From 558-mailing-list-admin@catarina.usc.edu  Sat Mar  1 08:12:38 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA04258
	for <pim-archive@lists.ietf.org>; Sat, 1 Mar 2003 08:12:37 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h21DE8l14787
	for <pim-archive@lists.ietf.org>; Sat, 1 Mar 2003 05:14:08 -0800 (PST)
	(envelope-from 558-mailing-list-admin@catarina.usc.edu)
Date: Sat, 01 Mar 2003 05:14:08 -0800
Message-ID: <20030301131408.10806.57553.Mailman@catarina.usc.edu>
Subject: catarina.usc.edu mailing list memberships reminder
From: mailman-owner@catarina.usc.edu
To: pim-archive@ietf.org
X-No-Archive: yes
X-Ack: no
Sender: 558-mailing-list-admin@catarina.usc.edu
Errors-To: 558-mailing-list-admin@catarina.usc.edu
X-BeenThere: 558-mailing-list@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk

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

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

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

If you have questions, problems, comments, etc, send them to
mailman-owner@catarina.usc.edu.  Thanks!

Passwords for pim-archive@lists.ietf.org:

List                                     Password // URL
----                                     --------  
pim@catarina.usc.edu                     kimoit    
http://catarina.usc.edu/mailman/options/pim/pim-archive%40lists.ietf.org


From pim-admin@catarina.usc.edu  Sat Mar  1 18:23:41 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA20513
	for <pim-archive@lists.ietf.org>; Sat, 1 Mar 2003 18:23:39 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h21NL3l20628;
	Sat, 1 Mar 2003 15:21:04 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from exch-cluster-01.force10networks.com (corp.force10networks.com [206.54.51.125])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h21NJ9l20593
	for <pim@catarina.usc.edu>; Sat, 1 Mar 2003 15:19:09 -0800 (PST)
	(envelope-from svallepa@force10networks.com)
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C2E048.F2FD1910"
Message-ID: <7A38462BE63B224DA3A758F9E325A0C9440C42@EXCH-CLUSTER-01.force10networks.com>
Thread-Topic: (S,G) Assert state machine in Winner state.
Thread-Index: AcLgSPL93G5kAItTTtaLXu3LWIoCVg==
From: "Sridhar Vallepalli" <svallepa@force10networks.com>
To: <pim@catarina.usc.edu>
Subject: [pim] (S,G) Assert state machine in Winner state.
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Sat, 1 Mar 2003 15:19:03 -0800

This is a multi-part message in MIME format.

------_=_NextPart_001_01C2E048.F2FD1910
Content-Type: text/plain;
	charset="utf-8"
Content-Transfer-Encoding: base64
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------_=_NextPart_001_01C2E048.F2FD1910
Content-Type: text/html;
	charset="utf-8"
Content-Transfer-Encoding: base64
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------_=_NextPart_001_01C2E048.F2FD1910--
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Sun Mar  2 15:36:36 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA18106
	for <pim-archive@lists.ietf.org>; Sun, 2 Mar 2003 15:36:36 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h22KY5l32849;
	Sun, 2 Mar 2003 12:34:05 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from mmm.icir.org (ppp3.ICSI.Berkeley.EDU [192.150.186.243])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h22KV3l32816
	for <pim@catarina.usc.edu>; Sun, 2 Mar 2003 12:31:03 -0800 (PST)
	(envelope-from mjh@mmm.icir.org)
Received: from mmm.icir.org (localhost [127.0.0.1])
	by mmm.icir.org (8.12.6/8.12.6) with ESMTP id h22KV3BE016713;
	Sun, 2 Mar 2003 12:31:04 -0800 (PST)
From: Mark Handley <mjh@icir.org>
X-Organisation: ICIR
To: "Sridhar Vallepalli" <svallepa@force10networks.com>
cc: pim@catarina.usc.edu
Subject: Re: [pim] (S,G) Assert state machine in Winner state. 
In-reply-to: Your message of "Sat, 01 Mar 2003 15:19:03 PST."
             <7A38462BE63B224DA3A758F9E325A0C9440C42@EXCH-CLUSTER-01.force10networks.com> 
Message-ID: <16712.1046637063@mmm.icir.org>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Sun, 02 Mar 2003 12:31:03 -0800


We made a policy decision not to optimize corner cases - PIM-SM is too
complex already.  This is a corner case, and not considered worthy of
optimization.

 - Mark
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar  3 16:53:26 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA14296
	for <pim-archive@lists.ietf.org>; Mon, 3 Mar 2003 16:53:26 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h23LmPl48213;
	Mon, 3 Mar 2003 13:48:26 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from hotmail.com (f96.law15.hotmail.com [64.4.23.96])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h23Lj4l48177
	for <pim@catarina.usc.edu>; Mon, 3 Mar 2003 13:45:04 -0800 (PST)
	(envelope-from dsa0000@hotmail.com)
Received: from mail pickup service by hotmail.com with Microsoft SMTPSVC;
	 Mon, 3 Mar 2003 13:44:59 -0800
Received: from 64.163.203.180 by lw15fd.law15.hotmail.msn.com with HTTP;
	Mon, 03 Mar 2003 21:44:58 GMT
X-Originating-IP: [64.163.203.180]
From: "dsa main" <dsa0000@hotmail.com>
To: pim@catarina.usc.edu
Mime-Version: 1.0
Content-Type: text/plain; format=flowed
Message-ID: <F96cNdbUu2o8q3p9Pcw0001f020@hotmail.com>
X-OriginalArrivalTime: 03 Mar 2003 21:44:59.0066 (UTC) FILETIME=[233F51A0:01C2E1CE]
Subject: [pim] PIM snooping resources
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Mon, 03 Mar 2003 21:44:58 +0000

Hi All,

I am looking for some pointers on PIM snooping...

Is there any documents on PIM snooping implementation ?

Regards
DSA







_________________________________________________________________
MSN 8 with e-mail virus protection service: 2 months FREE* 
http://join.msn.com/?page=features/virus

_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar  3 20:10:43 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA19952
	for <pim-archive@lists.ietf.org>; Mon, 3 Mar 2003 20:10:42 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2418kl50334;
	Mon, 3 Mar 2003 17:08:46 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from mail.gmx.net (mail.gmx.net [213.165.64.20])
	by catarina.usc.edu (8.11.6/8.11.6) with SMTP id h23NK7l49175
	for <pim@catarina.usc.edu>; Mon, 3 Mar 2003 15:20:07 -0800 (PST)
	(envelope-from gerd.besch@gmx.net)
Received: (qmail 17190 invoked by uid 0); 3 Mar 2003 23:19:55 -0000
Received: from p5091657A.dip.t-dialin.net (HELO gmx.net) (80.145.101.122)
  by mail.gmx.net (mp015-rz3) with SMTP; 3 Mar 2003 23:19:55 -0000
Message-ID: <3E63E394.4070508@gmx.net>
From: Gerd Besch <gerd.besch@gmx.net>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.0.2) Gecko/20030208 Netscape/7.02
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: pim@catarina.usc.edu
Subject: Re: [pim] PIM snooping resources
References: <F96cNdbUu2o8q3p9Pcw0001f020@hotmail.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Tue, 04 Mar 2003 00:21:56 +0100
Content-Transfer-Encoding: 7bit

  Hi DSA,

I worte my thesis about the following subject 'Analysis and engineering 
of the infrastructure required to support multicast traffic at an 
Internet Exchange point' 
<http://besch.gmxhome.de/eng/thesis-gerd-besch-preface.html>

Abstract:
"This thesis discusses using 'PIM-snooping' as a solution for handling
multicast traffic at a Multicast Internet Exchange Point (MIX) on a shared
Gigabit-Ethernet switch fabric. Existing MIXs essentially treat multicast
traffic as broadcast traffic. Switch ports are flooded due to the lack of
information on how to efficiently replicate multicast traffic only onto 
ports
where receivers for specific multicast groups exist.
Protocol Independent Multicast Join/Prune messages can be exploited to
provide necessary multicast containment."

If you're interested to receive a copy (PDF) of my thesis just drop me a 
line.

cheers,
Gerd

ps: real-name postings would be really coooooool! ;))))))

dsa main wrote:

> Hi All,
>
> I am looking for some pointers on PIM snooping...
>
> Is there any documents on PIM snooping implementation ?
>
> Regards
> DSA




_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar  3 20:18:13 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA20077
	for <pim-archive@lists.ietf.org>; Mon, 3 Mar 2003 20:18:11 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h241H5l50462;
	Mon, 3 Mar 2003 17:17:05 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from possum.icir.org (possum.icir.org [192.150.187.67])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h241GYl50444
	for <pim@catarina.usc.edu>; Mon, 3 Mar 2003 17:16:35 -0800 (PST)
	(envelope-from pavlin@possum.icir.org)
Received: from possum.icir.org (localhost [127.0.0.1])
	by possum.icir.org (8.12.3/8.12.3) with ESMTP id h241GYmo025584;
	Mon, 3 Mar 2003 17:16:34 -0800 (PST)
	(envelope-from pavlin@possum.icir.org)
Message-Id: <200303040116.h241GYmo025584@possum.icir.org>
To: Gerd Besch <gerd.besch@gmx.net>
cc: pim@catarina.usc.edu
Subject: Re: [pim] PIM snooping resources 
In-Reply-To: Message from Gerd Besch <gerd.besch@gmx.net> 
   of "Tue, 04 Mar 2003 00:21:56 +0100." <3E63E394.4070508@gmx.net> 
From: Pavlin Radoslavov <pavlin@icir.org>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Mon, 03 Mar 2003 17:16:34 -0800

>   Hi DSA,
> 
> I worte my thesis about the following subject 'Analysis and engineering 
> of the infrastructure required to support multicast traffic at an 
> Internet Exchange point' 
> <http://besch.gmxhome.de/eng/thesis-gerd-besch-preface.html>
> 
> Abstract:
> "This thesis discusses using 'PIM-snooping' as a solution for handling
> multicast traffic at a Multicast Internet Exchange Point (MIX) on a shared
> Gigabit-Ethernet switch fabric. Existing MIXs essentially treat multicast
> traffic as broadcast traffic. Switch ports are flooded due to the lack of
> information on how to efficiently replicate multicast traffic only onto 
> ports
> where receivers for specific multicast groups exist.
> Protocol Independent Multicast Join/Prune messages can be exploited to
> provide necessary multicast containment."
> 
> If you're interested to receive a copy (PDF) of my thesis just drop me a 
> line.

What about just putting a copy of your thesis on the web.
Thus you will get more people to download it and read it :)

Regards,
Pavlin

> 
> cheers,
> Gerd
> 
> ps: real-name postings would be really coooooool! ;))))))
> 
> dsa main wrote:
> 
> > Hi All,
> >
> > I am looking for some pointers on PIM snooping...
> >
> > Is there any documents on PIM snooping implementation ?
> >
> > Regards
> > DSA
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar  3 21:13:53 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA21060
	for <pim-archive@lists.ietf.org>; Mon, 3 Mar 2003 21:13:53 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h242Arl51163;
	Mon, 3 Mar 2003 18:10:53 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from hotmail.com (f116.law15.hotmail.com [64.4.23.116])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2429bl51129
	for <pim@catarina.usc.edu>; Mon, 3 Mar 2003 18:09:37 -0800 (PST)
	(envelope-from dsa0000@hotmail.com)
Received: from mail pickup service by hotmail.com with Microsoft SMTPSVC;
	 Mon, 3 Mar 2003 18:09:32 -0800
Received: from 64.163.203.180 by lw15fd.law15.hotmail.msn.com with HTTP;
	Tue, 04 Mar 2003 02:09:32 GMT
X-Originating-IP: [64.163.203.180]
From: "dsa main" <dsa0000@hotmail.com>
To: gerd.besch@gmx.net, pim@catarina.usc.edu
Subject: Re: [pim] PIM snooping resources
Mime-Version: 1.0
Content-Type: text/plain; format=flowed
Message-ID: <F116ixqDCZx4FR0rmTI00024ef3@hotmail.com>
X-OriginalArrivalTime: 04 Mar 2003 02:09:32.0835 (UTC) FILETIME=[18C03F30:01C2E1F3]
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Tue, 04 Mar 2003 02:09:32 +0000

Hi Gerd,

Thank you for the info. I went through your thesis. It is pretty good.
I also came across an ietf draft 
"http://www.potaroo.net/ietf/ids/draft-tsenevir-pim-sm-snoop-00.txt" - 
expired ?? I think an up-to-date draft similar to the igmp snooping one will 
be cool..

Assume a generic scenario where MR1-MR4 are multicast routers which
support IGMPv2/v3 and PIMSNOOP is a snooping device.

          MR2
           |
           |
MR1----PIMSNOOP----MR4
           |
           |
          MR4

On each interface is it needed to keep track of all the PIM-SM join and 
prune messages like Join (*, *, RP), Prune (*, *, RP), Join (*, G),
Prune (*, G), Join (S, G, rpt), Prune(S, G, rpt), Join (S, G)
and  Prune (S, G) for building the snooping table ?? What all assumptions 
can be made to make this simple ?

Also is it needed to lookup on (S, G) rather than G when it comes to
switching/forwarding a multicast packet based on the snooping table ?

Regards
DSA






>From: Gerd Besch <gerd.besch@gmx.net>
>To: pim@catarina.usc.edu
>Subject: Re: [pim] PIM snooping resources
>Date: Tue, 04 Mar 2003 00:21:56 +0100
>
>  Hi DSA,
>
>I worte my thesis about the following subject 'Analysis and engineering of 
>the infrastructure required to support multicast traffic at an Internet 
>Exchange point' 
><http://besch.gmxhome.de/eng/thesis-gerd-besch-preface.html>
>
>Abstract:
>"This thesis discusses using 'PIM-snooping' as a solution for handling
>multicast traffic at a Multicast Internet Exchange Point (MIX) on a shared
>Gigabit-Ethernet switch fabric. Existing MIXs essentially treat multicast
>traffic as broadcast traffic. Switch ports are flooded due to the lack of
>information on how to efficiently replicate multicast traffic only onto 
>ports
>where receivers for specific multicast groups exist.
>Protocol Independent Multicast Join/Prune messages can be exploited to
>provide necessary multicast containment."
>
>If you're interested to receive a copy (PDF) of my thesis just drop me a 
>line.
>
>cheers,
>Gerd
>
>ps: real-name postings would be really coooooool! ;))))))
>
>dsa main wrote:
>
>>Hi All,
>>
>>I am looking for some pointers on PIM snooping...
>>
>>Is there any documents on PIM snooping implementation ?
>>
>>Regards
>>DSA
>
>
>
>
>_______________________________________________
>pim mailing list
>pim@catarina.usc.edu
>http://catarina.usc.edu/mailman/listinfo/pim


_________________________________________________________________
Tired of spam? Get advanced junk mail protection with MSN 8. 
http://join.msn.com/?page=features/junkmail

_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Tue Mar  4 07:35:07 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA15427
	for <pim-archive@lists.ietf.org>; Tue, 4 Mar 2003 07:35:06 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h24CXAu06198;
	Tue, 4 Mar 2003 04:33:13 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h24CVJu06169
	for <pim@catarina.usc.edu>; Tue, 4 Mar 2003 04:31:24 -0800 (PST)
	(envelope-from nsyracus@cnri.reston.va.us)
Received: from CNRI.Reston.VA.US (localhost [127.0.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA14839;
	Tue, 4 Mar 2003 07:29:16 -0500 (EST)
Message-Id: <200303041229.HAA14839@ietf.org>
Mime-Version: 1.0
Content-Type: Multipart/Mixed; Boundary="NextPart"
To: IETF-Announce: ;
Cc: pim@catarina.usc.edu
From: Internet-Drafts@ietf.org
Reply-to: Internet-Drafts@ietf.org
Subject: [pim] I-D ACTION:draft-ietf-pim-sm-bsr-03.txt,.ps
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Tue, 04 Mar 2003 07:29:15 -0500

--NextPart

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

	Title		: Bootstrap Router (BSR) Mechanism for PIM Sparse Mode
	Author(s)	: B. Fenner et al.
	Filename	: draft-ietf-pim-sm-bsr-03.txt,.ps
	Pages		: 35
	Date		: 2003-3-3
	
This document specifies the Bootstrap Router (BSR) mechanism
for Protocol Independent Multicast - Sparse Mode (PIM-SM).
BSR is one way that a PIM-SM router can learn the set of
group-to-RP mappings required in order to function.

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-pim-sm-bsr-03.txt

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

Internet-Drafts are also available by anonymous FTP. Login with the username
"anonymous" and a password of your e-mail address. After logging in,
type "cd internet-drafts" and then
	"get draft-ietf-pim-sm-bsr-03.txt".

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


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

Send a message to:
	mailserv@ietf.org.
In the body type:
	"FILE /internet-drafts/draft-ietf-pim-sm-bsr-03.txt".
	
NOTE:	The mail server at ietf.org can return the document in
	MIME-encoded form by using the "mpack" utility.  To use this
	feature, insert the command "ENCODING mime" before the "FILE"
	command.  To decode the response(s), you will need "munpack" or
	a MIME-compliant mail reader.  Different MIME-compliant mail readers
	exhibit different behavior, especially when dealing with
	"multipart" MIME messages (i.e. documents which have been split
	up into multiple messages), so check your local documentation on
	how to manipulate these messages.
		
		
Below is the data which will enable a MIME compliant mail reader
implementation to automatically retrieve the ASCII version of the
Internet-Draft.

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

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

Content-Type: text/plain
Content-ID:	<2003-3-3144942.I-D@ietf.org>

ENCODING mime
FILE /internet-drafts/draft-ietf-pim-sm-bsr-03.txt

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

Content-Type: text/plain
Content-ID:	<2003-3-3144942.I-D@ietf.org>

--OtherAccess--

--NextPart--


_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Tue Mar  4 08:39:10 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA20523
	for <pim-archive@lists.ietf.org>; Tue, 4 Mar 2003 08:39:10 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h24DbNu07082;
	Tue, 4 Mar 2003 05:37:23 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from multicasttech.com (IDENT:root@lennon.multicasttech.com [63.105.122.7])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h24DaEu07062
	for <pim@catarina.usc.edu>; Tue, 4 Mar 2003 05:36:14 -0800 (PST)
	(envelope-from tme@multicasttech.com)
Received: from [68.100.8.94] (account marshall_eubanks HELO localhost)
  by multicasttech.com (CommuniGate Pro SMTP 3.4.8)
  with ESMTP-TLS id 1617561; Tue, 04 Mar 2003 08:36:13 -0500
Subject: Re: [pim] PIM snooping resources
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v482)
Cc: gerd.besch@gmx.net, pim@catarina.usc.edu
To: "dsa main" <dsa0000@hotmail.com>
From: Marshall Eubanks <tme@multicasttech.com>
In-Reply-To: <F116ixqDCZx4FR0rmTI00024ef3@hotmail.com>
Message-Id: <446990D2-4E46-11D7-A5CB-003065BA697E@multicasttech.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.482)
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Tue, 4 Mar 2003 08:36:13 -0500
Content-Transfer-Encoding: 7bit

PIM snooping is of most use at exchanges, where traffic is routed 
between AS's.
Interdomain multicast traffic is (from the router viewpoint) _all_ SSM - 
i.e., there are
only (S,G) joins across a domain boundary. Therefore it is my contention 
that PIM snooping is
best accomplished by focusing on (S,G) traffic only.

This also means that the market for PIM snooping is likely to be much 
smaller than
for IGMP snooping.

On Monday, March 3, 2003, at 09:09  PM, dsa main wrote:

> Hi Gerd,
>
> Thank you for the info. I went through your thesis. It is pretty good.
> I also came across an ietf draft 
> "http://www.potaroo.net/ietf/ids/draft-tsenevir-pim-sm-snoop-00.txt" - 
> expired ?? I think an up-to-date draft similar to the igmp snooping one 
> will be cool..
>
> Assume a generic scenario where MR1-MR4 are multicast routers which
> support IGMPv2/v3 and PIMSNOOP is a snooping device.
>
>          MR2
>           |
>           |
> MR1----PIMSNOOP----MR4
>           |
>           |
>          MR4
>

I may be missing something, but I do not see the need for PIM snooping 
in this scenario.
(Or is PIMSNOOP attached to MR2 and MR4 - which is not how the ascii art 
looks here ?)

BTW, I know that Foundry has implemented a type of PIM snooping. If 
there are others I would like to know
about them.

                                   Regards
                                  Marshall Eubanks


> On each interface is it needed to keep track of all the PIM-SM join and 
> prune messages like Join (*, *, RP), Prune (*, *, RP), Join (*, G),
> Prune (*, G), Join (S, G, rpt), Prune(S, G, rpt), Join (S, G)
> and  Prune (S, G) for building the snooping table ?? What all 
> assumptions can be made to make this simple ?
>
> Also is it needed to lookup on (S, G) rather than G when it comes to
> switching/forwarding a multicast packet based on the snooping table ?
>
> Regards
> DSA
>
>
>
>
>
>
>> From: Gerd Besch <gerd.besch@gmx.net>
>> To: pim@catarina.usc.edu
>> Subject: Re: [pim] PIM snooping resources
>> Date: Tue, 04 Mar 2003 00:21:56 +0100
>>
>>  Hi DSA,
>>
>> I worte my thesis about the following subject 'Analysis and 
>> engineering of the infrastructure required to support multicast 
>> traffic at an Internet Exchange point' 
>> <http://besch.gmxhome.de/eng/thesis-gerd-besch-preface.html>
>>
>> Abstract:
>> "This thesis discusses using 'PIM-snooping' as a solution for handling
>> multicast traffic at a Multicast Internet Exchange Point (MIX) on a 
>> shared
>> Gigabit-Ethernet switch fabric. Existing MIXs essentially treat 
>> multicast
>> traffic as broadcast traffic. Switch ports are flooded due to the lack 
>> of
>> information on how to efficiently replicate multicast traffic only 
>> onto ports
>> where receivers for specific multicast groups exist.
>> Protocol Independent Multicast Join/Prune messages can be exploited to
>> provide necessary multicast containment."
>>
>> If you're interested to receive a copy (PDF) of my thesis just drop me 
>> a line.
>>
>> cheers,
>> Gerd
>>
>> ps: real-name postings would be really coooooool! ;))))))
>>
>> dsa main wrote:
>>
>>> Hi All,
>>>
>>> I am looking for some pointers on PIM snooping...
>>>
>>> Is there any documents on PIM snooping implementation ?
>>>
>>> Regards
>>> DSA
>>
>>
>>
>>
>> _______________________________________________
>> pim mailing list
>> pim@catarina.usc.edu
>> http://catarina.usc.edu/mailman/listinfo/pim
>
>
> _________________________________________________________________
> Tired of spam? Get advanced junk mail protection with MSN 8. 
> http://join.msn.com/?page=features/junkmail
>
> _______________________________________________
> pim mailing list
> pim@catarina.usc.edu
> http://catarina.usc.edu/mailman/listinfo/pim
>


T.M. Eubanks
Multicast Technologies, Inc.
Phone : 703-293-9601       Fax     : 703-293-9609
e-mail : tme@multicasttech.com
http://www.multicasttech.com

Test your network for multicast :
http://www.multicasttech.com/mt/
  Status of Multicast on the Web  :
  http://www.multicasttech.com/status/index.html

_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Tue Mar  4 12:31:27 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA02747
	for <pim-archive@lists.ietf.org>; Tue, 4 Mar 2003 12:31:26 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h24HTlu09818;
	Tue, 4 Mar 2003 09:29:47 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from postal2.lbl.gov (postal2.lbl.gov [131.243.248.26])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h24HSTu09797
	for <pim@catarina.usc.edu>; Tue, 4 Mar 2003 09:28:29 -0800 (PST)
	(envelope-from GRBell@lbl.gov)
Received: from postal2.lbl.gov (localhost [127.0.0.1])
	by postal2.lbl.gov (8.11.2/8.11.2) with ESMTP id h24HSSI26918
	for <pim@catarina.usc.edu>; Tue, 4 Mar 2003 09:28:28 -0800 (PST)
Received: from lbl.gov (gravity.lbl.gov [131.243.254.30])
	by postal2.lbl.gov (8.11.2/8.11.2) with ESMTP id h24HSPN26905;
	Tue, 4 Mar 2003 09:28:25 -0800 (PST)
Message-ID: <3E64E23A.30209@lbl.gov>
From: Gregory Bell <GRBell@lbl.gov>
User-Agent: Mozilla/5.0 (X11; U; FreeBSD i386; en-US; rv:1.1) Gecko/20021114
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Marshall Eubanks <tme@multicasttech.com>
CC: dsa main <dsa0000@hotmail.com>, gerd.besch@gmx.net, pim@catarina.usc.edu
Subject: Re: [pim] PIM snooping resources
References: <446990D2-4E46-11D7-A5CB-003065BA697E@multicasttech.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Tue, 04 Mar 2003 09:28:26 -0800
Content-Transfer-Encoding: 7bit

Marshall Eubanks wrote:

> BTW, I know that Foundry has implemented a type of PIM snooping. If 
> there are others I would like to know
> about them.

You might look at Cisco's RGMP.  I think it bears roughly the same 
relationship to PIM snooping that CGMP bears to IGMP snooping (ie, 
neighboring routers do some of the work in each case).

PIM snooping might be useful in the enterprise as well.  If you have a 
layer-two core interconnecting half a dozen PIM routers, you might want 
to avoid flooding on the switch.

- Greg


-- 
Gregory Bell
Network Systems Engineer
Lawrence Berkeley National Laboratory
1 Cyclotron Road, Mail Stop 50E0101
Berkeley, CA 94720-8140 USA
(510) 486-6817
GRBell@lbl.gov


_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Tue Mar  4 21:45:31 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA19632
	for <pim-archive@lists.ietf.org>; Tue, 4 Mar 2003 21:45:30 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h252gmu19591;
	Tue, 4 Mar 2003 18:42:48 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from exch-cluster-01.force10networks.com (corp.force10networks.com [206.54.51.125])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h252dgu19520
	for <pim@catarina.usc.edu>; Tue, 4 Mar 2003 18:39:42 -0800 (PST)
	(envelope-from svallepa@force10networks.com)
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C2E2C0.73BE2CC6"
Message-ID: <7A38462BE63B224DA3A758F9E325A0C9440C46@EXCH-CLUSTER-01.force10networks.com>
Thread-Topic: setting SPT bit when RPF'(S,G) is different from RPF'(*,G) on same iif.
Thread-Index: AcLiwHO9p2KL2OGMQXi6r7C5cpJwsQ==
From: "Sridhar Vallepalli" <svallepa@force10networks.com>
To: <pim@catarina.usc.edu>
Subject: [pim] setting SPT bit when RPF'(S,G) is different from RPF'(*,G) on same iif.
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Tue, 4 Mar 2003 18:39:32 -0800

This is a multi-part message in MIME format.

------_=_NextPart_001_01C2E2C0.73BE2CC6
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Under certain conditions in the below topology SPT bit will never be set
on router R1.=20

                                            +----------+
                                            |             |
                                            |     RP    |
                                            +----+----+
                                                    | 2
=20
---------+----------------+-------------------+--------------
                              |   1
|  1
                    +------+-----+
+-----+-----+
                    |                |                                |
|
                    |      R1      |                                |
R2     |
                    +------------+
+------------+
                           |     2
|  2
                          R
S


1) Receiver joins group G at R1

R1 state : (*,G)  iif =3D 1 oif =3D 2

2) Join sent by R1 to RP

RP state : (*,G) iif=3Dregvif oif =3D 2

3) Source start sending traffic R2 sends register packets.

R2 state : (S,G) iif=3D2 oif =3D register interface SPT bit =3D TRUE

4) RP receives Register packet forwards data based on (*,G) entry and
generates (S,G) join towards source

RP state : (S,G) iif =3D 2 oif =3D NULL SPT bit =3D FALSE
             : (*,G) iif=3Dregvif oif =3D 2

5) R1 receives data packet and triggers a SPT switchover right away

R1 state : (S,G) iif =3D 1 oif =3D2 SPT bit =3D FALSE
             : (*,G)  iif =3D 1 oif =3D2

6) (S,G) join received by R2=20

R2 state : (S,G) iif=3D2 oif =3D (register interface, 1)   SPT bit =3D =
TRUE=20
Traffic starts flowing on interface 1 and also registered to RP

7) Traffic seen on interface 2 of RP and SPT is set, further register
packets are discarded and register stop is sent.

In this case there is a possibility that there will never be a duplicate
traffic flowing on to the lan causing Assert condition.
Because of not seeing any (S,G) assert packets R1 will never set its SPT
bit to TRUE, as its RPF'(S,G) !=3D RPF'(*,G) and RPF interface for S and
RP are same.

Is there a way that this can be avoided, or is this an acceptable case??
Please comment.

