenticated bits=0) by lakerest.net (8.14.3/8.14.3) with ESMTP id n7BDSF7G089245 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Tue, 11 Aug 2009 09:28:18 -0400 (EDT) (envelope-from randall@lakerest.net)
DKIM-Signature: a=rsa-sha1; c=simple/simple; d=lakerest.net; s=mail; t=1249997299; h=Cc:Message-Id:From:To:In-Reply-To:Content-Type: Content-Transfer-Encoding:Mime-Version:Subject:Date:References: X-Mailer; b=U5fq8gzrlhB2JPwdTqrddbO+kTLm1FEgphQbakFVWlXkXM8ft6eLU7/ yj16/fb8WS73mOXPhyKC4KE4Hj5fkdw==
Message-Id: <A2E2E510-8BE3-4F64-AEC9-B8E0FBDCDAE3@lakerest.net>
From: Randy Stewart <randall@lakerest.net>
To: =?ISO-8859-1?Q?Michael_T=FCxen?= <Michael.Tuexen@lurchi.franken.de>
In-Reply-To: <2E52235B-E074-49A2-BAE0-113D4668BECE@lurchi.franken.de>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed; delsp=yes
Content-Transfer-Encoding: quoted-printable
Mime-Version: 1.0 (Apple Message framework v936)
Date: Tue, 11 Aug 2009 06:28:14 -0700
References: <20090810.171420.104076446.nishida@sfc.wide.ad.jp> <A996D8C5-D745-4D14-BD6A-FF2C2EA5D67F@lurchi.franken.de> <20090811.172943.209950324.nishida@sfc.wide.ad.jp> <2E52235B-E074-49A2-BAE0-113D4668BECE@lurchi.franken.de>
X-Mailer: Apple Mail (2.936)
Cc: tsvwg@ietf.org
Subject: Re: [tsvwg] Q about path failure detection in SCTP
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 11 Aug 2009 13:28:28 -0000

On Aug 11, 2009, at 2:15 AM, Michael T=FCxen wrote:

> On Aug 11, 2009, at 10:29 AM, Yoshifumi Nishida wrote:
>
>>
>> From: Michael T=FCxen <Michael.Tuexen@lurchi.franken.de>
>> Subject: Re: [tsvwg] Q about path failure detection in SCTP
>> Date: Tue, 11 Aug 2009 09:43:22 +0200
>> Message-ID: <A996D8C5-D745-4D14-BD6A-FF2C2EA5D67F@lurchi.franken.de>
>>
>>>> This is because sctp sender tries to retransmit data packets =20
>>>> (usualy)
>>>> 5 times to the
>>>> primary path before SCTP considers it is unavailable. In my
>>>> understanding, this is
>>>> a normal behavior of SCTP since RFC4960 specifies that an endpoint
>>>> should always
>>>> transmit to the primary path.
>>> Have you you tried to reduce the number of retransmissions before
>>> considering
>>> a path as failed? This setting does have an impact on the failover
>>> time.
>>
>> Do you mean reducing Path.Max.Retrans? If so, yes. that's an option.
>> However, it might cause path flapping. also the recommended value =20
>> in RFC4960 is 5.
>> I think it would be better to have more sophisticated way for this.
>>
>>>> I just would like to know if there is any reason that we don't =20
>>>> adopt
>>>> this kind of
>>>> approach in the standard. It seems to me there is no major =20
>>>> drawback in
>>>> this approach.
>>>> Or am I misunderstanding something?
>>> Getting this adopted needs an analysis if it does not hurt the
>>> network,
>>> as the CMT stuff is not standardized due due to congestion control
>>> issues.
>>> However, this might change in the near future, since there seems =20
>>> to be
>>> a way to resolve this issue.
>>> However, if these technical point is resolved it needs people =20
>>> willing
>>> to write the ID and the TSVWG to adopt it as a WG item.
>>
>> My rough idea is something like this.
>>
>> 1) If timeout happens on the primary path, set the status of this =20
>> path to "suspicious"
> So a path having experienced a congestion indication (timeout) is =20
> suspicious?


And one could end up with bouncing paths...


I believe Armando's work actually had this "suspicious" state but what
he did is switch the primary to the new one... and did not switch back.
That kept the bouncing effect.... which by the way might limit =20
throughput...

R

>>
>> 2) If the status of the primary path is "suspicious", send data to =20=

>> the secondary,
>>  while send keep-alive to the primary.
>>
>> 3) If keep-alive ack is received from the primary, set the status =20
>> of the primay to
>>  "normal" and back to normal data transmission.
>>  If keep-alive cannot be received, increase error counter on each =20
>> timeout.
>>  If the counter exceeds Path.Max.Retrans, change primary
>>
>> As you suggest, we need to analyze the impact of 2), but I tend to =20=

>> think
>> it's mostly OK in many cases..
>>
>> Thanks,
>> --
>> Yoshifumi Nishida
>> nishida@sfc.wide.ad.jp
>>
>

-----
Randall Stewart
randall@lakerest.net





From nishida@sfc.wide.ad.jp  Tue Aug 11 11:58:42 2009
Return-Path: <nishida@sfc.wide.ad.jp>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 6024B3A6FF7 for <tsvwg@core3.amsl.com>; Tue, 11 Aug 2009 11:58:42 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 2.881
X-Spam-Level: **
X-Spam-Status: No, score=2.881 tagged_above=-999 required=5 tests=[AWL=1.977,  BAYES_00=-2.599, HELO_EQ_JP=1.244, HOST_EQ_JP=1.265, RELAY_IS_203=0.994]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id bHZCSd0bVL8Q for <tsvwg@core3.amsl.com>; Tue, 11 Aug 2009 11:58:41 -0700 (PDT)
Received: from mail.sfc.wide.ad.jp (mail.sfc.wide.ad.jp [203.178.142.146]) by core3.amsl.com (Postfix) with ESMTP id 820513A6FF2 for <tsvwg@ietf.org>; Tue, 11 Aug 2009 11:58:41 -0700 (PDT)
Received: from localhost (cpu.sfc.wide.ad.jp [IPv6:2001:200:0:8803:203:178:142:143]) by mail.sfc.wide.ad.jp (Postfix) with ESMTP id 866584C0AC; Wed, 12 Aug 2009 03:57:40 +0900 (JST)
Date: Wed, 12 Aug 2009 03:57:40 +0900 (JST)
Message-Id: <20090812.035740.189643280.nishida@sfc.wide.ad.jp>
To: randall@lakerest.net
From: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>
In-Reply-To: <A2E2E510-8BE3-4F64-AEC9-B8E0FBDCDAE3@lakerest.net>
References: <20090811.172943.209950324.nishida@sfc.wide.ad.jp> <2E52235B-E074-49A2-BAE0-113D4668BECE@lurchi.franken.de> <A2E2E510-8BE3-4F64-AEC9-B8E0FBDCDAE3@lakerest.net>
X-Mailer: Mew version 6.2 on Emacs 22.1 / Mule 5.0 (SAKAKI)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Cc: Michael.Tuexen@lurchi.franken.de, tsvwg@ietf.org
Subject: Re: [tsvwg] Q about path failure detection in SCTP
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 11 Aug 2009 18:58:42 -0000

From: Randy Stewart <randall@lakerest.net>
Subject: Re: [tsvwg] Q about path failure detection in SCTP
Date: Tue, 11 Aug 2009 06:28:14 -0700
Message-ID: <A2E2E510-8BE3-4F64-AEC9-B8E0FBDCDAE3@lakerest.net>

 > >> My rough idea is something like this.
 > >>
 > >> 1) If timeout happens on the primary path, set the status of this path
 > >> to "suspicious"

 > > So a path having experienced a congestion indication (timeout) is
 > > suspicious?
 > 
 > And one could end up with bouncing paths...

Yes, and my point is I would like to discuss bouncing more.

Let's say, the primary path is a bit lossy and causes timeout (congestions or whatever).
In case of normal SCTP, retransmit packets will be transmited to secondary.

   "Furthermore, when its peer is multi-homed, an endpoint SHOULD try to
   retransmit a chunk that timed out to an active destination transport
   address that is different from the last destination address to which
   the DATA chunk was sent."

Do we call this bouncing?  Probably not.
What I'm going to propose is, when packet is retransmitted to the secondary, heart-beat is 
transmitted to the primary. If we can receive heart-beat ACK, then use primary for the next 
transmission. Otherwise, use secondary for the next transmission. 
I tend to believe this approach does not cause much bouncing effects. It behaves mostly 
like the current SCTP in case of congestion. In addition, by sending small heart-beat 
instead of data chunk to potentially congested path, it can reduce the risk of congestion 
collapse.

Thanks,
--
Yoshifumi Nishida
nishida@sfc.wide.ad.jp

From dlehmann@ulticom.com  Tue Aug 11 12:29:54 2009
Return-Path: <dlehmann@ulticom.com>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 059D13A6FE8 for <tsvwg@core3.amsl.com>; Tue, 11 Aug 2009 12:29:54 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.599
X-Spam-Level: 
X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id N1u8M4j08YWB for <tsvwg@core3.amsl.com>; Tue, 11 Aug 2009 12:29:53 -0700 (PDT)
Received: from bw.ulticom.com (bw.ulticom.com [208.255.120.38]) by core3.amsl.com (Postfix) with ESMTP id 0D34A3A635F for <tsvwg@ietf.org>; Tue, 11 Aug 2009 12:29:00 -0700 (PDT)
Received: from colby.ulticom.com (colby.ulticom.com [192.73.206.10]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by bw.ulticom.com (BorderWare MXtreme Mail Firewall) with ESMTP id 7A2762495E; Tue, 11 Aug 2009 15:27:47 -0400 (EDT)
Received: from MTLEXVS01.ulticom.com (mtlex01.ulticom.com [172.16.40.5]) by colby.ulticom.com (8.13.4/8.12.10) with ESMTP id n7BJRebX025576; Tue, 11 Aug 2009 15:27:47 -0400 (EDT)
X-MimeOLE: Produced By Microsoft Exchange V6.5
Content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
Date: Tue, 11 Aug 2009 15:27:39 -0400
Message-ID: <A51D8ACD861B7E41BFC7FE5C64BE96480DC42BA7@MTLEXVS01.ulticom.com>
In-Reply-To: <20090812.035740.189643280.nishida@sfc.wide.ad.jp>
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Thread-Topic: [tsvwg] Q about path failure detection in SCTP
Thread-Index: AcoatfAMI1ZgXk8BS5uwP8gfp2AbYwAAL59A
References: <20090811.172943.209950324.nishida@sfc.wide.ad.jp><2E52235B-E074-49A2-BAE0-113D4668BECE@lurchi.franken.de><A2E2E510-8BE3-4F64-AEC9-B8E0FBDCDAE3@lakerest.net> <20090812.035740.189643280.nishida@sfc.wide.ad.jp>
From: "David Lehmann" <dlehmann@ulticom.com>
To: "Yoshifumi Nishida" <nishida@sfc.wide.ad.jp>, <randall@lakerest.net>
Received-SPF: none
Cc: Michael.Tuexen@lurchi.franken.de, tsvwg@ietf.org
Subject: Re: [tsvwg] Q about path failure detection in SCTP
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 11 Aug 2009 19:29:54 -0000

Yoshifumi raised a good point.

Waiting for the primary destination to time out doesn't fly with telecom
apps.  Our implementation has being doing something similar for years to
what Yoshifumi has suggested.  Our algorithm checks the primary path.
If the error count is zero, use it.  Otherwise, look for a path with the
least amount of errors.  As Randall mentioned, the spec says "SHOULD";
so it is RFC compliant.

As soon as the error is cleared, the primary will automatically be used,
given the above algorithm. (When an alternate path is used, that doesn't
change the primary.  The primary can only be changed by the user.)

Yes, a heartbeat will speed recovery of the primary path. (I think it
was Armando who suggested this to me at the UofD bake off.) Fast
recovery of the primary path seems to help throughput possibly because
data is being sent back on the primary path which will have a larger
cwnd.

We have seen no down side for this approach.  In tests, when we pulled
the Ethernet cable used for the primary path, we hardly notice a blip in
the throughput.

--
David Lehmann
Ulticom, Inc.
856-787-2952

> -----Original Message-----
> From: tsvwg-bounces@ietf.org [mailto:tsvwg-bounces@ietf.org] On Behalf
> Of Yoshifumi Nishida
> Sent: Tuesday, August 11, 2009 2:58 PM
> To: randall@lakerest.net
> Cc: Michael.Tuexen@lurchi.franken.de; tsvwg@ietf.org
> Subject: Re: [tsvwg] Q about path failure detection in SCTP
>=20
>=20
> From: Randy Stewart <randall@lakerest.net>
> Subject: Re: [tsvwg] Q about path failure detection in SCTP
> Date: Tue, 11 Aug 2009 06:28:14 -0700
> Message-ID: <A2E2E510-8BE3-4F64-AEC9-B8E0FBDCDAE3@lakerest.net>
>=20
>  > >> My rough idea is something like this.
>  > >>
>  > >> 1) If timeout happens on the primary path, set the status of
this
> path
>  > >> to "suspicious"
>=20
>  > > So a path having experienced a congestion indication (timeout) is
>  > > suspicious?
>  >
>  > And one could end up with bouncing paths...
>=20
> Yes, and my point is I would like to discuss bouncing more.
>=20
> Let's say, the primary path is a bit lossy and causes timeout
> (congestions or whatever).
> In case of normal SCTP, retransmit packets will be transmited to
> secondary.
>=20
>    "Furthermore, when its peer is multi-homed, an endpoint SHOULD try
to
>    retransmit a chunk that timed out to an active destination
transport
>    address that is different from the last destination address to
which
>    the DATA chunk was sent."
>=20
> Do we call this bouncing?  Probably not.
> What I'm going to propose is, when packet is retransmitted to the
> secondary, heart-beat is
> transmitted to the primary. If we can receive heart-beat ACK, then use
> primary for the next
> transmission. Otherwise, use secondary for the next transmission.
> I tend to believe this approach does not cause much bouncing effects.
It
> behaves mostly
> like the current SCTP in case of congestion. In addition, by sending
> small heart-beat
> instead of data chunk to potentially congested path, it can reduce the
> risk of congestion
> collapse.
>=20
> Thanks,
> --
> Yoshifumi Nishida
> nishida@sfc.wide.ad.jp