Thanks
Sridhar



------_=_NextPart_001_01C2E2C0.73BE2CC6
Content-Type: text/html;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Dus-ascii">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
6.0.6249.1">
<TITLE>setting SPT bit when RPF'(S,G) is different from RPF'(*,G) on =
same iif.</TITLE>
</HEAD>
<BODY>
<!-- Converted from text/rtf format -->

<P><FONT SIZE=3D2 FACE=3D"Arial">Under certain conditions in the below =
topology SPT bit will never be set on router R1. </FONT>
</P>

<P><FONT SIZE=3D2 =
FACE=3D"Arial">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =
+----------+</FONT>

<BR><FONT SIZE=3D2 =
FACE=3D"Arial">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =
|&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=
 |</FONT>

<BR><FONT SIZE=3D2 =
FACE=3D"Arial">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =
|&nbsp;&nbsp;&nbsp;&nbsp; RP&nbsp;&nbsp;&nbsp; |</FONT>

<BR><FONT SIZE=3D2 =
FACE=3D"Arial">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =
+----+----+</FONT>

<BR><FONT SIZE=3D2 =
FACE=3D"Arial">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp; | 2</FONT>

<BR><FONT SIZE=3D2 =
FACE=3D"Arial">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =
---------+----------------+-------------------+--------------</FONT>

<BR><FONT SIZE=3D2 =
FACE=3D"Arial">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; |&nbsp;&nbsp; =
1&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=
bsp;&nbsp;&nbsp;&nbsp; |&nbsp; 1</FONT>

<BR><FONT SIZE=3D2 =
FACE=3D"Arial">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =
+------+-----+&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; +-----+-----+</FONT>

<BR><FONT SIZE=3D2 =
FACE=3D"Arial">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =
|&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=
&nbsp;&nbsp;&nbsp; =
|&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =
|&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=
&nbsp;&nbsp; |</FONT>

<BR><FONT SIZE=3D2 =
FACE=3D"Arial">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =
|&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; R1&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =
|&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =
|&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; R2&nbsp;&nbsp;&nbsp;&nbsp; |</FONT>

<BR><FONT SIZE=3D2 =
FACE=3D"Arial">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =
+------------+&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; +------------+</FONT>

<BR><FONT SIZE=3D2 =
FACE=3D"Arial">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;&nbsp; |&nbsp;&nbsp;&nbsp;&nbsp; =
2&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=
bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; |&nbsp; 2</FONT>

<BR><FONT SIZE=3D2 =
FACE=3D"Arial">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp; =
R&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=
bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp; S</FONT>
</P>
<BR>

<P><FONT SIZE=3D2 FACE=3D"Arial">1) Receiver joins group G at R1</FONT>
</P>

<P><FONT SIZE=3D2 FACE=3D"Arial">R1 state : (*,G)&nbsp; iif =3D 1 oif =
=3D 2</FONT>
</P>

<P><FONT SIZE=3D2 FACE=3D"Arial">2) Join sent by R1 to RP</FONT>
</P>

<P><FONT SIZE=3D2 FACE=3D"Arial">RP state : (*,G) iif=3Dregvif oif =3D =
2</FONT>
</P>

<P><FONT SIZE=3D2 FACE=3D"Arial">3) Source start sending traffic R2 =
sends register packets.</FONT>
</P>

<P><FONT SIZE=3D2 FACE=3D"Arial">R2 state : (S,G) iif=3D2 oif =3D =
register interface SPT bit =3D TRUE</FONT>
</P>

<P><FONT SIZE=3D2 FACE=3D"Arial">4) RP receives Register packet forwards =
data based on (*,G) entry and generates (S,G) join towards source</FONT>
</P>

<P><FONT SIZE=3D2 FACE=3D"Arial">RP state : (S,G) iif =3D 2 oif =3D NULL =
SPT bit =3D FALSE</FONT>

<BR><FONT SIZE=3D2 =
FACE=3D"Arial">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp; : (*,G) iif=3Dregvif oif =3D 2</FONT>
</P>

<P><FONT SIZE=3D2 FACE=3D"Arial">5) R1 receives data packet and triggers =
a SPT switchover right away</FONT>
</P>

<P><FONT SIZE=3D2 FACE=3D"Arial">R1 state : (S,G) iif =3D 1 oif =3D2 SPT =
bit =3D FALSE</FONT>

<BR><FONT SIZE=3D2 =
FACE=3D"Arial">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp; : (*,G)&nbsp; iif =3D 1 oif =3D2</FONT>
</P>

<P><FONT SIZE=3D2 FACE=3D"Arial">6) (S,G) join received by R2 </FONT>
</P>

<P><FONT SIZE=3D2 FACE=3D"Arial">R2 state : (S,G) iif=3D2 oif =3D =
(register interface, 1)&nbsp;&nbsp; SPT bit =3D TRUE </FONT>

<BR><FONT SIZE=3D2 FACE=3D"Arial">Traffic starts flowing on interface 1 =
and also registered to RP</FONT>
</P>

<P><FONT SIZE=3D2 FACE=3D"Arial">7) Traffic seen on interface 2 of RP =
and SPT is set, further register packets are discarded and register stop =
is sent.</FONT>
</P>

<P><FONT SIZE=3D2 FACE=3D"Arial">In this case there is a possibility =
that there will never be a duplicate traffic flowing on to the lan =
causing Assert condition.</FONT></P>

<P><FONT SIZE=3D2 FACE=3D"Arial">Because of not seeing any (S,G) assert =
packets R1 will never set its SPT bit to TRUE, as its RPF'(S,G) !=3D =
RPF'(*,G) and RPF interface for S and RP are same.</FONT></P>

<P><FONT SIZE=3D2 FACE=3D"Arial">Is there a way that this can be =
avoided, or is this an acceptable case?? Please comment.</FONT>
</P>

<P><FONT SIZE=3D2 FACE=3D"Arial">Thanks<BR>
Sridhar</FONT>
</P>
<BR>

</BODY>
</HTML>
------_=_NextPart_001_01C2E2C0.73BE2CC6--
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Wed Mar  5 13:27:47 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA20149
	for <pim-archive@lists.ietf.org>; Wed, 5 Mar 2003 13:27:46 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h25IPvu35985;
	Wed, 5 Mar 2003 10:25:57 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from aardvark.icir.org (aardvark.icir.org [192.150.187.20])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h25INlu35926
	for <pim@catarina.usc.edu>; Wed, 5 Mar 2003 10:23:47 -0800 (PST)
	(envelope-from mjh@aardvark.icir.org)
Received: from aardvark.icir.org (localhost [127.0.0.1])
	by aardvark.icir.org (8.12.3/8.12.3) with ESMTP id h25INlXH002858
	for <pim@catarina.usc.edu>; Wed, 5 Mar 2003 10:23:47 -0800 (PST)
	(envelope-from mjh@aardvark.icir.org)
From: Mark Handley <mjh@icir.org>
X-Organisation: ICIR
To: pim@catarina.usc.edu
Subject: Re: [pim] I-D ACTION:draft-ietf-pim-sm-bsr-03.txt,.ps 
In-reply-to: Your message of "Tue, 04 Mar 2003 07:29:15 EST."
             <200303041229.HAA14839@ietf.org> 
Message-ID: <2857.1046888627@aardvark.icir.org>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Wed, 05 Mar 2003 10:23:47 -0800


>A New Internet-Draft is available from the on-line Internet-Drafts directories
>.
>This draft is a work item of the Protocol Independent Multicast Working Group 
>of the IETF.
>
>	Title		: Bootstrap Router (BSR) Mechanism for PIM Sparse Mode
>	Author(s)	: B. Fenner et al.
>	Filename	: draft-ietf-pim-sm-bsr-03.txt,.ps
>	Pages		: 35
>	Date		: 2003-3-3
>	
>This document specifies the Bootstrap Router (BSR) mechanism
>for Protocol Independent Multicast - Sparse Mode (PIM-SM).
>BSR is one way that a PIM-SM router can learn the set of
>group-to-RP mappings required in order to function.


This is essentially a re-issue of the PIM BSR draft which had expired.
I appreciate that it needs more work to make it more readable, but I
want to draw your attention to one important change.

This change fixes a problem with Bootstrap Message flooding, where
BSMs could be blackholed.  Previously a BSM was not forwarded out of
the interface on which it was received.  In section 3, the text on
forwarding BSMs now reads:

     Forward BSM
          A bootstrap message that passes the Bootstrap Message
          Processing Checks is forwarded out of all interfaces with PIM
          neighbors (including the interface it is received on), except
          where this would cause the BSM to cross an admin-scope
          boundary for the scope zone indicated in the message.  The
          source IP address of the message is the forwarding router's IP
          address on the interface the message is being forwarded from,
          the destination address is ALL-PIM-ROUTERS, and the TTL of the
          message is set to 1.

          As an optimation, a router MAY choose not to forward a BSM out
          of the interface the message was received on if that interface
          is a point-to-point interface.  On interfaces with multiple
          PIM neighbors, a router MUST forward an accepted BSM onto the
          interface that BSM was received on, but if the number of PIM
          neighbors on that interface is large, it MAY delay forwarding
          a BSM onto that interface by a small randomized interval to
          prevent message implosion.

          Rationale: A BSM needs to be forwarded onto the interface the
          message was received on (in addition to the other interfaces)
          because the routers on a LAN may not have consistent routing
          information.  If three routers on a LAN are A, B, and C, and
          at router B RPF(BSR)==A and at router C RPF(BSR)==B, then
          router A originally forwards the BSM onto the LAN, but router
          C will only accept it when router B re-forwards the message
          onto the LAN.

Cheers,
	Mark
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Wed Mar  5 14:21:46 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA23141
	for <pim-archive@lists.ietf.org>; Wed, 5 Mar 2003 14:21:45 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h25JKOu37138;
	Wed, 5 Mar 2003 11:20:25 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from cefni.aber.ac.uk (cefni.aber.ac.uk [144.124.16.40])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h25JIgu37073
	for <pim@catarina.usc.edu>; Wed, 5 Mar 2003 11:18:43 -0800 (PST)
	(envelope-from dap@aber.ac.uk)
Received: from moin.dcs.aber.ac.uk ([193.60.11.36] helo=aber.ac.uk)
	by cefni.aber.ac.uk with esmtp (Exim 3.35 #3)
	id 18qePa-00075R-02; Wed, 05 Mar 2003 19:18:42 +0000
X-Mailer: exmh version 2.0.2 2/24/98
To: pim@catarina.usc.edu
cc: dap@aber.ac.uk, axs@aber.ac.uk
From: Dave Price <dave.price@aber.ac.uk>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Message-ID: <17819.1046891921@aber.ac.uk>
X-UWA-Originating-IP: 193.60.11.36
Subject: [pim] specifying static RP in pimd
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Wed, 05 Mar 2003 19:18:41 +0000

Dear All,

	I'm just in the middle of setting up a Linux 8.0
box to run pimd to give me a PIM-SM router to use
inside our sites routed network. I can't currently
see how I can set "static RP".

Could some helpful sole offer advice please.

Thanks,

Dave Price
	---------------------------------------------------------
	| David Price, Computer Science				|
	|							|
	|  Computer Science, University of Wales, Aberystwyth,	|
	|  Penglais Campus, Aberystwyth, Ceredigion, SY23 3DB	|
	|                                                       |
	| Email: dap@aber.ac.uk WWW: http://www.aber.ac.uk/~dap |
	|  Phone: +44 1970 622428   FAX: +44 1970 622455	|
	---------------------------------------------------------


_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Fri Mar  7 07:04:52 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA23296
	for <pim-archive@lists.ietf.org>; Fri, 7 Mar 2003 07:04:51 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h27C2Mu82743;
	Fri, 7 Mar 2003 04:02:22 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h27BuCu82606
	for <pim@catarina.usc.edu>; Fri, 7 Mar 2003 03:56:13 -0800 (PST)
	(envelope-from nsyracus@cnri.reston.va.us)
Received: from CNRI.Reston.VA.US (localhost [127.0.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA21442;
	Fri, 7 Mar 2003 06:54:07 -0500 (EST)
Message-Id: <200303071154.GAA21442@ietf.org>
Mime-Version: 1.0
Content-Type: Multipart/Mixed; Boundary="NextPart"
To: IETF-Announce: ;
Cc: pim@catarina.usc.edu
From: Internet-Drafts@ietf.org
Reply-to: Internet-Drafts@ietf.org
Subject: [pim] I-D ACTION:draft-ietf-pim-sm-v2-new-07.txt,.ps
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Fri, 07 Mar 2003 06:54:06 -0500

--NextPart

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

	Title		: Protocol Independent Multicast - Sparse Mode PIM-SM): 
                          Protocol Specification (Revised)
	Author(s)	: B. Fenner, M. Handley, H. Holbrook, I. Kouvelas
	Filename	: draft-ietf-pim-sm-v2-new-07.txt,.ps
	Pages		: 138
	Date		: 2003-3-6
	
This document specifies Protocol Independent Multicast -
Sparse Mode (PIM-SM).  PIM-SM is a multicast routing protocol
that can use the underlying unicast routing information base
or a separate multicast-capable routing information base.  It
builds unidirectional shared trees rooted at a Rendezvous
Point (RP) per group, and optionally creates shortest-path
trees per source.

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-pim-sm-v2-new-07.txt

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

Internet-Drafts are also available by anonymous FTP. Login with the username
"anonymous" and a password of your e-mail address. After logging in,
type "cd internet-drafts" and then
	"get draft-ietf-pim-sm-v2-new-07.txt".

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


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

Send a message to:
	mailserv@ietf.org.
In the body type:
	"FILE /internet-drafts/draft-ietf-pim-sm-v2-new-07.txt".
	
NOTE:	The mail server at ietf.org can return the document in
	MIME-encoded form by using the "mpack" utility.  To use this
	feature, insert the command "ENCODING mime" before the "FILE"
	command.  To decode the response(s), you will need "munpack" or
	a MIME-compliant mail reader.  Different MIME-compliant mail readers
	exhibit different behavior, especially when dealing with
	"multipart" MIME messages (i.e. documents which have been split
	up into multiple messages), so check your local documentation on
	how to manipulate these messages.
		
		
Below is the data which will enable a MIME compliant mail reader
implementation to automatically retrieve the ASCII version of the
Internet-Draft.

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

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

Content-Type: text/plain
Content-ID:	<2003-3-6125354.I-D@ietf.org>

ENCODING mime
FILE /internet-drafts/draft-ietf-pim-sm-v2-new-07.txt

--OtherAccess
Content-Type: Message/External-body;
	name="draft-ietf-pim-sm-v2-new-07.txt";
	site="ftp.ietf.org";
	access-type="anon-ftp";
	directory="internet-drafts"

Content-Type: text/plain
Content-ID:	<2003-3-6125354.I-D@ietf.org>

--OtherAccess--

--NextPart--


_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Fri Mar 14 22:37:32 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA27208
	for <pim-archive@lists.ietf.org>; Fri, 14 Mar 2003 22:37:31 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2F3Yqu90696;
	Fri, 14 Mar 2003 19:34:52 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from v6.hitachi.co.jp ([133.145.167.4])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2F3Tmu90587
	for <pim@catarina.usc.edu>; Fri, 14 Mar 2003 19:29:49 -0800 (PST)
	(envelope-from suz@crl.hitachi.co.jp)
Received: from s30.crl.hitachi.co.jp (galilei.v6.hitachi.co.jp [IPv6:3ffe:501:1011::1])
	by v6.hitachi.co.jp (8.12.8/8.11.6) with ESMTP id h2F3cAJY074654;
	Sat, 15 Mar 2003 12:38:10 +0900 (JST)
	(envelope-from suz@crl.hitachi.co.jp)
Message-ID: <x7vfyl71fk.wl%suz@crl.hitachi.co.jp>
From: SUZUKI Shinsuke <suz@crl.hitachi.co.jp>
To: lwei@redback.com, pusateri@juniper.net
Cc: pim@catarina.usc.edu, jinmei@isl.rdc.toshiba.co.jp
Subject: Re: [pim] Call for agenda items for PIM WG in San Francisco
X-cite: xcite 1.33
In-Reply-To: <x74r7g5va6.wl@s30.crl.hitachi.co.jp>
References: <200302052347.h15NlgV02935@liming-pc.redback.com>
	<x74r7g5va6.wl@s30.crl.hitachi.co.jp>
User-Agent: User-Agent: Wanderlust/2.11.3 (Wonderwall) Emacs/21.1 Mule/5.0 (SAKAKI)
Organization: Network Architecture Research Dept., Central Research Laboratory, Hitachi, Ltd, Japan
MIME-Version: 1.0 (generated by SEMI 1.14.4 - "Hosorogi")
Content-Type: text/plain; charset=US-ASCII
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Sat, 15 Mar 2003 12:29:03 +0900

Hello all,

> >>>>> On Wed, 5 Feb 2003 15:47:42 -0800 (PST)
> >>>>> lwei@redback.com(Liming Wei)  said:

> I'd like to make a presentation regarding PIM RPF failure when a
> router has multiple addresses on a link.
> (now writing a 00-draft, and it'll be available soon)

I'm very sorry to mention that I cannot join IETF meeting, because my
company suspended my US travel because of the expected US-Iraq war....

So Jinmei-san (co-author of this draft), instead of me, will make a
presentation at PIM WG.

Sorry for an abrupt absense...
----
SUZUKI, Shinsuke @ Hitachi / KAME Project
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Sun Mar 16 06:56:02 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA23152
	for <pim-archive@lists.ietf.org>; Sun, 16 Mar 2003 06:55:59 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2GBrku29197;
	Sun, 16 Mar 2003 03:53:47 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from netcore.fi (netcore.fi [193.94.160.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2GBpuu29154
	for <pim@catarina.usc.edu>; Sun, 16 Mar 2003 03:51:56 -0800 (PST)
	(envelope-from pekkas@netcore.fi)
Received: from localhost (pekkas@localhost)
	by netcore.fi (8.11.6/8.11.6) with ESMTP id h2GBpos22218
	for <pim@catarina.usc.edu>; Sun, 16 Mar 2003 13:51:50 +0200
From: Pekka Savola <pekkas@netcore.fi>
To: pim@catarina.usc.edu
Message-ID: <Pine.LNX.4.44.0303161350030.21991-100000@netcore.fi>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [pim] suz-pim-upstream-detection-00 comments
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Sun, 16 Mar 2003 13:51:49 +0200 (EET)

Hi,

A quick re-read of the doc brought up a few issues, mostly editorial in 
nature.  This doc looks quite good.  Comments below.

non-editorial:
--------------
   Address-List:  The list of the IP addresses of the router on the link
      where this Hello message is advertised.  Each IP address is
      encoded in Encoded-Unicast format [1].

==> does the list really include *all* addresses, also link-locals (multiple
if configured)?  It would seem to me that if the router has only one
link-local address, there is no need to advertise that (as it will be
learned from PIM adjacency). Did I miss something?


editorial:
----------

            PIM upstream detection among multiple addresses
                draft-suz-pim-upstream-detection-00.txt
==> upper case words

   This memo describes a PIM routing problem owing to the reverse path
==> s/PIM/Protocol Independent Multicast (PIM)/

   Suppose a router-A has address-A1 and address-A2 on a link and it
   establishes a PIM neighboring with a router-B using address-A1.

==> s/neighboring/adjacency/ (or what's the correct word, I'm not sure)

2. Possible solutions
==> sections also have to be uppercased, s/sol/Sol/

   There is operational workaround for this problem.
==> s/is/is an/ (or were there multiple workarounds..)

   The problem occurring when a RP shares a subnet prefix with

==> s/a RP/an RP/

   Length :  The Length of the Address-List field in byte.
==> s/byte/bytes/

   address.  If the link-layer addresses coincides, the corresponding
==> s/coincides/coincide/

   When using the new PIM hello option described in Section Section 2.2,
==> s/Section //

   [3]  Marques, P. and F. Dupont, "Use of BGP-4 Multiprotocol
==> the last names in the refs are in wrong order, s/F. Dupont/Dupont, F./
(same for the rest)

-- 
Pekka Savola                 "You each name yourselves king, yet the
Netcore Oy                    kingdom bleeds."
Systems. Networks. Security. -- George R.R. Martin: A Clash of Kings

_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar 17 04:24:11 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA08067
	for <pim-archive@lists.ietf.org>; Mon, 17 Mar 2003 04:24:11 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2H9Jxu52750;
	Mon, 17 Mar 2003 01:20:00 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from atlantis.lkn.e-technik.tu-muenchen.de (atlantis.lkn.e-technik.tu-muenchen.de [129.187.9.137])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2H9IGu52709
	for <pim@catarina.usc.edu>; Mon, 17 Mar 2003 01:18:16 -0800 (PST)
	(envelope-from serdar@eikon.tum.de)
Received: from lkn.e-technik.tu-muenchen.de (aperol.LKN.E-Technik.TU-Muenchen.DE [10.152.4.5])
	by atlantis.lkn.e-technik.tu-muenchen.de (8.12.6/8.12.6/SuSE Linux 0.6) with ESMTP id h2H9IECl026386
	for <pim@catarina.usc.edu>; Mon, 17 Mar 2003 10:18:14 +0100
Received: from eikon.tum.de (princeton [10.152.4.66])
	by lkn.e-technik.tu-muenchen.de (8.12.6/8.12.6/SuSE Linux 0.6) with ESMTP id h2H9IDHD002364
	for <pim@catarina.usc.edu>; Mon, 17 Mar 2003 10:18:13 +0100
Message-ID: <3E75902D.1080509@eikon.tum.de>
From: Serdar Uezuemcue <serdar@eikon.tum.de>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.4) Gecko/20011126 Netscape6/6.2.1
X-Accept-Language: en-us
MIME-Version: 1.0
To: pim@catarina.usc.edu
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Subject: [pim] PIM Table
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Mon, 17 Mar 2003 10:06:53 +0100
Content-Transfer-Encoding: 7bit

Hi,


is that a file in which i can see the existent Multicast Groups and 
Senders ?

And which is the Sender for this Multicast Group ?


I get the Group-Addres and have to find the Sender for this Group .

( In my Multicast Application , i have only one Sender. )


Thanks ,
Serdar

_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar 17 14:15:36 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA25706
	for <pim-archive@lists.ietf.org>; Mon, 17 Mar 2003 14:15:35 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2HJEcu64382;
	Mon, 17 Mar 2003 11:14:38 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from possum.icir.org (possum.icir.org [192.150.187.67])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2HJAfu64310
	for <pim@catarina.usc.edu>; Mon, 17 Mar 2003 11:10:42 -0800 (PST)
	(envelope-from pavlin@possum.icir.org)
Received: from possum.icir.org (localhost [127.0.0.1])
	by possum.icir.org (8.12.3/8.12.3) with ESMTP id h2HJAfix092674;
	Mon, 17 Mar 2003 11:10:41 -0800 (PST)
	(envelope-from pavlin@possum.icir.org)
Message-Id: <200303171910.h2HJAfix092674@possum.icir.org>
To: Serdar Uezuemcue <serdar@eikon.tum.de>
cc: pim@catarina.usc.edu
Subject: Re: [pim] PIM Table 
In-Reply-To: Message from Serdar Uezuemcue <serdar@eikon.tum.de> 
   of "Mon, 17 Mar 2003 10:06:53 +0100." <3E75902D.1080509@eikon.tum.de> 
From: Pavlin Radoslavov <pavlin@icir.org>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Mon, 17 Mar 2003 11:10:41 -0800

> is that a file in which i can see the existent Multicast Groups and 
> Senders ?
> 
> And which is the Sender for this Multicast Group ?

The vendor-specific answer is to check for example the CLI commands
for your router.
On UNIX, you can try "netstat -g" or "cat /proc/net/ip_mr_cache"
(the second one for Linux).

One vendor-independent answer is to run tcpdump for the specific
multicast group :)

Regards,
Pavlin

> I get the Group-Addres and have to find the Sender for this Group .
> 
> ( In my Multicast Application , i have only one Sender. )
> 
> 
> Thanks ,
> Serdar
> 
> _______________________________________________
> pim mailing list
> pim@catarina.usc.edu
> http://catarina.usc.edu/mailman/listinfo/pim

_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Tue Mar 18 21:05:15 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA11923
	for <pim-archive@lists.ietf.org>; Tue, 18 Mar 2003 21:05:15 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2J241u99563;
	Tue, 18 Mar 2003 18:04:01 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from prattle.redback.com (hiddenuser@prattle.redback.com [155.53.12.9])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2J20Qu99484
	for <pim@catarina.usc.edu>; Tue, 18 Mar 2003 18:00:26 -0800 (PST)
	(envelope-from lwei@redback.com)
Received: from liming-pc.redback.com (liming-pc.redback.com [155.53.36.120])
	by prattle.redback.com (Postfix) with ESMTP id 70385ECC0B
	for <pim@catarina.usc.edu>; Tue, 18 Mar 2003 18:00:25 -0800 (PST)
Received: (from lwei@localhost)
	by liming-pc.redback.com (8.11.3nb1/8.8.8/null redback bsdclient) id h2J20Pf08821;
	Tue, 18 Mar 2003 18:00:25 -0800 (PST)
Message-Id: <200303190200.h2J20Pf08821@liming-pc.redback.com>
From: Liming Wei <lwei@redback.com>
To: pim@catarina.usc.edu
Reply-To: lwei@redback.com
Subject: [pim] San Francisco PIM WG meeting minutes
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Tue, 18 Mar 2003 18:00:25 -0800 (PST)

Folks,

We forgot to ask for a volunteer to take notes. Here is a note
I typed quickly while my memory is still fresh. Additions/updates
are welcome.


Working group: PIM
Location: San Francisco Hilton
Time:  3/18/2003, 3:45pm
Note Taker: Liming Wei

1. Hugh did an update on PIM SM spec. Updated the security section
   of draft. (*,*,RP) will stay in spec, for future use, to fix up
   one problem involving (S,G)RPT prune.
2. Mark Handley on PIM BSR flooding problem on LANs with inconsistent
   RPF information across different routers.
   Change of rule is to forward BSM out all interfaces including
   the RPF interface where the BSM was received on.
3. Dino presented proposal on Anycast RP with PIM. The proposal attempts
   to provide anycast RP functio without using MSDP.
   Gist is to form a set of RPs configured with the same RP address
   (any cast address). Each DR sends registers to the closest RP,
   which will forward to the other RPs in the RP set. RPs accept
   these relayed PIM registers, but do not forward relayed PIM registers.
   Register-stop mesages between RPs are optional, reason is register
   stops happen between the DR and the first RP already. Two interoperable
   implementations from cisco and Procket have been done.

   A few comments were made about this proposal. One issue raised was
   packet losses resulted from loss of one PIM register (Mark H).
   Dino responded with such event is recovered upon the next register,
   and one lives with such occasional losses in the beginning. Another
   suggestion (Bill F?) was to do the relays via multicast forwarding.
   Dino's response was to forcus on the simplicity, and to avoid
   introduction of new machinary. Etc.

   Question to working group, does the working group want to take on
   this as a working group item ?

4. Jimei presented proposals on correctly matching a global address
   to a link-local IPv6 address for PIM rpf checks. Options include
   introducing a new PIM hello options that will list all addresses
   on a PIM neighor's interface; configuration based; etc.
   Also questioned if working group will take this as a working group
   item.

   Mark commented after a short discussion with a few spec coauthers
   that they are open to the idea of using a new PIM hello option,
   and to explicitly document this RPF address issue in the base spec.

5. Tom asked the working group for possible feedbacks on the PIM dense
   mode draft and the new PIM MIB draft. 2 people raised hands.
   Tom asked working group members to read and comment on the
   drafts, as usual.

I briefly conversed with Tom after the WG meeting that we
think we should continue the design discussions for 2 and 3 above
on the working group mailing list.

On the subject of the new PIM dense mode draft, Tom and I felt that
the best course of action is to move PIM dense mode draft to either
informational or experimental RFC, and get done with it. We will
request the new drafters to get the PIM dense mode draft into
a form suitable for submission. Still we need document reviews
from the working group members.


-Liming
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Wed Mar 19 10:14:50 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA23158
	for <pim-archive@lists.ietf.org>; Wed, 19 Mar 2003 10:14:49 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JFBqu14995;
	Wed, 19 Mar 2003 07:11:52 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from relay2.alcatel.be (alc239.alcatel.be [195.207.101.239])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JFAsu14971
	for <pim@catarina.usc.edu>; Wed, 19 Mar 2003 07:10:55 -0800 (PST)
	(envelope-from prashant.jhingran@alcatel.be)
Received: from bemail04.net.alcatel.be (localhost [127.0.0.1])
	by relay2.alcatel.be (8.10.1/8.11.4) with ESMTP id h2JFAZg11459;
	Wed, 19 Mar 2003 16:10:35 +0100 (MET)
Received: from alcatel.be ([138.203.193.132])
          by bemail04.net.alcatel.be (Lotus Domino Release 5.0.11)
          with ESMTP id 2003031916103437:6072 ;
          Wed, 19 Mar 2003 16:10:34 +0100 
From: jhingran <prashant.jhingran@alcatel.be>
Message-ID: <3E788869.5572FAF1@alcatel.be>
From: Prashant_JHINGRAN/BE/ALCATEL@ALCATEL
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.6 sun4m)
X-Accept-Language: en
MIME-Version: 1.0
To: lwei@redback.com, pim@catarina.usc.edu
Subject: Re: [pim] San Francisco PIM WG meeting minutes
References: <200303190200.h2J20Pf08821@liming-pc.redback.com>
X-MIMETrack: Itemize by SMTP Server on BEMAIL04/BE/ALCATEL(Release 5.0.11  |July 24, 2002) at
 03/19/2003 16:10:34,
	Serialize by Router on BEMAIL04/BE/ALCATEL(Release 5.0.11  |July 24, 2002) at
 03/19/2003 16:10:35,
	Serialize complete at 03/19/2003 16:10:35
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset=us-ascii
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Wed, 19 Mar 2003 16:10:34 +0100
Content-Transfer-Encoding: 7bit

Hi Liming,

  Can you pls explain me what is meant by
   "Each DR sends registers to the closest RP,
   which will forward to the other RPs in the RP set",
    as far as I understand, it doen't makes much sense to relay register
messages to "the other RPs in the RP set"?

Do you mean the following by the above statement:

Consider a network
                                                                    n-routers

---------------------//---------RP(a)----------------
R1
-SDR
|                                                                R2-SDR
 -------------------------------RP(b)---//-----------

n -routers

Here both the RPs share the same RP address (e.g.   the loopbacks  )
Now if RP(a) receives a register message from R2, then it needs to do is a
unicast route lookup and
relay the reg to the other anycast RP(b).


 If yes then this might be an overhead for pim as it needs to keep track of
the unicast topology.

regards,
Prashant



Liming Wei wrote:

> Folks,
>
> We forgot to ask for a volunteer to take notes. Here is a note
> I typed quickly while my memory is still fresh. Additions/updates
> are welcome.
>
> Working group: PIM
> Location: San Francisco Hilton
> Time:  3/18/2003, 3:45pm
> Note Taker: Liming Wei
>
> 1. Hugh did an update on PIM SM spec. Updated the security section
>    of draft. (*,*,RP) will stay in spec, for future use, to fix up
>    one problem involving (S,G)RPT prune.
> 2. Mark Handley on PIM BSR flooding problem on LANs with inconsistent
>    RPF information across different routers.
>    Change of rule is to forward BSM out all interfaces including
>    the RPF interface where the BSM was received on.
> 3. Dino presented proposal on Anycast RP with PIM. The proposal attempts
>    to provide anycast RP functio without using MSDP.
>    Gist is to form a set of RPs configured with the same RP address
>    (any cast address). Each DR sends registers to the closest RP,
>    which will forward to the other RPs in the RP set. RPs accept
>    these relayed PIM registers, but do not forward relayed PIM registers.
>    Register-stop mesages between RPs are optional, reason is register
>    stops happen between the DR and the first RP already. Two interoperable
>    implementations from cisco and Procket have been done.
>
>    A few comments were made about this proposal. One issue raised was
>    packet losses resulted from loss of one PIM register (Mark H).
>    Dino responded with such event is recovered upon the next register,
>    and one lives with such occasional losses in the beginning. Another
>    suggestion (Bill F?) was to do the relays via multicast forwarding.
>    Dino's response was to forcus on the simplicity, and to avoid
>    introduction of new machinary. Etc.
>
>    Question to working group, does the working group want to take on
>    this as a working group item ?
>
> 4. Jimei presented proposals on correctly matching a global address
>    to a link-local IPv6 address for PIM rpf checks. Options include
>    introducing a new PIM hello options that will list all addresses
>    on a PIM neighor's interface; configuration based; etc.
>    Also questioned if working group will take this as a working group
>    item.
>
>    Mark commented after a short discussion with a few spec coauthers
>    that they are open to the idea of using a new PIM hello option,
>    and to explicitly document this RPF address issue in the base spec.
>
> 5. Tom asked the working group for possible feedbacks on the PIM dense
>    mode draft and the new PIM MIB draft. 2 people raised hands.
>    Tom asked working group members to read and comment on the
>    drafts, as usual.
>
> I briefly conversed with Tom after the WG meeting that we
> think we should continue the design discussions for 2 and 3 above
> on the working group mailing list.
>
> On the subject of the new PIM dense mode draft, Tom and I felt that
> the best course of action is to move PIM dense mode draft to either
> informational or experimental RFC, and get done with it. We will
> request the new drafters to get the PIM dense mode draft into
> a form suitable for submission. Still we need document reviews
> from the working group members.
>
> -Liming
> _______________________________________________
> pim mailing list
> pim@catarina.usc.edu
> http://catarina.usc.edu/mailman/listinfo/pim

_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Wed Mar 19 13:31:34 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA29447
	for <pim-archive@lists.ietf.org>; Wed, 19 Mar 2003 13:31:33 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JIUqu18742;
	Wed, 19 Mar 2003 10:30:52 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from motgate.mot.com (motgate.mot.com [129.188.136.100])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JITUu18700
	for <pim@catarina.usc.edu>; Wed, 19 Mar 2003 10:29:30 -0800 (PST)
	(envelope-from Chris.Cormier@motorola.com)
Received: from pobox3.mot.com (pobox3.mot.com [10.64.251.242])
	by motgate.mot.com (Motorola/Motgate) with ESMTP id h2JITT3x017338
	for <pim@catarina.usc.edu>; Wed, 19 Mar 2003 11:29:29 -0700 (MST)
Received: [from il06exm05.corp.mot.com (il06exm05.corp.mot.com [10.0.111.5]) by pobox3.mot.com (MOT-pobox3 2.0) with ESMTP id LAA27483 for <pim@catarina.usc.edu>; Wed, 19 Mar 2003 11:27:17 -0700 (MST)]
Received: by il06exm05.corp.mot.com with Internet Mail Service (5.5.2656.59)
	id <HALK9W95>; Wed, 19 Mar 2003 12:28:43 -0600
Message-ID: <FB578B06F252D511B95B009027E3267B067620DE@il06exm04.corp.mot.com>
From: Cormier Chris-ACC125 <Chris.Cormier@motorola.com>
To: pim@catarina.usc.edu
Subject: RE: [pim] San Francisco PIM WG meeting minutes
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2656.59)
Content-Type: text/plain
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Wed, 19 Mar 2003 12:28:38 -0600

two cents on BSR flooding out IIF.

I still believe that this should be optional. I would say that there are
more cases where we know for sure what our RPF to BSR will be than cases
where there will be inconsistent RPF routes. The bootstrap packet is bloated
as it is, any solution that would "require" more copies of this message
would not be a good thing.

-Chris Cormier

"2. Mark Handley on PIM BSR flooding problem on LANs with inconsistent
   RPF information across different routers.
   Change of rule is to forward BSM out all interfaces including
   the RPF interface where the BSM was received on."
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Wed Mar 19 15:34:22 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA06580
	for <pim-archive@lists.ietf.org>; Wed, 19 Mar 2003 15:34:21 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JKXsu21130;
	Wed, 19 Mar 2003 12:33:54 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from eve.icir.org (wl-130-157.wireless.ietf56.ietf.org [130.129.130.157])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JKWau21095
	for <pim@catarina.usc.edu>; Wed, 19 Mar 2003 12:32:36 -0800 (PST)
	(envelope-from mjh@eve.icir.org)
Received: from eve.icir.org (localhost [127.0.0.1])
	by eve.icir.org (8.12.7/8.12.2) with ESMTP id h2JKWVbC002826;
	Wed, 19 Mar 2003 12:32:31 -0800 (PST)
From: Mark Handley <mjh@icir.org>
X-Organisation: ICIR
To: Cormier Chris-ACC125 <Chris.Cormier@motorola.com>
cc: pim@catarina.usc.edu
Subject: Re: [pim] San Francisco PIM WG meeting minutes 
In-reply-to: Your message of "Wed, 19 Mar 2003 12:28:38 CST."
             <FB578B06F252D511B95B009027E3267B067620DE@il06exm04.corp.mot.com> 
Message-ID: <2825.1048105951@eve.icir.org>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Wed, 19 Mar 2003 12:32:31 -0800


>two cents on BSR flooding out IIF.
>
>I still believe that this should be optional. I would say that there are
>more cases where we know for sure what our RPF to BSR will be than cases
>where there will be inconsistent RPF routes. The bootstrap packet is bloated
>as it is, any solution that would "require" more copies of this message
>would not be a good thing.
>
>-Chris Cormier
>
>"2. Mark Handley on PIM BSR flooding problem on LANs with inconsistent
>   RPF information across different routers.
>   Change of rule is to forward BSM out all interfaces including
>   the RPF interface where the BSM was received on."


I think we had consensus at the meeting on making sending out of the
IIF optional.  I believe it should default to on so that it works by
default, but I'm fine with people disabling it where it's safe and the
bandwidth causes an issue.

 - Mark
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Wed Mar 19 16:49:15 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA10961
	for <pim-archive@lists.ietf.org>; Wed, 19 Mar 2003 16:49:14 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JLldu22641;
	Wed, 19 Mar 2003 13:47:39 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from prattle.redback.com (hiddenuser@prattle.redback.com [155.53.12.9])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JLkuu22608
	for <pim@catarina.usc.edu>; Wed, 19 Mar 2003 13:46:56 -0800 (PST)
	(envelope-from lwei@redback.com)
Received: from liming-pc.redback.com (liming-pc.redback.com [155.53.36.120])
	by prattle.redback.com (Postfix) with ESMTP
	id 88D1093B565; Wed, 19 Mar 2003 13:46:55 -0800 (PST)
Received: (from lwei@localhost)
	by liming-pc.redback.com (8.11.3nb1/8.8.8/null redback bsdclient) id h2JLktZ11481;
	Wed, 19 Mar 2003 13:46:55 -0800 (PST)
Message-Id: <200303192146.h2JLktZ11481@liming-pc.redback.com>
From: Liming Wei <lwei@redback.com>
To: prashant.jhingran@alcatel.be
Cc: pim@catarina.usc.edu
In-reply-to: <3E788869.5572FAF1@alcatel.be> (message from jhingran on Wed, 19
	Mar 2003 16:10:34 +0100)
Subject: Re: [pim] San Francisco PIM WG meeting minutes
Reply-To: lwei@redback.com
References: <200303190200.h2J20Pf08821@liming-pc.redback.com> <3E788869.5572FAF1@alcatel.be>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Wed, 19 Mar 2003 13:46:55 -0800 (PST)

Jhingran,

I am just the note taker:-).

But to respond to your observation, The unicast route lookup is
different from having to "keep track of unicast topology". Please keep
in mind that we are talking about an anycast RP mechanism, and we are
trying to replace MSDP with this mechaism. So a comparison with MSDP
would provide a better angle. 

-Liming
----------
> From: jhingran <prashant.jhingran@alcatel.be>
> 
> Hi Liming,
> 
>   Can you pls explain me what is meant by
>    "Each DR sends registers to the closest RP,
>    which will forward to the other RPs in the RP set",
>     as far as I understand, it doen't makes much sense to relay register
> messages to "the other RPs in the RP set"?
> 
> Do you mean the following by the above statement:
> 
> Consider a network
>                                                                     n-routers
> 
> ---------------------//---------RP(a)----------------
> R1
> -SDR
> |                                                                R2-SDR
>  -------------------------------RP(b)---//-----------
> 
> n -routers
> 
> Here both the RPs share the same RP address (e.g.   the loopbacks  )
> Now if RP(a) receives a register message from R2, then it needs to do is a
> unicast route lookup and
> relay the reg to the other anycast RP(b).
> 
> 
>  If yes then this might be an overhead for pim as it needs to keep track of
> the unicast topology.
> 
> regards,
> Prashant
> 
> 
> 
> Liming Wei wrote:
> 
> > Folks,
> >
> > We forgot to ask for a volunteer to take notes. Here is a note
> > I typed quickly while my memory is still fresh. Additions/updates
> > are welcome.
> >
> > Working group: PIM
> > Location: San Francisco Hilton
> > Time:  3/18/2003, 3:45pm
> > Note Taker: Liming Wei
> >
> > 1. Hugh did an update on PIM SM spec. Updated the security section
> >    of draft. (*,*,RP) will stay in spec, for future use, to fix up
> >    one problem involving (S,G)RPT prune.
> > 2. Mark Handley on PIM BSR flooding problem on LANs with inconsistent
> >    RPF information across different routers.
> >    Change of rule is to forward BSM out all interfaces including
> >    the RPF interface where the BSM was received on.
> > 3. Dino presented proposal on Anycast RP with PIM. The proposal attempts
> >    to provide anycast RP functio without using MSDP.
> >    Gist is to form a set of RPs configured with the same RP address
> >    (any cast address). Each DR sends registers to the closest RP,
> >    which will forward to the other RPs in the RP set. RPs accept
> >    these relayed PIM registers, but do not forward relayed PIM registers.
> >    Register-stop mesages between RPs are optional, reason is register
> >    stops happen between the DR and the first RP already. Two interoperable
> >    implementations from cisco and Procket have been done.
> >
> >    A few comments were made about this proposal. One issue raised was
> >    packet losses resulted from loss of one PIM register (Mark H).
> >    Dino responded with such event is recovered upon the next register,
> >    and one lives with such occasional losses in the beginning. Another
> >    suggestion (Bill F?) was to do the relays via multicast forwarding.
> >    Dino's response was to forcus on the simplicity, and to avoid
> >    introduction of new machinary. Etc.
> >
> >    Question to working group, does the working group want to take on
> >    this as a working group item ?
> >
> > 4. Jimei presented proposals on correctly matching a global address
> >    to a link-local IPv6 address for PIM rpf checks. Options include
> >    introducing a new PIM hello options that will list all addresses
> >    on a PIM neighor's interface; configuration based; etc.
> >    Also questioned if working group will take this as a working group
> >    item.
> >
> >    Mark commented after a short discussion with a few spec coauthers
> >    that they are open to the idea of using a new PIM hello option,
> >    and to explicitly document this RPF address issue in the base spec.
> >
> > 5. Tom asked the working group for possible feedbacks on the PIM dense
> >    mode draft and the new PIM MIB draft. 2 people raised hands.
> >    Tom asked working group members to read and comment on the
> >    drafts, as usual.
> >
> > I briefly conversed with Tom after the WG meeting that we
> > think we should continue the design discussions for 2 and 3 above
> > on the working group mailing list.
> >
> > On the subject of the new PIM dense mode draft, Tom and I felt that
> > the best course of action is to move PIM dense mode draft to either
> > informational or experimental RFC, and get done with it. We will
> > request the new drafters to get the PIM dense mode draft into
> > a form suitable for submission. Still we need document reviews
> > from the working group members.
> >
> > -Liming
> > _______________________________________________
> > pim mailing list
> > pim@catarina.usc.edu
> > http://catarina.usc.edu/mailman/listinfo/pim
> 
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Wed Mar 19 16:53:27 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA11246
	for <pim-archive@lists.ietf.org>; Wed, 19 Mar 2003 16:53:26 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JLq8u22753;
	Wed, 19 Mar 2003 13:52:08 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from dmz2.procket.com (dmz2.procket.com [65.174.124.37])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JLp9u22727
	for <pim@catarina.usc.edu>; Wed, 19 Mar 2003 13:51:09 -0800 (PST)
	(envelope-from John.Zwiebel@procket.com)
Received: from miata.procket.com (zeus-d-1.procket.com [65.174.124.60])
	by dmz2.procket.com (Postfix) with ESMTP
	id 254E33444D; Wed, 19 Mar 2003 13:04:09 -0800 (PST)
Received: from exchangefe1.na.procket.com (exchangefe1.na.procket.com [10.1.7.251])
	by miata.procket.com (8.12.1/8.12.1) with ESMTP id h2JLp4YB014852;
	Wed, 19 Mar 2003 13:51:04 -0800 (PST)
Received: from procket.com ([10.1.1.1]) by exchangefe1.na.procket.com with Microsoft SMTPSVC(5.0.2195.5329);
	 Wed, 19 Mar 2003 13:51:03 -0800
Subject: Re: [pim] San Francisco PIM WG meeting minutes
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v551)
Cc: pim@catarina.usc.edu
To: lwei@redback.com
From: John Zwiebel <jzwiebel@procket.com>
In-Reply-To: <200303190200.h2J20Pf08821@liming-pc.redback.com>
Message-Id: <BAE51AE0-5A54-11D7-B534-0003936A6062@procket.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.551)
X-OriginalArrivalTime: 19 Mar 2003 21:51:04.0152 (UTC) FILETIME=[A3770580:01C2EE61]
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Wed, 19 Mar 2003 13:49:59 -0800
Content-Transfer-Encoding: 7bit


On Tuesday, March 18, 2003, at 06:00 PM, Liming Wei wrote:

> Folks,
>
> We forgot to ask for a volunteer to take notes. Here is a note
> I typed quickly while my memory is still fresh. Additions/updates
> are welcome.
>
>
> Working group: PIM
> Location: San Francisco Hilton
> Time:  3/18/2003, 3:45pm
> Note Taker: Liming Wei

> 3. Dino presented proposal on Anycast RP with PIM.

>    A few comments were made about this proposal. One issue raised was
>    packet losses resulted from loss of one PIM register (Mark H).

Bill Nickless was very much in favor of -not- encapsulating data in the
registers between anycast-RPs but only using NULL registers to alert
the other RPs in the group that they should join to the SPT.

Mark's point would then be exacerbated since it would guarantee that the
first packet would be lost.

I (jzwiebel) was unable to gain microphone access to ask the following
(Note that this is really a "whole different discussion").

   "How realistic is it to expect the network to be able to just 
'handle' all
    mpackets dumped on to the network and -not- expect loss?"

This is a question that has been hanging around since the beginning of
multicasting and it has consistently been ignored.  Many, if not most,
multicast apps used in the interdomain environment do not care whether
or not the first packets are lost.  Many applications in the intradomain
environment (although I can't say most) do not care about the first
series of packets.  Because of this we have not addressed any sort of
request-to-send/cleared-to-send procedure or a handshake procedure
between sources and receivers.

As we move from ASM to SSM and bi-dir, we are going to still be faced
with this same question.  How does one coordinate between the
source and all the receivers to establish some kind of a "start" 
message?

IMHO, if such a message does not exist, any debate concerning the loss
of the first, second, third or even 100th data packet (meaning 1 through
100) is useless.

One possible message would be similar to typing "AT" on your modem.
If such a message did exist, then 60 seconds (or more) worth of "AT" 
packets
could be sent out to coordinate the start of the interesting data.  This
crude example has its own problems in that now you delay delivery of
data for an extended period.

But the point is that unless we recognize that dropping the first data
packets is not always important, and that it is going to be nearly 
impossible
to fix as long as we have to depend on PIM registers and MSDP, we
are going to continue voicing our prejudices concerning these first
few packets and not working toward solving the problem.

(Although the problem could be identified as "you dropped the first
packet", there are many folks who are unwilling to view that as a
problem so lets move on.)

>
> 4. Jimei presented proposals on correctly matching a global address
>    to a link-local IPv6 address for PIM rpf checks.

Someone suggested that this problem is not unique to PIM.  (Dino
had a rant about link-local addresses being a "bug" -- I agree with
him that they are a pain).  It may be that this problem should be
solved by one of the IPv6 working groups.

_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Wed Mar 19 16:55:21 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA11321
	for <pim-archive@lists.ietf.org>; Wed, 19 Mar 2003 16:55:20 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JLt5u22853;
	Wed, 19 Mar 2003 13:55:06 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from prattle.redback.com (hiddenuser@prattle.redback.com [155.53.12.9])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JLsCu22811
	for <pim@catarina.usc.edu>; Wed, 19 Mar 2003 13:54:12 -0800 (PST)
	(envelope-from lwei@redback.com)
Received: from liming-pc.redback.com (liming-pc.redback.com [155.53.36.120])
	by prattle.redback.com (Postfix) with ESMTP
	id E8F6AA1C6E0; Wed, 19 Mar 2003 13:54:11 -0800 (PST)
Received: (from lwei@localhost)
	by liming-pc.redback.com (8.11.3nb1/8.8.8/null redback bsdclient) id h2JLsB411529;
	Wed, 19 Mar 2003 13:54:11 -0800 (PST)
Message-Id: <200303192154.h2JLsB411529@liming-pc.redback.com>
From: Liming Wei <lwei@redback.com>
To: Chris.Cormier@motorola.com
Cc: pim@catarina.usc.edu
In-reply-to: <FB578B06F252D511B95B009027E3267B067620DE@il06exm04.corp.mot.com>
	(message from Cormier Chris-ACC125 on Wed, 19 Mar 2003 12:28:38 -0600)
Subject: Re: [pim] San Francisco PIM WG meeting minutes
Reply-To: lwei@redback.com
References:  <FB578B06F252D511B95B009027E3267B067620DE@il06exm04.corp.mot.com>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Wed, 19 Mar 2003 13:54:11 -0800 (PST)

Chris,

You just reminded me that there were suggestions during the meeting
that the rule could be changed to RPF only on the interface, but not
the RPF neighbor. Mark countered with the potential exponential
growth of duplicates in cascad d topologies with parallel paths sharing
multiple LANs over multiple hops. I quesitoned whether using holddown
timers would solve that problem, but don't remember what Mark responded.

The minute is too short to capture the whole presentation. Mark, can
you post your slides online somewhere ? In short, the primary goal was
to "fix" this blackholing of BSM's with as little change to the protocol
as possible (also as backward compatible  as possible). Of course,
different people may have different ideas of what "simple" and "little"
translate into.


-Liming
---------
> From: Cormier Chris-ACC125 <Chris.Cormier@motorola.com>
> 
> two cents on BSR flooding out IIF.
> 
> I still believe that this should be optional. I would say that there are
> more cases where we know for sure what our RPF to BSR will be than cases
> where there will be inconsistent RPF routes. The bootstrap packet is bloated
> as it is, any solution that would "require" more copies of this message
> would not be a good thing.
> 
> -Chris Cormier
> 
> "2. Mark Handley on PIM BSR flooding problem on LANs with inconsistent
>    RPF information across different routers.
>    Change of rule is to forward BSM out all interfaces including
>    the RPF interface where the BSM was received on."
> _______________________________________________
> pim mailing list
> pim@catarina.usc.edu
> http://catarina.usc.edu/mailman/listinfo/pim
> 
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Wed Mar 19 17:08:52 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA11830
	for <pim-archive@lists.ietf.org>; Wed, 19 Mar 2003 17:08:51 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JM8Cu23239;
	Wed, 19 Mar 2003 14:08:12 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from eve.icir.org (wl-130-157.wireless.ietf56.ietf.org [130.129.130.157])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JM7Eu23199
	for <pim@catarina.usc.edu>; Wed, 19 Mar 2003 14:07:14 -0800 (PST)
	(envelope-from mjh@eve.icir.org)
Received: from eve.icir.org (localhost [127.0.0.1])
	by eve.icir.org (8.12.7/8.12.2) with ESMTP id h2JM78bC003512;
	Wed, 19 Mar 2003 14:07:08 -0800 (PST)
From: Mark Handley <mjh@icir.org>
X-Organisation: ICIR
To: John Zwiebel <jzwiebel@procket.com>
cc: lwei@redback.com, pim@catarina.usc.edu
Subject: Re: [pim] San Francisco PIM WG meeting minutes 
In-reply-to: Your message of "Wed, 19 Mar 2003 13:49:59 PST."
             <BAE51AE0-5A54-11D7-B534-0003936A6062@procket.com> 
Message-ID: <3511.1048111628@eve.icir.org>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Wed, 19 Mar 2003 14:07:08 -0800


>I (jzwiebel) was unable to gain microphone access to ask the following
>(Note that this is really a "whole different discussion").
>
>   "How realistic is it to expect the network to be able to just 
>'handle' all
>    mpackets dumped on to the network and -not- expect loss?"
>
>This is a question that has been hanging around since the beginning of
>multicasting and it has consistently been ignored.  Many, if not most,
>multicast apps used in the interdomain environment do not care whether
>or not the first packets are lost.  Many applications in the intradomain
>environment (although I can't say most) do not care about the first
>series of packets.  Because of this we have not addressed any sort of
>request-to-send/cleared-to-send procedure or a handshake procedure
>between sources and receivers.

The multicast world seems to be shaping up as follows:

 - streaming applications (video broadcast, conferencing, file
   transfer) seem to be a good match with SSM.

 - discovery applications (svrloc, Apple Rendezvous, etc) are only
   served by ASM, but don't generally need really large scope.

In the light of this, the apps that are well suited to ASM and can't
be supported by SSM tend to be low rate bursty apps.  

For the streaming apps, there's probably not a big issue if a bunch of
packets in the first few hundred milliseconds get lost, but SSM isn't
likely to lose those packets anyway unless the receiver joins late.

For the discovery apps, it *really* matters if you don't get the first
few packets through.  

Thus the apps that really *need* ASM will be the ones that get hurt
most of if you drop the first few packets.  Of course if they start up
and immediately send at a high rate, all bets are off.

Cheers,		
	Mark
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Wed Mar 19 17:15:29 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA12079
	for <pim-archive@lists.ietf.org>; Wed, 19 Mar 2003 17:15:28 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JMFDu23423;
	Wed, 19 Mar 2003 14:15:13 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from prattle.redback.com (hiddenuser@prattle.redback.com [155.53.12.9])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JMDxu23378
	for <pim@catarina.usc.edu>; Wed, 19 Mar 2003 14:13:59 -0800 (PST)
	(envelope-from lwei@redback.com)
Received: from liming-pc.redback.com (liming-pc.redback.com [155.53.36.120])
	by prattle.redback.com (Postfix) with ESMTP
	id A49571AB001; Wed, 19 Mar 2003 14:13:58 -0800 (PST)
Received: (from lwei@localhost)
	by liming-pc.redback.com (8.11.3nb1/8.8.8/null redback bsdclient) id h2JMDwq11638;
	Wed, 19 Mar 2003 14:13:58 -0800 (PST)
Message-Id: <200303192213.h2JMDwq11638@liming-pc.redback.com>
From: Liming Wei <lwei@redback.com>
To: jzwiebel@procket.com
Cc: pim@catarina.usc.edu
In-reply-to: <BAE51AE0-5A54-11D7-B534-0003936A6062@procket.com> (message from
	John Zwiebel on Wed, 19 Mar 2003 13:49:59 -0800)
Subject: Re: [pim] San Francisco PIM WG meeting minutes
Reply-To: lwei@redback.com
References:  <BAE51AE0-5A54-11D7-B534-0003936A6062@procket.com>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Wed, 19 Mar 2003 14:13:58 -0800 (PST)

Welcome back Z.

On the first issue, the key is to realize that this is a replacement
for MSDP. I think a sensible choice is to accept the fact majority
of these RP-relayed registers will be received, and for the minority
of packets lossed, the design allows a silent period util the next
periodic register occurs. I believe this is the same as the MSDP
behavior.

On the second issue about multiple addresses, I spoke with some one
working on OSPF (let me not mention the name here) afterwards. I do
think there is significant benefit in using a PIM option to convey all
addresses. Relying on a collection of other mechaisms (like neihgbor
discovery etc) does not appeal to me to be good design choices. I
think Mark and company made a good choice on using a new PIM option.

-Liming
------------
> From: John Zwiebel <jzwiebel@procket.com>
> 
> On Tuesday, March 18, 2003, at 06:00 PM, Liming Wei wrote:
> 
> > Folks,
> >
> > We forgot to ask for a volunteer to take notes. Here is a note
> > I typed quickly while my memory is still fresh. Additions/updates
> > are welcome.
> >
> >
> > Working group: PIM
> > Location: San Francisco Hilton
> > Time:  3/18/2003, 3:45pm
> > Note Taker: Liming Wei
> 
> > 3. Dino presented proposal on Anycast RP with PIM.
> 
> >    A few comments were made about this proposal. One issue raised was
> >    packet losses resulted from loss of one PIM register (Mark H).
> 
> Bill Nickless was very much in favor of -not- encapsulating data in the
> registers between anycast-RPs but only using NULL registers to alert
> the other RPs in the group that they should join to the SPT.
> 
> Mark's point would then be exacerbated since it would guarantee that the
> first packet would be lost.
> 
> I (jzwiebel) was unable to gain microphone access to ask the following
> (Note that this is really a "whole different discussion").
> 
>    "How realistic is it to expect the network to be able to just 
> 'handle' all
>     mpackets dumped on to the network and -not- expect loss?"
> 
> This is a question that has been hanging around since the beginning of
> multicasting and it has consistently been ignored.  Many, if not most,
> multicast apps used in the interdomain environment do not care whether
> or not the first packets are lost.  Many applications in the intradomain
> environment (although I can't say most) do not care about the first
> series of packets.  Because of this we have not addressed any sort of
> request-to-send/cleared-to-send procedure or a handshake procedure
> between sources and receivers.
> 
> As we move from ASM to SSM and bi-dir, we are going to still be faced
> with this same question.  How does one coordinate between the
> source and all the receivers to establish some kind of a "start" 
> message?
> 
> IMHO, if such a message does not exist, any debate concerning the loss
> of the first, second, third or even 100th data packet (meaning 1 through
> 100) is useless.
> 
> One possible message would be similar to typing "AT" on your modem.
> If such a message did exist, then 60 seconds (or more) worth of "AT" 
> packets
> could be sent out to coordinate the start of the interesting data.  This
> crude example has its own problems in that now you delay delivery of
> data for an extended period.
> 
> But the point is that unless we recognize that dropping the first data
> packets is not always important, and that it is going to be nearly 
> impossible
> to fix as long as we have to depend on PIM registers and MSDP, we
> are going to continue voicing our prejudices concerning these first
> few packets and not working toward solving the problem.
> 
> (Although the problem could be identified as "you dropped the first
> packet", there are many folks who are unwilling to view that as a
> problem so lets move on.)
> 
> >
> > 4. Jimei presented proposals on correctly matching a global address
> >    to a link-local IPv6 address for PIM rpf checks.
> 
> Someone suggested that this problem is not unique to PIM.  (Dino
> had a rant about link-local addresses being a "bug" -- I agree with
> him that they are a pain).  It may be that this problem should be
> solved by one of the IPv6 working groups.
> 
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Wed Mar 19 18:05:08 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA13823
	for <pim-archive@lists.ietf.org>; Wed, 19 Mar 2003 18:05:07 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JN4Ru24440;
	Wed, 19 Mar 2003 15:04:27 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from motgate3.mot.com (motgate3.mot.com [144.189.100.103])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JN3Wu24409
	for <pim@catarina.usc.edu>; Wed, 19 Mar 2003 15:03:32 -0800 (PST)
	(envelope-from Chris.Cormier@motorola.com)
Received: from mothost.mot.com (mothost.mot.com [129.188.137.101])
	by motgate3.mot.com (Motorola/Motgate3) with ESMTP id h2JN2oT8014780
	for <pim@catarina.usc.edu>; Wed, 19 Mar 2003 16:02:50 -0700 (MST)
Received: [from il06exm05.corp.mot.com (il06exm05.corp.mot.com [10.0.111.5]) by mothost.mot.com (MOT-pobox 2.0) with ESMTP id QAA06566 for <pim@catarina.usc.edu>; Wed, 19 Mar 2003 16:03:31 -0700 (MST)]
Received: by il06exm05.corp.mot.com with Internet Mail Service (5.5.2656.59)
	id <HALK90S0>; Wed, 19 Mar 2003 17:03:31 -0600
Message-ID: <FB578B06F252D511B95B009027E3267B067620E6@il06exm04.corp.mot.com>
From: Cormier Chris-ACC125 <Chris.Cormier@motorola.com>
To: "'lwei@redback.com'" <lwei@redback.com>
Cc: pim@catarina.usc.edu
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2656.59)
Content-Type: text/plain
Subject: [pim] Multiple address RPF solution
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Wed, 19 Mar 2003 17:03:29 -0600

We have faced similar issues with RPF toward RP when using a non-neighbor IP
address for sending BSR messages. This hello option solution appeals to me
here as well as the original problem statement. I would rather not push this
off to v6 only solution as there may be benefits in a v4 world.
-chris

>On the second issue about multiple addresses, I spoke with some one
>working on OSPF (let me not mention the name here) afterwards. I do
>think there is significant benefit in using a PIM option to convey all
>addresses. Relying on a collection of other mechaisms (like neihgbor
>discovery etc) does not appeal to me to be good design choices. I
>think Mark and company made a good choice on using a new PIM option.
>
>-Liming
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Wed Mar 19 18:38:24 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA16090
	for <pim-archive@lists.ietf.org>; Wed, 19 Mar 2003 18:38:24 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JNcCu25152;
	Wed, 19 Mar 2003 15:38:12 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from mcs.anl.gov (cliff.mcs.anl.gov [140.221.9.17])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JNbUu25125
	for <pim@catarina.usc.edu>; Wed, 19 Mar 2003 15:37:30 -0800 (PST)
	(envelope-from nickless@mcs.anl.gov)
Received: from newtwinkle.mcs.anl.gov (vpnclient18106.mcs.anl.gov [140.221.18.106])
	by mcs.anl.gov (8.11.6/8.9.3) with ESMTP id h2JNbDL174942;
	Wed, 19 Mar 2003 17:37:13 -0600
Message-Id: <5.1.1.6.2.20030319160416.027f6a60@pop.mcs.anl.gov>
X-Sender: nickless@pop.mcs.anl.gov
X-Mailer: QUALCOMM Windows Eudora Version 5.1.1
To: John Zwiebel <jzwiebel@procket.com>
From: Bill Nickless <nickless@mcs.anl.gov>
Cc: lwei@redback.com, pim@catarina.usc.edu,
        mboned@network-services.uoregon.edu
In-Reply-To: <BAE51AE0-5A54-11D7-B534-0003936A6062@procket.com>
References: <200303190200.h2J20Pf08821@liming-pc.redback.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Subject: [pim] Discovery Apps [Was: San Francisco PIM WG meeting minutes]
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Wed, 19 Mar 2003 17:37:09 -0600

At 01:49 PM 3/19/2003 -0800, John Zwiebel wrote:

>I (jzwiebel) was unable to gain microphone access to ask the following
>(Note that this is really a "whole different discussion").
>
>   "How realistic is it to expect the network to be able to just 'handle' all
>    mpackets dumped on to the network and -not- expect loss?"

Remember that IP multicast, as defined in RFC-1112, was originally intended 
to give IP applications access to the broadcast-nature of a Metcalfe 
Ethernet.  RFC-1112 doesn't say anything other than there "may" be gateways 
to move IP multicast packets between Ethernets.

I think we've learned four lessons since 1989:

  <A> There is a class of application that wants one-to-many
      distribution of IP packets, even when the sources and
      receivers are on different Ethernet (and other) LANs.

  <B> What's appropriate on a Metcalfe Ethernet--transmit at will
      because everyone will hear you anyway--doesn't scale to an
      Internetwork.

  <C> IP Multicast deployments that use source-based distribution
      trees can and do scale to large Internetworks.

  <D> Source-rooted distribution trees should be maintained so
      long as either (1) there's traffic emitted by that source
      to interested receivers or (2) there is expressed receiver
      interest in receiving traffic from that source.

Mark Handley wrote:

>  - discovery applications (svrloc, Apple Rendezvous, etc) are only
>    served by ASM, but don't generally need really large scope.

[...]

>For the discovery apps, it *really* matters if you don't get the first
>few packets through.
>
>Thus the apps that really *need* ASM will be the ones that get hurt
>most of if you drop the first few packets.  Of course if they start up
>and immediately send at a high rate, all bets are off.

I would characterize discovery apps slightly differently---

For the discovery apps, it *really* matters that packets get through, even 
if they're at a low frequency.

One problem with certain discovery apps is that they don't emit packets 
with a high enough frequency.  This violates Lesson <D> above.  The 
source-rooted distribution trees don't exist long enough to carry the 
packets from the low-frequency source to all the interested receivers.

Data encapsulation in PIM Register or MSDP SA messages violates Lesson <B>.

My position:

1. Deprecate data encapsulation in PIM Register, MSDP SA, and all
    other such spreading-knowledge-of-active-source messages.

2. Apply lessons <A> through <D> above to multicast applications.
    Specifically, require applications to either (i) use SSM or (ii)
    maintain a minimum frequency of sourced packets.


===
Bill Nickless    http://www.mcs.anl.gov/people/nickless      +1 630 252 7390
PGP:0E 0F 16 80 C5 B1 69 52 E1 44 1A A5 0E 1B 74 F7     nickless@mcs.anl.gov

_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Wed Mar 19 18:50:26 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA16453
	for <pim-archive@lists.ietf.org>; Wed, 19 Mar 2003 18:50:25 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JNo7u25429;
	Wed, 19 Mar 2003 15:50:07 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from hotmail.com (oe59.law9.hotmail.com [64.4.8.194])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JNnwu25405
	for <pim@catarina.usc.edu>; Wed, 19 Mar 2003 15:49:58 -0800 (PST)
	(envelope-from desai_purvi@hotmail.com)
Received: from mail pickup service by hotmail.com with Microsoft SMTPSVC;
	 Wed, 19 Mar 2003 15:49:52 -0800
Received: from 63.126.135.11 by oe59.law9.hotmail.com with DAV;
	Wed, 19 Mar 2003 23:49:52 +0000
X-Originating-IP: [63.126.135.11]
X-Originating-Email: [desai_purvi@hotmail.com]
From: "Purvi" <desai_purvi@hotmail.com>
To: <lwei@redback.com>, <pim@catarina.usc.edu>
References: <200303190200.h2J20Pf08821@liming-pc.redback.com>
Subject: Re: [pim] San Francisco PIM WG meeting minutes
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 6.00.2600.0000
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000
Message-ID: <OE5961LEdyt3cew8k690002574f@hotmail.com>
X-OriginalArrivalTime: 19 Mar 2003 23:49:52.0908 (UTC) FILETIME=[3C88DCC0:01C2EE72]
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Wed, 19 Mar 2003 15:46:29 -0800
Content-Transfer-Encoding: 7bit

I have not gone through Anycast proposal, so much of my doubt and the
solution here might seem beaten up by you all but I am wondering about
initial packet drop and also about register-stop messages between RPs
being optional. and would also like to know about the feasibility of the
behavior mentioned in point 2 of this email.