From Michael.Tuexen@lurchi.franken.de  Tue Aug 11 12:50:42 2009
Return-Path: <Michael.Tuexen@lurchi.franken.de>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 058083A6F67 for <tsvwg@core3.amsl.com>; Tue, 11 Aug 2009 12:50:42 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 2.184
X-Spam-Level: **
X-Spam-Status: No, score=2.184 tagged_above=-999 required=5 tests=[AWL=0.146,  BAYES_00=-2.599, HOST_EQ_DIP_TDIAL=2.144, HOST_MISMATCH_NET=0.311,  MIME_8BIT_HEADER=0.3, RCVD_IN_PBL=0.905, RCVD_IN_SORBS_DUL=0.877, RDNS_DYNAMIC=0.1]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id vZVdA4ewGal4 for <tsvwg@core3.amsl.com>; Tue, 11 Aug 2009 12:50:41 -0700 (PDT)
Received: from mail-n.franken.de (drew.ipv6.franken.de [IPv6:2001:638:a02:a001:20e:cff:fe4a:feaa]) by core3.amsl.com (Postfix) with ESMTP id 10D2E3A6F7E for <tsvwg@ietf.org>; Tue, 11 Aug 2009 12:50:39 -0700 (PDT)
Received: from [192.168.1.100] (p508FE5BB.dip.t-dialin.net [80.143.229.187]) by mail-n.franken.de (Postfix) with ESMTP id 56B531C0C0BCC; Tue, 11 Aug 2009 21:50:39 +0200 (CEST)
Message-Id: <932B0EC1-B00C-4F0D-9876-F0F37D079524@lurchi.franken.de>
From: =?ISO-8859-1?Q?Michael_T=FCxen?= <Michael.Tuexen@lurchi.franken.de>
To: David Lehmann <dlehmann@ulticom.com>
In-Reply-To: <A51D8ACD861B7E41BFC7FE5C64BE96480DC42BA7@MTLEXVS01.ulticom.com>
Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes
Content-Transfer-Encoding: 7bit
Mime-Version: 1.0 (Apple Message framework v936)
Date: Tue, 11 Aug 2009 21:50:38 +0200
References: <20090811.172943.209950324.nishida@sfc.wide.ad.jp><2E52235B-E074-49A2-BAE0-113D4668BECE@lurchi.franken.de><A2E2E510-8BE3-4F64-AEC9-B8E0FBDCDAE3@lakerest.net> <20090812.035740.189643280.nishida@sfc.wide.ad.jp> <A51D8ACD861B7E41BFC7FE5C64BE96480DC42BA7@MTLEXVS01.ulticom.com>
X-Mailer: Apple Mail (2.936)
Cc: tsvwg@ietf.org, randall@lakerest.net
Subject: Re: [tsvwg] Q about path failure detection in SCTP
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 11 Aug 2009 19:50:42 -0000

On Aug 11, 2009, at 9:27 PM, David Lehmann wrote:

> Yoshifumi raised a good point.
>
> Waiting for the primary destination to time out doesn't fly with  
> telecom
> apps.  Our implementation has being doing something similar for  
> years to
> what Yoshifumi has suggested.  Our algorithm checks the primary path.
> If the error count is zero, use it.  Otherwise, look for a path with  
> the
> least amount of errors.  As Randall mentioned, the spec says "SHOULD";
> so it is RFC compliant.
This might be a good idea in a telco environment where the different
paths share no common link. But if for example all paths share a single
link and this one is responsable for the loss observed, switching
to a different path is counter productive.

The best way to handle it is to use all paths simultaneously, but
that is not yet standardized.
>
> As soon as the error is cleared, the primary will automatically be  
> used,
> given the above algorithm. (When an alternate path is used, that  
> doesn't
> change the primary.  The primary can only be changed by the user.)
>
> Yes, a heartbeat will speed recovery of the primary path. (I think it
> was Armando who suggested this to me at the UofD bake off.) Fast
> recovery of the primary path seems to help throughput possibly because
> data is being sent back on the primary path which will have a larger
> cwnd.
>
> We have seen no down side for this approach.  In tests, when we pulled
> the Ethernet cable used for the primary path, we hardly notice a  
> blip in
> the throughput.
>
> --
> David Lehmann
> Ulticom, Inc.
> 856-787-2952
>
>> -----Original Message-----
>> From: tsvwg-bounces@ietf.org [mailto:tsvwg-bounces@ietf.org] On  
>> Behalf
>> Of Yoshifumi Nishida
>> Sent: Tuesday, August 11, 2009 2:58 PM
>> To: randall@lakerest.net
>> Cc: Michael.Tuexen@lurchi.franken.de; tsvwg@ietf.org
>> Subject: Re: [tsvwg] Q about path failure detection in SCTP
>>
>>
>> From: Randy Stewart <randall@lakerest.net>
>> Subject: Re: [tsvwg] Q about path failure detection in SCTP
>> Date: Tue, 11 Aug 2009 06:28:14 -0700
>> Message-ID: <A2E2E510-8BE3-4F64-AEC9-B8E0FBDCDAE3@lakerest.net>
>>
>>>>> My rough idea is something like this.
>>>>>
>>>>> 1) If timeout happens on the primary path, set the status of
> this
>> path
>>>>> to "suspicious"
>>
>>>> So a path having experienced a congestion indication (timeout) is
>>>> suspicious?
>>>
>>> And one could end up with bouncing paths...
>>
>> Yes, and my point is I would like to discuss bouncing more.
>>
>> Let's say, the primary path is a bit lossy and causes timeout
>> (congestions or whatever).
>> In case of normal SCTP, retransmit packets will be transmited to
>> secondary.
>>
>>   "Furthermore, when its peer is multi-homed, an endpoint SHOULD try
> to
>>   retransmit a chunk that timed out to an active destination
> transport
>>   address that is different from the last destination address to
> which
>>   the DATA chunk was sent."
>>
>> Do we call this bouncing?  Probably not.
>> What I'm going to propose is, when packet is retransmitted to the
>> secondary, heart-beat is
>> transmitted to the primary. If we can receive heart-beat ACK, then  
>> use
>> primary for the next
>> transmission. Otherwise, use secondary for the next transmission.
>> I tend to believe this approach does not cause much bouncing effects.
> It
>> behaves mostly
>> like the current SCTP in case of congestion. In addition, by sending
>> small heart-beat
>> instead of data chunk to potentially congested path, it can reduce  
>> the
>> risk of congestion
>> collapse.
>>
>> Thanks,
>> --
>> Yoshifumi Nishida
>> nishida@sfc.wide.ad.jp
>


From nishida@sfc.wide.ad.jp  Tue Aug 11 22:02:52 2009
Return-Path: <nishida@sfc.wide.ad.jp>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 182D83A6A43 for <tsvwg@core3.amsl.com>; Tue, 11 Aug 2009 22:02:52 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 2.729
X-Spam-Level: **
X-Spam-Status: No, score=2.729 tagged_above=-999 required=5 tests=[AWL=1.825,  BAYES_00=-2.599, HELO_EQ_JP=1.244, HOST_EQ_JP=1.265, RELAY_IS_203=0.994]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 2JqZ98HmxjmT for <tsvwg@core3.amsl.com>; Tue, 11 Aug 2009 22:02:51 -0700 (PDT)
Received: from mail.sfc.wide.ad.jp (mail.sfc.wide.ad.jp [203.178.142.146]) by core3.amsl.com (Postfix) with ESMTP id 395153A6855 for <tsvwg@ietf.org>; Tue, 11 Aug 2009 22:02:50 -0700 (PDT)
Received: from localhost (cpu.sfc.wide.ad.jp [IPv6:2001:200:0:8803:203:178:142:143]) by mail.sfc.wide.ad.jp (Postfix) with ESMTP id 683554C173; Wed, 12 Aug 2009 14:01:34 +0900 (JST)
Date: Wed, 12 Aug 2009 14:01:34 +0900 (JST)
Message-Id: <20090812.140134.183090419.nishida@sfc.wide.ad.jp>
To: Michael.Tuexen@lurchi.franken.de
From: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>
In-Reply-To: <932B0EC1-B00C-4F0D-9876-F0F37D079524@lurchi.franken.de>
References: <20090812.035740.189643280.nishida@sfc.wide.ad.jp> <A51D8ACD861B7E41BFC7FE5C64BE96480DC42BA7@MTLEXVS01.ulticom.com> <932B0EC1-B00C-4F0D-9876-F0F37D079524@lurchi.franken.de>
X-Mailer: Mew version 6.2 on Emacs 22.1 / Mule 5.0 (SAKAKI)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable
Cc: tsvwg@ietf.org, randall@lakerest.net
Subject: Re: [tsvwg] Q about path failure detection in SCTP
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 12 Aug 2009 05:02:52 -0000

From: Michael T=FCxen <Michael.Tuexen@lurchi.franken.de>
Subject: Re: [tsvwg] Q about path failure detection in SCTP
Date: Tue, 11 Aug 2009 21:50:38 +0200
Message-ID: <932B0EC1-B00C-4F0D-9876-F0F37D079524@lurchi.franken.de>

 > On Aug 11, 2009, at 9:27 PM, David Lehmann wrote:
 > =

 > > Yoshifumi raised a good point.
 > >
 > > Waiting for the primary destination to time out doesn't fly with
 > > telecom
 > > apps.  Our implementation has being doing something similar for ye=
ars
 > > to
 > > what Yoshifumi has suggested.  Our algorithm checks the primary pa=
th.
 > > If the error count is zero, use it.  Otherwise, look for a path wi=
th
 > > the
 > > least amount of errors.  As Randall mentioned, the spec says "SHOU=
LD";
 > > so it is RFC compliant.

 > This might be a good idea in a telco environment where the different=

 > paths share no common link. But if for example all paths share a
 > single
 > link and this one is responsable for the loss observed, switching
 > to a different path is counter productive.

I think it's not very harmful even if all paths share a single link.

If all paths share a single link and the shared link is congested, =

even with normal SCTP, it tries to retransmit data to the non-primary p=
aths.
Hence, the data go toward the shared congested link anyhow.

The slight difference between the current SCTP and my proposal is heart=
-beat
packets transmitted to the primary path. Yes, this is a negative factor=

in this approach, but I tend to believe it's mostly acceptable.

CMT might be the ideal goal, but I would like to think what we can do o=
n =

the current SCTP with minimal change.

Thanks,
--
Yoshifumi Nishida
nishida@sfc.wide.ad.jp


From lars.eggert@nokia.com  Tue Aug 11 23:58:01 2009
Return-Path: <lars.eggert@nokia.com>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 787B93A692A for <tsvwg@core3.amsl.com>; Tue, 11 Aug 2009 23:58:01 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.422
X-Spam-Level: 
X-Spam-Status: No, score=-2.422 tagged_above=-999 required=5 tests=[AWL=0.177,  BAYES_00=-2.599]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 3y97sHUAsxwG for <tsvwg@core3.amsl.com>; Tue, 11 Aug 2009 23:58:00 -0700 (PDT)
Received: from mail.fit.nokia.com (mail.fit.nokia.com [195.148.124.195]) by core3.amsl.com (Postfix) with ESMTP id 69BB03A6821 for <tsvwg@ietf.org>; Tue, 11 Aug 2009 23:58:00 -0700 (PDT)
Received: from [10.0.1.5] (cs95024.pp.htv.fi [212.90.95.24]) (authenticated bits=0) by mail.fit.nokia.com (8.14.3/8.14.3) with ESMTP id n7C6ukOx030071 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT) for <tsvwg@ietf.org>; Wed, 12 Aug 2009 09:56:46 +0300 (EEST) (envelope-from lars.eggert@nokia.com)
Message-Id: <F46ED6BF-5DB6-414A-9E81-9F68D73EA525@nokia.com>
From: Lars Eggert <lars.eggert@nokia.com>
To: tsvwg <tsvwg@ietf.org>
In-Reply-To: <20090811081501.A2F2C3A68F7@core3.amsl.com>
Content-Type: multipart/signed; boundary=Apple-Mail-10--977417048; micalg=sha1; protocol="application/pkcs7-signature"
Mime-Version: 1.0 (Apple Message framework v936)
Date: Wed, 12 Aug 2009 09:56:46 +0300
References: <20090811081501.A2F2C3A68F7@core3.amsl.com>
X-Mailer: Apple Mail (2.936)
X-Greylist: Sender succeeded SMTP AUTH, not delayed by milter-greylist-4.2.2 (mail.fit.nokia.com [195.148.124.194]); Wed, 12 Aug 2009 09:56:46 +0300 (EEST)
Subject: Re: [tsvwg] I-D Action:draft-ietf-tsvwg-iana-ports-02.txt
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 12 Aug 2009 06:58:01 -0000

--Apple-Mail-10--977417048
Content-Type: text/plain;
	charset=US-ASCII;
	format=flowed;
	delsp=yes
Content-Transfer-Encoding: 7bit

Hi, WG,

On 2009-8-11, at 11:15, Internet-Drafts@ietf.org wrote:
> 	Title           : Internet Assigned Numbers Authority (IANA)  
> Procedures for the Management of the Transport Protocol Port Number  
> and Service Name Registry
> 	Author(s)       : M. Cotton, et al.
> 	Filename        : draft-ietf-tsvwg-iana-ports-02.txt
> 	Pages           : 24
> 	Date            : 2009-08-11

this is a pretty significant update based on discussions during the  
last two IETFs. In my personal opinion, this document is nearing  
completion and would benefit from careful review.

Lars
--Apple-Mail-10--977417048
Content-Disposition: attachment;
	filename=smime.p7s
Content-Type: application/pkcs7-signature;
	name=smime.p7s
Content-Transfer-Encoding: base64
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==

--Apple-Mail-10--977417048--

From dlehmann@ulticom.com  Wed Aug 12 05:58:33 2009
Return-Path: <dlehmann@ulticom.com>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 7BDFB3A6B30 for <tsvwg@core3.amsl.com>; Wed, 12 Aug 2009 05:58:33 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.599
X-Spam-Level: 
X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id vvYSUrH62zYO for <tsvwg@core3.amsl.com>; Wed, 12 Aug 2009 05:58:32 -0700 (PDT)
Received: from bw.ulticom.com (bw.ulticom.com [208.255.120.38]) by core3.amsl.com (Postfix) with ESMTP id 4E2403A6805 for <tsvwg@ietf.org>; Wed, 12 Aug 2009 05:58:32 -0700 (PDT)
Received: from colby.ulticom.com (colby.ulticom.com [192.73.206.10]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by bw.ulticom.com (BorderWare MXtreme Mail Firewall) with ESMTP id 259757E3FE for <tsvwg@ietf.org>; Wed, 12 Aug 2009 08:57:01 -0400 (EDT)
Received: from MTLEXVS01.ulticom.com (mtlex01.ulticom.com [172.16.40.5]) by colby.ulticom.com (8.13.4/8.12.10) with ESMTP id n7CCut0C025117 for <tsvwg@ietf.org>; Wed, 12 Aug 2009 08:57:01 -0400 (EDT)
X-MimeOLE: Produced By Microsoft Exchange V6.5
Content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Date: Wed, 12 Aug 2009 08:56:54 -0400
Message-ID: <A51D8ACD861B7E41BFC7FE5C64BE96480DC42D51@MTLEXVS01.ulticom.com>
In-Reply-To: <932B0EC1-B00C-4F0D-9876-F0F37D079524@lurchi.franken.de>
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Thread-Topic: [tsvwg] Q about path failure detection in SCTP
Thread-Index: AcoavQREI5t3GAotTaGW4FRb56eVRQAjTauw
References: <20090811.172943.209950324.nishida@sfc.wide.ad.jp><2E52235B-E074-49A2-BAE0-113D4668BECE@lurchi.franken.de><A2E2E510-8BE3-4F64-AEC9-B8E0FBDCDAE3@lakerest.net> <20090812.035740.189643280.nishida@sfc.wide.ad.jp> <A51D8ACD861B7E41BFC7FE5C64BE96480DC42BA7@MTLEXVS01.ulticom.com> <932B0EC1-B00C-4F0D-9876-F0F37D079524@lurchi.franken.de>
From: "David Lehmann" <dlehmann@ulticom.com>
To: <tsvwg@ietf.org>
Received-SPF: none
Subject: Re: [tsvwg] Q about path failure detection in SCTP
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 12 Aug 2009 12:58:33 -0000

I would think that most apps (not just telco apps) would benefit from =
and _expect_ a quick (i.e. hardly noticeable) failover when using =
multihomed endpoints.