1) Is it that first RP just forwards the register packet received from DR?
    If it is then out of many similar cases consider a case where first RP
    receives register from DR and since it does not have any  state for G
    (*,g) or (*,*,rp), it sends register stop immediately.
    And relayed register message cannot reach other RPs . So
    all receivers which are interested have to wait for next 60 seconds.
And
   even after  that in next register cycle, same thing can be repeated.

   This looks like big loophole in the protocol. PIM SM atleast tried to
   reduce problem by  making DR send register message till it receives stop,
   BUT in this case the initial packet problem is outrightly ignored.
   (is this over simplification ?? or I have got something wrong in my
    understanding)

2) I have this question above because  I can think of more complicated but a
   little more robust behaviour where first RP waits for register stop from
all
   RPs or have some max time for which it waits for stop
   and times out before it stops sending register packet to other RPs.
    Ofcourse many variant of this solution are possible but tweaking the
behaviour.

   And this registering and register from DR can be completely decoupled and
   some optimizations like sending just null registers are always possible.
   With some latitude, I can compare this to heirarchical model. one source
   lets one RP know and that RP registers to other RPs till needed.

-Regards,
Purvi


----- Original Message -----
From: "Liming Wei" <lwei@redback.com>
To: <pim@catarina.usc.edu>
Sent: Tuesday, March 18, 2003 6:00 PM
Subject: [pim] San Francisco PIM WG meeting minutes


> Folks,
>
> We forgot to ask for a volunteer to take notes. Here is a note
> I typed quickly while my memory is still fresh. Additions/updates
> are welcome.
>
>
> Working group: PIM
> Location: San Francisco Hilton
> Time:  3/18/2003, 3:45pm
> Note Taker: Liming Wei
>
> 1. Hugh did an update on PIM SM spec. Updated the security section
>    of draft. (*,*,RP) will stay in spec, for future use, to fix up
>    one problem involving (S,G)RPT prune.
> 2. Mark Handley on PIM BSR flooding problem on LANs with inconsistent
>    RPF information across different routers.
>    Change of rule is to forward BSM out all interfaces including
>    the RPF interface where the BSM was received on.
> 3. Dino presented proposal on Anycast RP with PIM. The proposal attempts
>    to provide anycast RP functio without using MSDP.
>    Gist is to form a set of RPs configured with the same RP address
>    (any cast address). Each DR sends registers to the closest RP,
>    which will forward to the other RPs in the RP set. RPs accept
>    these relayed PIM registers, but do not forward relayed PIM registers.
>    Register-stop mesages between RPs are optional, reason is register
>    stops happen between the DR and the first RP already. Two interoperable
>    implementations from cisco and Procket have been done.
>
>    A few comments were made about this proposal. One issue raised was
>    packet losses resulted from loss of one PIM register (Mark H).
>    Dino responded with such event is recovered upon the next register,
>    and one lives with such occasional losses in the beginning. Another
>    suggestion (Bill F?) was to do the relays via multicast forwarding.
>    Dino's response was to forcus on the simplicity, and to avoid
>    introduction of new machinary. Etc.
>
>    Question to working group, does the working group want to take on
>    this as a working group item ?
>
> 4. Jimei presented proposals on correctly matching a global address
>    to a link-local IPv6 address for PIM rpf checks. Options include
>    introducing a new PIM hello options that will list all addresses
>    on a PIM neighor's interface; configuration based; etc.
>    Also questioned if working group will take this as a working group
>    item.
>
>    Mark commented after a short discussion with a few spec coauthers
>    that they are open to the idea of using a new PIM hello option,
>    and to explicitly document this RPF address issue in the base spec.
>
> 5. Tom asked the working group for possible feedbacks on the PIM dense
>    mode draft and the new PIM MIB draft. 2 people raised hands.
>    Tom asked working group members to read and comment on the
>    drafts, as usual.
>
> I briefly conversed with Tom after the WG meeting that we
> think we should continue the design discussions for 2 and 3 above
> on the working group mailing list.
>
> On the subject of the new PIM dense mode draft, Tom and I felt that
> the best course of action is to move PIM dense mode draft to either
> informational or experimental RFC, and get done with it. We will
> request the new drafters to get the PIM dense mode draft into
> a form suitable for submission. Still we need document reviews
> from the working group members.
>
>
> -Liming
> _______________________________________________
> pim mailing list
> pim@catarina.usc.edu
> http://catarina.usc.edu/mailman/listinfo/pim
>
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Wed Mar 19 19:06:57 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA16946
	for <pim-archive@lists.ietf.org>; Wed, 19 Mar 2003 19:06:56 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2K06Mu25815;
	Wed, 19 Mar 2003 16:06:22 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from hotmail.com (oe51.law9.hotmail.com [64.4.8.40])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2K05nu25795
	for <pim@catarina.usc.edu>; Wed, 19 Mar 2003 16:05:49 -0800 (PST)
	(envelope-from desai_purvi@hotmail.com)
Received: from mail pickup service by hotmail.com with Microsoft SMTPSVC;
	 Wed, 19 Mar 2003 16:05:44 -0800
Received: from 63.126.135.11 by oe51.law9.hotmail.com with DAV;
	Thu, 20 Mar 2003 00:05:44 +0000
X-Originating-IP: [63.126.135.11]
X-Originating-Email: [desai_purvi@hotmail.com]
From: "Purvi" <desai_purvi@hotmail.com>
To: <pim@catarina.usc.edu>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 6.00.2600.0000
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000
Message-ID: <OE514aJ6pkNzPZrIkyP00025649@hotmail.com>
X-OriginalArrivalTime: 20 Mar 2003 00:05:44.0531 (UTC) FILETIME=[73BF0230:01C2EE74]
Subject: [pim] Re: San Francisco PIM WG meeting minutes
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Wed, 19 Mar 2003 16:02:21 -0800
Content-Transfer-Encoding: 7bit

Regarding the parallel paths sharing multiple LANS -
How do routers handle a case where there are multiple
interfaces on same subnet and other routers on that LAN can
have any of these multiple interfaces as the next hop for the BSR.
I think currently it is needed that router  forwards same BSM
on all interfaces of the same subnet , since it doesnt know which
interface IP, would had been picked by other routers for RPF to BSR.
This is clearly an unoptimized case if this is the only way to circumvent
this problem. Is there a way to avoid this flooding on multiple
interfaces on same subnet.

I would think that this could had been optimized if there was a notion of
one router id as other protocols have.

Regards,
Purvi

Chris,

You just reminded me that there were suggestions during the meeting
that the rule could be changed to RPF only on the interface, but not
the RPF neighbor. Mark countered with the potential exponential
growth of duplicates in cascad d topologies with parallel paths sharing
multiple LANs over multiple hops. I quesitoned whether using holddown
timers would solve that problem, but don't remember what Mark responded.

The minute is too short to capture the whole presentation. Mark, can
you post your slides online somewhere ? In short, the primary goal was
to "fix" this blackholing of BSM's with as little change to the protocol
as possible (also as backward compatible  as possible). Of course,
different people may have different ideas of what "simple" and "little"
translate into.


-Liming
---------
> From: Cormier Chris-ACC125 <Chris.Cormier@motorola.com>
>
> two cents on BSR flooding out IIF.
>
> I still believe that this should be optional. I would say that there are
> more cases where we know for sure what our RPF to BSR will be than cases
> where there will be inconsistent RPF routes. The bootstrap packet is
bloated
> as it is, any solution that would "require" more copies of this message
> would not be a good thing.
>
> -Chris Cormier
>
> "2. Mark Handley on PIM BSR flooding problem on LANs with inconsistent
>    RPF information across different routers.
>    Change of rule is to forward BSM out all interfaces including
>    the RPF interface where the BSM was received on."
> _______________________________________________
> pim mailing list
> pim@catarina.usc.edu
> http://catarina.usc.edu/mailman/listinfo/pim
>
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Wed Mar 19 19:17:33 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA17330
	for <pim-archive@lists.ietf.org>; Wed, 19 Mar 2003 19:17:31 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2K0HCu26068;
	Wed, 19 Mar 2003 16:17:12 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from tyholt.uninett.no (tyholt.uninett.no [158.38.60.10])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2K0GZu26042
	for <pim@catarina.usc.edu>; Wed, 19 Mar 2003 16:16:35 -0800 (PST)
	(envelope-from Stig.Venaas@uninett.no)
Received: from sverresborg.uninett.no (sverresborg.uninett.no [IPv6:2001:700:e000:0:290:27ff:fe50:6bfa])
	by tyholt.uninett.no (8.12.8/8.12.6) with ESMTP id h2K0GYgJ026361;
	Thu, 20 Mar 2003 01:16:34 +0100
Received: (from venaas@localhost)
	by sverresborg.uninett.no (8.11.6/8.11.2) id h2K0GXf24386;
	Thu, 20 Mar 2003 01:16:33 +0100
X-Authentication-Warning: sverresborg.uninett.no: venaas set sender to Stig.Venaas@uninett.no using -f
From: Stig Venaas <Stig.Venaas@uninett.no>
To: Purvi <desai_purvi@hotmail.com>
Cc: pim@catarina.usc.edu
Subject: Re: [pim] Re: San Francisco PIM WG meeting minutes
Message-ID: <20030320011633.A24380@sverresborg.uninett.no>
References: <OE514aJ6pkNzPZrIkyP00025649@hotmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <OE514aJ6pkNzPZrIkyP00025649@hotmail.com>; from desai_purvi@hotmail.com on Wed, Mar 19, 2003 at 04:02:21PM -0800
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Thu, 20 Mar 2003 01:16:33 +0100

On Wed, Mar 19, 2003 at 04:02:21PM -0800, Purvi wrote:
> Regarding the parallel paths sharing multiple LANS -
> How do routers handle a case where there are multiple
> interfaces on same subnet and other routers on that LAN can
> have any of these multiple interfaces as the next hop for the BSR.
> I think currently it is needed that router  forwards same BSM
> on all interfaces of the same subnet , since it doesnt know which
> interface IP, would had been picked by other routers for RPF to BSR.

One could perhaps use the discussed hello option? A router could
send BSMs on just one of the interfaces provided that hello messages
include all addresses the router has on the LAN.

Just a quick idea, might be some issues I haven't thought of,

Stig
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Wed Mar 19 20:05:46 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA18906
	for <pim-archive@lists.ietf.org>; Wed, 19 Mar 2003 20:05:44 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2K15Nu27098;
	Wed, 19 Mar 2003 17:05:23 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from merlot.juniper.net (natint.juniper.net [207.17.136.129])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2JM4ku23133
	for <pim@catarina.usc.edu>; Wed, 19 Mar 2003 14:04:46 -0800 (PST)
	(envelope-from pusateri@juniper.net)
Received: from juniper.net (garnet.juniper.net [172.17.28.17])
	by merlot.juniper.net (8.11.3/8.11.3) with ESMTP id h2JM4bS40873;
	Wed, 19 Mar 2003 14:04:37 -0800 (PST)
	(envelope-from pusateri@juniper.net)
Message-Id: <200303192204.h2JM4bS40873@merlot.juniper.net>
To: lwei@redback.com
cc: Chris.Cormier@motorola.com, pim@catarina.usc.edu, pusateri@juniper.net
Subject: Re: [pim] San Francisco PIM WG meeting minutes 
In-Reply-To: Message from Liming Wei <lwei@redback.com> 
   of "Wed, 19 Mar 2003 13:54:11 PST." <200303192154.h2JLsB411529@liming-pc.redback.com> 
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <36997.1048111477.1@juniper.net>
From: Tom Pusateri <pusateri@juniper.net>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Wed, 19 Mar 2003 14:04:37 -0800

Actually, anyone who presented slides yesterday needs to send them
to me for submission to the minutes. Please do so right away.

Thanks,
Tom

In message <200303192154.h2JLsB411529@liming-pc.redback.com> you write:
>Chris,
>
>You just reminded me that there were suggestions during the meeting
>that the rule could be changed to RPF only on the interface, but not
>the RPF neighbor. Mark countered with the potential exponential
>growth of duplicates in cascad d topologies with parallel paths sharing
>multiple LANs over multiple hops. I quesitoned whether using holddown
>timers would solve that problem, but don't remember what Mark responded.
>
>The minute is too short to capture the whole presentation. Mark, can
>you post your slides online somewhere ? In short, the primary goal was
>to "fix" this blackholing of BSM's with as little change to the protocol
>as possible (also as backward compatible  as possible). Of course,
>different people may have different ideas of what "simple" and "little"
>translate into.
>
>
>-Liming
>---------
>> From: Cormier Chris-ACC125 <Chris.Cormier@motorola.com>
>> 
>> two cents on BSR flooding out IIF.
>> 
>> I still believe that this should be optional. I would say that there are
>> more cases where we know for sure what our RPF to BSR will be than cases
>> where there will be inconsistent RPF routes. The bootstrap packet is bloated
>> as it is, any solution that would "require" more copies of this message
>> would not be a good thing.
>> 
>> -Chris Cormier
>> 
>> "2. Mark Handley on PIM BSR flooding problem on LANs with inconsistent
>>    RPF information across different routers.
>>    Change of rule is to forward BSM out all interfaces including
>>    the RPF interface where the BSM was received on."
>> _______________________________________________
>> pim mailing list
>> pim@catarina.usc.edu
>> http://catarina.usc.edu/mailman/listinfo/pim
>> 
>_______________________________________________
>pim mailing list
>pim@catarina.usc.edu
>http://catarina.usc.edu/mailman/listinfo/pim
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Wed Mar 19 20:19:13 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA19336
	for <pim-archive@lists.ietf.org>; Wed, 19 Mar 2003 20:19:11 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2K1FKu27328;
	Wed, 19 Mar 2003 17:15:20 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from hotmail.com (oe17.law9.hotmail.com [64.4.8.121])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2K1EVu27299
	for <pim@catarina.usc.edu>; Wed, 19 Mar 2003 17:14:31 -0800 (PST)
	(envelope-from desai_purvi@hotmail.com)
Received: from mail pickup service by hotmail.com with Microsoft SMTPSVC;
	 Wed, 19 Mar 2003 17:14:26 -0800
Received: from 63.126.135.11 by oe17.law9.hotmail.com with DAV;
	Thu, 20 Mar 2003 01:14:24 +0000
X-Originating-IP: [63.126.135.11]
X-Originating-Email: [desai_purvi@hotmail.com]
From: "Purvi" <desai_purvi@hotmail.com>
To: "Stig Venaas" <Stig.Venaas@uninett.no>
Cc: <pim@catarina.usc.edu>
References: <OE514aJ6pkNzPZrIkyP00025649@hotmail.com> <20030320011633.A24380@sverresborg.uninett.no>
Subject: Re: [pim] Re: San Francisco PIM WG meeting minutes
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 6.00.2600.0000
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000
Message-ID: <OE17xniMRQETEk37euX0002af18@hotmail.com>
X-OriginalArrivalTime: 20 Mar 2003 01:14:26.0348 (UTC) FILETIME=[0C8A5EC0:01C2EE7E]
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Wed, 19 Mar 2003 17:11:01 -0800
Content-Transfer-Encoding: 7bit

Right, I was thinking in this line, but workaround
was to send messages on all routers.
Rpf check followed by check in address list does solve this problem
So if this draft is extended further, this means that a router can send one
hello
for all the interfaces on that subnet. This sounds like hello
messages of other protocols with one router id.

Regards,
-Purvi


----- Original Message -----
From: "Stig Venaas" <Stig.Venaas@uninett.no>
To: "Purvi" <desai_purvi@hotmail.com>
Cc: <pim@catarina.usc.edu>
Sent: Wednesday, March 19, 2003 4:16 PM
Subject: Re: [pim] Re: San Francisco PIM WG meeting minutes


> On Wed, Mar 19, 2003 at 04:02:21PM -0800, Purvi wrote:
> > Regarding the parallel paths sharing multiple LANS -
> > How do routers handle a case where there are multiple
> > interfaces on same subnet and other routers on that LAN can
> > have any of these multiple interfaces as the next hop for the BSR.
> > I think currently it is needed that router  forwards same BSM
> > on all interfaces of the same subnet , since it doesnt know which
> > interface IP, would had been picked by other routers for RPF to BSR.
>
> One could perhaps use the discussed hello option? A router could
> send BSMs on just one of the interfaces provided that hello messages
> include all addresses the router has on the LAN.
>
> Just a quick idea, might be some issues I haven't thought of,
>
> Stig
>
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Wed Mar 19 20:30:02 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA19655
	for <pim-archive@lists.ietf.org>; Wed, 19 Mar 2003 20:30:01 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2K1T7u27630;
	Wed, 19 Mar 2003 17:29:07 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from eve.icir.org (wl-130-157.wireless.ietf56.ietf.org [130.129.130.157])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2K1Sbu27606
	for <pim@catarina.usc.edu>; Wed, 19 Mar 2003 17:28:37 -0800 (PST)
	(envelope-from mjh@eve.icir.org)
Received: from eve.icir.org (localhost [127.0.0.1])
	by eve.icir.org (8.12.7/8.12.2) with ESMTP id h2K1SCbC006405;
	Wed, 19 Mar 2003 17:28:27 -0800 (PST)
From: Mark Handley <mjh@icir.org>
X-Organisation: ICIR
To: Tom Pusateri <pusateri@juniper.net>
cc: pim@catarina.usc.edu
Subject: Re: [pim] San Francisco PIM WG meeting minutes 
In-reply-to: Your message of "Wed, 19 Mar 2003 14:04:37 PST."
             <200303192204.h2JM4bS40873@merlot.juniper.net> 
Message-ID: <6404.1048123692@eve.icir.org>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Wed, 19 Mar 2003 17:28:12 -0800


>Actually, anyone who presented slides yesterday needs to send them
>to me for submission to the minutes. Please do so right away.

http://www.icir.org/mjh/tmp/bsr-sf.pdf
http://www.icir.org/mjh/tmp/bsr-sf.ppt

 - Mark
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Wed Mar 19 21:11:43 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA20911
	for <pim-archive@lists.ietf.org>; Wed, 19 Mar 2003 21:11:42 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2K2AQu28510;
	Wed, 19 Mar 2003 18:10:26 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from v6.hitachi.co.jp (galilei.v6.hitachi.co.jp [133.145.167.4])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2K28pu28470
	for <pim@catarina.usc.edu>; Wed, 19 Mar 2003 18:08:51 -0800 (PST)
	(envelope-from suz@crl.hitachi.co.jp)
Received: from s30.crl.hitachi.co.jp (galilei.v6.hitachi.co.jp [IPv6:3ffe:501:1011::1])
	by v6.hitachi.co.jp (8.12.8/8.11.6) with ESMTP id h2K2HKJY010058;
	Thu, 20 Mar 2003 11:17:20 +0900 (JST)
	(envelope-from suz@crl.hitachi.co.jp)
Message-ID: <x74r5yx01n.wl%suz@crl.hitachi.co.jp>
From: SUZUKI Shinsuke <suz@crl.hitachi.co.jp>
To: Tom Pusateri <pusateri@juniper.net>
Cc: lwei@redback.com, Chris.Cormier@motorola.com, pim@catarina.usc.edu,
        jinmei@isl.rdc.toshiba.co.jp
Subject: Re: [pim] San Francisco PIM WG meeting minutes 
X-cite: xcite 1.33
In-Reply-To: <200303192204.h2JM4bS40873@merlot.juniper.net>
References: <lwei@redback.com>
	<200303192154.h2JLsB411529@liming-pc.redback.com>
	<200303192204.h2JM4bS40873@merlot.juniper.net>
User-Agent: User-Agent: Wanderlust/2.11.3 (Wonderwall) Emacs/21.1 Mule/5.0 (SAKAKI)
Organization: Network Architecture Research Dept., Central Research Laboratory, Hitachi, Ltd, Japan
MIME-Version: 1.0 (generated by SEMI 1.14.4 - "Hosorogi")
Content-Type: text/plain; charset=US-ASCII
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Thu, 20 Mar 2003 11:08:04 +0900

>>>>> On Wed, 19 Mar 2003 14:04:37 -0800
>>>>> pusateri@juniper.net(Tom Pusateri)  said:

> Actually, anyone who presented slides yesterday needs to send them
> to me for submission to the minutes. Please do so right away.

Here's jinmei-san's presentation (converted from magicpoint to html).
	http://www.kame.net/~suz/pim-upstream-detection/index.html

Thanks,
----
SUZUKI, Shinsuke @ Hitachi / KAME Project
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Thu Mar 20 13:47:04 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA29054
	for <pim-archive@lists.ietf.org>; Thu, 20 Mar 2003 13:47:03 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2KIiBu47487;
	Thu, 20 Mar 2003 10:44:11 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from sj-core-5.cisco.com (sj-core-5.cisco.com [171.71.177.238])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2KIglu47440
	for <pim@catarina.usc.edu>; Thu, 20 Mar 2003 10:42:47 -0800 (PST)
	(envelope-from jmeylor@cisco.com)
Received: from mira-sjc5-b.cisco.com (IDENT:mirapoint@mira-sjc5-b.cisco.com [171.71.163.14])
	by sj-core-5.cisco.com (8.12.6/8.12.6) with ESMTP id h2KIgZHp011271;
	Thu, 20 Mar 2003 10:42:36 -0800 (PST)
Received: from cisco.com (jmeylor-frame5.cisco.com [10.19.252.142])
	by mira-sjc5-b.cisco.com (Mirapoint Messaging Server MOS 3.2.1-GA)
	with ESMTP id AFH87571;
	Thu, 20 Mar 2003 10:41:16 -0800 (PST)
Message-ID: <3E7A0B9A.AB3D344@cisco.com>
From: John Meylor <jmeylor@cisco.com>
X-Mailer: Mozilla 4.76 [en]C-CCK-MCD   (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Bill Nickless <nickless@mcs.anl.gov>
CC: John Zwiebel <jzwiebel@procket.com>, lwei@redback.com,
        pim@catarina.usc.edu, mboned@network-services.uoregon.edu
References: <200303190200.h2J20Pf08821@liming-pc.redback.com> <5.1.1.6.2.20030319160416.027f6a60@pop.mcs.anl.gov>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Subject: [pim] Re: Discovery Apps [Was: San Francisco PIM WG meeting minutes]
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Thu, 20 Mar 2003 10:42:34 -0800
Content-Transfer-Encoding: 7bit

Bill Nickless wrote:
> 
> At 01:49 PM 3/19/2003 -0800, John Zwiebel wrote:
> 
> >I (jzwiebel) was unable to gain microphone access to ask the following
> >(Note that this is really a "whole different discussion").
> >
> >   "How realistic is it to expect the network to be able to just 'handle' all
> >    mpackets dumped on to the network and -not- expect loss?"
> 
> Remember that IP multicast, as defined in RFC-1112, was originally intended
> to give IP applications access to the broadcast-nature of a Metcalfe
> Ethernet.  RFC-1112 doesn't say anything other than there "may" be gateways
> to move IP multicast packets between Ethernets.
> 
> I think we've learned four lessons since 1989:
> 
>   <A> There is a class of application that wants one-to-many
>       distribution of IP packets, even when the sources and
>       receivers are on different Ethernet (and other) LANs.
> 
>   <B> What's appropriate on a Metcalfe Ethernet--transmit at will
>       because everyone will hear you anyway--doesn't scale to an
>       Internetwork.
> 
>   <C> IP Multicast deployments that use source-based distribution
>       trees can and do scale to large Internetworks.
> 
>   <D> Source-rooted distribution trees should be maintained so
>       long as either (1) there's traffic emitted by that source
>       to interested receivers or (2) there is expressed receiver
>       interest in receiving traffic from that source.
> 
> Mark Handley wrote:
> 
> >  - discovery applications (svrloc, Apple Rendezvous, etc) are only
> >    served by ASM, but don't generally need really large scope.
> 
> [...]
> 
> >For the discovery apps, it *really* matters if you don't get the first
> >few packets through.
> >
> >Thus the apps that really *need* ASM will be the ones that get hurt
> >most of if you drop the first few packets.  Of course if they start up
> >and immediately send at a high rate, all bets are off.
> 
> I would characterize discovery apps slightly differently---
> 
> For the discovery apps, it *really* matters that packets get through, even
> if they're at a low frequency.
> 
> One problem with certain discovery apps is that they don't emit packets
> with a high enough frequency.  This violates Lesson <D> above.  The
> source-rooted distribution trees don't exist long enough to carry the
> packets from the low-frequency source to all the interested receivers.
> 
> Data encapsulation in PIM Register or MSDP SA messages violates Lesson <B>.
> 
> My position:
> 
> 1. Deprecate data encapsulation in PIM Register, 

what would you do for SM deployments which leverage the shared
tree?

> MSDP SA, and all
>     other such spreading-knowledge-of-active-source messages.

registers in true shared-tree SM mode are moving more than
information about active sources, they are moving data to
the distribution tree.   There are signficant deployments
using this well described, long established model.

I agree SSM is a much simplier model for the network, 
- we need ubiquitous v3 support in apps/stacks/l2 devices,

and that it could even be the network service for dynamic 
multisource groups, 
- we some out-of-band dynamic source discovery element

nothing is stopping either of those except market pressure

but we won't get any closer to these SSM based models
by breaking currently working ASM models.

 john

> 
> 2. Apply lessons <A> through <D> above to multicast applications.
>     Specifically, require applications to either (i) use SSM or (ii)
>     maintain a minimum frequency of sourced packets.
> 
> ===
> Bill Nickless    http://www.mcs.anl.gov/people/nickless      +1 630 252 7390
> PGP:0E 0F 16 80 C5 B1 69 52 E1 44 1A A5 0E 1B 74 F7     nickless@mcs.anl.gov
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Thu Mar 20 14:31:56 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA01151
	for <pim-archive@lists.ietf.org>; Thu, 20 Mar 2003 14:31:56 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2KJSBu48368;
	Thu, 20 Mar 2003 11:28:11 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from mcs.anl.gov (cliff.mcs.anl.gov [140.221.9.17])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2KJRhu48343
	for <pim@catarina.usc.edu>; Thu, 20 Mar 2003 11:27:43 -0800 (PST)
	(envelope-from nickless@mcs.anl.gov)
Received: from newtwinkle.mcs.anl.gov (vpnclient18104.mcs.anl.gov [140.221.18.104])
	by mcs.anl.gov (8.11.6/8.9.3) with ESMTP id h2KJRNL108250;
	Thu, 20 Mar 2003 13:27:24 -0600
Message-Id: <5.1.1.6.2.20030320125003.02e2ee18@pop.mcs.anl.gov>
X-Sender: nickless@pop.mcs.anl.gov
X-Mailer: QUALCOMM Windows Eudora Version 5.1.1
To: John Meylor <jmeylor@cisco.com>
From: Bill Nickless <nickless@mcs.anl.gov>
Cc: pim@catarina.usc.edu, mboned@network-services.uoregon.edu
In-Reply-To: <3E7A0B9A.AB3D344@cisco.com>
References: <200303190200.h2J20Pf08821@liming-pc.redback.com>
 <5.1.1.6.2.20030319160416.027f6a60@pop.mcs.anl.gov>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Subject: [pim] Re: Discovery Apps [Was: San Francisco PIM WG meeting minutes]
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Thu, 20 Mar 2003 13:27:18 -0600

Earlier, Bill Nickless wrote:

 > My position:
 >
 > 1. Deprecate data encapsulation in PIM Register,

At 10:42 AM 3/20/2003 -0800, John Meylor wrote:
>what would you do for SM deployments which leverage the shared
>tree?

Let them continue to exist and operate like they are.  The existence of PIM 
Register encapsulation in current deployments does not offend me.  :-)

> > MSDP SA, and all
> >     other such spreading-knowledge-of-active-source messages.
>
>registers in true shared-tree SM mode are moving more than
>information about active sources, they are moving data to
>the distribution tree.   There are significant deployments
>using this well described, long established model.

Moving data in the control plane is inherently a bad idea.  Granted, we've 
been doing it for many years, but it's still a bad idea.

>I agree SSM is a much simplier model for the network,
>- we need ubiquitous v3 support in apps/stacks/l2 devices,

"Deploy IGMPv3" is not a sufficient transition strategy.

We're only setting ourselves up for failure by making IGMPv3 snooping the 
cornerstone of a transition to SSM.  IEEE 802.1 bridges should be left to 
operate on IEEE 802 frames.  IP Multicast, like other IP protocols, must be 
media agnostic and not place requirements on underlying transport.

At least three approaches have been described that could support IP 
Multicast operation on IEEE 802 compliant networks, without requiring the 
IEEE 802.1 devices to understand IGMP/IP or MLD/IP:

  - IEEE 802.1p GARP/GMRP provides a way for IP-speaking hosts to control
    how Ethernet multicast frame distribution is constrained.  (I've
    written about that approach at some length in
    draft-ietf-mboned-iesg-gap-analysis-01.txt .)

  - Radia Perlman and Keyur Patel have outlined a concept for
    using IP unicast encapsulation to traverse subnetworks that
    otherwise aren't configured for IP multicast forwarding.
    (See draft-keyur-pim-host-extensions-00.txt .)  Their draft
    defines receiver behavior, but could be easily extended to
    cover source extension.

  - Dave Thaler, Mohit Talwar, Lorenzo Vicisano, and Dirk Ooms
    have also described an automatic tunneling mechanism for SSM
    deployment.
    http://www.ietf.org/proceedings/01mar/I-D/mboned-auto-multicast-00.txt

>and that it could even be the network service for dynamic
>multisource groups,
>- we some out-of-band dynamic source discovery element

Agreed.  Google works for finding lots of sources, or at least finding 
application servers that can dynamically spread knowledge of active sources.

>nothing is stopping either of those except market pressure

One clear advantage of the Perlman/Patel approach is that it restructures 
market pressures:

  - It bypasses the energy barrier of IGMPv3 deployment at all edges.
    This allows applications to operate correctly, demonstrating demand.

  - When applications operate in the Perlman/Patel mode of unicast
    encapsulation, each individual stream consumes network bandwidth.
    This may motivate customer networks to deploy SSM closer to their
    networks, in order to optimize the use of their bandwidth towards
    the Internet core.

  - It allows core network operators to leverage their existing support
    for SSM-mode routing.

For this reason, I think a transition approach in the spirit of 
Perlman/Patel should be encouraged.

>but we won't get any closer to these SSM based models
>by breaking currently working ASM models.

Again, I'm not advocating that we break existing deployments.  What I am 
advocating is:

  - Making a conscious decision to refine the model of what we are
    trying to support, so that what we support can be scalably deployed.

  - Defining transition mechanisms to enable SSM applications without
    requiring end-to-end IGMPv3 deployment.


===
Bill Nickless    http://www.mcs.anl.gov/people/nickless      +1 630 252 7390
PGP:0E 0F 16 80 C5 B1 69 52 E1 44 1A A5 0E 1B 74 F7     nickless@mcs.anl.gov

_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Thu Mar 20 16:48:12 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA09266
	for <pim-archive@lists.ietf.org>; Thu, 20 Mar 2003 16:48:10 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2KLlbu51546;
	Thu, 20 Mar 2003 13:47:37 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from tyholt.uninett.no (tyholt.uninett.no [158.38.60.10])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2KLkqu51515
	for <pim@catarina.usc.edu>; Thu, 20 Mar 2003 13:46:52 -0800 (PST)
	(envelope-from Stig.Venaas@uninett.no)
Received: from sverresborg.uninett.no (sverresborg.uninett.no [IPv6:2001:700:e000:0:290:27ff:fe50:6bfa])
	by tyholt.uninett.no (8.12.8/8.12.6) with ESMTP id h2KLkngJ001565;
	Thu, 20 Mar 2003 22:46:49 +0100
Received: (from venaas@localhost)
	by sverresborg.uninett.no (8.11.6/8.11.2) id h2KLknN26170;
	Thu, 20 Mar 2003 22:46:49 +0100
X-Authentication-Warning: sverresborg.uninett.no: venaas set sender to Stig.Venaas@uninett.no using -f
From: Stig Venaas <Stig.Venaas@uninett.no>
To: shep@shepfarm.com
Cc: John Meylor <jmeylor@cisco.com>, Bill Nickless <nickless@mcs.anl.gov>,
        John Zwiebel <jzwiebel@procket.com>, lwei@redback.com,
        pim@catarina.usc.edu, mboned@network-services.uoregon.edu
Message-ID: <20030320224649.D26144@sverresborg.uninett.no>
References: <3E7A0B9A.AB3D344@cisco.com> <Pine.LNX.4.21.0303201314300.9646-100000@shepfarm.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.LNX.4.21.0303201314300.9646-100000@shepfarm.com>; from shep@shepfarm.com on Thu, Mar 20, 2003 at 01:23:13PM -0800
Subject: [pim] Re: Discovery Apps [Was: San Francisco PIM WG meeting minutes]
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Thu, 20 Mar 2003 22:46:49 +0100

On Thu, Mar 20, 2003 at 01:23:13PM -0800, shep@shepfarm.com wrote:
> > information about active sources, they are moving data to
> > the distribution tree.   There are signficant deployments
> > using this well described, long established model.
> 
> All intra-domain from what I've seen. Is this true John? Or are you aware
> of any deployments using ASM for anything OTHER THAN source discovery in
> the interdomain space?

Sure there is. Not that much used, but there are cases where say one
university streams something that someone in another university want
to see. And I guess one good example is the IETF multicast sessions (:
Some of us have also done videoconferences with mbone tools between
domains.

Stig
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Thu Mar 20 16:55:28 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA09555
	for <pim-archive@lists.ietf.org>; Thu, 20 Mar 2003 16:55:27 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2KLs5u51707;
	Thu, 20 Mar 2003 13:54:05 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from mcs.anl.gov (cliff.mcs.anl.gov [140.221.9.17])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2KLrju51687
	for <pim@catarina.usc.edu>; Thu, 20 Mar 2003 13:53:45 -0800 (PST)
	(envelope-from nickless@mcs.anl.gov)
Received: from newtwinkle.mcs.anl.gov (vpnclient18104.mcs.anl.gov [140.221.18.104])
	by mcs.anl.gov (8.11.6/8.9.3) with ESMTP id h2KLrAL60654;
	Thu, 20 Mar 2003 15:53:10 -0600
Message-Id: <5.1.1.6.2.20030320155121.02407e38@pop.mcs.anl.gov>
X-Sender: nickless@pop.mcs.anl.gov
X-Mailer: QUALCOMM Windows Eudora Version 5.1.1
To: Stig Venaas <Stig.Venaas@uninett.no>
From: Bill Nickless <nickless@mcs.anl.gov>
Cc: shep@shepfarm.com, John Meylor <jmeylor@cisco.com>,
        Bill Nickless <nickless@mcs.anl.gov>,
        John Zwiebel <jzwiebel@procket.com>, lwei@redback.com,
        pim@catarina.usc.edu, mboned@network-services.uoregon.edu
In-Reply-To: <20030320224649.D26144@sverresborg.uninett.no>
References: <Pine.LNX.4.21.0303201314300.9646-100000@shepfarm.com>
 <3E7A0B9A.AB3D344@cisco.com>
 <Pine.LNX.4.21.0303201314300.9646-100000@shepfarm.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Subject: [pim] Re: Discovery Apps [Was: San Francisco PIM WG meeting minutes]
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Thu, 20 Mar 2003 15:53:08 -0600

At 10:46 PM 3/20/2003 +0100, Stig Venaas wrote:
>Sure there is. Not that much used, but there are cases where say one
>university streams something that someone in another university want
>to see. And I guess one good example is the IETF multicast sessions (:
>Some of us have also done videoconferences with mbone tools between
>domains.

Streaming media applications work just fine when the network spreads the 
knowledge of active sources, to support interested receiver networks 
joining source-rooted distribution trees.

Streaming applications don't require the network to spread the initial 
packets of each stream in the control plane.

===
Bill Nickless    http://www.mcs.anl.gov/people/nickless      +1 630 252 7390
PGP:0E 0F 16 80 C5 B1 69 52 E1 44 1A A5 0E 1B 74 F7     nickless@mcs.anl.gov

_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Thu Mar 20 17:00:32 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA09775
	for <pim-archive@lists.ietf.org>; Thu, 20 Mar 2003 17:00:30 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2KM06u51892;
	Thu, 20 Mar 2003 14:00:06 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from tyholt.uninett.no (tyholt.uninett.no [158.38.60.10])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2KLxJu51856
	for <pim@catarina.usc.edu>; Thu, 20 Mar 2003 13:59:19 -0800 (PST)
	(envelope-from Stig.Venaas@uninett.no)
Received: from sverresborg.uninett.no (sverresborg.uninett.no [IPv6:2001:700:e000:0:290:27ff:fe50:6bfa])
	by tyholt.uninett.no (8.12.8/8.12.6) with ESMTP id h2KLxIgJ003321;
	Thu, 20 Mar 2003 22:59:18 +0100
Received: (from venaas@localhost)
	by sverresborg.uninett.no (8.11.6/8.11.2) id h2KLxHI26181;
	Thu, 20 Mar 2003 22:59:17 +0100
X-Authentication-Warning: sverresborg.uninett.no: venaas set sender to Stig.Venaas@uninett.no using -f
From: Stig Venaas <Stig.Venaas@uninett.no>
To: Bill Nickless <nickless@mcs.anl.gov>
Cc: shep@shepfarm.com, John Meylor <jmeylor@cisco.com>,
        John Zwiebel <jzwiebel@procket.com>, lwei@redback.com,
        pim@catarina.usc.edu, mboned@network-services.uoregon.edu
Message-ID: <20030320225917.A26177@sverresborg.uninett.no>
References: <Pine.LNX.4.21.0303201314300.9646-100000@shepfarm.com> <3E7A0B9A.AB3D344@cisco.com> <Pine.LNX.4.21.0303201314300.9646-100000@shepfarm.com> <20030320224649.D26144@sverresborg.uninett.no> <5.1.1.6.2.20030320155121.02407e38@pop.mcs.anl.gov>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <5.1.1.6.2.20030320155121.02407e38@pop.mcs.anl.gov>; from nickless@mcs.anl.gov on Thu, Mar 20, 2003 at 03:53:08PM -0600
Subject: [pim] Re: Discovery Apps [Was: San Francisco PIM WG meeting minutes]
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Thu, 20 Mar 2003 22:59:17 +0100

On Thu, Mar 20, 2003 at 03:53:08PM -0600, Bill Nickless wrote:
> At 10:46 PM 3/20/2003 +0100, Stig Venaas wrote:
> >Sure there is. Not that much used, but there are cases where say one
> >university streams something that someone in another university want
> >to see. And I guess one good example is the IETF multicast sessions (:
> >Some of us have also done videoconferences with mbone tools between
> >domains.
> 
> Streaming media applications work just fine when the network spreads the 
> knowledge of active sources, to support interested receiver networks 
> joining source-rooted distribution trees.

Agreed, I also dislike the control plane being used for data transport.

Stig
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Thu Mar 20 17:07:26 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA09987
	for <pim-archive@lists.ietf.org>; Thu, 20 Mar 2003 17:07:25 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2KM7Du52067;
	Thu, 20 Mar 2003 14:07:13 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from mcs.anl.gov (cliff.mcs.anl.gov [140.221.9.17])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2KM6ou52049
	for <pim@catarina.usc.edu>; Thu, 20 Mar 2003 14:06:50 -0800 (PST)
	(envelope-from nickless@mcs.anl.gov)
Received: from newtwinkle.mcs.anl.gov (vpnclient18104.mcs.anl.gov [140.221.18.104])
	by mcs.anl.gov (8.11.6/8.9.3) with ESMTP id h2KM5rL89078;
	Thu, 20 Mar 2003 16:05:54 -0600
Message-Id: <5.1.1.6.2.20030320155912.025cd268@pop.mcs.anl.gov>
X-Sender: nickless@pop.mcs.anl.gov
X-Mailer: QUALCOMM Windows Eudora Version 5.1.1
To: <shep@shepfarm.com>
From: Bill Nickless <nickless@mcs.anl.gov>
Cc: Stig Venaas <Stig.Venaas@uninett.no>, John Meylor <jmeylor@cisco.com>,
        Bill Nickless <nickless@mcs.anl.gov>,
        John Zwiebel <jzwiebel@procket.com>, lwei@redback.com,
        pim@catarina.usc.edu, mboned@network-services.uoregon.edu
In-Reply-To: <Pine.LNX.4.21.0303201348320.9646-100000@shepfarm.com>
References: <20030320224649.D26144@sverresborg.uninett.no>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Subject: [pim] Re: Discovery Apps [Was: San Francisco PIM WG meeting minutes]
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Thu, 20 Mar 2003 16:05:33 -0600

At 01:56 PM 3/20/2003 -0800, shep@shepfarm.com wrote:

>Going back to John's concern about the current deployed base of ASM
>intra-domain networks.. Thinking about this somemore, I think here too we
>can separate network-based source discovery (NBSD) from shared-tree
>distribution.
>
>In these enterprise networks, I believe the applications currently use
>NBSD because they can. There really is nothing advantages about it. What
>the DO want and need is shared-tree distribution. In fact, in the most
>recent deployments, they engineer the network to ensure ALL sources stay
>on the shared tree - so they don't bother 'discovering' any sources: they
>just take them all down the shared-tree.

In the context of these deployments....

Can you characterize what the relative value accrues to having multiple 
sources sharing a distribution tree?

More specifically: are there enough unique sources to preclude static 
configuration of multiple source-rooted distribution trees?

What is the "tipping point" between when a shared tree should be used 
versus multiple source-rooted trees?

>So am I far enough off topic yet? ;-)
>
>Greg

That's never stopped me before!  :-P


===
Bill Nickless    http://www.mcs.anl.gov/people/nickless      +1 630 252 7390
PGP:0E 0F 16 80 C5 B1 69 52 E1 44 1A A5 0E 1B 74 F7     nickless@mcs.anl.gov

_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Thu Mar 20 17:08:18 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA10007
	for <pim-archive@lists.ietf.org>; Thu, 20 Mar 2003 17:08:16 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2KM83u52112;
	Thu, 20 Mar 2003 14:08:03 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from tyholt.uninett.no (tyholt.uninett.no [158.38.60.10])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2KM7Su52081
	for <pim@catarina.usc.edu>; Thu, 20 Mar 2003 14:07:28 -0800 (PST)
	(envelope-from Stig.Venaas@uninett.no)
Received: from sverresborg.uninett.no (sverresborg.uninett.no [IPv6:2001:700:e000:0:290:27ff:fe50:6bfa])
	by tyholt.uninett.no (8.12.8/8.12.6) with ESMTP id h2KM7QgJ004042;
	Thu, 20 Mar 2003 23:07:26 +0100
Received: (from venaas@localhost)
	by sverresborg.uninett.no (8.11.6/8.11.2) id h2KM7Qk26193;
	Thu, 20 Mar 2003 23:07:26 +0100
X-Authentication-Warning: sverresborg.uninett.no: venaas set sender to Stig.Venaas@uninett.no using -f
From: Stig Venaas <Stig.Venaas@uninett.no>
To: shep@shepfarm.com
Cc: John Meylor <jmeylor@cisco.com>, Bill Nickless <nickless@mcs.anl.gov>,
        John Zwiebel <jzwiebel@procket.com>, lwei@redback.com,
        pim@catarina.usc.edu, mboned@network-services.uoregon.edu
Message-ID: <20030320230726.B26177@sverresborg.uninett.no>
References: <20030320224649.D26144@sverresborg.uninett.no> <Pine.LNX.4.21.0303201348320.9646-100000@shepfarm.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.LNX.4.21.0303201348320.9646-100000@shepfarm.com>; from shep@shepfarm.com on Thu, Mar 20, 2003 at 01:56:10PM -0800
Subject: [pim] Re: Discovery Apps [Was: San Francisco PIM WG meeting minutes]
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Thu, 20 Mar 2003 23:07:26 +0100

On Thu, Mar 20, 2003 at 01:56:10PM -0800, shep@shepfarm.com wrote:
> On Thu, 20 Mar 2003, Stig Venaas wrote:
> > Sure there is. Not that much used, but there are cases where say one
> > university streams something that someone in another university want
> > to see. And I guess one good example is the IETF multicast sessions (:
> > Some of us have also done videoconferences with mbone tools between
> > domains.
> 
> This IS a PRIME example of using ASM for network-based source discovery
> ONLY. There are no inter-domain shared trees.

Oh okay, I see what you're asking for now. Ignore my mail please.

Stig
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Thu Mar 20 17:28:27 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA10411
	for <pim-archive@lists.ietf.org>; Thu, 20 Mar 2003 17:28:26 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2KMSCu52601;
	Thu, 20 Mar 2003 14:28:12 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from dmz1.procket.com (dmz1.procket.com [65.174.124.36])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2KMRju52577
	for <pim@catarina.usc.edu>; Thu, 20 Mar 2003 14:27:45 -0800 (PST)
	(envelope-from John.Zwiebel@procket.com)
Received: from miata.procket.com (zeus-d-1.procket.com [65.174.124.60])
	by dmz1.procket.com (Postfix) with ESMTP
	id D682023C63; Thu, 20 Mar 2003 14:49:15 -0800 (PST)
Received: from exchangefe1.na.procket.com (exchangefe1.na.procket.com [10.1.7.251])
	by miata.procket.com (8.12.1/8.12.1) with ESMTP id h2KMReYB017708;
	Thu, 20 Mar 2003 14:27:40 -0800 (PST)
Received: from procket.com ([10.1.1.1]) by exchangefe1.na.procket.com with Microsoft SMTPSVC(5.0.2195.5329);
	 Thu, 20 Mar 2003 14:27:40 -0800
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v551)
Cc: John Zwiebel <jzwiebel@procket.com>, Bill Nickless <nickless@mcs.anl.gov>,
        Stig Venaas <Stig.Venaas@uninett.no>, John Meylor <jmeylor@cisco.com>,
        lwei@redback.com, pim@catarina.usc.edu,
        mboned@network-services.uoregon.edu
To: <shep@shepfarm.com>
From: John Zwiebel <jzwiebel@procket.com>
In-Reply-To: <Pine.LNX.4.21.0303201408310.9646-100000@shepfarm.com>
Message-Id: <2DFA58BF-5B23-11D7-A90A-00039307689E@procket.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.551)
X-OriginalArrivalTime: 20 Mar 2003 22:27:40.0065 (UTC) FILETIME=[EABE8110:01C2EF2F]
Subject: [pim] Re: Discovery Apps [Was: San Francisco PIM WG meeting minutes]
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Thu, 20 Mar 2003 14:27:48 -0800
Content-Transfer-Encoding: 7bit


On Thursday, March 20, 2003, at 02:09 PM, <shep@shepfarm.com> wrote:

>
> S,G state explosion for many-to-many applications is the trigger.

There is also the question of the diameter of the multicast group.
IMHO, interdomain multicast is not going to ever get to the size
that a shared-tree will be useful.  But since one should never
say never (again), the need for share-trees in the interdomain
space is not going to happen for years.

In the enterprise space the need is already here.

I believe the applications requiring the shared-trees are more
"deterministic" in that the location of the sources and receivers
is well known and under fairly strict control, which limits the
need for shared-trees to the enterprise.

Interdomain traffic is much more ad hoc in that there is little
control over the location of the sources and receivers.  At the
same time, little need for coordination between sources (which
may exist in the enterprise space.)

but we are drifting quite afar from the need to figure out how to
coordinate between sources and receivers on when the data stream
will begin.

>
> Greg
>
> On Thu, 20 Mar 2003, Bill Nickless wrote:
>
>> At 01:56 PM 3/20/2003 -0800, shep@shepfarm.com wrote:
>>
>>> Going back to John's concern about the current deployed base of ASM
>>> intra-domain networks.. Thinking about this somemore, I think here 
>>> too we
>>> can separate network-based source discovery (NBSD) from shared-tree
>>> distribution.
>>>
>>> In these enterprise networks, I believe the applications currently 
>>> use
>>> NBSD because they can. There really is nothing advantages about it. 
>>> What
>>> the DO want and need is shared-tree distribution. In fact, in the 
>>> most
>>> recent deployments, they engineer the network to ensure ALL sources 
>>> stay
>>> on the shared tree - so they don't bother 'discovering' any sources: 
>>> they
>>> just take them all down the shared-tree.
>>
>> In the context of these deployments....
>>
>> Can you characterize what the relative value accrues to having 
>> multiple
>> sources sharing a distribution tree?
>>
>> More specifically: are there enough unique sources to preclude static
>> configuration of multiple source-rooted distribution trees?
>>
>> What is the "tipping point" between when a shared tree should be used
>> versus multiple source-rooted trees?
>>
>>> So am I far enough off topic yet? ;-)
>>>
>>> Greg
>>
>> That's never stopped me before!  :-P
>>
>>
>> ===
>> Bill Nickless    http://www.mcs.anl.gov/people/nickless      +1 630 
>> 252 7390
>> PGP:0E 0F 16 80 C5 B1 69 52 E1 44 1A A5 0E 1B 74 F7     
>> nickless@mcs.anl.gov
>>
>

_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Thu Mar 20 20:32:12 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA17713
	for <pim-archive@lists.ietf.org>; Thu, 20 Mar 2003 20:32:11 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2L1Uru56077;
	Thu, 20 Mar 2003 17:30:53 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from shepfarm.com (IDENT:root@shepfarm.com [198.58.3.10])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2KM9Lu52176
	for <pim@catarina.usc.edu>; Thu, 20 Mar 2003 14:09:21 -0800 (PST)
	(envelope-from shep@shepfarm.com)
Received: from shepfarm.com (IDENT:shep@shepfarm.com [198.58.3.10])
	by shepfarm.com (8.11.3/8.11.3) with ESMTP id h2KM92t09883;
	Thu, 20 Mar 2003 14:09:02 -0800
From: <shep@shepfarm.com>
To: Bill Nickless <nickless@mcs.anl.gov>
cc: Stig Venaas <Stig.Venaas@uninett.no>, John Meylor <jmeylor@cisco.com>,
        John Zwiebel <jzwiebel@procket.com>, lwei@redback.com,
        pim@catarina.usc.edu, mboned@network-services.uoregon.edu
In-Reply-To: <5.1.1.6.2.20030320155912.025cd268@pop.mcs.anl.gov>
Message-ID: <Pine.LNX.4.21.0303201408310.9646-100000@shepfarm.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [pim] Re: Discovery Apps [Was: San Francisco PIM WG meeting minutes]
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Thu, 20 Mar 2003 14:09:02 -0800 (PST)


S,G state explosion for many-to-many applications is the trigger.

Greg

On Thu, 20 Mar 2003, Bill Nickless wrote:

> At 01:56 PM 3/20/2003 -0800, shep@shepfarm.com wrote:
> 
> >Going back to John's concern about the current deployed base of ASM
> >intra-domain networks.. Thinking about this somemore, I think here too we
> >can separate network-based source discovery (NBSD) from shared-tree
> >distribution.
> >
> >In these enterprise networks, I believe the applications currently use
> >NBSD because they can. There really is nothing advantages about it. What
> >the DO want and need is shared-tree distribution. In fact, in the most
> >recent deployments, they engineer the network to ensure ALL sources stay
> >on the shared tree - so they don't bother 'discovering' any sources: they
> >just take them all down the shared-tree.
> 
> In the context of these deployments....
> 
> Can you characterize what the relative value accrues to having multiple 
> sources sharing a distribution tree?
> 
> More specifically: are there enough unique sources to preclude static 
> configuration of multiple source-rooted distribution trees?
> 
> What is the "tipping point" between when a shared tree should be used 
> versus multiple source-rooted trees?
> 
> >So am I far enough off topic yet? ;-)
> >
> >Greg
> 
> That's never stopped me before!  :-P
> 
> 
> ===
> Bill Nickless    http://www.mcs.anl.gov/people/nickless      +1 630 252 7390
> PGP:0E 0F 16 80 C5 B1 69 52 E1 44 1A A5 0E 1B 74 F7     nickless@mcs.anl.gov
> 

_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Thu Mar 20 20:32:21 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA17731
	for <pim-archive@lists.ietf.org>; Thu, 20 Mar 2003 20:32:20 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2L1Ueu56045;
	Thu, 20 Mar 2003 17:30:41 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from shepfarm.com (IDENT:root@shepfarm.com [198.58.3.10])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2KLNfu51078
	for <pim@catarina.usc.edu>; Thu, 20 Mar 2003 13:23:41 -0800 (PST)
	(envelope-from shep@shepfarm.com)
Received: from shepfarm.com (IDENT:shep@shepfarm.com [198.58.3.10])
	by shepfarm.com (8.11.3/8.11.3) with ESMTP id h2KLNDt09670;
	Thu, 20 Mar 2003 13:23:13 -0800
From: <shep@shepfarm.com>
To: John Meylor <jmeylor@cisco.com>
cc: Bill Nickless <nickless@mcs.anl.gov>, John Zwiebel <jzwiebel@procket.com>,
        lwei@redback.com, pim@catarina.usc.edu,
        mboned@network-services.uoregon.edu
In-Reply-To: <3E7A0B9A.AB3D344@cisco.com>
Message-ID: <Pine.LNX.4.21.0303201314300.9646-100000@shepfarm.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [pim] Re: Discovery Apps [Was: San Francisco PIM WG meeting minutes]
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Thu, 20 Mar 2003 13:23:13 -0800 (PST)


> information about active sources, they are moving data to
> the distribution tree.   There are signficant deployments
> using this well described, long established model.

All intra-domain from what I've seen. Is this true John? Or are you aware
of any deployments using ASM for anything OTHER THAN source discovery in
the interdomain space?

I think we need to disassociate ASM as it applies to
shared-tree distribution from network source-discovery. The intra-domain
deployments use both, but interdomain only uses the source-discovery
element of ASM. True? 
 
> I agree SSM is a much simplier model for the network, 
> - we need ubiquitous v3 support in apps/stacks/l2 devices,
> 
> and that it could even be the network service for dynamic 
> multisource groups, 
> - we some out-of-band dynamic source discovery element

There is nothing special about doing source-discovery in the network that
can't be accomplished at the application layer.

> nothing is stopping either of those except market pressure
> 
> but we won't get any closer to these SSM based models
> by breaking currently working ASM models.

Agreed. 

BUT, it was evident too me this week that over 90% of the efforts in the
mcast wg's I attended, was spent dealing with network-based source
discovery. The law of diminishing returns must apply sometime.

Greg

>  john
> 
> > 
> > 2. Apply lessons <A> through <D> above to multicast applications.
> >     Specifically, require applications to either (i) use SSM or (ii)
> >     maintain a minimum frequency of sourced packets.
> > 
> > ===
> > Bill Nickless    http://www.mcs.anl.gov/people/nickless      +1 630 252 7390
> > PGP:0E 0F 16 80 C5 B1 69 52 E1 44 1A A5 0E 1B 74 F7     nickless@mcs.anl.gov
> 

_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Thu Mar 20 20:32:24 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA17744
	for <pim-archive@lists.ietf.org>; Thu, 20 Mar 2003 20:32:22 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2L1Unu56061;
	Thu, 20 Mar 2003 17:30:49 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from shepfarm.com (IDENT:root@shepfarm.com [198.58.3.10])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2KLuXu51779
	for <pim@catarina.usc.edu>; Thu, 20 Mar 2003 13:56:33 -0800 (PST)
	(envelope-from shep@shepfarm.com)
Received: from shepfarm.com (IDENT:shep@shepfarm.com [198.58.3.10])
	by shepfarm.com (8.11.3/8.11.3) with ESMTP id h2KLuAt09806;
	Thu, 20 Mar 2003 13:56:10 -0800
From: <shep@shepfarm.com>
To: Stig Venaas <Stig.Venaas@uninett.no>
cc: John Meylor <jmeylor@cisco.com>, Bill Nickless <nickless@mcs.anl.gov>,
        John Zwiebel <jzwiebel@procket.com>, lwei@redback.com,
        pim@catarina.usc.edu, mboned@network-services.uoregon.edu
In-Reply-To: <20030320224649.D26144@sverresborg.uninett.no>
Message-ID: <Pine.LNX.4.21.0303201348320.9646-100000@shepfarm.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [pim] Re: Discovery Apps [Was: San Francisco PIM WG meeting minutes]
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Thu, 20 Mar 2003 13:56:10 -0800 (PST)

On Thu, 20 Mar 2003, Stig Venaas wrote:

> On Thu, Mar 20, 2003 at 01:23:13PM -0800, shep@shepfarm.com wrote:
> > > information about active sources, they are moving data to
> > > the distribution tree.   There are signficant deployments
> > > using this well described, long established model.
> > 
> > All intra-domain from what I've seen. Is this true John? Or are you aware
> > of any deployments using ASM for anything OTHER THAN source discovery in
> > the interdomain space?
> 
> Sure there is. Not that much used, but there are cases where say one
> university streams something that someone in another university want
> to see. And I guess one good example is the IETF multicast sessions (:
> Some of us have also done videoconferences with mbone tools between
> domains.

This IS a PRIME example of using ASM for network-based source discovery
ONLY. There are no inter-domain shared trees.

Going back to John's concern about the current deployed base of ASM
intra-domain networks.. Thinking about this somemore, I think here too we
can separate network-based source discovery (NBSD) from shared-tree
distribution.

In these enterprise networks, I believe the applications currently use
NBSD because they can. There really is nothing advantages about it. What
the DO want and need is shared-tree distribution. In fact, in the most
recent deployments, they engineer the network to ensure ALL sources stay
on the shared tree - so they don't bother 'discovering' any sources: they
just take them all down the shared-tree.
 
So am I far enough off topic yet? ;-)

Greg

> Stig
> 

_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Fri Mar 21 23:07:54 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA06842
	for <pim-archive@lists.ietf.org>; Fri, 21 Mar 2003 23:07:52 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2M46nu86448;
	Fri, 21 Mar 2003 20:06:49 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from shuttle.wide.toshiba.co.jp (shuttle.wide.toshiba.co.jp [202.249.10.124])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2M43Zu86387
	for <pim@catarina.usc.edu>; Fri, 21 Mar 2003 20:03:35 -0800 (PST)
	(envelope-from jinmei@isl.rdc.toshiba.co.jp)
Received: from localhost (unknown [3ffe:501:4819:2000:95e4:ac11:b66a:f62f])
	by shuttle.wide.toshiba.co.jp (Postfix) with ESMTP
	id BC49315210; Sat, 22 Mar 2003 13:04:01 +0900 (JST)
Message-ID: <y7vk7es2gjs.wl@ocean.jinmei.org>
From: JINMEI Tatuya / =?ISO-2022-JP?B?GyRCP0BMQEMjOkgbKEI=?= <jinmei@isl.rdc.toshiba.co.jp>
To: Tom Pusateri <pusateri@juniper.net>, lwei@redback.com,
        Chris.Cormier@motorola.com, pim@catarina.usc.edu
Subject: Re: [pim] San Francisco PIM WG meeting minutes 
In-Reply-To: <x74r5yx01n.wl%suz@crl.hitachi.co.jp>
References: <lwei@redback.com>
	 <200303192154.h2JLsB411529@liming-pc.redback.com>
	 <200303192204.h2JM4bS40873@merlot.juniper.net>
	 <x74r5yx01n.wl%suz@crl.hitachi.co.jp>
User-Agent: Wanderlust/2.6.1 (Upside Down) Emacs/21.2 Mule/5.0 (SAKAKI)
Organization: Research & Development Center, Toshiba Corp., Kawasaki, Japan.
MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya")
Content-Type: text/plain; charset=US-ASCII
X-Dispatcher: imput version 20000228(IM140)
Lines: 16
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Sat, 22 Mar 2003 13:04:07 +0900

>>>>> On Thu, 20 Mar 2003 11:08:04 +0900, 
>>>>> SUZUKI Shinsuke <suz@crl.hitachi.co.jp> said:

>> Actually, anyone who presented slides yesterday needs to send them
>> to me for submission to the minutes. Please do so right away.

> Here's jinmei-san's presentation (converted from magicpoint to html).
> 	http://www.kame.net/~suz/pim-upstream-detection/index.html

A pdf version is also available at:
http://www.jinmei.org/mgp/ietf56-pim/pim-upstream-detection.pdf

					JINMEI, Tatuya
					Communication Platform Lab.
					Corporate R&D Center, Toshiba Corp.
					jinmei@isl.rdc.toshiba.co.jp
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Sat Mar 22 07:31:20 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA25281
	for <pim-archive@lists.ietf.org>; Sat, 22 Mar 2003 07:31:19 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2MCTpu98489;
	Sat, 22 Mar 2003 04:29:51 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from shuttle.wide.toshiba.co.jp (shuttle.wide.toshiba.co.jp [202.249.10.124])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2MCShu98458
	for <pim@catarina.usc.edu>; Sat, 22 Mar 2003 04:28:43 -0800 (PST)
	(envelope-from jinmei@isl.rdc.toshiba.co.jp)
Received: from localhost (unknown [3ffe:501:4819:2000:95e4:ac11:b66a:f62f])
	by shuttle.wide.toshiba.co.jp (Postfix) with ESMTP
	id CEF9515210; Sat, 22 Mar 2003 21:29:11 +0900 (JST)
Message-ID: <y7vadfnwpno.wl@ocean.jinmei.org>
From: JINMEI Tatuya / =?ISO-2022-JP?B?GyRCP0BMQEMjOkgbKEI=?= <jinmei@isl.rdc.toshiba.co.jp>
To: John Zwiebel <jzwiebel@procket.com>
Cc: lwei@redback.com, pim@catarina.usc.edu
Subject: Re: [pim] San Francisco PIM WG meeting minutes
In-Reply-To: <BAE51AE0-5A54-11D7-B534-0003936A6062@procket.com>
References: <200303190200.h2J20Pf08821@liming-pc.redback.com>
	 <BAE51AE0-5A54-11D7-B534-0003936A6062@procket.com>
User-Agent: Wanderlust/2.6.1 (Upside Down) Emacs/21.2 Mule/5.0 (SAKAKI)
Organization: Research & Development Center, Toshiba Corp., Kawasaki, Japan.
MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya")
Content-Type: text/plain; charset=US-ASCII
X-Dispatcher: imput version 20000228(IM140)
Lines: 42
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Sat, 22 Mar 2003 21:29:15 +0900

>>>>> On Wed, 19 Mar 2003 13:49:59 -0800, 
>>>>> John Zwiebel <jzwiebel@procket.com> said:

>> 4. Jimei presented proposals on correctly matching a global address
>> to a link-local IPv6 address for PIM rpf checks.

> Someone suggested that this problem is not unique to PIM.

Correct (as I admitted at the presentation), but at the same time,
"protocol independence", a unique characteristic of PIM, is one the
major sources of the issue.

> (Dino
> had a rant about link-local addresses being a "bug" -- I agree with
> him that they are a pain).

I admit link-local addresses introduce a pain, but it also provides
benefits.  In fact, the IETF is introducing link-local addresses even
in the IPv4 world by the zeroconf wg.  There may still be people who
just think the notion of link-local is a bug and should be killed, but
the fact shows it is at least controversial whether we should have
link-local or not.  So, I'd be glad if we could concentrate on the
issue based on the fact we have link-local addresses, rather than
discuss the validity of having link-local here.

> It may be that this problem should be
> solved by one of the IPv6 working groups.

I don't mind to discuss this somewhere else if it is more appropriate
place than here.  However, as some other people have pointed out in
this list, the issued addressed in the draft is not necessarily
specific to IPv6 (we the authors realized this point, but mainly
focused on the IPv6 issue in the draft since this is the most
practical one we've seen).

I still believe the pim wg is the best place, and would really
appreciate if the wg allows us to discuss the issue here.