As far as being "counter productive", waiting 30 seconds to failover to =
an alternate path is counter productive.  I would venture to guess that =
this characteristic is unacceptable to most apps.

--
David Lehmann
Ulticom, Inc.
856-787-2952

> -----Original Message-----
> From: Michael T=FCxen [mailto:Michael.Tuexen@lurchi.franken.de]
> Sent: Tuesday, August 11, 2009 3:51 PM
> To: David Lehmann
> Cc: Yoshifumi Nishida; randall@lakerest.net; tsvwg@ietf.org
> Subject: Re: [tsvwg] Q about path failure detection in SCTP
>=20
> On Aug 11, 2009, at 9:27 PM, David Lehmann wrote:
>=20
> > Yoshifumi raised a good point.
> >
> > Waiting for the primary destination to time out doesn't fly with
> > telecom
> > apps.  Our implementation has being doing something similar for
> > years to
> > what Yoshifumi has suggested.  Our algorithm checks the primary =
path.
> > If the error count is zero, use it.  Otherwise, look for a path with
> > the
> > least amount of errors.  As Randall mentioned, the spec says =
"SHOULD";
> > so it is RFC compliant.
> This might be a good idea in a telco environment where the different
> paths share no common link. But if for example all paths share a =
single
> link and this one is responsable for the loss observed, switching
> to a different path is counter productive.
>=20
> The best way to handle it is to use all paths simultaneously, but
> that is not yet standardized.
> >
> > As soon as the error is cleared, the primary will automatically be
> > used,
> > given the above algorithm. (When an alternate path is used, that
> > doesn't
> > change the primary.  The primary can only be changed by the user.)
> >
> > Yes, a heartbeat will speed recovery of the primary path. (I think =
it
> > was Armando who suggested this to me at the UofD bake off.) Fast
> > recovery of the primary path seems to help throughput possibly =
because
> > data is being sent back on the primary path which will have a larger
> > cwnd.
> >
> > We have seen no down side for this approach.  In tests, when we =
pulled
> > the Ethernet cable used for the primary path, we hardly notice a
> > blip in
> > the throughput.
> >
> > --
> > David Lehmann
> > Ulticom, Inc.
> > 856-787-2952
> >
> >> -----Original Message-----
> >> From: tsvwg-bounces@ietf.org [mailto:tsvwg-bounces@ietf.org] On
> >> Behalf
> >> Of Yoshifumi Nishida
> >> Sent: Tuesday, August 11, 2009 2:58 PM
> >> To: randall@lakerest.net
> >> Cc: Michael.Tuexen@lurchi.franken.de; tsvwg@ietf.org
> >> Subject: Re: [tsvwg] Q about path failure detection in SCTP
> >>
> >>
> >> From: Randy Stewart <randall@lakerest.net>
> >> Subject: Re: [tsvwg] Q about path failure detection in SCTP
> >> Date: Tue, 11 Aug 2009 06:28:14 -0700
> >> Message-ID: <A2E2E510-8BE3-4F64-AEC9-B8E0FBDCDAE3@lakerest.net>
> >>
> >>>>> My rough idea is something like this.
> >>>>>
> >>>>> 1) If timeout happens on the primary path, set the status of
> > this
> >> path
> >>>>> to "suspicious"
> >>
> >>>> So a path having experienced a congestion indication (timeout) is
> >>>> suspicious?
> >>>
> >>> And one could end up with bouncing paths...
> >>
> >> Yes, and my point is I would like to discuss bouncing more.
> >>
> >> Let's say, the primary path is a bit lossy and causes timeout
> >> (congestions or whatever).
> >> In case of normal SCTP, retransmit packets will be transmited to
> >> secondary.
> >>
> >>   "Furthermore, when its peer is multi-homed, an endpoint SHOULD =
try
> > to
> >>   retransmit a chunk that timed out to an active destination
> > transport
> >>   address that is different from the last destination address to
> > which
> >>   the DATA chunk was sent."
> >>
> >> Do we call this bouncing?  Probably not.
> >> What I'm going to propose is, when packet is retransmitted to the
> >> secondary, heart-beat is
> >> transmitted to the primary. If we can receive heart-beat ACK, then
> >> use
> >> primary for the next
> >> transmission. Otherwise, use secondary for the next transmission.
> >> I tend to believe this approach does not cause much bouncing =
effects.
> > It
> >> behaves mostly
> >> like the current SCTP in case of congestion. In addition, by =
sending
> >> small heart-beat
> >> instead of data chunk to potentially congested path, it can reduce
> >> the
> >> risk of congestion
> >> collapse.
> >>
> >> Thanks,
> >> --
> >> Yoshifumi Nishida
> >> nishida@sfc.wide.ad.jp
> >


From randall@lakerest.net  Wed Aug 12 06:07:51 2009
Return-Path: <randall@lakerest.net>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 697083A69F8 for <tsvwg@core3.amsl.com>; Wed, 12 Aug 2009 06:07:51 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.449
X-Spam-Level: 
X-Spam-Status: No, score=-2.449 tagged_above=-999 required=5 tests=[AWL=0.150,  BAYES_00=-2.599]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id rUG0CTQoGjyf for <tsvwg@core3.amsl.com>; Wed, 12 Aug 2009 06:07:48 -0700 (PDT)
Received: from lakerest.net (unknown [IPv6:2001:240:585:2:213:d4ff:fef3:2d8d]) by core3.amsl.com (Postfix) with ESMTP id 252E33A6886 for <tsvwg@ietf.org>; Wed, 12 Aug 2009 06:07:47 -0700 (PDT)
Received: from [12.236.17.84] ([12.236.17.84]) (authenticated bits=0) by lakerest.net (8.14.3/8.14.3) with ESMTP id n7CD7Uq8068849 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Wed, 12 Aug 2009 09:07:33 -0400 (EDT) (envelope-from randall@lakerest.net)
DKIM-Signature: a=rsa-sha1; c=simple/simple; d=lakerest.net; s=mail; t=1250082454; h=Cc:Message-Id:From:To:In-Reply-To:Content-Type: Content-Transfer-Encoding:Mime-Version:Subject:Date:References: X-Mailer; b=QJu/oaK/YjOllr3yWMbge06M0URUMWfjQ7n1dB77rnnPcnbxyHRr2sh AThGRU/7na4Tdz1FdL55AgtpylXBb9g==
Message-Id: <F38D1A12-70EE-487B-8AF3-E22AC92CD708@lakerest.net>
From: Randy Stewart <randall@lakerest.net>
To: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>
In-Reply-To: <20090812.140134.183090419.nishida@sfc.wide.ad.jp>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed; delsp=yes
Content-Transfer-Encoding: quoted-printable
Mime-Version: 1.0 (Apple Message framework v936)
Date: Wed, 12 Aug 2009 06:07:29 -0700
References: <20090812.035740.189643280.nishida@sfc.wide.ad.jp> <A51D8ACD861B7E41BFC7FE5C64BE96480DC42BA7@MTLEXVS01.ulticom.com> <932B0EC1-B00C-4F0D-9876-F0F37D079524@lurchi.franken.de> <20090812.140134.183090419.nishida@sfc.wide.ad.jp>
X-Mailer: Apple Mail (2.936)
Cc: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>, tsvwg <tsvwg@ietf.org>
Subject: Re: [tsvwg] Q about path failure detection in SCTP
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 12 Aug 2009 13:07:51 -0000

On Aug 11, 2009, at 10:01 PM, Yoshifumi Nishida wrote:

>
> From: Michael T=FCxen <Michael.Tuexen@lurchi.franken.de>
> Subject: Re: [tsvwg] Q about path failure detection in SCTP
> Date: Tue, 11 Aug 2009 21:50:38 +0200
> Message-ID: <932B0EC1-B00C-4F0D-9876-F0F37D079524@lurchi.franken.de>
>
>> On Aug 11, 2009, at 9:27 PM, David Lehmann wrote:
>>
>>> Yoshifumi raised a good point.
>>>
>>> Waiting for the primary destination to time out doesn't fly with
>>> telecom
>>> apps.  Our implementation has being doing something similar for =20
>>> years
>>> to
>>> what Yoshifumi has suggested.  Our algorithm checks the primary =20
>>> path.
>>> If the error count is zero, use it.  Otherwise, look for a path with
>>> the
>>> least amount of errors.  As Randall mentioned, the spec says =20
>>> "SHOULD";
>>> so it is RFC compliant.
>
>> This might be a good idea in a telco environment where the different
>> paths share no common link. But if for example all paths share a
>> single
>> link and this one is responsable for the loss observed, switching
>> to a different path is counter productive.
>
> I think it's not very harmful even if all paths share a single link.
>
> If all paths share a single link and the shared link is congested,
> even with normal SCTP, it tries to retransmit data to the non-=20
> primary paths.
> Hence, the data go toward the shared congested link anyhow.
>
> The slight difference between the current SCTP and my proposal is =20
> heart-beat
> packets transmitted to the primary path. Yes, this is a negative =20
> factor
> in this approach, but I tend to believe it's mostly acceptable.

Actually, I believe we do something along these lines
when a congestion event occurs anyway. This came out
of Armando's work and was not in violation of the RFC... since
you can send a HB at any time ;-)

What I think you really want is what Armando studied in
his PHd Thesis. And showed to have no real harm.. i.e
a two level trigger... can't remember his terms for it..

Lets call it Trigger 1 and Trigger 2..

When you hit T1 you switch from your active to your
secondary until a HB is ack'd on the old primary.
He might have even stayed on the secondary.. I know
he looked at this. Since when you switch you pretty
much enter slow start and start ramping up the secondary
path... and switching back slows you again...

If you hit T2 with the old primary it is like hitting
the normal threshold.. i.e. you tell the user its unreachable.

He played with different values for the values of T1 and T2 and I =20
believe
setting T1 to 1 (which is what you are proposing) was not shown to
be harmful.

You really should google his work.. in fact maybe he will chime in
here since I know he still watches this list...

Armando? You out there (probably just to busy at BBN ;-D)?


Anyway, I think doing what you propose is within the bounds of the spec
since what it goes against is in the SHOULD department.

It might be worth writing up a separate I-D documenting how it
works and why you can try it. I would think one would want to
pitch it has experimental.. I would also think one should offer
to have Armando on board as a co-author if he is interested.

Of course once the M-TCP folks figure out the right congestion control,
I will simply turn-on by default SCTP-CMT in FreeBSD after we put in
whatever the CC is they come up with...  that is, in my opinion, the
only good thing to come out of the M-TCP work ;-)

R


>
> CMT might be the ideal goal, but I would like to think what we can =20
> do on
> the current SCTP with minimal change.
>
> Thanks,
> --
> Yoshifumi Nishida
> nishida@sfc.wide.ad.jp
>

-----
Randall Stewart
randall@lakerest.net





From Michael.Tuexen@lurchi.franken.de  Wed Aug 12 08:04:55 2009
Return-Path: <Michael.Tuexen@lurchi.franken.de>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 600463A6837 for <tsvwg@core3.amsl.com>; Wed, 12 Aug 2009 08:04:55 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.524
X-Spam-Level: 
X-Spam-Status: No, score=-1.524 tagged_above=-999 required=5 tests=[AWL=-0.776, BAYES_00=-2.599, FH_RELAY_NODNS=1.451, MIME_8BIT_HEADER=0.3, RDNS_NONE=0.1]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id KNaEtMig7n9V for <tsvwg@core3.amsl.com>; Wed, 12 Aug 2009 08:04:44 -0700 (PDT)
Received: from mail-n.franken.de (drew.ipv6.franken.de [IPv6:2001:638:a02:a001:20e:cff:fe4a:feaa]) by core3.amsl.com (Postfix) with ESMTP id A65E83A68EA for <tsvwg@ietf.org>; Wed, 12 Aug 2009 08:04:28 -0700 (PDT)
Received: from [10.0.1.102] (unknown [194.95.73.188]) by mail-n.franken.de (Postfix) with ESMTP id BFD491C0C0BCC; Wed, 12 Aug 2009 17:04:28 +0200 (CEST)
Message-Id: <7BE8A9D7-3278-4D02-92A1-0AD60D60911F@lurchi.franken.de>
From: =?ISO-8859-1?Q?Michael_T=FCxen?= <Michael.Tuexen@lurchi.franken.de>
To: David Lehmann <dlehmann@ulticom.com>
In-Reply-To: <A51D8ACD861B7E41BFC7FE5C64BE96480DC42D51@MTLEXVS01.ulticom.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed; delsp=yes
Content-Transfer-Encoding: quoted-printable
Mime-Version: 1.0 (Apple Message framework v936)
Date: Wed, 12 Aug 2009 17:04:28 +0200
References: <20090811.172943.209950324.nishida@sfc.wide.ad.jp><2E52235B-E074-49A2-BAE0-113D4668BECE@lurchi.franken.de><A2E2E510-8BE3-4F64-AEC9-B8E0FBDCDAE3@lakerest.net> <20090812.035740.189643280.nishida@sfc.wide.ad.jp> <A51D8ACD861B7E41BFC7FE5C64BE96480DC42BA7@MTLEXVS01.ulticom.com> <932B0EC1-B00C-4F0D-9876-F0F37D079524@lurchi.franken.de> <A51D8ACD861B7E41BFC7FE5C64BE96480DC42D51@MTLEXVS01.ulticom.com>
X-Mailer: Apple Mail (2.936)
Cc: tsvwg@ietf.org
Subject: Re: [tsvwg] Q about path failure detection in SCTP
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 12 Aug 2009 15:04:55 -0000

On Aug 12, 2009, at 2:56 PM, David Lehmann wrote:

> I would think that most apps (not just telco apps) would benefit =20
> from and _expect_ a quick (i.e. hardly noticeable) failover when =20
> using multihomed endpoints.
I agree. But this needs diverse paths...
>
> As far as being "counter productive", waiting 30 seconds to failover =20=

> to an alternate path is counter productive.  I would venture to =20
> guess that this characteristic is unacceptable to most apps.
What I mean by counter productive is the following:
Assume multiple paths which share a link which is responsable for the =20=

message loss.
So switching paths doesn't help. But if you switch you always have a =20
new (small)
congestion window...

If you know that you have diverse paths (like in telco setups) you can =20=

reduce
the threshold for switching the primary path and it will help. But if =20=

you
do not know what the setup is, switching due to a single timeout might =20=

be
suboptimal, since you do not know if the alternate path provides better
service.

Best regards
Michael
>
> --
> David Lehmann
> Ulticom, Inc.
> 856-787-2952
>
>> -----Original Message-----
>> From: Michael T=FCxen [mailto:Michael.Tuexen@lurchi.franken.de]
>> Sent: Tuesday, August 11, 2009 3:51 PM
>> To: David Lehmann
>> Cc: Yoshifumi Nishida; randall@lakerest.net; tsvwg@ietf.org
>> Subject: Re: [tsvwg] Q about path failure detection in SCTP
>>
>> On Aug 11, 2009, at 9:27 PM, David Lehmann wrote:
>>
>>> Yoshifumi raised a good point.
>>>
>>> Waiting for the primary destination to time out doesn't fly with
>>> telecom
>>> apps.  Our implementation has being doing something similar for
>>> years to
>>> what Yoshifumi has suggested.  Our algorithm checks the primary =20
>>> path.
>>> If the error count is zero, use it.  Otherwise, look for a path with
>>> the
>>> least amount of errors.  As Randall mentioned, the spec says =20
>>> "SHOULD";
>>> so it is RFC compliant.
>> This might be a good idea in a telco environment where the different
>> paths share no common link. But if for example all paths share a =20
>> single
>> link and this one is responsable for the loss observed, switching
>> to a different path is counter productive.
>>
>> The best way to handle it is to use all paths simultaneously, but
>> that is not yet standardized.
>>>
>>> As soon as the error is cleared, the primary will automatically be
>>> used,
>>> given the above algorithm. (When an alternate path is used, that
>>> doesn't
>>> change the primary.  The primary can only be changed by the user.)
>>>
>>> Yes, a heartbeat will speed recovery of the primary path. (I think =20=

>>> it
>>> was Armando who suggested this to me at the UofD bake off.) Fast
>>> recovery of the primary path seems to help throughput possibly =20
>>> because
>>> data is being sent back on the primary path which will have a larger
>>> cwnd.
>>>
>>> We have seen no down side for this approach.  In tests, when we =20
>>> pulled
>>> the Ethernet cable used for the primary path, we hardly notice a
>>> blip in
>>> the throughput.
>>>
>>> --
>>> David Lehmann
>>> Ulticom, Inc.
>>> 856-787-2952
>>>
>>>> -----Original Message-----
>>>> From: tsvwg-bounces@ietf.org [mailto:tsvwg-bounces@ietf.org] On
>>>> Behalf
>>>> Of Yoshifumi Nishida
>>>> Sent: Tuesday, August 11, 2009 2:58 PM
>>>> To: randall@lakerest.net
>>>> Cc: Michael.Tuexen@lurchi.franken.de; tsvwg@ietf.org
>>>> Subject: Re: [tsvwg] Q about path failure detection in SCTP
>>>>
>>>>
>>>> From: Randy Stewart <randall@lakerest.net>
>>>> Subject: Re: [tsvwg] Q about path failure detection in SCTP
>>>> Date: Tue, 11 Aug 2009 06:28:14 -0700
>>>> Message-ID: <A2E2E510-8BE3-4F64-AEC9-B8E0FBDCDAE3@lakerest.net>
>>>>
>>>>>>> My rough idea is something like this.
>>>>>>>
>>>>>>> 1) If timeout happens on the primary path, set the status of
>>> this
>>>> path
>>>>>>> to "suspicious"
>>>>
>>>>>> So a path having experienced a congestion indication (timeout) is
>>>>>> suspicious?
>>>>>
>>>>> And one could end up with bouncing paths...
>>>>
>>>> Yes, and my point is I would like to discuss bouncing more.
>>>>
>>>> Let's say, the primary path is a bit lossy and causes timeout
>>>> (congestions or whatever).
>>>> In case of normal SCTP, retransmit packets will be transmited to
>>>> secondary.
>>>>
>>>>  "Furthermore, when its peer is multi-homed, an endpoint SHOULD try
>>> to
>>>>  retransmit a chunk that timed out to an active destination
>>> transport
>>>>  address that is different from the last destination address to
>>> which
>>>>  the DATA chunk was sent."
>>>>
>>>> Do we call this bouncing?  Probably not.
>>>> What I'm going to propose is, when packet is retransmitted to the
>>>> secondary, heart-beat is
>>>> transmitted to the primary. If we can receive heart-beat ACK, then
>>>> use
>>>> primary for the next
>>>> transmission. Otherwise, use secondary for the next transmission.
>>>> I tend to believe this approach does not cause much bouncing =20
>>>> effects.
>>> It
>>>> behaves mostly
>>>> like the current SCTP in case of congestion. In addition, by =20
>>>> sending
>>>> small heart-beat
>>>> instead of data chunk to potentially congested path, it can reduce
>>>> the
>>>> risk of congestion
>>>> collapse.
>>>>
>>>> Thanks,
>>>> --
>>>> Yoshifumi Nishida
>>>> nishida@sfc.wide.ad.jp
>>>
>
>


From alcortesm.uc3m@gmail.com  Wed Aug 12 09:02:25 2009
Return-Path: <alcortesm.uc3m@gmail.com>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id C66813A6B99 for <tsvwg@core3.amsl.com>; Wed, 12 Aug 2009 09:02:25 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.077
X-Spam-Level: 
X-Spam-Status: No, score=-3.077 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, GB_I_LETTER=-2, J_CHICKENPOX_21=0.6, MIME_8BIT_HEADER=0.3]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id U0KF8YmCxJHf for <tsvwg@core3.amsl.com>; Wed, 12 Aug 2009 09:02:24 -0700 (PDT)
Received: from mail-ew0-f214.google.com (mail-ew0-f214.google.com [209.85.219.214]) by core3.amsl.com (Postfix) with ESMTP id 6928F3A6BB2 for <tsvwg@ietf.org>; Wed, 12 Aug 2009 09:02:24 -0700 (PDT)
Received: by ewy10 with SMTP id 10so120582ewy.37 for <tsvwg@ietf.org>; Wed, 12 Aug 2009 09:01:37 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:message-id:subject:from:to:cc :content-type:content-transfer-encoding; bh=v7ShNy5L1zr35NWeSRWXFwA5JPP0a512nfusFT3Hc3w=; b=fBs1oey4bzafpzsv5Nww4vrU0hfy7JAcQfSsSkYsan94cqP+qoQ+wOUu8qHpAju/IP Mb0TmM3Hhg9bO9BghwYkBbmgcEef4MH6X/lO56kyFU1nF9Le6AXBZtfAJgiCiN0kVHNA AkTzEHms0GsaXNTlPLXjqNVpQWCx3Snjo/XHs=
DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type :content-transfer-encoding; b=eEupJz/5sYrwW1jhYh3rOO0YoTFnoCy3bFIjjbFn409+MiEMVeAHYtsbJAf1GVX7lB hyuDI1qaCFA9EbZMppeI61AyzFzegl+lJ5+vP/HtxB2E5KC0L7zFOYRDJpu924jzgyX+ cVSJFaYwGJhY+FFHua3IygR38e5cB/Y4eQZDw=
MIME-Version: 1.0
Sender: alcortesm.uc3m@gmail.com
Received: by 10.216.11.84 with SMTP id 62mr39145wew.148.1250092896753; Wed, 12  Aug 2009 09:01:36 -0700 (PDT)
In-Reply-To: <F38D1A12-70EE-487B-8AF3-E22AC92CD708@lakerest.net>
References: <20090812.035740.189643280.nishida@sfc.wide.ad.jp> <A51D8ACD861B7E41BFC7FE5C64BE96480DC42BA7@MTLEXVS01.ulticom.com> <932B0EC1-B00C-4F0D-9876-F0F37D079524@lurchi.franken.de> <20090812.140134.183090419.nishida@sfc.wide.ad.jp> <F38D1A12-70EE-487B-8AF3-E22AC92CD708@lakerest.net>
Date: Wed, 12 Aug 2009 18:01:36 +0200
X-Google-Sender-Auth: 272306a65cea44ac
Message-ID: <b53c10010908120901k5d10bd96ya720e10d05c9118b@mail.gmail.com>
From: =?UTF-8?Q?Alberto_Cort=C3=A9s?= <alcortes@it.uc3m.es>
To: Randy Stewart <randall@lakerest.net>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Cc: Michael Tuexen <Michael.Tuexen@lurchi.franken.de>, tsvwg <tsvwg@ietf.org>
Subject: Re: [tsvwg] Q about path failure detection in SCTP
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 12 Aug 2009 16:02:25 -0000

On Wed, Aug 12, 2009 at 3:07 PM, Randy Stewart<randall@lakerest.net> wrote:
> Actually, I believe we do something along these lines
> when a congestion event occurs anyway. This came out
> of Armando's work and was not in violation of the RFC... since
> you can send a HB at any time ;-)
>
> What I think you really want is what Armando studied in
> his PHd Thesis. And showed to have no real harm.. i.e
> a two level trigger... can't remember his terms for it..
>
> Lets call it Trigger 1 and Trigger 2..
>
> When you hit T1 you switch from your active to your
> secondary until a HB is ack'd on the old primary.
> He might have even stayed on the secondary.. I know
> he looked at this. Since when you switch you pretty
> much enter slow start and start ramping up the secondary
> path... and switching back slows you again...
>
> If you hit T2 with the old primary it is like hitting
> the normal threshold.. i.e. you tell the user its unreachable.
>
> He played with different values for the values of T1 and T2 and I believe
> setting T1 to 1 (which is what you are proposing) was not shown to
> be harmful.
>
> You really should google his work.. in fact maybe he will chime in
> here since I know he still watches this list...
>
> Armando? You out there (probably just to busy at BBN ;-D)?
>
>
> Anyway, I think doing what you propose is within the bounds of the spec
> since what it goes against is in the SHOULD department.
>
> It might be worth writing up a separate I-D documenting how it
> works and why you can try it. I would think one would want to
> pitch it has experimental.. I would also think one should offer
> to have Armando on board as a co-author if he is interested.
>
> Of course once the M-TCP folks figure out the right congestion control,
> I will simply turn-on by default SCTP-CMT in FreeBSD after we put in
> whatever the CC is they come up with... =C2=A0that is, in my opinion, the
> only good thing to come out of the M-TCP work ;-)
>
> R

I have explore the use of HBs to reduce failure detection delay,
please take a look at:

"Decoupling path failure detection from congestion control to improve
SCTP failovers"
Cortes, A.   Garcia-Rubio, C.   Campo, C.   Marin, A.   Almenarez, F.
 Diaz, D.
Dept. of Telematic Eng.,, Univ. Carlos III of Madrid, Leganes;

This paper appears in: Communications Letters, IEEE
Publication Date: November 2008
Volume: 12,  Issue: 11
On page(s): 858-860
ISSN: 1089-7798
INSPEC Accession Number: 10346278
Digital Object Identifier: 10.1109/LCOMM.2008.080884
Current Version Published: 2008-11-25
Abstract
SCTP handover is too slow to be useful as a mobility solution for most
applications. With quicker handovers, SCTP will be an interesting
solution to mobility. SCTP uses retransmissions as probes for path
failure detection. Congestion control forces slow paced
retransmissions but quick failovers need fast paced probes. Our
solution is to stop using retransmissions as probes. Instead, we
propose active path monitoring using unreliable heartbeats.
Applications can tune this algorithm to meet their own requirements.
We show data from a simple experiment in a real environment.


--=20
Alberto Cort=C3=A9s
Telematic Engineering Dept. at UC3M

From nishida@sfc.wide.ad.jp  Thu Aug 13 01:27:55 2009
Return-Path: <nishida@sfc.wide.ad.jp>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 6D5D73A6960 for <tsvwg@core3.amsl.com>; Thu, 13 Aug 2009 01:27:55 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 2.598
X-Spam-Level: **
X-Spam-Status: No, score=2.598 tagged_above=-999 required=5 tests=[AWL=1.694,  BAYES_00=-2.599, HELO_EQ_JP=1.244, HOST_EQ_JP=1.265, RELAY_IS_203=0.994]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id oXHQqSPDfkch for <tsvwg@core3.amsl.com>; Thu, 13 Aug 2009 01:27:54 -0700 (PDT)
Received: from mail.sfc.wide.ad.jp (mail.sfc.wide.ad.jp [203.178.142.146]) by core3.amsl.com (Postfix) with ESMTP id 6630E3A6856 for <tsvwg@ietf.org>; Thu, 13 Aug 2009 01:27:54 -0700 (PDT)
Received: from localhost (cpu.sfc.wide.ad.jp [IPv6:2001:200:0:8803:203:178:142:143]) by mail.sfc.wide.ad.jp (Postfix) with ESMTP id 9B66A4C173; Thu, 13 Aug 2009 17:25:39 +0900 (JST)
Date: Thu, 13 Aug 2009 17:25:39 +0900 (JST)
Message-Id: <20090813.172539.200390062.nishida@sfc.wide.ad.jp>
To: randall@lakerest.net
From: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>
In-Reply-To: <F38D1A12-70EE-487B-8AF3-E22AC92CD708@lakerest.net>
References: <932B0EC1-B00C-4F0D-9876-F0F37D079524@lurchi.franken.de> <20090812.140134.183090419.nishida@sfc.wide.ad.jp> <F38D1A12-70EE-487B-8AF3-E22AC92CD708@lakerest.net>
X-Mailer: Mew version 6.2 on Emacs 22.1 / Mule 5.0 (SAKAKI)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable
Cc: Michael.Tuexen@lurchi.franken.de, tsvwg@ietf.org
Subject: Re: [tsvwg] Q about path failure detection in SCTP
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 13 Aug 2009 08:27:55 -0000

From: Randy Stewart <randall@lakerest.net>
Subject: Re: [tsvwg] Q about path failure detection in SCTP
Date: Wed, 12 Aug 2009 06:07:29 -0700
Message-ID: <F38D1A12-70EE-487B-8AF3-E22AC92CD708@lakerest.net>

 > > From: Michael T=FCxen <Michael.Tuexen@lurchi.franken.de>
 > > Subject: Re: [tsvwg] Q about path failure detection in SCTP
 > > Date: Tue, 11 Aug 2009 21:50:38 +0200
 > > Message-ID: <932B0EC1-B00C-4F0D-9876-F0F37D079524@lurchi.franken.d=
e>
 > >
 > >> On Aug 11, 2009, at 9:27 PM, David Lehmann wrote:
 > >>
 > >>> Yoshifumi raised a good point.
 > >>>
 > >>> Waiting for the primary destination to time out doesn't fly with=