					JINMEI, Tatuya
					Communication Platform Lab.
					Corporate R&D Center, Toshiba Corp.
					jinmei@isl.rdc.toshiba.co.jp
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Sat Mar 22 07:38:17 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA25333
	for <pim-archive@lists.ietf.org>; Sat, 22 Mar 2003 07:38:16 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2MCc5u98679;
	Sat, 22 Mar 2003 04:38:05 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from shuttle.wide.toshiba.co.jp (shuttle.wide.toshiba.co.jp [202.249.10.124])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2MCbWu98649
	for <pim@catarina.usc.edu>; Sat, 22 Mar 2003 04:37:32 -0800 (PST)
	(envelope-from jinmei@isl.rdc.toshiba.co.jp)
Received: from localhost (unknown [3ffe:501:4819:2000:95e4:ac11:b66a:f62f])
	by shuttle.wide.toshiba.co.jp (Postfix) with ESMTP
	id 7F65615210; Sat, 22 Mar 2003 21:38:01 +0900 (JST)
Message-ID: <y7v8yv7wp8y.wl@ocean.jinmei.org>
From: JINMEI Tatuya / =?ISO-2022-JP?B?GyRCP0BMQEMjOkgbKEI=?= <jinmei@isl.rdc.toshiba.co.jp>
To: lwei@redback.com
Cc: pim@catarina.usc.edu
Subject: Re: [pim] San Francisco PIM WG meeting minutes
In-Reply-To: <200303192213.h2JMDwq11638@liming-pc.redback.com>
References: <BAE51AE0-5A54-11D7-B534-0003936A6062@procket.com>
	 <200303192213.h2JMDwq11638@liming-pc.redback.com>
User-Agent: Wanderlust/2.6.1 (Upside Down) Emacs/21.2 Mule/5.0 (SAKAKI)
Organization: Research & Development Center, Toshiba Corp., Kawasaki, Japan.
MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya")
Content-Type: text/plain; charset=US-ASCII
X-Dispatcher: imput version 20000228(IM140)
Lines: 29
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Sat, 22 Mar 2003 21:38:05 +0900

>>>>> On Wed, 19 Mar 2003 14:13:58 -0800 (PST), 
>>>>> Liming Wei <lwei@redback.com> said:

> On the second issue about multiple addresses, I spoke with some one
> working on OSPF (let me not mention the name here) afterwards. I do
> think there is significant benefit in using a PIM option to convey all
> addresses. Relying on a collection of other mechaisms (like neihgbor
> discovery etc) does not appeal to me to be good design choices. I
> think Mark and company made a good choice on using a new PIM option.

Thanks, (as I said at the presentation) though we do not stick to a
particular solution, especially the new hello option, at this moment,
there seems to be some interest in pursuing the idea.

So, what should we do on this?  We're willing to either

- to revise our draft with more details of the hello option, or
- to contribute some text on this issue and the option to the main PIM
  specification
(or whatever else we can)

Would you chairs suggest a proper course of action, please?

Thanks,

					JINMEI, Tatuya
					Communication Platform Lab.
					Corporate R&D Center, Toshiba Corp.
					jinmei@isl.rdc.toshiba.co.jp
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Sat Mar 22 11:59:27 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA29849
	for <pim-archive@lists.ietf.org>; Sat, 22 Mar 2003 11:59:27 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2MGwgu03592;
	Sat, 22 Mar 2003 08:58:42 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from dino-pc.procket.com (zeus-ext.procket.com [65.174.124.29])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2MGvcu03556
	for <pim@catarina.usc.edu>; Sat, 22 Mar 2003 08:57:38 -0800 (PST)
	(envelope-from dino@procket.com)
Received: (from dino@localhost)
	by dino-pc.procket.com (8.9.3/8.9.3) id JAA21988;
	Sat, 22 Mar 2003 09:03:01 -0800
Message-Id: <200303221703.JAA21988@dino-pc.procket.com>
X-Authentication-Warning: dino-pc.procket.com: dino set sender to dino@dino-pc.procket.com using -f
From: Dino Farinacci <dino@procket.com>
To: jinmei@isl.rdc.toshiba.co.jp
CC: jzwiebel@procket.com, lwei@redback.com, pim@catarina.usc.edu
In-reply-to: <y7vadfnwpno.wl@ocean.jinmei.org> (message from JINMEI Tatuya / 
	$B?@L@C#:H(B on Sat, 22 Mar 2003 21:29:15 +0900)
Subject: Re: [pim] San Francisco PIM WG meeting minutes
References: <200303190200.h2J20Pf08821@liming-pc.redback.com>
	 <BAE51AE0-5A54-11D7-B534-0003936A6062@procket.com> <y7vadfnwpno.wl@ocean.jinmei.org>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Sat, 22 Mar 2003 09:03:01 -0800

>> I admit link-local addresses introduce a pain, but it also provides
>> benefits.  In fact, the IETF is introducing link-local addresses even
>> in the IPv4 world by the zeroconf wg.  There may still be people who
>> just think the notion of link-local is a bug and should be killed, but
>> the fact shows it is at least controversial whether we should have
>> link-local or not.  So, I'd be glad if we could concentrate on the
>> issue based on the fact we have link-local addresses, rather than
>> discuss the validity of having link-local here.

    Can you explain what the real benefits are of link-local addresses.

Dino
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Sat Mar 22 16:29:11 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA05434
	for <pim-archive@lists.ietf.org>; Sat, 22 Mar 2003 16:29:11 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2MLShu08534;
	Sat, 22 Mar 2003 13:28:43 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from prattle.redback.com (hiddenuser@prattle.redback.com [155.53.12.9])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2MLRwu08507
	for <pim@catarina.usc.edu>; Sat, 22 Mar 2003 13:27:58 -0800 (PST)
	(envelope-from lwei@redback.com)
Received: from liming-pc.redback.com (liming-pc.redback.com [155.53.36.120])
	by prattle.redback.com (Postfix) with ESMTP
	id EDD102C63C5; Sat, 22 Mar 2003 13:27:57 -0800 (PST)
Received: (from lwei@localhost)
	by liming-pc.redback.com (8.11.3nb1/8.8.8/null redback bsdclient) id h2MLRvK24849;
	Sat, 22 Mar 2003 13:27:57 -0800 (PST)
Message-Id: <200303222127.h2MLRvK24849@liming-pc.redback.com>
From: Liming Wei <lwei@redback.com>
To: jinmei@isl.rdc.toshiba.co.jp
Cc: pim@catarina.usc.edu
In-reply-to: <y7v8yv7wp8y.wl@ocean.jinmei.org> (message from JINMEI Tatuya /
	=?ISO-2022-JP?B?GyRCP0BMQEMjOkgbKEI=?= on Sat, 22 Mar 2003 21:38:05
	+0900)
Subject: Re: [pim] San Francisco PIM WG meeting minutes
Reply-To: lwei@redback.com
References: <BAE51AE0-5A54-11D7-B534-0003936A6062@procket.com>
	 <200303192213.h2JMDwq11638@liming-pc.redback.com> <y7v8yv7wp8y.wl@ocean.jinmei.org>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Sat, 22 Mar 2003 13:27:57 -0800 (PST)

Jinmei,

Here is what I think:

1) There exists multple addresses on an interface in IPv6;
2) We need to specify how PIM works under such scenarios.

Else, it will be a mess in the field when it comes to
interoperability. In short, a solution on 2) should be discussed
in the PIM WG.

About the disadvantages (or benefits) of multiple addresses in IPv6,
it is interesting (boring after a couple of messages in thread) to be
informed of the activities elsewhere, but lengthy discussions here
will not change the outcome in IPv6's design, or zeroconf's design.

Can you coordinate with Mark et al on this issue ?

 Thanks,
-Liming

P.s. Somehow all your emails ended up in my spam folder (which I clean
up once every 2 weeks). I guess it is due to the following in your
email header:

Received: from localhost (unknown [3ffe:501:4819:2000:95e4:ac11:b66a:f62f])
	by shuttle.wide.toshiba.co.jp (Postfix) with ESMTP
	id 7F65615210; Sat, 22 Mar 2003 21:38:01 +0900 (JST)


X-Coding-System: undecided-unix
Mail-from: From jinmei@isl.rdc.toshiba.co.jp  Sat Mar 22 04:38:13 2003
Return-Path: <jinmei@isl.rdc.toshiba.co.jp>
Delivered-To: lwei@redback.com
Received: from pop.redback.com [155.53.12.56]
	by localhost with POP3 (fetchmail-5.9.13)
	for lwei@localhost (single-drop); Sat, 22 Mar 2003 04:38:13 -0800 (PST)
Received: from shuttle.wide.toshiba.co.jp (shuttle.wide.toshiba.co.jp [202.249.10.124])
	by prattle.redback.com (Postfix) with ESMTP id 32CAD21BB85
	for <lwei@redback.com>; Sat, 22 Mar 2003 04:37:32 -0800 (PST)
Received: from localhost (unknown [3ffe:501:4819:2000:95e4:ac11:b66a:f62f])
	by shuttle.wide.toshiba.co.jp (Postfix) with ESMTP
	id 7F65615210; Sat, 22 Mar 2003 21:38:01 +0900 (JST)
Date: Sat, 22 Mar 2003 21:38:05 +0900
Message-ID: <y7v8yv7wp8y.wl@ocean.jinmei.org>
From: JINMEI Tatuya / =?ISO-2022-JP?B?GyRCP0BMQEMjOkgbKEI=?= <jinmei@isl.rdc.toshiba.co.jp>
To: lwei@redback.com
Cc: pim@catarina.usc.edu
Subject: Re: [pim] San Francisco PIM WG meeting minutes
In-Reply-To: <200303192213.h2JMDwq11638@liming-pc.redback.com>
References: <BAE51AE0-5A54-11D7-B534-0003936A6062@procket.com>
	 <200303192213.h2JMDwq11638@liming-pc.redback.com>
User-Agent: Wanderlust/2.6.1 (Upside Down) Emacs/21.2 Mule/5.0 (SAKAKI)
Organization: Research & Development Center, Toshiba Corp., Kawasaki, Japan.
MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya")
Content-Type: text/plain; charset=US-ASCII
X-Dispatcher: imput version 20000228(IM140)
Lines: 29

>>>>> On Wed, 19 Mar 2003 14:13:58 -0800 (PST), 
>>>>> Liming Wei <lwei@redback.com> said:

> On the second issue about multiple addresses, I spoke with some one
> working on OSPF (let me not mention the name here) afterwards. I do
> think there is significant benefit in using a PIM option to convey all
> addresses. Relying on a collection of other mechaisms (like neihgbor
> discovery etc) does not appeal to me to be good design choices. I
> think Mark and company made a good choice on using a new PIM option.

Thanks, (as I said at the presentation) though we do not stick to a
particular solution, especially the new hello option, at this moment,
there seems to be some interest in pursuing the idea.

So, what should we do on this?  We're willing to either

- to revise our draft with more details of the hello option, or
- to contribute some text on this issue and the option to the main PIM
  specification
(or whatever else we can)

Would you chairs suggest a proper course of action, please?

Thanks,

					JINMEI, Tatuya
					Communication Platform Lab.
					Corporate R&D Center, Toshiba Corp.
					jinmei@isl.rdc.toshiba.co.jp

_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Sat Mar 22 21:21:11 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA11026
	for <pim-archive@lists.ietf.org>; Sat, 22 Mar 2003 21:21:09 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2N2Kbu14007;
	Sat, 22 Mar 2003 18:20:37 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from mcs.anl.gov (cliff.mcs.anl.gov [140.221.9.17])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2N2Jpu13974
	for <pim@catarina.usc.edu>; Sat, 22 Mar 2003 18:19:51 -0800 (PST)
	(envelope-from nickless@mcs.anl.gov)
Received: from newtwinkle.mcs.anl.gov (vpnclient18108.mcs.anl.gov [140.221.18.108])
	by mcs.anl.gov (8.11.6/8.9.3) with ESMTP id h2N2Jg429136;
	Sat, 22 Mar 2003 20:19:42 -0600
Message-Id: <5.1.1.6.2.20030322200815.01e5cc38@pop.mcs.anl.gov>
X-Sender: nickless@pop.mcs.anl.gov
X-Mailer: QUALCOMM Windows Eudora Version 5.1.1
To: lwei@redback.com
From: Bill Nickless <nickless@mcs.anl.gov>
Subject: Re: [pim] San Francisco PIM WG meeting minutes
Cc: pim@catarina.usc.edu
In-Reply-To: <200303222127.h2MLRvK24849@liming-pc.redback.com>
References: <y7v8yv7wp8y.wl@ocean.jinmei.org>
 <BAE51AE0-5A54-11D7-B534-0003936A6062@procket.com>
 <200303192213.h2JMDwq11638@liming-pc.redback.com>
 <y7v8yv7wp8y.wl@ocean.jinmei.org>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Sat, 22 Mar 2003 20:19:39 -0600

At 01:27 PM 3/22/2003 -0800, Liming Wei wrote:

>1) There exists multiple addresses on an interface in IPv6;
>2) We need to specify how PIM works under such scenarios.
>
>Else, it will be a mess in the field when it comes to
>interoperability. In short, a solution on 2) should be discussed
>in the PIM WG.

There are plenty of situations where interfaces are associated with 
multiple IPv4 addresses as well.  Some PIM implementations don't take this 
into account.  In my experience, this has led to operational deployment 
confusion and difficulty.

One specific example involves HSRP/VRRP.  Unicast reachability is satisfied 
because the "active" HSRP/VRRP speaker responds to IPv4 ARP for the virtual 
address.  However, multicast reachability is broken because the "active" 
HSRP/VRRP speaker fails to respond to IPv4 PIM Joins addressed to the 
virtual address.

The result is a situation where unicast routing works just fine, but 
multicast routing is broken.  Such incidents contribute to the perception 
that multicast routing is more fragile or somehow less reliable than 
unicast routing.

===
Bill Nickless    http://www.mcs.anl.gov/people/nickless      +1 630 252 7390
PGP:0E 0F 16 80 C5 B1 69 52 E1 44 1A A5 0E 1B 74 F7     nickless@mcs.anl.gov

_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar 24 15:22:19 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA02604
	for <pim-archive@lists.ietf.org>; Mon, 24 Mar 2003 15:22:18 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2OKGbu63889;
	Mon, 24 Mar 2003 12:16:37 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from prattle.redback.com (hiddenuser@prattle.redback.com [155.53.12.9])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2OKDKu63816
	for <pim@catarina.usc.edu>; Mon, 24 Mar 2003 12:13:20 -0800 (PST)
	(envelope-from lwei@redback.com)
Received: from liming-pc.redback.com (liming-pc.redback.com [155.53.36.120])
	by prattle.redback.com (Postfix) with ESMTP
	id 6B62C590560; Mon, 24 Mar 2003 12:13:18 -0800 (PST)
Received: (from lwei@localhost)
	by liming-pc.redback.com (8.11.3nb1/8.8.8/null redback bsdclient) id h2OKDIT28218;
	Mon, 24 Mar 2003 12:13:18 -0800 (PST)
Message-Id: <200303242013.h2OKDIT28218@liming-pc.redback.com>
From: Liming Wei <lwei@redback.com>
To: nickless@mcs.anl.gov
Cc: pim@catarina.usc.edu
In-reply-to: <5.1.1.6.2.20030322200815.01e5cc38@pop.mcs.anl.gov> (message from
	Bill Nickless on Sat, 22 Mar 2003 20:19:39 -0600)
Subject: Re: [pim] San Francisco PIM WG meeting minutes
Reply-To: lwei@redback.com
References: <y7v8yv7wp8y.wl@ocean.jinmei.org>
 <BAE51AE0-5A54-11D7-B534-0003936A6062@procket.com>
 <200303192213.h2JMDwq11638@liming-pc.redback.com>
 <y7v8yv7wp8y.wl@ocean.jinmei.org> <5.1.1.6.2.20030322200815.01e5cc38@pop.mcs.anl.gov>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Mon, 24 Mar 2003 12:13:18 -0800 (PST)

Bill,

I actually am thinking about cases invlving multiple addresses in
IPv4 envnronment too :-).

The HSRP or VRRP case sounds unusual. Why does a PIM joiner/pruner
have to rely on a "static default route" type of unicast routing ?

An implementation today can "solve" this problem by special-knob'ing
PIM to run from the VRRP active router only, and only "on" the virtual
address in PIM messages (as opposed to the physical addresses). Of
course, this is not as general/simple (robust?) as learning all
addresses via a PIM hello option.

-Liming
--------
> From: Bill Nickless <nickless@mcs.anl.gov>
> 
> At 01:27 PM 3/22/2003 -0800, Liming Wei wrote:
> 
> >1) There exists multiple addresses on an interface in IPv6;
> >2) We need to specify how PIM works under such scenarios.
> >
> >Else, it will be a mess in the field when it comes to
> >interoperability. In short, a solution on 2) should be discussed
> >in the PIM WG.
> 
> There are plenty of situations where interfaces are associated with 
> multiple IPv4 addresses as well.  Some PIM implementations don't take this 
> into account.  In my experience, this has led to operational deployment 
> confusion and difficulty.
> 
> One specific example involves HSRP/VRRP.  Unicast reachability is satisfied 
> because the "active" HSRP/VRRP speaker responds to IPv4 ARP for the virtual 
> address.  However, multicast reachability is broken because the "active" 
> HSRP/VRRP speaker fails to respond to IPv4 PIM Joins addressed to the 
> virtual address.
> 
> The result is a situation where unicast routing works just fine, but 
> multicast routing is broken.  Such incidents contribute to the perception 
> that multicast routing is more fragile or somehow less reliable than 
> unicast routing.
> 
> ===
> Bill Nickless    http://www.mcs.anl.gov/people/nickless      +1 630 252 7390
> PGP:0E 0F 16 80 C5 B1 69 52 E1 44 1A A5 0E 1B 74 F7     nickless@mcs.anl.gov
> 
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar 24 15:27:03 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA02678
	for <pim-archive@lists.ietf.org>; Mon, 24 Mar 2003 15:27:02 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2OKPAu64078;
	Mon, 24 Mar 2003 12:25:10 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from sj-msg-core-1.cisco.com (sj-msg-core-1.cisco.com [171.71.163.11])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2OKO1u64041
	for <pim@catarina.usc.edu>; Mon, 24 Mar 2003 12:24:01 -0800 (PST)
	(envelope-from nbhaskar@cisco.com)
Received: from cypher.cisco.com (cypher.cisco.com [171.69.11.143])
	by sj-msg-core-1.cisco.com (8.12.2/8.12.6) with ESMTP id h2OKNuhE016777;
	Mon, 24 Mar 2003 12:23:56 -0800 (PST)
From: Nidhi Bhaskar <nbhaskar@cisco.com>
To: JINMEI Tatuya / =?ISO-2022-JP?B?GyRCP0BMQEMjOkgbKEI=?= <jinmei@isl.rdc.toshiba.co.jp>
cc: John Zwiebel <jzwiebel@procket.com>, <lwei@redback.com>,
        <pim@catarina.usc.edu>
Subject: Re: [pim] San Francisco PIM WG meeting minutes
In-Reply-To: <y7vadfnwpno.wl@ocean.jinmei.org>
Message-ID: <Pine.GSO.4.44.0303241219360.4165-100000@cypher.cisco.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=X-UNKNOWN
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Mon, 24 Mar 2003 12:23:56 -0800 (PST)

Hi,


I think we should discuss this in the PIM WG as well. And I vote for
going with the PIM hello option proposal. But I am not opposed to a
moving to a generic solution, when it becomes available.

AFAIK, Inverse ND (the generic solution suggested for ipv6) is not
widely implemented -- we need some way to fix this in the meantime.
Another issue is that Inverse ND works only on links that have link
layer addresses. I'm not sure if that will be a limitation for PIM,
but would be nice not to have this dependency. Also it looks like the
problem might exist in ipv4 and we don't have a generic solution for
that.

Perhaps we can continue to use an option number from the >6500 space
and deprecate it once generic solutions are available.

I'd also like to point out that this problem is not really due to ipv6
link-local addresses, but multiple addresses. We'd still have to solve
this if global ipv6 addresses were used in hellos. I.e there is a
problem as long as there is more than one address on the i/f.

And finally, suggest renaming to something like 'multiple address
advertisement' or 'address advertisement' option. Seems more apt than
'upstream detection'... to me atleast :)


Thanks,
Nidhi



On Sat, 22 Mar 2003, JINMEI Tatuya / [ISO-2022-JP] $B?@L@C#:H(B wrote:

> >>>>> On Wed, 19 Mar 2003 13:49:59 -0800,
> >>>>> John Zwiebel <jzwiebel@procket.com> said:
>
> >> 4. Jimei presented proposals on correctly matching a global address
> >> to a link-local IPv6 address for PIM rpf checks.
>
> > Someone suggested that this problem is not unique to PIM.
>
> Correct (as I admitted at the presentation), but at the same time,
> "protocol independence", a unique characteristic of PIM, is one the
> major sources of the issue.
>
> > (Dino
> > had a rant about link-local addresses being a "bug" -- I agree with
> > him that they are a pain).
>
> I admit link-local addresses introduce a pain, but it also provides
> benefits.  In fact, the IETF is introducing link-local addresses even
> in the IPv4 world by the zeroconf wg.  There may still be people who
> just think the notion of link-local is a bug and should be killed, but
> the fact shows it is at least controversial whether we should have
> link-local or not.  So, I'd be glad if we could concentrate on the
> issue based on the fact we have link-local addresses, rather than
> discuss the validity of having link-local here.
>
> > It may be that this problem should be
> > solved by one of the IPv6 working groups.
>
> I don't mind to discuss this somewhere else if it is more appropriate
> place than here.  However, as some other people have pointed out in
> this list, the issued addressed in the draft is not necessarily
> specific to IPv6 (we the authors realized this point, but mainly
> focused on the IPv6 issue in the draft since this is the most
> practical one we've seen).
>
> I still believe the pim wg is the best place, and would really
> appreciate if the wg allows us to discuss the issue here.
>
> 					JINMEI, Tatuya
> 					Communication Platform Lab.
> 					Corporate R&D Center, Toshiba Corp.
> 					jinmei@isl.rdc.toshiba.co.jp
> _______________________________________________
> pim mailing list
> pim@catarina.usc.edu
> http://catarina.usc.edu/mailman/listinfo/pim
>


_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar 24 16:34:18 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA04368
	for <pim-archive@lists.ietf.org>; Mon, 24 Mar 2003 16:34:18 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2OLRku65362;
	Mon, 24 Mar 2003 13:27:46 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from mailhost.iprg.nokia.com (mailhost.iprg.nokia.com [205.226.5.12])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2OLQhu65328
	for <pim@catarina.usc.edu>; Mon, 24 Mar 2003 13:26:43 -0800 (PST)
	(envelope-from Rahul.Bahadur@nokia.com)
Received: from darkstar.iprg.nokia.com (darkstar.iprg.nokia.com [205.226.5.69])
	by mailhost.iprg.nokia.com (8.9.3/8.9.3-GLGS) with ESMTP id NAA17179;
	Mon, 24 Mar 2003 13:26:37 -0800 (PST)
Received: (from root@localhost)
	by darkstar.iprg.nokia.com (8.11.0/8.11.0-DARKSTAR) id h2OLQav07206;
	Mon, 24 Mar 2003 13:26:36 -0800
X-mProtect: <200303242126> Nokia Silicon Valley Messaging Protection
Received: from UNKNOWN (172.19.74.182, claiming to be "dadhcp-172019074182.americas.nokia.com")
	by darkstar.iprg.nokia.com smtpd1pfU1z; Mon, 24 Mar 2003 13:26:35 PST
Subject: Re: [pim] San Francisco PIM WG meeting minutes
From: Rahul Bahadur <Rahul.Bahadur@nokia.com>
To: lwei@redback.com
Cc: nickless@mcs.anl.gov, pim@catarina.usc.edu,
        Bahadur  "Rahul (IPRG/MtView)"
	 <Rahul.Bahadur@nokia.com>
In-Reply-To: <200303242013.h2OKDIT28218@liming-pc.redback.com>
References: <y7v8yv7wp8y.wl@ocean.jinmei.org>
	<BAE51AE0-5A54-11D7-B534-0003936A6062@procket.com>
	<200303192213.h2JMDwq11638@liming-pc.redback.com>
	<y7v8yv7wp8y.wl@ocean.jinmei.org>
	<5.1.1.6.2.20030322200815.01e5cc38@pop.mcs.anl.gov> 
	<200303242013.h2OKDIT28218@liming-pc.redback.com>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
X-Mailer: Ximian Evolution 1.0.3 (1.0.3-6) 
Message-Id: <1048541195.1844.27.camel@dadhcp-172019074182.americas.nokia.com>
Mime-Version: 1.0
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: 24 Mar 2003 13:26:35 -0800
Content-Transfer-Encoding: 7bit


Liming, Bill,

Maybe I can chime in here. The VRRP/HSRP situation is not that unusual -
several vendors want to use the same pair of routers to back-up both
unicast and multicast traffic. However, they do want this pair to run
any unicast protocols but do want to use PIM for multicast failover.
Since the routers neighboring the VRRP pair only have a default route
pointing to the virtual address (which never matches that of a PIM
neighbor) both sparse-mode and dense-mode break at different times. Only
problem here is that users are expecting certain behaviour from one
protocol based on their experience with another. Liming, your solution
noted below is the obvious workaround.

Rahul

On Mon, 2003-03-24 at 12:13, ext Liming Wei wrote:
> Bill,
> 
> I actually am thinking about cases invlving multiple addresses in
> IPv4 envnronment too :-).
> 
> The HSRP or VRRP case sounds unusual. Why does a PIM joiner/pruner
> have to rely on a "static default route" type of unicast routing ?
> 
> An implementation today can "solve" this problem by special-knob'ing
> PIM to run from the VRRP active router only, and only "on" the virtual
> address in PIM messages (as opposed to the physical addresses). Of
> course, this is not as general/simple (robust?) as learning all
> addresses via a PIM hello option.
> 
> -Liming
> --------
> > From: Bill Nickless <nickless@mcs.anl.gov>
> > 
> > At 01:27 PM 3/22/2003 -0800, Liming Wei wrote:
> > 
> > >1) There exists multiple addresses on an interface in IPv6;
> > >2) We need to specify how PIM works under such scenarios.
> > >
> > >Else, it will be a mess in the field when it comes to
> > >interoperability. In short, a solution on 2) should be discussed
> > >in the PIM WG.
> > 
> > There are plenty of situations where interfaces are associated with 
> > multiple IPv4 addresses as well.  Some PIM implementations don't take this 
> > into account.  In my experience, this has led to operational deployment 
> > confusion and difficulty.
> > 
> > One specific example involves HSRP/VRRP.  Unicast reachability is satisfied 
> > because the "active" HSRP/VRRP speaker responds to IPv4 ARP for the virtual 
> > address.  However, multicast reachability is broken because the "active" 
> > HSRP/VRRP speaker fails to respond to IPv4 PIM Joins addressed to the 
> > virtual address.
> > 
> > The result is a situation where unicast routing works just fine, but 
> > multicast routing is broken.  Such incidents contribute to the perception 
> > that multicast routing is more fragile or somehow less reliable than 
> > unicast routing.
> > 
> > ===
> > Bill Nickless    http://www.mcs.anl.gov/people/nickless      +1 630 252 7390
> > PGP:0E 0F 16 80 C5 B1 69 52 E1 44 1A A5 0E 1B 74 F7     nickless@mcs.anl.gov
> > 
> _______________________________________________
> pim mailing list
> pim@catarina.usc.edu
> http://catarina.usc.edu/mailman/listinfo/pim


_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar 24 16:54:50 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA05186
	for <pim-archive@lists.ietf.org>; Mon, 24 Mar 2003 16:54:50 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2OLrEu65917;
	Mon, 24 Mar 2003 13:53:14 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from dino-pc.procket.com (zeus-ext.procket.com [65.174.124.29])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2OLqou65895
	for <pim@catarina.usc.edu>; Mon, 24 Mar 2003 13:52:50 -0800 (PST)
	(envelope-from dino@procket.com)
Received: (from dino@localhost)
	by dino-pc.procket.com (8.9.3/8.9.3) id NAA29785;
	Mon, 24 Mar 2003 13:58:41 -0800
Message-Id: <200303242158.NAA29785@dino-pc.procket.com>
X-Authentication-Warning: dino-pc.procket.com: dino set sender to dino@dino-pc.procket.com using -f
From: Dino Farinacci <dino@procket.com>
To: nbhaskar@cisco.com
CC: jinmei@isl.rdc.toshiba.co.jp, jzwiebel@procket.com, lwei@redback.com,
        pim@catarina.usc.edu
In-reply-to: <Pine.GSO.4.44.0303241219360.4165-100000@cypher.cisco.com>
	(message from Nidhi Bhaskar on Mon, 24 Mar 2003 12:23:56 -0800 (PST))
Subject: Re: [pim] San Francisco PIM WG meeting minutes
References:  <Pine.GSO.4.44.0303241219360.4165-100000@cypher.cisco.com>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Mon, 24 Mar 2003 13:58:41 -0800

>> AFAIK, Inverse ND (the generic solution suggested for ipv6) is not
>> widely implemented -- we need some way to fix this in the meantime.
>> Another issue is that Inverse ND works only on links that have link
>> layer addresses. I'm not sure if that will be a limitation for PIM,
>> but would be nice not to have this dependency. Also it looks like the
>> problem might exist in ipv4 and we don't have a generic solution for
>> that.

    I think an even shorter-term solution is to send multiple Hellos with
    different source addresses. Then any router can do the right thing.

Dino
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar 24 17:23:48 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA06011
	for <pim-archive@lists.ietf.org>; Mon, 24 Mar 2003 17:23:46 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2OMMAu66539;
	Mon, 24 Mar 2003 14:22:10 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from motgate5.mot.com (motgate5.mot.com [144.189.100.105])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2OMLmu66519
	for <pim@catarina.usc.edu>; Mon, 24 Mar 2003 14:21:48 -0800 (PST)
	(envelope-from Chris.Cormier@motorola.com)
Received: from mothost.mot.com (mothost.mot.com [129.188.137.101])
	by motgate5.mot.com (Motorola/Motgate5) with ESMTP id h2OMLMHG026126
	for <pim@catarina.usc.edu>; Mon, 24 Mar 2003 15:21:23 -0700 (MST)
Received: [from il06exm05.corp.mot.com (il06exm05.corp.mot.com [10.0.111.5]) by mothost.mot.com (MOT-pobox 2.0) with ESMTP id PAA01519 for <pim@catarina.usc.edu>; Mon, 24 Mar 2003 15:21:42 -0700 (MST)]
Received: by il06exm05.corp.mot.com with Internet Mail Service (5.5.2656.59)
	id <HPN1QQY2>; Mon, 24 Mar 2003 16:21:02 -0600
Message-ID: <FB578B06F252D511B95B009027E3267B06762108@il06exm04.corp.mot.com>
From: Cormier Chris-ACC125 <Chris.Cormier@motorola.com>
To: "'Dino Farinacci'" <dino@procket.com>, nbhaskar@cisco.com
Cc: jinmei@isl.rdc.toshiba.co.jp, jzwiebel@procket.com, lwei@redback.com,
        pim@catarina.usc.edu
Subject: RE: [pim] San Francisco PIM WG meeting minutes
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2656.59)
Content-Type: text/plain
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Mon, 24 Mar 2003 16:21:01 -0600

Why wouldn't this be a long term solution?
-cgc

-----Original Message-----
From: Dino Farinacci [mailto:dino@procket.com]
Sent: Monday, March 24, 2003 1:59 PM
To: nbhaskar@cisco.com
Cc: jinmei@isl.rdc.toshiba.co.jp; jzwiebel@procket.com;
lwei@redback.com; pim@catarina.usc.edu
Subject: Re: [pim] San Francisco PIM WG meeting minutes


>> AFAIK, Inverse ND (the generic solution suggested for ipv6) is not
>> widely implemented -- we need some way to fix this in the meantime.
>> Another issue is that Inverse ND works only on links that have link
>> layer addresses. I'm not sure if that will be a limitation for PIM,
>> but would be nice not to have this dependency. Also it looks like the
>> problem might exist in ipv4 and we don't have a generic solution for
>> that.