 > >>> telecom
 > >>> apps.  Our implementation has being doing something similar for =
years
 > >>> to
 > >>> what Yoshifumi has suggested.  Our algorithm checks the primary =
path.
 > >>> If the error count is zero, use it.  Otherwise, look for a path =
with
 > >>> the
 > >>> least amount of errors.  As Randall mentioned, the spec says "SH=
OULD";
 > >>> so it is RFC compliant.
 > >
 > >> This might be a good idea in a telco environment where the differ=
ent
 > >> paths share no common link. But if for example all paths share a
 > >> single
 > >> link and this one is responsable for the loss observed, switching=

 > >> to a different path is counter productive.
 > >
 > > I think it's not very harmful even if all paths share a single lin=
k.
 > >
 > > If all paths share a single link and the shared link is congested,=

 > > even with normal SCTP, it tries to retransmit data to the non-prim=
ary
 > > paths.
 > > Hence, the data go toward the shared congested link anyhow.
 > >
 > > The slight difference between the current SCTP and my proposal is
 > > heart-beat
 > > packets transmitted to the primary path. Yes, this is a negative
 > > factor
 > > in this approach, but I tend to believe it's mostly acceptable.
 > =

 > Actually, I believe we do something along these lines
 > when a congestion event occurs anyway. This came out
 > of Armando's work and was not in violation of the RFC... since
 > you can send a HB at any time ;-)
 > =

 > What I think you really want is what Armando studied in
 > his PHd Thesis. And showed to have no real harm.. i.e
 > a two level trigger... can't remember his terms for it..
 > =

 > Lets call it Trigger 1 and Trigger 2..
 > =

 > When you hit T1 you switch from your active to your
 > secondary until a HB is ack'd on the old primary.
 > He might have even stayed on the secondary.. I know
 > he looked at this. Since when you switch you pretty
 > much enter slow start and start ramping up the secondary
 > path... and switching back slows you again...
 > =

 > If you hit T2 with the old primary it is like hitting
 > the normal threshold.. i.e. you tell the user its unreachable.
 > =

 > He played with different values for the values of T1 and T2 and I
 > believe
 > setting T1 to 1 (which is what you are proposing) was not shown to
 > be harmful.
 > =

 > You really should google his work.. in fact maybe he will chime in
 > here since I know he still watches this list...
 > =

 > Armando? You out there (probably just to busy at BBN ;-D)?
 > =

 > =

 > Anyway, I think doing what you propose is within the bounds of the
 > spec
 > since what it goes against is in the SHOULD department.
 > =

 > It might be worth writing up a separate I-D documenting how it
 > works and why you can try it. I would think one would want to
 > pitch it has experimental.. I would also think one should offer
 > to have Armando on board as a co-author if he is interested.
 > =

 > Of course once the M-TCP folks figure out the right congestion
 > control,
 > I will simply turn-on by default SCTP-CMT in FreeBSD after we put in=

 > whatever the CC is they come up with...  that is, in my opinion, the=

 > only good thing to come out of the M-TCP work ;-)

OK. I'll check Armando's work and other related stuff again and try to =
prepare I-D =

and see how many people have an interest.
Thanks for your advice. =

--
Yoshifumi Nishida
nishida@sfc.wide.ad.jp

From Scott.Brim@gmail.com  Fri Aug 14 12:23:00 2009
Return-Path: <Scott.Brim@gmail.com>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id DB5CB3A6DB0 for <tsvwg@core3.amsl.com>; Fri, 14 Aug 2009 12:23:00 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.38
X-Spam-Level: 
X-Spam-Status: No, score=-6.38 tagged_above=-999 required=5 tests=[AWL=0.219,  BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Bf+qQdSXGi11 for <tsvwg@core3.amsl.com>; Fri, 14 Aug 2009 12:23:00 -0700 (PDT)
Received: from sj-iport-2.cisco.com (sj-iport-2.cisco.com [171.71.176.71]) by core3.amsl.com (Postfix) with ESMTP id 3882E3A6DAE for <tsvwg@ietf.org>; Fri, 14 Aug 2009 12:23:00 -0700 (PDT)
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: ApoEAEJWhUqrR7MV/2dsb2JhbAC6K4cKkjCEGQU
X-IronPort-AV: E=Sophos;i="4.43,381,1246838400"; d="scan'208";a="195756691"
Received: from sj-dkim-1.cisco.com ([171.71.179.21]) by sj-iport-2.cisco.com with ESMTP; 14 Aug 2009 19:23:03 +0000
Received: from sj-core-3.cisco.com (sj-core-3.cisco.com [171.68.223.137]) by sj-dkim-1.cisco.com (8.12.11/8.12.11) with ESMTP id n7EJN3kj002533 for <tsvwg@ietf.org>; Fri, 14 Aug 2009 12:23:03 -0700
Received: from xbh-rtp-211.amer.cisco.com (xbh-rtp-211.cisco.com [64.102.31.102]) by sj-core-3.cisco.com (8.13.8/8.14.3) with ESMTP id n7EJMm3n011197 for <tsvwg@ietf.org>; Fri, 14 Aug 2009 19:23:03 GMT
Received: from xfe-rtp-201.amer.cisco.com ([64.102.31.38]) by xbh-rtp-211.amer.cisco.com with Microsoft SMTPSVC(6.0.3790.3959);  Fri, 14 Aug 2009 15:22:51 -0400
Received: from cisco.com ([10.86.240.126]) by xfe-rtp-201.amer.cisco.com with Microsoft SMTPSVC(6.0.3790.3959); Fri, 14 Aug 2009 15:22:51 -0400
Date: Fri, 14 Aug 2009 15:22:44 -0400
From: Scott Brim <Scott.Brim@gmail.com>
To: tsvwg@ietf.org
Message-ID: <20090814192244.GA12873@cisco.com>
References: <20090812.035740.189643280.nishida@sfc.wide.ad.jp> <A51D8ACD861B7E41BFC7FE5C64BE96480DC42BA7@MTLEXVS01.ulticom.com> <932B0EC1-B00C-4F0D-9876-F0F37D079524@lurchi.franken.de> <20090812.140134.183090419.nishida@sfc.wide.ad.jp> <F38D1A12-70EE-487B-8AF3-E22AC92CD708@lakerest.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <F38D1A12-70EE-487B-8AF3-E22AC92CD708@lakerest.net>
User-Agent: Mutt/1.5.19 (2009-01-05)
X-OriginalArrivalTime: 14 Aug 2009 19:22:51.0426 (UTC) FILETIME=[9D9BB820:01CA1D14]
Authentication-Results: sj-dkim-1; header.From=Scott.Brim@gmail.com; dkim=neutral
Subject: Re: [tsvwg] Q about path failure detection in SCTP
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 14 Aug 2009 19:23:00 -0000

Excerpts from Randy Stewart on Wed, Aug 12, 2009 06:07:29AM -0700:
> Of course once the M-TCP folks figure out the right congestion control,
> I will simply turn-on by default SCTP-CMT in FreeBSD after we put in
> whatever the CC is they come up with...  

and I assume you'll be there helping to test the algorithms.

> that is, in my opinion, the
> only good thing to come out of the M-TCP work ;-)

I see the wink in that smiley.

Scott

From doug.mtview@gmail.com  Fri Aug 14 15:19:24 2009
Return-Path: <doug.mtview@gmail.com>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 359D028C225 for <tsvwg@core3.amsl.com>; Fri, 14 Aug 2009 15:19:24 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.547
X-Spam-Level: 
X-Spam-Status: No, score=-1.547 tagged_above=-999 required=5 tests=[AWL=1.052,  BAYES_00=-2.599]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id L7nN2VdbQS+D for <tsvwg@core3.amsl.com>; Fri, 14 Aug 2009 15:19:23 -0700 (PDT)
Received: from mail-pz0-f178.google.com (mail-pz0-f178.google.com [209.85.222.178]) by core3.amsl.com (Postfix) with ESMTP id CB54C28C20B for <tsvwg@ietf.org>; Fri, 14 Aug 2009 15:18:54 -0700 (PDT)
Received: by pzk8 with SMTP id 8so1346983pzk.31 for <tsvwg@ietf.org>; Fri, 14 Aug 2009 15:18:54 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from :user-agent:mime-version:to:subject:content-type :content-transfer-encoding; bh=dUMrU1i9ysWilHw9X/CbLeW5g2guNGi1vPH/BbHgt+U=; b=T4OEGKWDr3PLfoSfsdvwGWL8IeVlm0HpGzfcH18r7izB1BJHBN68+L/fQoC14ynW6P zwXRm/2HlcfqvNwu5yYugWK+40h1Cr1UjXjfllLEMEl4MdXlPTAX/GF+7Q3q1LgUFrlb IkkuUo4ZNb811RiMEdGbxamXxfxXfFentImyg=
DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:user-agent:mime-version:to:subject :content-type:content-transfer-encoding; b=nRYDs4QJqkDPLoA+H153iGC92JoztiQezhJ9GfkX+eem6a3tDKyFJ6GM/Ephr7XNmP zGqkdv2HojDZYrcUEBDwaWfE1m8SKBj5MnAJfHM59Xe+Ee5Vh+QToVeUvgSFqMejdndN j/F3E+qbc2j3mDOmXgXQqr7J0wzRT+Ssbe1sE=
Received: by 10.115.101.12 with SMTP id d12mr2236092wam.179.1250288334617; Fri, 14 Aug 2009 15:18:54 -0700 (PDT)
Received: from SJC-Office-NAT-214.mail-abuse.org (SJC-Office-NAT-214.mail-abuse.org [168.61.10.214]) by mx.google.com with ESMTPS id k35sm2951064waf.29.2009.08.14.15.18.52 (version=TLSv1/SSLv3 cipher=RC4-MD5); Fri, 14 Aug 2009 15:18:53 -0700 (PDT)
Message-ID: <4A85E2CB.4060502@gmail.com>
Date: Fri, 14 Aug 2009 15:18:51 -0700
From: Doug Otis <doug.mtview@gmail.com>
User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.1.1) Gecko/20090715 Thunderbird/3.0b3
MIME-Version: 1.0
To: tsvwg@ietf.org
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Subject: [tsvwg] SCTP tie-tag info in state cookie?
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 14 Aug 2009 22:19:24 -0000

Is there an errata describing concerns related to obscuring tie-tag info?

-Doug


From randall@lakerest.net  Sat Aug 15 04:42:57 2009
Return-Path: <randall@lakerest.net>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id C57953A6A4B for <tsvwg@core3.amsl.com>; Sat, 15 Aug 2009 04:42:57 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.599
X-Spam-Level: 
X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id iTpiErRGtbej for <tsvwg@core3.amsl.com>; Sat, 15 Aug 2009 04:42:56 -0700 (PDT)
Received: from lakerest.net (unknown [IPv6:2001:240:585:2:213:d4ff:fef3:2d8d]) by core3.amsl.com (Postfix) with ESMTP id 3E2B63A697D for <tsvwg@ietf.org>; Sat, 15 Aug 2009 04:42:54 -0700 (PDT)
Received: from [192.168.1.70] (pool-71-125-103-120.atl01.dsl-w.verizon.net [71.125.103.120]) (authenticated bits=0) by lakerest.net (8.14.3/8.14.3) with ESMTP id n7FBgmPO056619 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Sat, 15 Aug 2009 07:42:54 -0400 (EDT) (envelope-from randall@lakerest.net)
DKIM-Signature: a=rsa-sha1; c=simple/simple; d=lakerest.net; s=mail; t=1250336574; h=Cc:Message-Id:From:To:In-Reply-To:Content-Type: Content-Transfer-Encoding:Mime-Version:Subject:Date:References: X-Mailer; b=MNzogmds7kN89qkdDChVXhCnfzQDjTYKJ3gG4Roj9C+jY1tFEN+JwF+ +UIus4OGDX1yFHRK1NLkNAiFA1V+F8w==
Message-Id: <48A80E8F-B574-445C-BAD1-633DBB71B812@lakerest.net>
From: Randy Stewart <randall@lakerest.net>
To: Scott Brim <Scott.Brim@gmail.com>
In-Reply-To: <20090814192244.GA12873@cisco.com>
Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes
Content-Transfer-Encoding: 7bit
Mime-Version: 1.0 (Apple Message framework v936)
Date: Sat, 15 Aug 2009 04:42:43 -0700
References: <20090812.035740.189643280.nishida@sfc.wide.ad.jp> <A51D8ACD861B7E41BFC7FE5C64BE96480DC42BA7@MTLEXVS01.ulticom.com> <932B0EC1-B00C-4F0D-9876-F0F37D079524@lurchi.franken.de> <20090812.140134.183090419.nishida@sfc.wide.ad.jp> <F38D1A12-70EE-487B-8AF3-E22AC92CD708@lakerest.net> <20090814192244.GA12873@cisco.com>
X-Mailer: Apple Mail (2.936)
Cc: tsvwg@ietf.org
Subject: Re: [tsvwg] Q about path failure detection in SCTP
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 15 Aug 2009 11:42:57 -0000

On Aug 14, 2009, at 12:22 PM, Scott Brim wrote:

> Excerpts from Randy Stewart on Wed, Aug 12, 2009 06:07:29AM -0700:
>> Of course once the M-TCP folks figure out the right congestion  
>> control,
>> I will simply turn-on by default SCTP-CMT in FreeBSD after we put in
>> whatever the CC is they come up with...
>
> and I assume you'll be there helping to test the algorithms.

Of course :-)


We will be glad to implement and test things in real O/S's

R
>
>> that is, in my opinion, the
>> only good thing to come out of the M-TCP work ;-)
>
> I see the wink in that smiley.
>
> Scott
>

-----
Randall Stewart
randall@lakerest.net





From Michael.Tuexen@lurchi.franken.de  Sun Aug 16 02:38:05 2009
Return-Path: <Michael.Tuexen@lurchi.franken.de>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 797433A6D4D for <tsvwg@core3.amsl.com>; Sun, 16 Aug 2009 02:38:05 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 2.038
X-Spam-Level: **
X-Spam-Status: No, score=2.038 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HOST_EQ_DIP_TDIAL=2.144, HOST_MISMATCH_NET=0.311, MIME_8BIT_HEADER=0.3, RCVD_IN_PBL=0.905, RCVD_IN_SORBS_DUL=0.877, RDNS_DYNAMIC=0.1]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id NW5ymOT80AN9 for <tsvwg@core3.amsl.com>; Sun, 16 Aug 2009 02:38:04 -0700 (PDT)
Received: from mail-n.franken.de (drew.ipv6.franken.de [IPv6:2001:638:a02:a001:20e:cff:fe4a:feaa]) by core3.amsl.com (Postfix) with ESMTP id A8A9B3A6B64 for <tsvwg@ietf.org>; Sun, 16 Aug 2009 02:38:04 -0700 (PDT)
Received: from [192.168.1.100] (p508FF5B8.dip.t-dialin.net [80.143.245.184]) by mail-n.franken.de (Postfix) with ESMTP id A1B4C1C0B4617; Sun, 16 Aug 2009 11:38:06 +0200 (CEST)
Message-Id: <2A131DEC-80B7-4BF4-B6F8-F67E00C1610F@lurchi.franken.de>
From: =?ISO-8859-1?Q?Michael_T=FCxen?= <Michael.Tuexen@lurchi.franken.de>
To: Doug Otis <doug.mtview@gmail.com>
In-Reply-To: <4A85E2CB.4060502@gmail.com>
Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes
Content-Transfer-Encoding: 7bit
Mime-Version: 1.0 (Apple Message framework v936)
Date: Sun, 16 Aug 2009 11:38:05 +0200
References: <4A85E2CB.4060502@gmail.com>
X-Mailer: Apple Mail (2.936)
Cc: tsvwg@ietf.org
Subject: Re: [tsvwg] SCTP tie-tag info in state cookie?
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 16 Aug 2009 09:38:05 -0000

Hi Doug,

I'm not sure what you mean...