    I think an even shorter-term solution is to send multiple Hellos with
    different source addresses. Then any router can do the right thing.

Dino
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar 24 17:31:18 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA06285
	for <pim-archive@lists.ietf.org>; Mon, 24 Mar 2003 17:31:17 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2OMSSu66707;
	Mon, 24 Mar 2003 14:28:28 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from dino-pc.procket.com (zeus-ext.procket.com [65.174.124.29])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2OMRpu66678
	for <pim@catarina.usc.edu>; Mon, 24 Mar 2003 14:27:51 -0800 (PST)
	(envelope-from dino@procket.com)
Received: (from dino@localhost)
	by dino-pc.procket.com (8.9.3/8.9.3) id OAA29955;
	Mon, 24 Mar 2003 14:33:48 -0800
Message-Id: <200303242233.OAA29955@dino-pc.procket.com>
X-Authentication-Warning: dino-pc.procket.com: dino set sender to dino@dino-pc.procket.com using -f
From: Dino Farinacci <dino@procket.com>
To: Chris.Cormier@motorola.com
CC: nbhaskar@cisco.com, jinmei@isl.rdc.toshiba.co.jp, jzwiebel@procket.com,
        lwei@redback.com, pim@catarina.usc.edu
In-reply-to: <FB578B06F252D511B95B009027E3267B06762108@il06exm04.corp.mot.com>
	(message from Cormier Chris-ACC125 on Mon, 24 Mar 2003 16:21:01 -0600)
Subject: Re: [pim] San Francisco PIM WG meeting minutes
References:  <FB578B06F252D511B95B009027E3267B06762108@il06exm04.corp.mot.com>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Mon, 24 Mar 2003 14:33:48 -0800

>> Why wouldn't this be a long term solution?
>> -cgc

    It certainly could. Anything that avoids protocols changes at the expense
    of extra Hello traffic should also be considered a long-term solution too.

Dino
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar 24 17:41:54 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA06584
	for <pim-archive@lists.ietf.org>; Mon, 24 Mar 2003 17:41:53 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2OMdBu66950;
	Mon, 24 Mar 2003 14:39:11 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from possum.icir.org (possum.icir.org [192.150.187.67])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2OMcKu66924
	for <pim@catarina.usc.edu>; Mon, 24 Mar 2003 14:38:20 -0800 (PST)
	(envelope-from pavlin@possum.icir.org)
Received: from possum.icir.org (localhost [127.0.0.1])
	by possum.icir.org (8.12.3/8.12.3) with ESMTP id h2OMcJix044976;
	Mon, 24 Mar 2003 14:38:19 -0800 (PST)
	(envelope-from pavlin@possum.icir.org)
Message-Id: <200303242238.h2OMcJix044976@possum.icir.org>
To: Dino Farinacci <dino@procket.com>
cc: nbhaskar@cisco.com, jinmei@isl.rdc.toshiba.co.jp, jzwiebel@procket.com,
        lwei@redback.com, pim@catarina.usc.edu
Subject: Re: [pim] San Francisco PIM WG meeting minutes 
In-Reply-To: Message from Dino Farinacci <dino@procket.com> 
   of "Mon, 24 Mar 2003 13:58:41 PST." <200303242158.NAA29785@dino-pc.procket.com> 
From: Pavlin Radoslavov <pavlin@icir.org>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Mon, 24 Mar 2003 14:38:19 -0800

> >> AFAIK, Inverse ND (the generic solution suggested for ipv6) is not
> >> widely implemented -- we need some way to fix this in the meantime.
> >> Another issue is that Inverse ND works only on links that have link
> >> layer addresses. I'm not sure if that will be a limitation for PIM,
> >> but would be nice not to have this dependency. Also it looks like the
> >> problem might exist in ipv4 and we don't have a generic solution for
> >> that.
> 
>     I think an even shorter-term solution is to send multiple Hellos with
>     different source addresses. Then any router can do the right thing.

This may not work for things like Prune overriding.
E.g., lets say that the upstream router addresses are A1 and
A2. If the downstream router sends a Join(*,G) with the upstream
router address set to A2, then when the upstream router has to send
PruneEcho on that interface, it may have to remember to put A2 as
the upstream router address rather than A1. Well, if another router
on that LAN uses A1 as the upstream router address for same G, then
that other router won't override the Prune or the PruneEcho. The
result: the interface with address A1 and A2 is pruned when it
should be in Joined state.

Indeed, I don't exclude the possibility that if someone does very
detailed analyses it may be found that the Prune overriding won't be
broken for some other reasons like Asserts or whatever, but the
point I want to emphasize is that the multiple Hellos makes the
protocol more fragile.
For example, one thing that will happen is that the number of the
neighbors will be miscalculated, and then when (*,G) Prune is
received, the PrunePending timer may be set to
J/P_OverrideInterval(I) instead of zero, because this behavior
depends on the number of neighbors.

(Here I should be "fair" to admit that approx a month ago when I
read draft-suz-pim-upstream-detection-00.txt, my first reaction was
that the multiple Hellos scheme may be a reasonable work-around that
wasn't mentioned in the draft, but now I see I was wrong :)

My preference is for the new Hello option rather than any
workarounds, because the new Hello option makes things clean and
explicit.

Pavlin
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar 24 17:50:33 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA06849
	for <pim-archive@lists.ietf.org>; Mon, 24 Mar 2003 17:50:31 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2OMm9u67205;
	Mon, 24 Mar 2003 14:48:09 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from sj-msg-core-1.cisco.com (sj-msg-core-1.cisco.com [171.71.163.11])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2OMl4u67167
	for <pim@catarina.usc.edu>; Mon, 24 Mar 2003 14:47:04 -0800 (PST)
	(envelope-from nbhaskar@cisco.com)
Received: from cypher.cisco.com (cypher.cisco.com [171.69.11.143])
	by sj-msg-core-1.cisco.com (8.12.2/8.12.6) with ESMTP id h2OMkwhE029494;
	Mon, 24 Mar 2003 14:46:58 -0800 (PST)
From: Nidhi Bhaskar <nbhaskar@cisco.com>
To: Dino Farinacci <dino@procket.com>
cc: jinmei@isl.rdc.toshiba.co.jp, <jzwiebel@procket.com>, <lwei@redback.com>,
        <pim@catarina.usc.edu>
Subject: Re: [pim] San Francisco PIM WG meeting minutes
In-Reply-To: <200303242158.NAA29785@dino-pc.procket.com>
Message-ID: <Pine.GSO.4.44.0303241412510.4165-100000@cypher.cisco.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Mon, 24 Mar 2003 14:46:58 -0800 (PST)

My only issue with this is the duration for which it might be
required... Given that it can have some side effects like turning off
PIM optimizations otherwise possible on 1 nbr links, increase nbr
processing, interfere with per nbr j/p pacing etc. Maybe others I
can't think of... These are probably only significant if we end up
using it for a long time.

Another issue -- it'll probably violate various ipv6 rfcs (+ the new
PIM draft) by using routable addreses in PIM messages :)

The hello option is almost as simple & avoids the above. It seems
cleaner without much overhead, so I have a preference for it.


Thanks,
Nidhi.

On Mon, 24 Mar 2003, Dino Farinacci wrote:

> >> AFAIK, Inverse ND (the generic solution suggested for ipv6) is not
> >> widely implemented -- we need some way to fix this in the meantime.
> >> Another issue is that Inverse ND works only on links that have link
> >> layer addresses. I'm not sure if that will be a limitation for PIM,
> >> but would be nice not to have this dependency. Also it looks like the
> >> problem might exist in ipv4 and we don't have a generic solution for
> >> that.
>
>     I think an even shorter-term solution is to send multiple Hellos with
>     different source addresses. Then any router can do the right thing.
>
> Dino
>



_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar 24 17:54:01 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA07000
	for <pim-archive@lists.ietf.org>; Mon, 24 Mar 2003 17:54:01 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2OMqFu67311;
	Mon, 24 Mar 2003 14:52:15 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from dino-pc.procket.com (zeus-ext.procket.com [65.174.124.29])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2OMp5u67283
	for <pim@catarina.usc.edu>; Mon, 24 Mar 2003 14:51:05 -0800 (PST)
	(envelope-from dino@procket.com)
Received: (from dino@localhost)
	by dino-pc.procket.com (8.9.3/8.9.3) id OAA30080;
	Mon, 24 Mar 2003 14:57:05 -0800
Message-Id: <200303242257.OAA30080@dino-pc.procket.com>
X-Authentication-Warning: dino-pc.procket.com: dino set sender to dino@dino-pc.procket.com using -f
From: Dino Farinacci <dino@procket.com>
To: pavlin@icir.org
CC: nbhaskar@cisco.com, jinmei@isl.rdc.toshiba.co.jp, jzwiebel@procket.com,
        lwei@redback.com, pim@catarina.usc.edu
In-reply-to: <200303242238.h2OMcJix044976@possum.icir.org> (message from
	Pavlin Radoslavov on Mon, 24 Mar 2003 14:38:19 -0800)
Subject: Re: [pim] San Francisco PIM WG meeting minutes
References:  <200303242238.h2OMcJix044976@possum.icir.org>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Mon, 24 Mar 2003 14:57:05 -0800

>> This may not work for things like Prune overriding.
>> E.g., lets say that the upstream router addresses are A1 and
>> A2. If the downstream router sends a Join(*,G) with the upstream
>> router address set to A2, then when the upstream router has to send
>> PruneEcho on that interface, it may have to remember to put A2 as
>> the upstream router address rather than A1. Well, if another router
>> on that LAN uses A1 as the upstream router address for same G, then
>> that other router won't override the Prune or the PruneEcho. The
>> result: the interface with address A1 and A2 is pruned when it
>> should be in Joined state.

    This all depends if you treat the LAN as multiple logical subnets and
    running the protocol on each subnet. Most implementations don't so this
    because they don't want the unicast and multicast protocol overhead. If
    they need to do it for some reason, they then configure VLANs. 

    This really isn't a big problem with IPv4 since there is one instance
    of a unicast routing protocol running and each router is configured with
    a primary address on the same subnet.

    As long as the PIM hello protocol and the unicast routing protocol use
    the same set of addresses (same subnet or link-local versus same-subnet) 
    there shouldn't be a problem.

    So, let's back off. What real problem are we trying to solve?

>> (Here I should be "fair" to admit that approx a month ago when I
>> read draft-suz-pim-upstream-detection-00.txt, my first reaction was
>> that the multiple Hellos scheme may be a reasonable work-around that
>> wasn't mentioned in the draft, but now I see I was wrong :)

    It is reasonable if, say you were running OSPF on both subnets on the
    same wire. Then you could have upstream routes using next-hops on the
    different subnets.

>> My preference is for the new Hello option rather than any
>> workarounds, because the new Hello option makes things clean and
>> explicit.

    My preference is to keep it the way it is. And if you have to do this,
    use VLANs. And for the IPv6 case, decide on a per LAN basis if all
    protocols will use link-local addresses for route table entries or 
    configured addresses for route table entries.

Dino
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar 24 18:12:40 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA08886
	for <pim-archive@lists.ietf.org>; Mon, 24 Mar 2003 18:12:39 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2ONBAu67770;
	Mon, 24 Mar 2003 15:11:10 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from dino-pc.procket.com (zeus-ext.procket.com [65.174.124.29])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2ONAGu67745
	for <pim@catarina.usc.edu>; Mon, 24 Mar 2003 15:10:16 -0800 (PST)
	(envelope-from dino@procket.com)
Received: (from dino@localhost)
	by dino-pc.procket.com (8.9.3/8.9.3) id PAA30152;
	Mon, 24 Mar 2003 15:16:11 -0800
Message-Id: <200303242316.PAA30152@dino-pc.procket.com>
X-Authentication-Warning: dino-pc.procket.com: dino set sender to dino@dino-pc.procket.com using -f
From: Dino Farinacci <dino@procket.com>
To: nbhaskar@cisco.com
CC: jinmei@isl.rdc.toshiba.co.jp, jzwiebel@procket.com, lwei@redback.com,
        pim@catarina.usc.edu
In-reply-to: <Pine.GSO.4.44.0303241412510.4165-100000@cypher.cisco.com>
	(message from Nidhi Bhaskar on Mon, 24 Mar 2003 14:46:58 -0800 (PST))
Subject: Re: [pim] San Francisco PIM WG meeting minutes
References:  <Pine.GSO.4.44.0303241412510.4165-100000@cypher.cisco.com>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Mon, 24 Mar 2003 15:16:11 -0800

>> required... Given that it can have some side effects like turning off
>> PIM optimizations otherwise possible on 1 nbr links, increase nbr

    Agree. But not sure why JP pacing is affected by this since it is a per
    neighbor issue.

>> Another issue -- it'll probably violate various ipv6 rfcs (+ the new
>> PIM draft) by using routable addreses in PIM messages :)

    Aren't we trying to improve the Internet with IPv6? I feel we are going
    in the opposite direction.

>> The hello option is almost as simple & avoids the above. It seems
>> cleaner without much overhead, so I have a preference for it.

    Right.    

Dino
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar 24 18:28:53 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA09634
	for <pim-archive@lists.ietf.org>; Mon, 24 Mar 2003 18:28:52 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2ONQHu68118;
	Mon, 24 Mar 2003 15:26:17 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from possum.icir.org (possum.icir.org [192.150.187.67])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2ONP6u68091
	for <pim@catarina.usc.edu>; Mon, 24 Mar 2003 15:25:06 -0800 (PST)
	(envelope-from pavlin@possum.icir.org)
Received: from possum.icir.org (localhost [127.0.0.1])
	by possum.icir.org (8.12.3/8.12.3) with ESMTP id h2ONOpix045464;
	Mon, 24 Mar 2003 15:24:51 -0800 (PST)
	(envelope-from pavlin@possum.icir.org)
Message-Id: <200303242324.h2ONOpix045464@possum.icir.org>
To: Dino Farinacci <dino@procket.com>
cc: pavlin@icir.org, nbhaskar@cisco.com, jinmei@isl.rdc.toshiba.co.jp,
        jzwiebel@procket.com, lwei@redback.com, pim@catarina.usc.edu
Subject: Re: [pim] San Francisco PIM WG meeting minutes 
In-Reply-To: Message from Dino Farinacci <dino@procket.com> 
   of "Mon, 24 Mar 2003 14:57:05 PST." <200303242257.OAA30080@dino-pc.procket.com> 
From: Pavlin Radoslavov <pavlin@icir.org>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Mon, 24 Mar 2003 15:24:51 -0800

> >> This may not work for things like Prune overriding.
> >> E.g., lets say that the upstream router addresses are A1 and
> >> A2. If the downstream router sends a Join(*,G) with the upstream
> >> router address set to A2, then when the upstream router has to send
> >> PruneEcho on that interface, it may have to remember to put A2 as
> >> the upstream router address rather than A1. Well, if another router
> >> on that LAN uses A1 as the upstream router address for same G, then
> >> that other router won't override the Prune or the PruneEcho. The
> >> result: the interface with address A1 and A2 is pruned when it
> >> should be in Joined state.
> 
>     This all depends if you treat the LAN as multiple logical subnets and
>     running the protocol on each subnet. Most implementations don't so this
>     because they don't want the unicast and multicast protocol overhead. If
>     they need to do it for some reason, they then configure VLANs. 
> 
>     This really isn't a big problem with IPv4 since there is one instance
>     of a unicast routing protocol running and each router is configured with
>     a primary address on the same subnet.

If you relay on this, you are making an assumption about the
behavior of the underlying unicast routing protocol, so PIM is not
protocol-independent anymore :)
Not to mention that if someone decides to use static IP routing with
IP aliases on an interface, then running PIM-SM on top of it will
add additional constrains on how the routing is configured.

>     As long as the PIM hello protocol and the unicast routing protocol use
>     the same set of addresses (same subnet or link-local versus same-subnet) 
>     there shouldn't be a problem.
> 
>     So, let's back off. What real problem are we trying to solve?

For me, it is making sure that PIM-SM works in an environment where
there is more one IP address per interface, but without making
specific assumptions about how the underlying unicast routing
works.

> >> My preference is for the new Hello option rather than any
> >> workarounds, because the new Hello option makes things clean and
> >> explicit.
> 
>     My preference is to keep it the way it is. And if you have to do this,
>     use VLANs. And for the IPv6 case, decide on a per LAN basis if all
>     protocols will use link-local addresses for route table entries or 
>     configured addresses for route table entries.

Yet another thing that can go wrong.
If it is kept the way it is, then you have always to explain this
issue to your customers, which is just more headaches for them (and
for your customer support :).

Regards,
Pavlin
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar 24 19:02:12 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA10399
	for <pim-archive@lists.ietf.org>; Mon, 24 Mar 2003 19:02:12 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2ONwVu68831;
	Mon, 24 Mar 2003 15:58:31 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from hotmail.com (oe53.law9.hotmail.com [64.4.8.46])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2ONv4u68786
	for <pim@catarina.usc.edu>; Mon, 24 Mar 2003 15:57:04 -0800 (PST)
	(envelope-from desai_purvi@hotmail.com)
Received: from mail pickup service by hotmail.com with Microsoft SMTPSVC;
	 Mon, 24 Mar 2003 15:56:59 -0800
Received: from 63.126.135.11 by oe53.law9.hotmail.com with DAV;
	Mon, 24 Mar 2003 23:56:59 +0000
X-Originating-IP: [63.126.135.11]
X-Originating-Email: [desai_purvi@hotmail.com]
From: "Purvi" <desai_purvi@hotmail.com>
To: "Dino Farinacci" <dino@procket.com>, "Pavlin Radoslavov" <pavlin@icir.org>
Cc: <pavlin@icir.org>, <nbhaskar@cisco.com>, <jinmei@isl.rdc.toshiba.co.jp>,
        <jzwiebel@procket.com>, <lwei@redback.com>, <pim@catarina.usc.edu>
References: <200303242324.h2ONOpix045464@possum.icir.org>
Subject: Re: [pim] San Francisco PIM WG meeting minutes
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 6.00.2600.0000
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000
Message-ID: <OE53NYH2J5aXwu1Ju2P00004717@hotmail.com>
X-OriginalArrivalTime: 24 Mar 2003 23:56:59.0352 (UTC) FILETIME=[0EC7CD80:01C2F261]
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Mon, 24 Mar 2003 15:53:23 -0800
Content-Transfer-Encoding: 7bit


And its not just sending hello pkts on all interfaces of same subnet or
using
all ip addresses of the interface to facilitate joins but it also involves
sending
prune echos using all ip addresses on that subnet/interface and even
transmitting BSR messages using all ip addresses. And ipv4 n/ws do have this
problem.
Hello option does seem better to all these useless duplicates.

Purvi


----- Original Message -----
From: "Pavlin Radoslavov" <pavlin@icir.org>
To: "Dino Farinacci" <dino@procket.com>
Cc: <pavlin@icir.org>; <nbhaskar@cisco.com>; <jinmei@isl.rdc.toshiba.co.jp>;
<jzwiebel@procket.com>; <lwei@redback.com>; <pim@catarina.usc.edu>
Sent: Monday, March 24, 2003 3:24 PM
Subject: Re: [pim] San Francisco PIM WG meeting minutes


> > >> This may not work for things like Prune overriding.
> > >> E.g., lets say that the upstream router addresses are A1 and
> > >> A2. If the downstream router sends a Join(*,G) with the upstream
> > >> router address set to A2, then when the upstream router has to send
> > >> PruneEcho on that interface, it may have to remember to put A2 as
> > >> the upstream router address rather than A1. Well, if another router
> > >> on that LAN uses A1 as the upstream router address for same G, then
> > >> that other router won't override the Prune or the PruneEcho. The
> > >> result: the interface with address A1 and A2 is pruned when it
> > >> should be in Joined state.
> >
> >     This all depends if you treat the LAN as multiple logical subnets
and
> >     running the protocol on each subnet. Most implementations don't so
this
> >     because they don't want the unicast and multicast protocol overhead.
If
> >     they need to do it for some reason, they then configure VLANs.
> >
> >     This really isn't a big problem with IPv4 since there is one
instance
> >     of a unicast routing protocol running and each router is configured
with
> >     a primary address on the same subnet.
>
> If you relay on this, you are making an assumption about the
> behavior of the underlying unicast routing protocol, so PIM is not
> protocol-independent anymore :)
> Not to mention that if someone decides to use static IP routing with
> IP aliases on an interface, then running PIM-SM on top of it will
> add additional constrains on how the routing is configured.
>
> >     As long as the PIM hello protocol and the unicast routing protocol
use
> >     the same set of addresses (same subnet or link-local versus
same-subnet)
> >     there shouldn't be a problem.
> >
> >     So, let's back off. What real problem are we trying to solve?
>
> For me, it is making sure that PIM-SM works in an environment where
> there is more one IP address per interface, but without making
> specific assumptions about how the underlying unicast routing
> works.
>
> > >> My preference is for the new Hello option rather than any
> > >> workarounds, because the new Hello option makes things clean and
> > >> explicit.
> >
> >     My preference is to keep it the way it is. And if you have to do
this,
> >     use VLANs. And for the IPv6 case, decide on a per LAN basis if all
> >     protocols will use link-local addresses for route table entries or
> >     configured addresses for route table entries.
>
> Yet another thing that can go wrong.
> If it is kept the way it is, then you have always to explain this
> issue to your customers, which is just more headaches for them (and
> for your customer support :).
>
> Regards,
> Pavlin
> _______________________________________________
> pim mailing list
> pim@catarina.usc.edu
> http://catarina.usc.edu/mailman/listinfo/pim
>
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar 24 19:27:39 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA11030
	for <pim-archive@lists.ietf.org>; Mon, 24 Mar 2003 19:27:37 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2P0OAu69426;
	Mon, 24 Mar 2003 16:24:10 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from dino-pc.procket.com (zeus-ext.procket.com [65.174.124.29])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2P0N4u69392
	for <pim@catarina.usc.edu>; Mon, 24 Mar 2003 16:23:04 -0800 (PST)
	(envelope-from dino@procket.com)
Received: (from dino@localhost)
	by dino-pc.procket.com (8.9.3/8.9.3) id QAA30376;
	Mon, 24 Mar 2003 16:29:05 -0800
Message-Id: <200303250029.QAA30376@dino-pc.procket.com>
X-Authentication-Warning: dino-pc.procket.com: dino set sender to dino@dino-pc.procket.com using -f
From: Dino Farinacci <dino@procket.com>
To: pavlin@icir.org
CC: pavlin@icir.org, nbhaskar@cisco.com, jinmei@isl.rdc.toshiba.co.jp,
        jzwiebel@procket.com, lwei@redback.com, pim@catarina.usc.edu
In-reply-to: <200303242324.h2ONOpix045464@possum.icir.org> (message from
	Pavlin Radoslavov on Mon, 24 Mar 2003 15:24:51 -0800)
Subject: Re: [pim] San Francisco PIM WG meeting minutes
References:  <200303242324.h2ONOpix045464@possum.icir.org>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Mon, 24 Mar 2003 16:29:05 -0800

>> If you relay on this, you are making an assumption about the
>> behavior of the underlying unicast routing protocol, so PIM is not
>> protocol-independent anymore :)
>> Not to mention that if someone decides to use static IP routing with
>> IP aliases on an interface, then running PIM-SM on top of it will
>> add additional constrains on how the routing is configured.

    Pavlin, this is simply PIM looking in the unicast routing table for a
    next-hop to a source or RP. If you make Hellos consistent with the
    routing table next-hop, there isn't a problem. 

>> For me, it is making sure that PIM-SM works in an environment where
>> there is more one IP address per interface, but without making
>> specific assumptions about how the underlying unicast routing
>> works.

    PIM in and of itself is not making any assumptions. You just interpreted
    it that way.

>> Yet another thing that can go wrong.
>> If it is kept the way it is, then you have always to explain this
>> issue to your customers, which is just more headaches for them (and
>> for your customer support :).

    PIM over VLANs are in wide-spread use today. I consider it a done-deal.

Dino 
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar 24 19:29:04 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA11080
	for <pim-archive@lists.ietf.org>; Mon, 24 Mar 2003 19:29:03 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2P0P4u69470;
	Mon, 24 Mar 2003 16:25:04 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from sj-msg-core-1.cisco.com (sj-msg-core-1.cisco.com [171.71.163.11])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2P0Oxu69446
	for <pim@catarina.usc.edu>; Mon, 24 Mar 2003 16:24:59 -0800 (PST)
	(envelope-from nbhaskar@cisco.com)
Received: from cypher.cisco.com (cypher.cisco.com [171.69.11.143])
	by sj-msg-core-1.cisco.com (8.12.2/8.12.6) with ESMTP id h2P0OshE007515;
	Mon, 24 Mar 2003 16:24:54 -0800 (PST)
From: Nidhi Bhaskar <nbhaskar@cisco.com>
To: Dino Farinacci <dino@procket.com>
cc: jinmei@isl.rdc.toshiba.co.jp, <jzwiebel@procket.com>, <lwei@redback.com>,
        <pim@catarina.usc.edu>
Subject: Re: [pim] San Francisco PIM WG meeting minutes
In-Reply-To: <200303242316.PAA30152@dino-pc.procket.com>
Message-ID: <Pine.GSO.4.44.0303241524260.4165-100000@cypher.cisco.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Mon, 24 Mar 2003 16:24:54 -0800 (PST)

> >> required... Given that it can have some side effects like turning off
> >> PIM optimizations otherwise possible on 1 nbr links, increase nbr
>
>     Agree. But not sure why JP pacing is affected by this since it is a per
>     neighbor issue.

I was thinking if you wanted to send j/ps at a certain rate to each
nbr, having say 3 nbrs that map to one physical box would mean you'd
send at 3 times what you intended to. But I think others have
pointed out other more basic problems...

>
> >> Another issue -- it'll probably violate various ipv6 rfcs (+ the new
> >> PIM draft) by using routable addreses in PIM messages :)
>
>     Aren't we trying to improve the Internet with IPv6? I feel we are going
>     in the opposite direction.

Well, deprecating site-locals was one step in the "right" :) direction
this ietf. Perhaps we'll see more as ipv6 gets wider deployment.


Nidhi.

> >> The hello option is almost as simple & avoids the above. It seems
> >> cleaner without much overhead, so I have a preference for it.
>
>     Right.
>
> Dino
>






_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar 24 20:31:29 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA12591
	for <pim-archive@lists.ietf.org>; Mon, 24 Mar 2003 20:31:29 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2P1Tgu70862;
	Mon, 24 Mar 2003 17:29:42 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from merlot.juniper.net (natint.juniper.net [207.17.136.129])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2OL2fu64867
	for <pim@catarina.usc.edu>; Mon, 24 Mar 2003 13:02:41 -0800 (PST)
	(envelope-from pusateri@juniper.net)
Received: from juniper.net (garnet.juniper.net [172.17.28.17])
	by merlot.juniper.net (8.11.3/8.11.3) with ESMTP id h2OL2PS30772;
	Mon, 24 Mar 2003 13:02:25 -0800 (PST)
	(envelope-from pusateri@juniper.net)
Message-Id: <200303242102.h2OL2PS30772@merlot.juniper.net>
To: Nidhi Bhaskar <nbhaskar@cisco.com>
cc: JINMEI Tatuya / =?ISO-2022-JP?B?GyRCP0BMQEMjOkgbKEI=?= <jinmei@isl.rdc.toshiba.co.jp>,
        John Zwiebel <jzwiebel@procket.com>, lwei@redback.com,
        pim@catarina.usc.edu, pusateri@juniper.net
Subject: Re: [pim] San Francisco PIM WG meeting minutes 
In-Reply-To: Message from Nidhi Bhaskar <nbhaskar@cisco.com> 
   of "Mon, 24 Mar 2003 12:23:56 PST." <Pine.GSO.4.44.0303241219360.4165-100000@cypher.cisco.com> 
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <15609.1048539745.1@juniper.net>
From: Tom Pusateri <pusateri@juniper.net>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Mon, 24 Mar 2003 13:02:25 -0800

In message <Pine.GSO.4.44.0303241219360.4165-100000@cypher.cisco.com> you write
:
>Hi,
>
>
>I think we should discuss this in the PIM WG as well. And I vote for
>going with the PIM hello option proposal. But I am not opposed to a
>moving to a generic solution, when it becomes available.

This seems like the best solution to me as well.
When I said that I thought we had solved the problem internally,
it turns out we just required static nexthops to have link local
addresses which doesn't really solve the problem in all cases.

>AFAIK, Inverse ND (the generic solution suggested for ipv6) is not
>widely implemented -- we need some way to fix this in the meantime.
>Another issue is that Inverse ND works only on links that have link
>layer addresses. I'm not sure if that will be a limitation for PIM,
>but would be nice not to have this dependency. Also it looks like the
>problem might exist in ipv4 and we don't have a generic solution for
>that.
>
>Perhaps we can continue to use an option number from the >6500 space
>and deprecate it once generic solutions are available.
>
>I'd also like to point out that this problem is not really due to ipv6
>link-local addresses, but multiple addresses. We'd still have to solve
>this if global ipv6 addresses were used in hellos. I.e there is a
>problem as long as there is more than one address on the i/f.

I think its entirely due to IPv6 link-local addresses.

In v4, you can have multiple addresses but you can only have a single
address on a subnet. This allows you to consistenly have the same
neighbor address as upstream router address.

With v6, this isn't true because you have a link local address and
possibly also one or more global addresses. So your neighbor will always be
the link local but your upstream router address may be any of them.

Thanks,
Tom

>And finally, suggest renaming to something like 'multiple address
>advertisement' or 'address advertisement' option. Seems more apt than
>'upstream detection'... to me atleast :)
>
>
>Thanks,
>Nidhi
>
>
>
>On Sat, 22 Mar 2003, JINMEI Tatuya / [ISO-2022-JP] $B?@L@C#:H(B wrote:
>
>> >>>>> On Wed, 19 Mar 2003 13:49:59 -0800,
>> >>>>> John Zwiebel <jzwiebel@procket.com> said:
>>
>> >> 4. Jimei presented proposals on correctly matching a global address
>> >> to a link-local IPv6 address for PIM rpf checks.
>>
>> > Someone suggested that this problem is not unique to PIM.
>>
>> Correct (as I admitted at the presentation), but at the same time,
>> "protocol independence", a unique characteristic of PIM, is one the
>> major sources of the issue.
>>
>> > (Dino
>> > had a rant about link-local addresses being a "bug" -- I agree with
>> > him that they are a pain).
>>
>> I admit link-local addresses introduce a pain, but it also provides
>> benefits.  In fact, the IETF is introducing link-local addresses even
>> in the IPv4 world by the zeroconf wg.  There may still be people who
>> just think the notion of link-local is a bug and should be killed, but
>> the fact shows it is at least controversial whether we should have
>> link-local or not.  So, I'd be glad if we could concentrate on the
>> issue based on the fact we have link-local addresses, rather than
>> discuss the validity of having link-local here.
>>
>> > It may be that this problem should be
>> > solved by one of the IPv6 working groups.
>>
>> I don't mind to discuss this somewhere else if it is more appropriate
>> place than here.  However, as some other people have pointed out in
>> this list, the issued addressed in the draft is not necessarily
>> specific to IPv6 (we the authors realized this point, but mainly
>> focused on the IPv6 issue in the draft since this is the most
>> practical one we've seen).
>>
>> I still believe the pim wg is the best place, and would really
>> appreciate if the wg allows us to discuss the issue here.
>>
>> 					JINMEI, Tatuya
>> 					Communication Platform Lab.
>> 					Corporate R&D Center, Toshiba Corp.
>> 					jinmei@isl.rdc.toshiba.co.jp
>> _______________________________________________
>> pim mailing list
>> pim@catarina.usc.edu
>> http://catarina.usc.edu/mailman/listinfo/pim
>>
>
>
>_______________________________________________
>pim mailing list
>pim@catarina.usc.edu
>http://catarina.usc.edu/mailman/listinfo/pim
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar 24 20:32:24 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA12664
	for <pim-archive@lists.ietf.org>; Mon, 24 Mar 2003 20:32:23 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2P1Tmu70877;
	Mon, 24 Mar 2003 17:29:48 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from ms-smtp-01.southeast.rr.com (ms-smtp-01.southeast.rr.com [24.93.67.82])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2OLANu65019
	for <pim@catarina.usc.edu>; Mon, 24 Mar 2003 13:10:23 -0800 (PST)
	(envelope-from bkhabs@nc.rr.com)
Received: from mail5.nc.rr.com (fe5 [24.93.67.52])
	by ms-smtp-01.southeast.rr.com (8.12.5/8.12.2) with ESMTP id h2OL6JgA019041;
	Mon, 24 Mar 2003 16:06:19 -0500 (EST)
Received: from nc.rr.com ([66.26.252.37]) by mail5.nc.rr.com  with Microsoft SMTPSVC(5.5.1877.757.75);
	 Mon, 24 Mar 2003 16:07:51 -0500
Message-ID: <3E7F73BF.4040000@nc.rr.com>
From: Brian Haberman <bkhabs@nc.rr.com>
Organization: Me? Organized??
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.1) Gecko/20020826
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Nidhi Bhaskar <nbhaskar@cisco.com>
CC: JINMEI Tatuya <jinmei@isl.rdc.toshiba.co.jp>,
        John Zwiebel <jzwiebel@procket.com>, lwei@redback.com,
        pim@catarina.usc.edu
Subject: Re: [pim] San Francisco PIM WG meeting minutes
References: <Pine.GSO.4.44.0303241219360.4165-100000@cypher.cisco.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Mon, 24 Mar 2003 16:08:15 -0500
Content-Transfer-Encoding: 7bit

Nidhi,
      After thinking about the multiple address issue in v4,
I agree with your suggestion and think we should look at a
PIM option to deal with the problem.

Regards,
Brian

Nidhi Bhaskar wrote:
> Hi,
> 
> 
> I think we should discuss this in the PIM WG as well. And I vote for
> going with the PIM hello option proposal. But I am not opposed to a
> moving to a generic solution, when it becomes available.
> 
> AFAIK, Inverse ND (the generic solution suggested for ipv6) is not
> widely implemented -- we need some way to fix this in the meantime.
> Another issue is that Inverse ND works only on links that have link
> layer addresses. I'm not sure if that will be a limitation for PIM,
> but would be nice not to have this dependency. Also it looks like the
> problem might exist in ipv4 and we don't have a generic solution for
> that.
> 
> Perhaps we can continue to use an option number from the >6500 space
> and deprecate it once generic solutions are available.
> 
> I'd also like to point out that this problem is not really due to ipv6
> link-local addresses, but multiple addresses. We'd still have to solve
> this if global ipv6 addresses were used in hellos. I.e there is a
> problem as long as there is more than one address on the i/f.
> 
> And finally, suggest renaming to something like 'multiple address
> advertisement' or 'address advertisement' option. Seems more apt than
> 'upstream detection'... to me atleast :)
> 
> 
> Thanks,
> Nidhi

_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Mon Mar 24 23:58:33 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA18177
	for <pim-archive@lists.ietf.org>; Mon, 24 Mar 2003 23:58:32 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2P4tau74887;
	Mon, 24 Mar 2003 20:55:36 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from dino-pc.procket.com (zeus-ext.procket.com [65.174.124.29])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2P4sBu74846
	for <pim@catarina.usc.edu>; Mon, 24 Mar 2003 20:54:11 -0800 (PST)
	(envelope-from dino@procket.com)
Received: (from dino@localhost)
	by dino-pc.procket.com (8.9.3/8.9.3) id VAA00966;
	Mon, 24 Mar 2003 21:00:08 -0800
Message-Id: <200303250500.VAA00966@dino-pc.procket.com>
X-Authentication-Warning: dino-pc.procket.com: dino set sender to dino@dino-pc.procket.com using -f
From: Dino Farinacci <dino@procket.com>
To: nbhaskar@cisco.com
CC: jinmei@isl.rdc.toshiba.co.jp, jzwiebel@procket.com, lwei@redback.com,
        pim@catarina.usc.edu
In-reply-to: <Pine.GSO.4.44.0303241524260.4165-100000@cypher.cisco.com>
	(message from Nidhi Bhaskar on Mon, 24 Mar 2003 16:24:54 -0800 (PST))
Subject: Re: [pim] San Francisco PIM WG meeting minutes
References:  <Pine.GSO.4.44.0303241524260.4165-100000@cypher.cisco.com>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Mon, 24 Mar 2003 21:00:08 -0800

>> I was thinking if you wanted to send j/ps at a certain rate to each
>> nbr, having say 3 nbrs that map to one physical box would mean you'd
>> send at 3 times what you intended to. But I think others have
>> pointed out other more basic problems...

    Only if they were RPF neighbor for your multicast routes.

>> Well, deprecating site-locals was one step in the "right" :) direction
>> this ietf. Perhaps we'll see more as ipv6 gets wider deployment.

    Anyone for deprecating link-locals?  ;-)

Dino
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Tue Mar 25 02:27:01 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA02344
	for <pim-archive@lists.ietf.org>; Tue, 25 Mar 2003 02:27:00 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2P7OVu77801;
	Mon, 24 Mar 2003 23:24:31 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from orange.kame.net (orange.kame.net [203.178.141.194])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2P7Nxu77776
	for <pim@catarina.usc.edu>; Mon, 24 Mar 2003 23:23:59 -0800 (PST)
	(envelope-from suz@crl.hitachi.co.jp)
Received: from s30.crl.hitachi.co.jp (orange.kame.net [3ffe:501:4819:2000:203:47ff:fea5:3085])
	by orange.kame.net (Postfix) with ESMTP
	id 6BC917070; Tue, 25 Mar 2003 16:23:57 +0900 (JST)
Message-ID: <x7r88voqoq.wl%suz@crl.hitachi.co.jp>
From: SUZUKI Shinsuke <suz@crl.hitachi.co.jp>
To: Dino Farinacci <dino@procket.com>
Cc: jinmei@isl.rdc.toshiba.co.jp, jzwiebel@procket.com, lwei@redback.com,
        pim@catarina.usc.edu
X-cite: xcite 1.33
In-Reply-To: <200303221703.JAA21988@dino-pc.procket.com>
References: <200303190200.h2J20Pf08821@liming-pc.redback.com>
	<BAE51AE0-5A54-11D7-B534-0003936A6062@procket.com>
	<y7vadfnwpno.wl@ocean.jinmei.org>
User-Agent: User-Agent: Wanderlust/2.11.3 (Wonderwall) Emacs/21.1 Mule/5.0 (SAKAKI)
Organization: Network Architecture Research Dept., Central Research Laboratory, Hitachi, Ltd, Japan
MIME-Version: 1.0 (generated by SEMI 1.14.4 - "Hosorogi")
Content-Type: text/plain; charset=US-ASCII
Subject: [pim] 'link-local' in IPv4?
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Tue, 25 Mar 2003 16:23:17 +0900

Hello all,

>>>>> On Sat, 22 Mar 2003 09:03:01 -0800
>>>>> dino@procket.com(Dino Farinacci)  said:

> >> I admit link-local addresses introduce a pain, but it also provides
> >> benefits.  In fact, the IETF is introducing link-local addresses even
> >> in the IPv4 world by the zeroconf wg.  There may still be people who
> >> just think the notion of link-local is a bug and should be killed, but
> >> the fact shows it is at least controversial whether we should have
> >> link-local or not.  So, I'd be glad if we could concentrate on the
> >> issue based on the fact we have link-local addresses, rather than
> >> discuss the validity of having link-local here.
> 
>     Can you explain what the real benefits are of link-local addresses.

I feel the "link-local" keyword might be one of the reason of the
long thread.
(Although the problem to be resolve in this draft is caused by the
 multiple address assignment on one link, not specific to link-local
 address)


Actually we intentionally used the word "link-local" based on the
following expression in draft-ietf-pim-sm-v2-new-07.txt.

> 4.10.  PIM Packet Formats
> (snip)
> PIM messages are either unicast (e.g.  Registers and RegisterStop), or
> multicast with TTL 1 to the `ALL-PIM-ROUTERS' group (e.g. Join/Prune,
> Asserts, etc.).  The source address used for unicast messages is a
> domain-wide reachable address; the source address used for multicast
> messages is the link-local address of the interface on which the message
                  ~~~~~~~~~
> is being sent.

In case of IPv6, it's explicitly defined in RFC2373, but AFAIK there's
no explicit definition in IPv4.  So can anyone clarify this definition?

If I misunderstand the intention of "link-local" this draft, I'd like
to change the wording in our draft based on that.

Thanks,
----
SUZUKI, Shinsuke @ Hitachi / KAME Project
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Tue Mar 25 02:54:53 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA03078
	for <pim-archive@lists.ietf.org>; Tue, 25 Mar 2003 02:54:52 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2P7mDu78346;
	Mon, 24 Mar 2003 23:48:13 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from dino-pc.procket.com (zeus-ext.procket.com [65.174.124.29])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2P7l0u78312
	for <pim@catarina.usc.edu>; Mon, 24 Mar 2003 23:47:00 -0800 (PST)
	(envelope-from dino@procket.com)
Received: (from dino@localhost)
	by dino-pc.procket.com (8.9.3/8.9.3) id XAA01735;
	Mon, 24 Mar 2003 23:53:02 -0800
Message-Id: <200303250753.XAA01735@dino-pc.procket.com>
X-Authentication-Warning: dino-pc.procket.com: dino set sender to dino@dino-pc.procket.com using -f
From: Dino Farinacci <dino@procket.com>
To: suz@crl.hitachi.co.jp
CC: jinmei@isl.rdc.toshiba.co.jp, jzwiebel@procket.com, lwei@redback.com,
        pim@catarina.usc.edu
In-reply-to: <x7r88voqoq.wl%suz@crl.hitachi.co.jp> (message from SUZUKI
	Shinsuke on Tue, 25 Mar 2003 16:23:17 +0900)
References: <200303190200.h2J20Pf08821@liming-pc.redback.com>
	<BAE51AE0-5A54-11D7-B534-0003936A6062@procket.com>
	<y7vadfnwpno.wl@ocean.jinmei.org> <x7r88voqoq.wl%suz@crl.hitachi.co.jp>
Subject: [pim] Re: 'link-local' in IPv4?
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Mon, 24 Mar 2003 23:53:02 -0800

>> I feel the "link-local" keyword might be one of the reason of the
>> long thread.
>> (Although the problem to be resolve in this draft is caused by the

    I'm using the definition according to RFC 2373, which is specific to IPv6
    only.

>> In case of IPv6, it's explicitly defined in RFC2373, but AFAIK there's
>> no explicit definition in IPv4.  So can anyone clarify this definition?

    All IPv4 non-multicast addresses are routable addresses with some very few
    exceptions (i.e. 127.0.0.1, 255.255.255.255, directed broadcast, etc).

Dino



    
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Tue Mar 25 13:32:45 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA24506
	for <pim-archive@lists.ietf.org>; Tue, 25 Mar 2003 13:32:44 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2PIVpu90961;
	Tue, 25 Mar 2003 10:31:51 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from motgate2.mot.com (motgate2.mot.com [136.182.1.10])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2PIUau90931
	for <pim@catarina.usc.edu>; Tue, 25 Mar 2003 10:30:37 -0800 (PST)
	(envelope-from Chris.Cormier@motorola.com)
Received: from pobox3.mot.com (pobox3.mot.com [10.64.251.242])
	by motgate2.mot.com (Motorola/Motgate2) with ESMTP id h2PIUdxd002139
	for <pim@catarina.usc.edu>; Tue, 25 Mar 2003 11:30:39 -0700 (MST)
Received: [from il06exm06.corp.mot.com (il06exm06.corp.mot.com [10.0.111.6]) by pobox3.mot.com (MOT-pobox3 2.0) with ESMTP id LAA22553 for <pim@catarina.usc.edu>; Tue, 25 Mar 2003 11:28:15 -0700 (MST)]
Received: by il06exm06.corp.mot.com with Internet Mail Service (5.5.2656.59)
	id <HPPHQRAZ>; Tue, 25 Mar 2003 12:29:49 -0600
Message-ID: <FB578B06F252D511B95B009027E3267B0676210E@il06exm04.corp.mot.com>
From: Cormier Chris-ACC125 <Chris.Cormier@motorola.com>
To: "'Dino Farinacci'" <dino@procket.com>, suz@crl.hitachi.co.jp
Cc: jinmei@isl.rdc.toshiba.co.jp, jzwiebel@procket.com, lwei@redback.com,
        pim@catarina.usc.edu
Subject: RE: [pim] Re: 'link-local' in IPv4?
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2656.59)
Content-Type: text/plain
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Tue, 25 Mar 2003 12:29:48 -0600

How about addressing a "box level" IP address that is routable but not on
any local subnet?
-chris

-----Original Message-----
From: Dino Farinacci [mailto:dino@procket.com]
Sent: Monday, March 24, 2003 11:53 PM
To: suz@crl.hitachi.co.jp
Cc: jinmei@isl.rdc.toshiba.co.jp; jzwiebel@procket.com;
lwei@redback.com; pim@catarina.usc.edu
Subject: [pim] Re: 'link-local' in IPv4?


>> I feel the "link-local" keyword might be one of the reason of the
>> long thread.
>> (Although the problem to be resolve in this draft is caused by the

    I'm using the definition according to RFC 2373, which is specific to
IPv6
    only.

>> In case of IPv6, it's explicitly defined in RFC2373, but AFAIK there's
>> no explicit definition in IPv4.  So can anyone clarify this definition?

    All IPv4 non-multicast addresses are routable addresses with some very
few
    exceptions (i.e. 127.0.0.1, 255.255.255.255, directed broadcast, etc).

Dino



    
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Tue Mar 25 14:45:18 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA27865
	for <pim-archive@lists.ietf.org>; Tue, 25 Mar 2003 14:45:17 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2PJgbu92321;
	Tue, 25 Mar 2003 11:42:38 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from dino-pc.procket.com (zeus-ext.procket.com [65.174.124.29])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2PJfMu92289
	for <pim@catarina.usc.edu>; Tue, 25 Mar 2003 11:41:22 -0800 (PST)
	(envelope-from dino@procket.com)
Received: (from dino@localhost)
	by dino-pc.procket.com (8.9.3/8.9.3) id LAA03930;
	Tue, 25 Mar 2003 11:47:24 -0800
Message-Id: <200303251947.LAA03930@dino-pc.procket.com>
X-Authentication-Warning: dino-pc.procket.com: dino set sender to dino@dino-pc.procket.com using -f
From: Dino Farinacci <dino@procket.com>
To: Chris.Cormier@motorola.com
CC: suz@crl.hitachi.co.jp, jinmei@isl.rdc.toshiba.co.jp, jzwiebel@procket.com,
        lwei@redback.com, pim@catarina.usc.edu
In-reply-to: <FB578B06F252D511B95B009027E3267B0676210E@il06exm04.corp.mot.com>
	(message from Cormier Chris-ACC125 on Tue, 25 Mar 2003 12:29:48 -0600)
Subject: Re: [pim] Re: 'link-local' in IPv4?
References:  <FB578B06F252D511B95B009027E3267B0676210E@il06exm04.corp.mot.com>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Tue, 25 Mar 2003 11:47:24 -0800

>> How about addressing a "box level" IP address that is routable but not on
>> any local subnet?
>> -chris

    Unicast protocols would have to do the same. I think this would be too
    big a change to IS-IS/OSPF/BGP for v6.

Dino
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Tue Mar 25 14:48:19 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA28032
	for <pim-archive@lists.ietf.org>; Tue, 25 Mar 2003 14:48:17 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2PJl4u92441;
	Tue, 25 Mar 2003 11:47:04 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from mcs.anl.gov (cliff.mcs.anl.gov [140.221.9.17])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2PJkku92418
	for <pim@catarina.usc.edu>; Tue, 25 Mar 2003 11:46:46 -0800 (PST)
	(envelope-from nickless@mcs.anl.gov)
Received: from newtwinkle.mcs.anl.gov (vpnclient18104.mcs.anl.gov [140.221.18.104])
	by mcs.anl.gov (8.11.6/8.9.3) with ESMTP id h2PJkd4101900;
	Tue, 25 Mar 2003 13:46:39 -0600
Message-Id: <5.1.1.6.2.20030325133451.01926360@pop.mcs.anl.gov>
X-Sender: nickless@pop.mcs.anl.gov
X-Mailer: QUALCOMM Windows Eudora Version 5.1.1
To: lwei@redback.com
From: Bill Nickless <nickless@mcs.anl.gov>
Subject: Re: [pim] San Francisco PIM WG meeting minutes
Cc: pim@catarina.usc.edu
In-Reply-To: <200303242013.h2OKDIT28218@liming-pc.redback.com>
References: <5.1.1.6.2.20030322200815.01e5cc38@pop.mcs.anl.gov>
 <y7v8yv7wp8y.wl@ocean.jinmei.org>
 <BAE51AE0-5A54-11D7-B534-0003936A6062@procket.com>
 <200303192213.h2JMDwq11638@liming-pc.redback.com>
 <y7v8yv7wp8y.wl@ocean.jinmei.org>
 <5.1.1.6.2.20030322200815.01e5cc38@pop.mcs.anl.gov>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Tue, 25 Mar 2003 13:44:10 -0600

At 12:13 PM 3/24/2003 -0800, Liming Wei wrote:
>Bill,
>
>I actually am thinking about cases involving multiple addresses in
>IPv4 environment too :-).
>
>The HSRP or VRRP case sounds unusual. Why does a PIM joiner/pruner
>have to rely on a "static default route" type of unicast routing ?

I've run across a half dozen different situations where departments of an 
institution hate each other, but one still has to depend on the other for 
Internet routing.  Typically, this is a computer science department and a 
computing center.

So they define a DMZ LAN where both sides have routers and both sides 
install static routes toward the other.  Because a dynamic routing protocol 
would require too much trust, one or both sides will configure HSRP/VRRP to 
provide redundancy.

This approach works fine with unicast routing because of how ARP/ND work.

It breaks some existing PIM implementations.

>An implementation today can "solve" this problem by special-knob'ing
>PIM to run from the VRRP active router only, and only "on" the virtual
>address in PIM messages (as opposed to the physical addresses). Of
>course, this is not as general/simple (robust?) as learning all
>addresses via a PIM hello option.

Consider this simple rule:

If a PIM speaker will respond to ARP/ND for a given address, then the PIM 
speaker should process PIM joins/prunes for that address.

This would cover link-local addresses, HSRP/VRRP, and secondary addresses.

Of course, proxy-ARP would be an exception to this rule.

===
Bill Nickless    http://www.mcs.anl.gov/people/nickless      +1 630 252 7390
PGP:0E 0F 16 80 C5 B1 69 52 E1 44 1A A5 0E 1B 74 F7     nickless@mcs.anl.gov

_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Tue Mar 25 14:54:38 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA28373
	for <pim-archive@lists.ietf.org>; Tue, 25 Mar 2003 14:54:37 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2PJr5u92612;
	Tue, 25 Mar 2003 11:53:05 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from motgate.mot.com (motgate.mot.com [129.188.136.100])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2PJqDu92577
	for <pim@catarina.usc.edu>; Tue, 25 Mar 2003 11:52:13 -0800 (PST)
	(envelope-from Chris.Cormier@motorola.com)
Received: from mothost.mot.com (mothost.mot.com [129.188.137.101])
	by motgate.mot.com (Motorola/Motgate) with ESMTP id h2PJqCFG007022
	for <pim@catarina.usc.edu>; Tue, 25 Mar 2003 12:52:13 -0700 (MST)
Received: [from il06exm06.corp.mot.com (il06exm06.corp.mot.com [10.0.111.6]) by mothost.mot.com (MOT-pobox 2.0) with ESMTP id MAA27195 for <pim@catarina.usc.edu>; Tue, 25 Mar 2003 12:52:05 -0700 (MST)]
Received: by il06exm06.corp.mot.com with Internet Mail Service (5.5.2656.59)
	id <HPPHQTF9>; Tue, 25 Mar 2003 13:52:12 -0600
Message-ID: <FB578B06F252D511B95B009027E3267B0676210F@il06exm04.corp.mot.com>
From: Cormier Chris-ACC125 <Chris.Cormier@motorola.com>
To: "'Dino Farinacci'" <dino@procket.com>,
        Cormier Chris-ACC125
	 <Chris.Cormier@motorola.com>
Cc: suz@crl.hitachi.co.jp, jinmei@isl.rdc.toshiba.co.jp, jzwiebel@procket.com,
        lwei@redback.com, pim@catarina.usc.edu
Subject: RE: [pim] Re: 'link-local' in IPv4?
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2656.59)
Content-Type: text/plain
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Tue, 25 Mar 2003 13:52:12 -0600

Unicast routing to these addresses is not a big deal. The fact that these
addresses are not local on any subnet is, when it comes to PIM neighbors.
-cgc

-----Original Message-----
From: Dino Farinacci [mailto:dino@procket.com]
Sent: Tuesday, March 25, 2003 11:47 AM
To: Chris Cormier
Cc: suz@crl.hitachi.co.jp; jinmei@isl.rdc.toshiba.co.jp;
jzwiebel@procket.com; lwei@redback.com; pim@catarina.usc.edu
Subject: Re: [pim] Re: 'link-local' in IPv4?


>> How about addressing a "box level" IP address that is routable but not on
>> any local subnet?
>> -chris

    Unicast protocols would have to do the same. I think this would be too
    big a change to IS-IS/OSPF/BGP for v6.

Dino
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Tue Mar 25 16:17:13 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA03996
	for <pim-archive@lists.ietf.org>; Tue, 25 Mar 2003 16:17:12 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2PLF7u94279;
	Tue, 25 Mar 2003 13:15:07 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from dino-pc.procket.com (zeus-ext.procket.com [65.174.124.29])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2PLDnu94235
	for <pim@catarina.usc.edu>; Tue, 25 Mar 2003 13:13:49 -0800 (PST)
	(envelope-from dino@procket.com)
Received: (from dino@localhost)
	by dino-pc.procket.com (8.9.3/8.9.3) id NAA04612;
	Tue, 25 Mar 2003 13:19:55 -0800
Message-Id: <200303252119.NAA04612@dino-pc.procket.com>
X-Authentication-Warning: dino-pc.procket.com: dino set sender to dino@dino-pc.procket.com using -f
From: Dino Farinacci <dino@procket.com>
To: Chris.Cormier@motorola.com
CC: Chris.Cormier@motorola.com, suz@crl.hitachi.co.jp,
        jinmei@isl.rdc.toshiba.co.jp, jzwiebel@procket.com, lwei@redback.com,
        pim@catarina.usc.edu
In-reply-to: <FB578B06F252D511B95B009027E3267B0676210F@il06exm04.corp.mot.com>
	(message from Cormier Chris-ACC125 on Tue, 25 Mar 2003 13:52:12 -0600)
Subject: Re: [pim] Re: 'link-local' in IPv4?
References:  <FB578B06F252D511B95B009027E3267B0676210F@il06exm04.corp.mot.com>
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Tue, 25 Mar 2003 13:19:55 -0800

>> Unicast routing to these addresses is not a big deal. The fact that these
>> addresses are not local on any subnet is, when it comes to PIM neighbors.
>> -cgc

    Yes, it is. You are changing the next-hop selection for each route that
    OSPF, IS-IS, and BGP add to the routing table.

Dino
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Wed Mar 26 04:33:35 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA19305
	for <pim-archive@lists.ietf.org>; Wed, 26 Mar 2003 04:33:34 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2Q9W8u08341;
	Wed, 26 Mar 2003 01:32:08 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from shuttle.wide.toshiba.co.jp (shuttle.wide.toshiba.co.jp [202.249.10.124])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2Q9Twu08280
	for <pim@catarina.usc.edu>; Wed, 26 Mar 2003 01:29:58 -0800 (PST)
	(envelope-from jinmei@isl.rdc.toshiba.co.jp)
Received: from localhost (kame201.kame.net [203.178.141.201])
	by shuttle.wide.toshiba.co.jp (Postfix) with ESMTP
	id 4E9EB15248; Wed, 26 Mar 2003 18:29:57 +0900 (JST)
Message-ID: <y7v4r5qscem.wl@ocean.jinmei.org>
From: JINMEI Tatuya / =?ISO-2022-JP?B?GyRCP0BMQEMjOkgbKEI=?= <jinmei@isl.rdc.toshiba.co.jp>
To: Pavlin Radoslavov <pavlin@icir.org>
Cc: Dino Farinacci <dino@procket.com>, nbhaskar@cisco.com,
        jzwiebel@procket.com, lwei@redback.com, pim@catarina.usc.edu
Subject: Re: [pim] San Francisco PIM WG meeting minutes 
In-Reply-To: <200303242238.h2OMcJix044976@possum.icir.org>
References: <dino@procket.com>
	 <200303242158.NAA29785@dino-pc.procket.com>
	 <200303242238.h2OMcJix044976@possum.icir.org>
User-Agent: Wanderlust/2.6.1 (Upside Down) Emacs/21.2 Mule/5.0 (SAKAKI)
Organization: Research & Development Center, Toshiba Corp., Kawasaki, Japan.
MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya")
Content-Type: text/plain; charset=US-ASCII
X-Dispatcher: imput version 20000228(IM140)
Lines: 62
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Wed, 26 Mar 2003 18:30:25 +0900

>>>>> On Mon, 24 Mar 2003 14:38:19 -0800, 
>>>>> Pavlin Radoslavov <pavlin@icir.org> said:

>> >> AFAIK, Inverse ND (the generic solution suggested for ipv6) is not
>> >> widely implemented -- we need some way to fix this in the meantime.
>> >> Another issue is that Inverse ND works only on links that have link
>> >> layer addresses. I'm not sure if that will be a limitation for PIM,
>> >> but would be nice not to have this dependency. Also it looks like the
>> >> problem might exist in ipv4 and we don't have a generic solution for
>> >> that.
>> 
>> I think an even shorter-term solution is to send multiple Hellos with
>> different source addresses. Then any router can do the right thing.

> This may not work for things like Prune overriding.
> E.g., lets say that the upstream router addresses are A1 and
> A2. If the downstream router sends a Join(*,G) with the upstream
> router address set to A2, then when the upstream router has to send
> PruneEcho on that interface, it may have to remember to put A2 as
> the upstream router address rather than A1. Well, if another router
> on that LAN uses A1 as the upstream router address for same G, then
> that other router won't override the Prune or the PruneEcho. The
> result: the interface with address A1 and A2 is pruned when it
> should be in Joined state.

Yes, this can happen.

A similar scenario can also happen for assert messages.  If an assert
message is sent from an address of an upstream router which is
different from the upstream address for a downstream router, the
downstream router will not switch the upstream router to the assert
winner.  In order to avoid this, the upstream router would have to
send assert messages using multiple source addresses, just like it
does so for hello messages.

In general, the trade-off between the multiple-source vs hello option
is:

- the multiple-source approach may have a less impact on the PIM
  specification (and implementations), than the hello option.
- the multiple-source approach will require more messages on the wire
  than the hello option approach.
- the multiple-source approach does not provide mapping among multiple
  upstream addresses.  Thus, we need a special handling whenever we
  need such mapping.  The prune-echo and assert cases are examples.

> (Here I should be "fair" to admit that approx a month ago when I
> read draft-suz-pim-upstream-detection-00.txt, my first reaction was
> that the multiple Hellos scheme may be a reasonable work-around that
> wasn't mentioned in the draft, but now I see I was wrong :)

I noticed your message, but had not found time to make a response
before the presentation (sorry about that).  I added the yet-another
solution in the presentation slides, but I could not explain it due to
the time limitation.

Thanks,

					JINMEI, Tatuya
					Communication Platform Lab.
					Corporate R&D Center, Toshiba Corp.
					jinmei@isl.rdc.toshiba.co.jp
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


From pim-admin@catarina.usc.edu  Wed Mar 26 04:35:44 2003
Received: from catarina.usc.edu (catarina.usc.edu [204.57.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA19339
	for <pim-archive@lists.ietf.org>; Wed, 26 Mar 2003 04:35:42 -0500 (EST)
Received: from catarina.usc.edu (localhost.localdomain [127.0.0.1])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2Q9Y7u08404;
	Wed, 26 Mar 2003 01:34:07 -0800 (PST)
	(envelope-from pim-admin@catarina.usc.edu)
Received: from shuttle.wide.toshiba.co.jp (shuttle.wide.toshiba.co.jp [202.249.10.124])
	by catarina.usc.edu (8.11.6/8.11.6) with ESMTP id h2Q9WXu08359
	for <pim@catarina.usc.edu>; Wed, 26 Mar 2003 01:32:34 -0800 (PST)
	(envelope-from jinmei@isl.rdc.toshiba.co.jp)
Received: from localhost (kame201.kame.net [203.178.141.201])
	by shuttle.wide.toshiba.co.jp (Postfix) with ESMTP
	id 3F91615248; Wed, 26 Mar 2003 18:32:39 +0900 (JST)
Message-ID: <y7v3clasca1.wl@ocean.jinmei.org>
From: JINMEI Tatuya / =?ISO-2022-JP?B?GyRCP0BMQEMjOkgbKEI=?= <jinmei@isl.rdc.toshiba.co.jp>
To: lwei@redback.com
Cc: pim@catarina.usc.edu
Subject: Re: [pim] San Francisco PIM WG meeting minutes
In-Reply-To: <200303222127.h2MLRvK24849@liming-pc.redback.com>
References: <BAE51AE0-5A54-11D7-B534-0003936A6062@procket.com>
	 <200303192213.h2JMDwq11638@liming-pc.redback.com>
	 <y7v8yv7wp8y.wl@ocean.jinmei.org>
	 <200303222127.h2MLRvK24849@liming-pc.redback.com>
User-Agent: Wanderlust/2.6.1 (Upside Down) Emacs/21.2 Mule/5.0 (SAKAKI)
Organization: Research & Development Center, Toshiba Corp., Kawasaki, Japan.
MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya")
Content-Type: text/plain; charset=US-ASCII
X-Dispatcher: imput version 20000228(IM140)
Lines: 29
Sender: pim-admin@catarina.usc.edu
Errors-To: pim-admin@catarina.usc.edu
X-BeenThere: pim@catarina.usc.edu
X-Mailman-Version: 2.0.13
Precedence: bulk
List-Unsubscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=unsubscribe>
List-Id: <pim.catarina.usc.edu>
List-Post: <mailto:pim@catarina.usc.edu>
List-Help: <mailto:pim-request@catarina.usc.edu?subject=help>
List-Subscribe: <http://catarina.usc.edu/mailman/listinfo/pim>,
	<mailto:pim-request@catarina.usc.edu?subject=subscribe>
List-Archive: <http://catarina.usc.edu/pipermail/pim/>
Date: Wed, 26 Mar 2003 18:33:10 +0900

>>>>> On Sat, 22 Mar 2003 13:27:57 -0800 (PST), 
>>>>> Liming Wei <lwei@redback.com> said:

> 1) There exists multple addresses on an interface in IPv6;
> 2) We need to specify how PIM works under such scenarios.

> Else, it will be a mess in the field when it comes to
> interoperability. In short, a solution on 2) should be discussed
> in the PIM WG.

(snip)

> Can you coordinate with Mark et al on this issue ?

Sure, then shall we privately contact Mark for the next step, or
continue the discussion here in the ML for a while?

> About the disadvantages (or benefits) of multiple addresses in IPv6,
> it is interesting (boring after a couple of messages in thread) to be
> informed of the activities elsewhere, but lengthy discussions here
> will not change the outcome in IPv6's design, or zeroconf's design.

I agree, and sorry for my lengthy comments on this.  I'll refrain from
mentioning the off-topic in this list.

					JINMEI, Tatuya
					Communication Platform Lab.
					Corporate R&D Center, Toshiba Corp.
					jinmei@isl.rdc.toshiba.co.jp
_______________________________________________
pim mailing list
pim@catarina.usc.edu
http://catarina.usc.edu/mailman/listinfo/pim