Possibly you are talking about the issue described in section 2.34 of
http://www.ietf.org/rfc/rfc4460.txt
This issue is addressed in
http://www.ietf.org/rfc/rfc4960.txt

Best regards
Michael

On Aug 15, 2009, at 12:18 AM, Doug Otis wrote:

>
> Is there an errata describing concerns related to obscuring tie-tag  
> info?
>
> -Doug
>
>


From doug.mtview@gmail.com  Mon Aug 17 15:18:07 2009
Return-Path: <doug.mtview@gmail.com>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id E78BE3A6FB5 for <tsvwg@core3.amsl.com>; Mon, 17 Aug 2009 15:18:07 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.607
X-Spam-Level: 
X-Spam-Status: No, score=-1.607 tagged_above=-999 required=5 tests=[AWL=0.692,  BAYES_00=-2.599, MIME_8BIT_HEADER=0.3]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id RiK-BStYmOhs for <tsvwg@core3.amsl.com>; Mon, 17 Aug 2009 15:18:07 -0700 (PDT)
Received: from wa-out-1112.google.com (wa-out-1112.google.com [209.85.146.179]) by core3.amsl.com (Postfix) with ESMTP id 3BA603A6FB4 for <tsvwg@ietf.org>; Mon, 17 Aug 2009 15:18:07 -0700 (PDT)
Received: by wa-out-1112.google.com with SMTP id v27so416929wah.5 for <tsvwg@ietf.org>; Mon, 17 Aug 2009 15:18:08 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from :user-agent:mime-version:to:cc:subject:references:in-reply-to :content-type:content-transfer-encoding; bh=z+s0VX1MOHhYhP16eaDoJ0K54XRNUUfUCVusmc4qmQM=; b=q6kWqswTmRJJs8ixn6ODEGrjQxdNGtNvDaDuePEwdhJ8vGFzkZe73IKNWccO4sIjRJ qfUyuzjx3DvKIvVCmyjFEtL+nrxfnt9HLaZjwGuUqqqZ0coEz1Wo2r1JToW3+2MwOq/4 c4dG05qmfdrXfERFU93NiLesh32FQ6HbDX5I0=
DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:user-agent:mime-version:to:cc:subject :references:in-reply-to:content-type:content-transfer-encoding; b=QFV74CYVgbKuW1vgS9NnSwjEUd/qPYsk/fL9YSCVq1IuQ2gNcbTTsfEWt6QWPpV1ml 4vnoHObJp1UIT8vFHtG5e0+AxTaxhqHPUiPnIAsNyWUmX+gS/KRsTzOKeG4U+bXL4eAz kfOHDtFUM3ge+XSqMt2RK0xfM/zirKfkxSq4Q=
Received: by 10.114.6.25 with SMTP id 25mr5126346waf.25.1250547488719; Mon, 17 Aug 2009 15:18:08 -0700 (PDT)
Received: from SJC-Office-NAT-214.mail-abuse.org (SJC-Office-NAT-214.mail-abuse.org [168.61.10.214]) by mx.google.com with ESMTPS id m6sm11181351wag.33.2009.08.17.15.18.06 (version=TLSv1/SSLv3 cipher=RC4-MD5); Mon, 17 Aug 2009 15:18:07 -0700 (PDT)
Message-ID: <4A89D71D.5050901@gmail.com>
Date: Mon, 17 Aug 2009 15:18:05 -0700
From: Doug Otis <doug.mtview@gmail.com>
User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.1.1) Gecko/20090715 Thunderbird/3.0b3
MIME-Version: 1.0
To: =?ISO-8859-1?Q?Michael_T=FCxen?= <Michael.Tuexen@lurchi.franken.de>
References: <4A85E2CB.4060502@gmail.com> <2A131DEC-80B7-4BF4-B6F8-F67E00C1610F@lurchi.franken.de>
In-Reply-To: <2A131DEC-80B7-4BF4-B6F8-F67E00C1610F@lurchi.franken.de>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 8bit
Cc: tsvwg@ietf.org
Subject: Re: [tsvwg] SCTP tie-tag info in state cookie?
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 17 Aug 2009 22:18:08 -0000

On 8/16/09 2:38 AM, Michael Tüxen wrote:
> Hi Doug,
>
> I'm not sure what you mean...
>
> Possibly you are talking about the issue described in section 2.34 of
> http://www.ietf.org/rfc/rfc4460.txt
> This issue is addressed in
> http://www.ietf.org/rfc/rfc4960.txt
>
> Best regards
> Michael
>
> On Aug 15, 2009, at 12:18 AM, Doug Otis wrote:
>
>>
>> Is there an errata describing concerns related to obscuring tie-tag info?

That answers the concern.  I should have seen this.  Thanks.

-Doug

From fernando@gont.com.ar  Tue Aug 18 04:18:07 2009
Return-Path: <fernando@gont.com.ar>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 0C1E53A6A11 for <tsvwg@core3.amsl.com>; Tue, 18 Aug 2009 04:18:07 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.185
X-Spam-Level: 
X-Spam-Status: No, score=-1.185 tagged_above=-999 required=5 tests=[BAYES_40=-0.185, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id iagAPFVmu8Zy for <tsvwg@core3.amsl.com>; Tue, 18 Aug 2009 04:18:06 -0700 (PDT)
Received: from smtp1.xmundo.net (smtp1.xmundo.net [201.216.232.80]) by core3.amsl.com (Postfix) with ESMTP id 33BD23A68F3 for <tsvwg@ietf.org>; Tue, 18 Aug 2009 04:18:05 -0700 (PDT)
Received: from venus.xmundo.net (venus.xmundo.net [201.216.232.56]) by smtp1.xmundo.net (Postfix) with ESMTP id 8BA176B65DD; Tue, 18 Aug 2009 08:10:34 -0300 (ART)
Received: from [192.168.0.151] (129-130-17-190.fibertel.com.ar [190.17.130.129]) (authenticated bits=0) by venus.xmundo.net (8.14.1/8.14.1) with ESMTP id n7IBARFJ028234; Tue, 18 Aug 2009 08:10:28 -0300
Message-ID: <4A8A8C25.6050707@gont.com.ar>
Date: Tue, 18 Aug 2009 08:10:29 -0300
From: Fernando Gont <fernando@gont.com.ar>
User-Agent: Thunderbird 2.0.0.22 (Windows/20090605)
MIME-Version: 1.0
To: tsvwg <tsvwg@ietf.org>
X-Enigmail-Version: 0.95.7
OpenPGP: id=D076FFF1
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
X-Greylist: Sender succeeded SMTP AUTH authentication, not delayed by milter-greylist-3.0 (venus.xmundo.net [201.216.232.56]); Tue, 18 Aug 2009 08:10:33 -0300 (ART)
Cc: Mark Allman <mallman@icir.org>
Subject: [tsvwg] port randomization I-D
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 18 Aug 2009 11:18:07 -0000

Hello, folks,

The last rev of the port randomization I-D was meant to address the wglc
comments (which, IIRC, were all submitted by Mark Allman).

Does this rev address all of them?

If so, should the i-d be shipped to the IESG?

Thanks!

Kind regards,
-- 
Fernando Gont
e-mail: fernando@gont.com.ar || fgont@acm.org
PGP Fingerprint: 7809 84F5 322E 45C7 F1C9 3945 96EE A9EF D076 FFF1





From gorry@erg.abdn.ac.uk  Fri Aug 21 10:01:44 2009
Return-Path: <gorry@erg.abdn.ac.uk>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 8902F3A6AD2 for <tsvwg@core3.amsl.com>; Fri, 21 Aug 2009 10:01:44 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.559
X-Spam-Level: 
X-Spam-Status: No, score=-2.559 tagged_above=-999 required=5 tests=[AWL=0.040,  BAYES_00=-2.599]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id May54-+imhNj for <tsvwg@core3.amsl.com>; Fri, 21 Aug 2009 10:01:43 -0700 (PDT)
Received: from erg.abdn.ac.uk (dee.erg.abdn.ac.uk [IPv6:2001:630:241:204:203:baff:fe9a:8c9b]) by core3.amsl.com (Postfix) with ESMTP id 8DD063A6974 for <tsvwg@ietf.org>; Fri, 21 Aug 2009 10:01:42 -0700 (PDT)
Received: from Gorry-Fairhursts-Laptop-6.local (ra-gorry.erg.abdn.ac.uk [139.133.204.42]) (authenticated bits=0) by erg.abdn.ac.uk (8.13.4/8.13.4) with ESMTP id n7LH1hZX016088 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for <tsvwg@ietf.org>; Fri, 21 Aug 2009 18:01:44 +0100 (BST)
Message-ID: <4A8ED2F7.8070706@erg.abdn.ac.uk>
Date: Fri, 21 Aug 2009 18:01:43 +0100
From: Gorry Fairhurst <gorry@erg.abdn.ac.uk>
Organization: The University of Aberdeen is a charity registered in Scotland,  No SC013683. 
User-Agent: Thunderbird 2.0.0.22 (Macintosh/20090605)
MIME-Version: 1.0
To: tsvwg <tsvwg@ietf.org>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
X-ERG-MailScanner: Found to be clean
X-ERG-MailScanner-From: gorry@erg.abdn.ac.uk
Subject: [tsvwg] First cut at TSV minutes for comment by the working group.
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
Reply-To: gorry@erg.abdn.ac.uk
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 21 Aug 2009 17:01:44 -0000

The draft notes from the TSV meeting in Stockholm are at the URL below, 
could you please check these for accuracy and for actions?

Best wishes,

Gorry & James

TSVWG meeting
http://www.ietf.org/proceedings/75/minutes/tsvwg.txt


From flefauch@cisco.com  Mon Aug 24 04:26:56 2009
Return-Path: <flefauch@cisco.com>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 6F0EB3A6A18 for <tsvwg@core3.amsl.com>; Mon, 24 Aug 2009 04:26:56 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -8.256
X-Spam-Level: 
X-Spam-Status: No, score=-8.256 tagged_above=-999 required=5 tests=[AWL=-1.657, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id H3X4r5VXUJ0c for <tsvwg@core3.amsl.com>; Mon, 24 Aug 2009 04:26:55 -0700 (PDT)
Received: from sj-iport-1.cisco.com (sj-iport-1.cisco.com [171.71.176.70]) by core3.amsl.com (Postfix) with ESMTP id 5C3923A69AF for <tsvwg@ietf.org>; Mon, 24 Aug 2009 04:26:55 -0700 (PDT)
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: ApoEACQWkkqrR7PD/2dsb2JhbAC8Y4g3jiQFgjiBYopJ
X-IronPort-AV: E=Sophos;i="4.44,265,1249257600"; d="scan'208";a="232020230"
Received: from sj-dkim-3.cisco.com ([171.71.179.195]) by sj-iport-1.cisco.com with ESMTP; 24 Aug 2009 11:27:01 +0000
Received: from sj-core-3.cisco.com (sj-core-3.cisco.com [171.68.223.137]) by sj-dkim-3.cisco.com (8.12.11/8.12.11) with ESMTP id n7OBR1Y5016184;  Mon, 24 Aug 2009 04:27:01 -0700
Received: from xbh-ams-101.cisco.com (xbh-ams-101.cisco.com [144.254.74.71]) by sj-core-3.cisco.com (8.13.8/8.14.3) with ESMTP id n7OBQtTl024628; Mon, 24 Aug 2009 11:27:01 GMT
Received: from xfe-ams-102.cisco.com ([144.254.231.94]) by xbh-ams-101.cisco.com with Microsoft SMTPSVC(6.0.3790.3959);  Mon, 24 Aug 2009 13:26:58 +0200
Received: from ams-flefauch-8712.cisco.com ([10.55.161.195]) by xfe-ams-102.cisco.com with Microsoft SMTPSVC(6.0.3790.3959);  Mon, 24 Aug 2009 13:26:58 +0200
Message-Id: <789B4642-4935-4508-956E-6AFFFA02AECD@cisco.com>
From: Francois Le Faucheur IMAP <flefauch@cisco.com>
To: Gorry Fairhurst <gorry@erg.abdn.ac.uk>, "James (jmpolk) Polk" <jmpolk@cisco.com>
In-Reply-To: <4A8ED2F7.8070706@erg.abdn.ac.uk>
Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes
Content-Transfer-Encoding: 7bit
Mime-Version: 1.0 (Apple Message framework v935.3)
Date: Mon, 24 Aug 2009 13:26:56 +0200
References: <4A8ED2F7.8070706@erg.abdn.ac.uk>
X-Mailer: Apple Mail (2.935.3)
X-OriginalArrivalTime: 24 Aug 2009 11:26:58.0134 (UTC) FILETIME=[CAA69F60:01CA24AD]
X-TM-AS-Product-Ver: SMEX-8.0.0.1181-5.600.1016-16844.005
X-TM-AS-Result: No--22.024100-8.000000-31
X-TM-AS-User-Approved-Sender: No
X-TM-AS-User-Blocked-Sender: No
DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; l=2621; t=1251113221; x=1251977221; c=relaxed/simple; s=sjdkim3002; h=Content-Type:From:Subject:Content-Transfer-Encoding:MIME-Version; d=cisco.com; i=flefauch@cisco.com; z=From:=20Francois=20Le=20Faucheur=20IMAP=20<flefauch@cisco. com> |Subject:=20Re=3A=20[tsvwg]=20First=20cut=20at=20TSV=20minu tes=20for=20comment=20by=20the=20working=20group. |Sender:=20; bh=tXxM0W0UZy4CZdHjgpFB0JHUUCm4NvevyQMHdCiyZE8=; b=vcduvzqRIuycywWYR6ZvOmtpQLIKHOfLsg+JPjXpcQzzu6pekTwUKSh7Vr Pg7ngKxflH0yE0KVofZqm6v0bO1Mfr3S6AI5RPpj3eNsG3z0v7wDEHnGKxjd Vwsj+zYloY;
Authentication-Results: sj-dkim-3; header.From=flefauch@cisco.com; dkim=pass ( sig from cisco.com/sjdkim3002 verified; ); 
Cc: Magnus Westerlund <magnus.westerlund@ericsson.com>, tsvwg list IETF <tsvwg@ietf.org>
Subject: Re: [tsvwg] First cut at TSV minutes for comment by the working group.
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 24 Aug 2009 11:26:56 -0000

Gorry, James,

Regarding the following draft minute text:
"
James asked for the current status of the I-D on RSVP extensions for  
emergency services.
Magnus Westerlund (as AD): The RSVP extensions for emergency services  
I-D has been returned to the WG with comments. It will need to go  
through another WGLC and then IETF last call before resubmission.
James: Francois has been trying to clear the discusses.
Magnus Westerlund: This document had an usual state where 2/3 of the  
IESG did not vote, requiring a resubmission. The proxy RSVP documents  
have been stuck in "discuss" within the IESG, there has been some some  
progress, and some discussion of end-to-end usage.
"

This seems to mix together the discussion about the I-D on RSVP  
extensions for emergency and the I-Ds on RSVP extensions for RSVP  
Proxy. Also, it is not quite in line with my recollection of the  
discussion.

Regarding I-D on RSVP extensions for Emergency:
* as I explained at the mike during WG:
	o we have already issued (in May) a new version (draft-ietf-tsvwg- 
emergency-rsvp-12.txt) that aims at addressing the 3 changes requested  
by the IESG.
	o we have been waiting from IESG (Ron Bonica in particular) for  
confirmation that this addresses their concerns.
	o the new rev does not introduce change of technical nature (ie it  
basically only affects applicability) and is basically a subset of  
previous version. For this reason, I don't believe the new rev  
justifies another WG Last Call.
* Janet Gunn agreed at the mike during WG that the new version does  
not introduce technical changes and does not justify another WG LC.
* I thought Magnus agreed then that we would not go thorough another  
WG LC and would proceed when we get feedback from IESG.

Regarding I-D on RSVP Proxy:
* as mentioned on the mike (I think):
	o this is the I-D for which we are working on addressing the  
"discusses".
	o we will post on the list a description of the approach proposed to  
IESG to address the topic of "migration to end-to-end" based on the  
proposal and request feedback
	o we will issue a new rev including this approach

(as not mentioned at the mike, generally we would need more IESG  
responsiveness to be able to make better progress).

Francois


On 21 Aug 2009, at 19:01, Gorry Fairhurst wrote:

>
> The draft notes from the TSV meeting in Stockholm are at the URL  
> below, could you please check these for accuracy and for actions?
>
> Best wishes,
>
> Gorry & James
>
> TSVWG meeting
> http://www.ietf.org/proceedings/75/minutes/tsvwg.txt
>


From flefauch@cisco.com  Mon Aug 24 04:31:29 2009
Return-Path: <flefauch@cisco.com>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 7A3023A68A7 for <tsvwg@core3.amsl.com>; Mon, 24 Aug 2009 04:31:29 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -8.072
X-Spam-Level: 
X-Spam-Status: No, score=-8.072 tagged_above=-999 required=5 tests=[AWL=-1.473, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id mYbmnq2H1uaJ for <tsvwg@core3.amsl.com>; Mon, 24 Aug 2009 04:31:28 -0700 (PDT)
Received: from sj-iport-1.cisco.com (sj-iport-1.cisco.com [171.71.176.70]) by core3.amsl.com (Postfix) with ESMTP id 8FAE23A6A1E for <tsvwg@ietf.org>; Mon, 24 Aug 2009 04:31:28 -0700 (PDT)
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: ApoEAEwXkkqrR7O6/2dsb2JhbAC8X4g3jiUFgjiBYg
X-IronPort-AV: E=Sophos;i="4.44,265,1249257600"; d="scan'208";a="232022111"
Received: from sj-dkim-2.cisco.com ([171.71.179.186]) by sj-iport-1.cisco.com with ESMTP; 24 Aug 2009 11:31:34 +0000
Received: from sj-core-4.cisco.com (sj-core-4.cisco.com [171.68.223.138]) by sj-dkim-2.cisco.com (8.12.11/8.12.11) with ESMTP id n7OBVYZ9016936;  Mon, 24 Aug 2009 04:31:34 -0700
Received: from xbh-ams-101.cisco.com (xbh-ams-101.cisco.com [144.254.74.71]) by sj-core-4.cisco.com (8.13.8/8.14.3) with ESMTP id n7OBV2jr023108; Mon, 24 Aug 2009 11:31:34 GMT
Received: from xfe-ams-101.cisco.com ([144.254.231.93]) by xbh-ams-101.cisco.com with Microsoft SMTPSVC(6.0.3790.3959);  Mon, 24 Aug 2009 13:31:29 +0200
Received: from ams-flefauch-8712.cisco.com ([10.55.161.195]) by xfe-ams-101.cisco.com with Microsoft SMTPSVC(6.0.3790.3959);  Mon, 24 Aug 2009 13:31:28 +0200
Message-Id: <DF44EEB2-2BEA-44C1-8CCA-639C3197A687@cisco.com>
From: Francois Le Faucheur IMAP <flefauch@cisco.com>
To: Francois Le Faucheur IMAP <flefauch@cisco.com>
In-Reply-To: <789B4642-4935-4508-956E-6AFFFA02AECD@cisco.com>
Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes
Content-Transfer-Encoding: 7bit
Mime-Version: 1.0 (Apple Message framework v935.3)
Date: Mon, 24 Aug 2009 13:31:26 +0200
References: <4A8ED2F7.8070706@erg.abdn.ac.uk> <789B4642-4935-4508-956E-6AFFFA02AECD@cisco.com>
X-Mailer: Apple Mail (2.935.3)
X-OriginalArrivalTime: 24 Aug 2009 11:31:28.0728 (UTC) FILETIME=[6BEFFD80:01CA24AE]
DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; l=3107; t=1251113494; x=1251977494; c=relaxed/simple; s=sjdkim2002; h=Content-Type:From:Subject:Content-Transfer-Encoding:MIME-Version; d=cisco.com; i=flefauch@cisco.com; z=From:=20Francois=20Le=20Faucheur=20IMAP=20<flefauch@cisco. com> |Subject:=20Re=3A=20[tsvwg]=20First=20cut=20at=20TSV=20minu tes=20for=20comment=20by=20the=20working=20group. |Sender:=20; bh=MH5YhY1jP0P8xilVR2AQa37u4kzP6tMZsv1AqEeyb9A=; b=e1IaOphBf5Cp4g1Um8R6lRjz0r8MKvKFoz/DnIGmsImCgss6x3o1+z4Uvi MNoOXfSgszwMz8VHViwhs/dV4776gw+S4TKOX4BRjBwVEcr0mgO6zzMcC/GG /6Kee0pelu;
Authentication-Results: sj-dkim-2; header.From=flefauch@cisco.com; dkim=pass ( sig from cisco.com/sjdkim2002 verified; ); 
Cc: Gorry Fairhurst <gorry@erg.abdn.ac.uk>, Magnus Westerlund <magnus.westerlund@ericsson.com>, "James \(jmpolk\) Polk" <jmpolk@cisco.com>, tsvwg list IETF <tsvwg@ietf.org>
Subject: Re: [tsvwg] First cut at TSV minutes for comment by the working group.
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 24 Aug 2009 11:31:29 -0000

Gorry, James,

I just noticed that you already cover most of my points regarding RSVP  
extensions for emergency later in the minutes under item 4. So that's  
cool. But still, I suggest the discussion at the beginning should  
better separate the discussion on "emergency" vs "Proxy".

Francois

On 24 Aug 2009, at 13:26, Francois Le Faucheur IMAP wrote:

> Gorry, James,
>
> Regarding the following draft minute text:
> "
> James asked for the current status of the I-D on RSVP extensions for  
> emergency services.
> Magnus Westerlund (as AD): The RSVP extensions for emergency  
> services I-D has been returned to the WG with comments. It will need  
> to go through another WGLC and then IETF last call before  
> resubmission.
> James: Francois has been trying to clear the discusses.
> Magnus Westerlund: This document had an usual state where 2/3 of the  
> IESG did not vote, requiring a resubmission. The proxy RSVP  
> documents have been stuck in "discuss" within the IESG, there has  
> been some some progress, and some discussion of end-to-end usage.
> "
>
> This seems to mix together the discussion about the I-D on RSVP  
> extensions for emergency and the I-Ds on RSVP extensions for RSVP  
> Proxy. Also, it is not quite in line with my recollection of the  
> discussion.
>
> Regarding I-D on RSVP extensions for Emergency:
> * as I explained at the mike during WG:
> 	o we have already issued (in May) a new version (draft-ietf-tsvwg- 
> emergency-rsvp-12.txt) that aims at addressing the 3 changes  
> requested by the IESG.
> 	o we have been waiting from IESG (Ron Bonica in particular) for  
> confirmation that this addresses their concerns.
> 	o the new rev does not introduce change of technical nature (ie it  
> basically only affects applicability) and is basically a subset of  
> previous version. For this reason, I don't believe the new rev  
> justifies another WG Last Call.
> * Janet Gunn agreed at the mike during WG that the new version does  
> not introduce technical changes and does not justify another WG LC.
> * I thought Magnus agreed then that we would not go thorough another  
> WG LC and would proceed when we get feedback from IESG.
>
> Regarding I-D on RSVP Proxy:
> * as mentioned on the mike (I think):
> 	o this is the I-D for which we are working on addressing the  
> "discusses".
> 	o we will post on the list a description of the approach proposed  
> to IESG to address the topic of "migration to end-to-end" based on  
> the proposal and request feedback
> 	o we will issue a new rev including this approach
>
> (as not mentioned at the mike, generally we would need more IESG  
> responsiveness to be able to make better progress).
>
> Francois
>
>
> On 21 Aug 2009, at 19:01, Gorry Fairhurst wrote:
>
>>
>> The draft notes from the TSV meeting in Stockholm are at the URL  
>> below, could you please check these for accuracy and for actions?
>>
>> Best wishes,
>>
>> Gorry & James
>>
>> TSVWG meeting
>> http://www.ietf.org/proceedings/75/minutes/tsvwg.txt
>>
>


From lars.eggert@nokia.com  Mon Aug 24 06:53:53 2009
Return-Path: <lars.eggert@nokia.com>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 7122B28C166 for <tsvwg@core3.amsl.com>; Mon, 24 Aug 2009 06:53:53 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.599
X-Spam-Level: 
X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id MW-ZzLLcVryd for <tsvwg@core3.amsl.com>; Mon, 24 Aug 2009 06:53:52 -0700 (PDT)
Received: from mail.fit.nokia.com (mail.fit.nokia.com [195.148.124.195]) by core3.amsl.com (Postfix) with ESMTP id 5C81828C0F2 for <tsvwg@ietf.org>; Mon, 24 Aug 2009 06:53:52 -0700 (PDT)
Received: from wifi-visiteurs-107.sri.ucl.ac.be (wifi-visiteurs-107.sri.ucl.ac.be [130.104.202.235]) (authenticated bits=0) by mail.fit.nokia.com (8.14.3/8.14.3) with ESMTP id n7ODrnrC039777 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT) for <tsvwg@ietf.org>; Mon, 24 Aug 2009 16:53:50 +0300 (EEST) (envelope-from lars.eggert@nokia.com)
Message-Id: <3B0EA24C-E8C1-48EC-A233-BBD26D6DDEC6@nokia.com>
From: Lars Eggert <lars.eggert@nokia.com>
To: tsvwg <tsvwg@ietf.org>
In-Reply-To: <F46ED6BF-5DB6-414A-9E81-9F68D73EA525@nokia.com>
Content-Type: multipart/signed; boundary=Apple-Mail-34-84399805; micalg=sha1; protocol="application/pkcs7-signature"
Mime-Version: 1.0 (Apple Message framework v936)
Date: Mon, 24 Aug 2009 15:53:42 +0200
References: <20090811081501.A2F2C3A68F7@core3.amsl.com> <F46ED6BF-5DB6-414A-9E81-9F68D73EA525@nokia.com>
X-Mailer: Apple Mail (2.936)
Subject: Re: [tsvwg] I-D Action:draft-ietf-tsvwg-iana-ports-02.txt
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 24 Aug 2009 13:53:53 -0000

--Apple-Mail-34-84399805
Content-Type: text/plain;
	charset=US-ASCII;
	format=flowed;
	delsp=yes
Content-Transfer-Encoding: 7bit

Hi,

so, in case anyone needs some reading material, we would appreciate  
your thoughts on this one...

Lars

On 2009-8-12, at 8:56, Eggert Lars (Nokia-NRC/Espoo) wrote:
> On 2009-8-11, at 11:15, Internet-Drafts@ietf.org wrote:
>> 	Title           : Internet Assigned Numbers Authority (IANA)  
>> Procedures for the Management of the Transport Protocol Port Number  
>> and Service Name Registry
>> 	Author(s)       : M. Cotton, et al.
>> 	Filename        : draft-ietf-tsvwg-iana-ports-02.txt
>> 	Pages           : 24
>> 	Date            : 2009-08-11
>
> this is a pretty significant update based on discussions during the  
> last two IETFs. In my personal opinion, this document is nearing  
> completion and would benefit from careful review.
>
> Lars


--Apple-Mail-34-84399805
Content-Disposition: attachment;
	filename=smime.p7s
Content-Type: application/pkcs7-signature;
	name=smime.p7s
Content-Transfer-Encoding: base64
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--Apple-Mail-34-84399805--

From padman.m@gmail.com  Mon Aug 31 04:57:14 2009
Return-Path: <padman.m@gmail.com>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 9108C3A6991 for <tsvwg@core3.amsl.com>; Mon, 31 Aug 2009 04:57:14 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.51
X-Spam-Level: 
X-Spam-Status: No, score=-0.51 tagged_above=-999 required=5 tests=[BAYES_05=-1.11, J_CHICKENPOX_52=0.6]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 9bZLHD09Wkpr for <tsvwg@core3.amsl.com>; Mon, 31 Aug 2009 04:57:14 -0700 (PDT)
Received: from mail-px0-f181.google.com (mail-px0-f181.google.com [209.85.216.181]) by core3.amsl.com (Postfix) with ESMTP id 0ACE13A6A86 for <tsvwg@ietf.org>; Mon, 31 Aug 2009 04:57:14 -0700 (PDT)
Received: by pxi11 with SMTP id 11so42840pxi.17 for <tsvwg@ietf.org>; Mon, 31 Aug 2009 04:57:22 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:date:message-id:subject :from:to:content-type; bh=hJHmTW8c0iXHynswGfKQ1HHVyiXFROczuQ33plJo3B0=; b=QreADL95o6733xw2Dvg8FK3/RIFBrLrZENSR+n4l7KIMtv/KLfqfY/AuLEbDK7x6pi uqk5uJ9uSIida0++nPeyY2Q7uwtH8zNnjJKBE9iUpzwKhHKdTA35Gj2G9OrjlHp3WhQB H5VuEDkQoytwJYtbZudrmFX/dah4xz5FyF7PI=
DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:date:message-id:subject:from:to:content-type; b=aQi9I6ltf5VbZmVxPdQqI/GhjAPKmnR/7BDTyBfykbxTv2M43fNVek87aEeWKfHOoz vTfu7/sZxir1bwa5kpabfnuL+2v1AL/KaOabXq7dNaIWmat0I6MWrin682OPzKqQ5A92 Huq/Oxxvqq8D5vWnx0u9OBppKCtZupAZGvl4E=
MIME-Version: 1.0
Received: by 10.114.18.29 with SMTP id 29mr2470714war.147.1251719842620; Mon,  31 Aug 2009 04:57:22 -0700 (PDT)
Date: Mon, 31 Aug 2009 17:27:22 +0530
Message-ID: <816fa6970908310457o1c0a9fc9p284c7875389f5f6a@mail.gmail.com>
From: Padmalochan Moharana <padman.m@gmail.com>
To: tsvwg <tsvwg@ietf.org>
Content-Type: text/plain; charset=ISO-8859-1
Subject: [tsvwg] Query regarding Abort cause.
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 31 Aug 2009 11:57:14 -0000

Hi ,

Should one end point send Abort with cause "User-Initiated Abort"to
peer end on receive packet COOKIE-ECHO message with ERROR chunk (for
"Adoption layer indication" unrecognized parameter in INIT ACK).

If no then in  which case SCTP should sent Abort to peer with cause
"User-Initiated Abort".

Br,
Padmalochan

From randall@lakerest.net  Mon Aug 31 05:14:18 2009
Return-Path: <randall@lakerest.net>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 6B24628C235 for <tsvwg@core3.amsl.com>; Mon, 31 Aug 2009 05:14:18 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.999
X-Spam-Level: 
X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, J_CHICKENPOX_52=0.6]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id thC5PJLIcOQ0 for <tsvwg@core3.amsl.com>; Mon, 31 Aug 2009 05:14:17 -0700 (PDT)
Received: from lakerest.net (unknown [IPv6:2001:240:585:2:213:d4ff:fef3:2d8d]) by core3.amsl.com (Postfix) with ESMTP id 1EEF428C257 for <tsvwg@ietf.org>; Mon, 31 Aug 2009 05:13:08 -0700 (PDT)
Received: from [10.1.1.53] ([10.1.1.53]) (authenticated bits=0) by lakerest.net (8.14.3/8.14.3) with ESMTP id n7VCDB25048893 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Mon, 31 Aug 2009 08:13:11 -0400 (EDT) (envelope-from randall@lakerest.net)
DKIM-Signature: a=rsa-sha1; c=simple/simple; d=lakerest.net; s=mail; t=1251720791; h=Cc:Message-Id:From:To:In-Reply-To:Content-Type: Content-Transfer-Encoding:Mime-Version:Subject:Date:References: X-Mailer; b=bXSXCrgnt1MEmUEF54sUKU0by/Tl6SUwEE1THO5/LzuV96UwacqbsE7 bTtTjWq0f571w9cwL1thJqLRz0jaAWA==
Message-Id: <0FE59A98-03D2-4B60-9F56-E32A3FF2DE7E@lakerest.net>
From: Randy Stewart <randall@lakerest.net>
To: Padmalochan Moharana <padman.m@gmail.com>
In-Reply-To: <816fa6970908310457o1c0a9fc9p284c7875389f5f6a@mail.gmail.com>
Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes
Content-Transfer-Encoding: 7bit
Mime-Version: 1.0 (Apple Message framework v936)
Date: Mon, 31 Aug 2009 08:13:10 -0400
References: <816fa6970908310457o1c0a9fc9p284c7875389f5f6a@mail.gmail.com>
X-Mailer: Apple Mail (2.936)
Cc: tsvwg <tsvwg@ietf.org>
Subject: Re: [tsvwg] Query regarding Abort cause.
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 31 Aug 2009 12:14:18 -0000

On Aug 31, 2009, at 7:57 AM, Padmalochan Moharana wrote:

> Hi ,
>
> Should one end point send Abort with cause "User-Initiated Abort"to
> peer end on receive packet COOKIE-ECHO message with ERROR chunk (for
> "Adoption layer indication" unrecognized parameter in INIT ACK).

Adaption layer indications are just "advisory". They mean only
something to the upper layer above SCTP. So SCTP itself should not
be responding with an ABORT with such an error cause.

>
> If no then in  which case SCTP should sent Abort to peer with cause
> "User-Initiated Abort".

Where SCTP would respond with such a case is when the upper layer
accepted the connection (or received the association event), and then
did an ABORT by either closing the accepted socket with linger set to 0
OR (using the 1-M model) send a SCTP_ABORT in options, and possibly  
including
a cause reason that would also accompany the send.

This of course is speaking in terms of the socket API...

In any event SCTP itself should not take such a view... i.e. if I get
an unknown adaption layer indication.. since an adaption indication is
for the user of SCTP not SCTP itself..

Hope that helps

R


>
> Br,
> Padmalochan
>

-----
Randall Stewart
randall@lakerest.net





From zoltan.kiss@nsn.com  Mon Aug 31 08:17:14 2009
Return-Path: <zoltan.kiss@nsn.com>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 9D34D3A6E66 for <tsvwg@core3.amsl.com>; Mon, 31 Aug 2009 08:17:14 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.308
X-Spam-Level: 
X-Spam-Status: No, score=-0.308 tagged_above=-999 required=5 tests=[AWL=2.291,  BAYES_00=-2.599]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id NQkY5EO6jUWE for <tsvwg@core3.amsl.com>; Mon, 31 Aug 2009 08:17:13 -0700 (PDT)
Received: from demumfd002.nsn-inter.net (demumfd002.nsn-inter.net [217.115.75.234]) by core3.amsl.com (Postfix) with ESMTP id 7EF3D28C298 for <tsvwg@ietf.org>; Mon, 31 Aug 2009 08:17:13 -0700 (PDT)
Received: from demuprx016.emea.nsn-intra.net ([10.150.129.55]) by demumfd002.nsn-inter.net (8.12.11.20060308/8.12.11) with ESMTP id n7VFHLx4027718 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=FAIL) for <tsvwg@ietf.org>; Mon, 31 Aug 2009 17:17:21 +0200
Received: from demuexc024.nsn-intra.net (demuexc024.nsn-intra.net [10.159.32.11]) by demuprx016.emea.nsn-intra.net (8.12.11.20060308/8.12.11) with ESMTP id n7VFHLHn002837 for <tsvwg@ietf.org>; Mon, 31 Aug 2009 17:17:21 +0200
Received: from DEMUEXC035.nsn-intra.net ([10.150.128.26]) by demuexc024.nsn-intra.net with Microsoft SMTPSVC(6.0.3790.3959);  Mon, 31 Aug 2009 17:17:21 +0200
X-MimeOLE: Produced By Microsoft Exchange V6.5
Content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
Date: Mon, 31 Aug 2009 17:17:18 +0200
Message-ID: <79C4240C13B4C84B910850B96B1B43123FD716@DEMUEXC035.nsn-intra.net>
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Thread-Topic: Multihoming and source address selection
Thread-Index: AcoqTiFd5kX7pQeYQdq5ltf9XSNoPQ==
From: "Kiss, Zoltan (NSN - HU/Budapest)" <zoltan.kiss@nsn.com>
To: <tsvwg@ietf.org>
X-OriginalArrivalTime: 31 Aug 2009 15:17:21.0469 (UTC) FILETIME=[22E466D0:01CA2A4E]
Subject: [tsvwg] Multihoming and source address selection
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 31 Aug 2009 15:50:24 -0000

Hi folks,

I have some questions about SCTP multihoming. I'm testing SCTP MH
between 2 Linux-based machine, both are connected with 2 NIC to the
network, owning 2 separate IP on them, all from different subnets. When
I set up multihomed associations between them, they agreed about IP
addresses correctly during INIT, and start to use one pair of them as
"primary path". Actually this pair was used on both machines as
interfaces to default gateway. I will call them A and A', and the
secondary addresses as B and B', and I also have to notify, that each
NIC can access both addresses of the other side. It works properly, but
when one side sent a HEARTBEAT to the peer's secondary address, it
received the HB ACK from the primary address, and the HEARTBEATs were
always sent from the primary address. I quickly realized that this is a
problem of IP routing, so I configured it, so now packets to peer's
primary address are always sent from our primary address, and secondary
on secondary, respectively. But in this way I lost some redundancy,
because now primary addresses won't access secondaries and vice versa,
and I have to do additional configuring for static routing.
As I see, the root cause of all of these that SCTP layer doesn't specify
the source IP address when it passes packet to IP, which raised me the
question: is this good or not? I read through the RFC but it doesn't
specify this exactly, and I also couldn't decide it unambiguously. Using
only IP static routing has a disadvantage that we can use only disjunct
pairs of source and destination addresses, a reduced set of possible
paths, and we have to do extra configs, which is an another point of
failure. I think involve SCTP as well by means of specifying source IP
address for packets is better, therefore it could try and find every
possible pairs with HEARTBEAT, and we also don't need static routing.
But as I saw during my research, the opposite way is the trend. Is there
any point of the RFC where is this succeed from? Or is this just common
practice in SCTP implementations, based on real world experiences which
I don't know? Are my assumptions correct?

Regards,

Zoltan Kiss

From randall@lakerest.net  Mon Aug 31 13:37:32 2009
Return-Path: <randall@lakerest.net>
X-Original-To: tsvwg@core3.amsl.com
Delivered-To: tsvwg@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 56B1228C537 for <tsvwg@core3.amsl.com>; Mon, 31 Aug 2009 13:37:32 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.299
X-Spam-Level: 
X-Spam-Status: No, score=-2.299 tagged_above=-999 required=5 tests=[AWL=0.300,  BAYES_00=-2.599]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 8bUakEVU522u for <tsvwg@core3.amsl.com>; Mon, 31 Aug 2009 13:37:31 -0700 (PDT)
Received: from lakerest.net (unknown [IPv6:2001:240:585:2:213:d4ff:fef3:2d8d]) by core3.amsl.com (Postfix) with ESMTP id 1F28F28C245 for <tsvwg@ietf.org>; Mon, 31 Aug 2009 13:37:20 -0700 (PDT)
Received: from [10.1.1.53] ([10.1.1.53]) (authenticated bits=0) by lakerest.net (8.14.3/8.14.3) with ESMTP id n7VKbT57070600 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Mon, 31 Aug 2009 16:37:29 -0400 (EDT) (envelope-from randall@lakerest.net)
DKIM-Signature: a=rsa-sha1; c=simple/simple; d=lakerest.net; s=mail; t=1251751050; h=Cc:Message-Id:From:To:In-Reply-To:Content-Type: Content-Transfer-Encoding:Mime-Version:Subject:Date:References: X-Mailer; b=PkwU16dPQv4PX92ysVy4EyuV70B814cfh5W5ucbI2aM/BqpUnKG2jS/ j6M2svipexwCgTosKmN26KETQP6AnIw==
Message-Id: <3A004BA7-4765-4874-A04E-5879CC97A1E5@lakerest.net>
From: Randy Stewart <randall@lakerest.net>
To: "Kiss, Zoltan (NSN - HU/Budapest)" <zoltan.kiss@nsn.com>
In-Reply-To: <79C4240C13B4C84B910850B96B1B43123FD716@DEMUEXC035.nsn-intra.net>
Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes
Content-Transfer-Encoding: 7bit
Mime-Version: 1.0 (Apple Message framework v936)
Date: Mon, 31 Aug 2009 16:37:28 -0400
References: <79C4240C13B4C84B910850B96B1B43123FD716@DEMUEXC035.nsn-intra.net>
X-Mailer: Apple Mail (2.936)
Cc: tsvwg@ietf.org
Subject: Re: [tsvwg] Multihoming and source address selection
X-BeenThere: tsvwg@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Transport Area Working Group <tsvwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tsvwg>
List-Post: <mailto:tsvwg@ietf.org>
List-Help: <mailto:tsvwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tsvwg>, <mailto:tsvwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 31 Aug 2009 20:37:32 -0000

Zoltan:


Actually I would bet that Linux IS specifying the src address.
I know that FreeBSD does.

Just because SCTP places B' in the source address DOES NOT mean
that the packet will go out the B' interface. This goes back to
routing.. as you correctly identify. Routing in host's generally
do not look at src address. Instead only the destination address
is looked at. In fact this is pretty much true in most of the internet
as well ;-)

I actually had in FreeBSD at one time a set of patches that would allow
SCTP to do "alternate routing". Which might be what you want...

This would allow you to enter two default addresses for example and
then when you hit a retransmission you could ask for a new route (or
N retransmissions) so that you could steer the packet out the other
interface (even with the original src address)....

What generally happened with this code (using your descriptions) is:

  Host
      A  --- IP1 Router ------ A'
      B  --- IP2 Router -------B'


You would set in two default routes i.e.

route add default ip1
route add default ip2

And then when SCTP gathered its routes into the routing table
it would get the first route for A' and IP1 would be the gateway.
It would next ask for a route for B' but would indicated it did not
want (if possible) the route via IP1. It would then get back the
route via IP2.

If one started having issues with retransmission, it would ask for
an alternate route and get the opposite...

But of course this code is long dead... I have been meaning to
circle back and resurrect it but alas a shortage or cycles ;-)
Maybe someday... ;-)

R

On Aug 31, 2009, at 11:17 AM, Kiss, Zoltan (NSN - HU/Budapest) wrote:

> Hi folks,
>
> I have some questions about SCTP multihoming. I'm testing SCTP MH
> between 2 Linux-based machine, both are connected with 2 NIC to the
> network, owning 2 separate IP on them, all from different subnets.  
> When
> I set up multihomed associations between them, they agreed about IP
> addresses correctly during INIT, and start to use one pair of them as
> "primary path". Actually this pair was used on both machines as
> interfaces to default gateway. I will call them A and A', and the
> secondary addresses as B and B', and I also have to notify, that each
> NIC can access both addresses of the other side. It works properly,  
> but
> when one side sent a HEARTBEAT to the peer's secondary address, it
> received the HB ACK from the primary address, and the HEARTBEATs were
> always sent from the primary address. I quickly realized that this  
> is a
> problem of IP routing, so I configured it, so now packets to peer's
> primary address are always sent from our primary address, and  
> secondary
> on secondary, respectively. But in this way I lost some redundancy,
> because now primary addresses won't access secondaries and vice versa,
> and I have to do additional configuring for static routing.
> As I see, the root cause of all of these that SCTP layer doesn't  
> specify
> the source IP address when it passes packet to IP, which raised me the
> question: is this good or not? I read through the RFC but it doesn't
> specify this exactly, and I also couldn't decide it unambiguously.  
> Using
> only IP static routing has a disadvantage that we can use only  
> disjunct
> pairs of source and destination addresses, a reduced set of possible
> paths, and we have to do extra configs, which is an another point of
> failure. I think involve SCTP as well by means of specifying source IP
> address for packets is better, therefore it could try and find every
> possible pairs with HEARTBEAT, and we also don't need static routing.
> But as I saw during my research, the opposite way is the trend. Is  
> there
> any point of the RFC where is this succeed from? Or is this just  
> common
> practice in SCTP implementations, based on real world experiences  
> which
> I don't know? Are my assumptions correct?
>
> Regards,
>
> Zoltan Kiss
>

-----
Randall Stewart
randall@lakerest.net




