
From internet-drafts@ietf.org  Tue Jan  1 20:49:08 2013
Return-Path: <internet-drafts@ietf.org>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0BC7221E8083; Tue,  1 Jan 2013 20:49:08 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.525
X-Spam-Level: 
X-Spam-Status: No, score=-102.525 tagged_above=-999 required=5 tests=[AWL=0.074, BAYES_00=-2.599, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id PmDDyoS1pvZW; Tue,  1 Jan 2013 20:49:07 -0800 (PST)
Received: from ietfa.amsl.com (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7251221F8C04; Tue,  1 Jan 2013 20:49:07 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
From: internet-drafts@ietf.org
To: i-d-announce@ietf.org
X-Test-IDTracker: no
X-IETF-IDTracker: 4.37
Message-ID: <20130102044907.24997.40697.idtracker@ietfa.amsl.com>
Date: Tue, 01 Jan 2013 20:49:07 -0800
Cc: trill@ietf.org
Subject: [trill] I-D Action: draft-ietf-trill-rbridge-vlan-mapping-08.txt
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 02 Jan 2013 04:49:08 -0000

A New Internet-Draft is available from the on-line Internet-Drafts director=
ies.
 This draft is a work item of the Transparent Interconnection of Lots of Li=
nks Working Group of the IETF.

	Title           : TRILL: Campus VLAN and Priority Regions
	Author(s)       : Radia Perlman
                          Anil Rijhsinghani
                          Donald Eastlake
                          Ayan Banerjee
                          Dinesh G. Dutt
	Filename        : draft-ietf-trill-rbridge-vlan-mapping-08.txt
	Pages           : 19
	Date            : 2013-01-01

Abstract:
   Within a TRILL campus, the VLAN and priority of TRILL encapsulated
   frames is preserved. However, in some cases it may be desired that
   data VLAN and/or priority be mapped at the boundary between regions
   of such a campus. This document describes an optional RBridge feature
   to provide this function.



The IETF datatracker status page for this draft is:
https://datatracker.ietf.org/doc/draft-ietf-trill-rbridge-vlan-mapping

There's also a htmlized version available at:
http://tools.ietf.org/html/draft-ietf-trill-rbridge-vlan-mapping-08

A diff from the previous version is available at:
http://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-trill-rbridge-vlan-mapping-08


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


From Kesava_Vijaya_Krupak@Dell.com  Thu Jan  3 02:48:56 2013
Return-Path: <Kesava_Vijaya_Krupak@Dell.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id ACE1521F8715 for <trill@ietfa.amsl.com>; Thu,  3 Jan 2013 02:48:56 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -110.598
X-Spam-Level: 
X-Spam-Status: No, score=-110.598 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-8, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ucbGtLeCu+rm for <trill@ietfa.amsl.com>; Thu,  3 Jan 2013 02:48:56 -0800 (PST)
Received: from ausxippc101.us.dell.com (ausxippc101.us.dell.com [143.166.85.207]) by ietfa.amsl.com (Postfix) with ESMTP id 16A4921F844F for <trill@ietf.org>; Thu,  3 Jan 2013 02:48:55 -0800 (PST)
X-LoopCount0: from 10.170.28.40
X-IronPort-AV: E=Sophos;i="4.84,402,1355119200"; d="scan'208,217";a="46624544"
From: <Kesava_Vijaya_Krupak@Dell.com>
To: <trill@ietf.org>
Date: Thu, 3 Jan 2013 16:18:42 +0530
Thread-Topic: New version : TRILL Smart Endnodes draft
Thread-Index: Ac3pn3YBNeqnV+E/RhyPSQ5orD2kkA==
Message-ID: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US
Content-Type: multipart/alternative; boundary="_000_D5A6F3355F664C40AFB65BB1277D8D450186C7F7ADMAAX7MCDC101A_"
MIME-Version: 1.0
Cc: d3e3e3@gmail.com, radiaperlman@gmail.com, hu.fangwei@zte.com.cn
Subject: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 03 Jan 2013 10:48:56 -0000

--_000_D5A6F3355F664C40AFB65BB1277D8D450186C7F7ADMAAX7MCDC101A_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Hi all,

A new version of the Trill Smart Endnodes draft has been posted.

http://datatracker.ietf.org/doc/draft-perlman-trill-smart-endnodes

Kindly let us know your comments.

Thanks,
kvk

--_000_D5A6F3355F664C40AFB65BB1277D8D450186C7F7ADMAAX7MCDC101A_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micr=
osoft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" xmlns=3D"http:=
//www.w3.org/TR/REC-html40"><head><meta http-equiv=3DContent-Type content=
=3D"text/html; charset=3Dus-ascii"><meta name=3DGenerator content=3D"Micros=
oft Word 14 (filtered medium)"><style><!--
/* Font Definitions */
@font-face
	{font-family:SimSun;
	panose-1:2 1 6 0 3 1 1 1 1 1;}
@font-face
	{font-family:SimSun;
	panose-1:2 1 6 0 3 1 1 1 1 1;}
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:"\@SimSun";
	panose-1:2 1 6 0 3 1 1 1 1 1;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:11.0pt;
	font-family:"Calibri","sans-serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
span.EmailStyle17
	{mso-style-type:personal-compose;
	font-family:"Calibri","sans-serif";
	color:windowtext;}
.MsoChpDefault
	{mso-style-type:export-only;
	font-family:"Calibri","sans-serif";}
@page WordSection1
	{size:8.5in 11.0in;
	margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
	{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]--></head><body lang=3DEN-US link=3Dblue vli=
nk=3Dpurple><div class=3DWordSection1><p class=3DMsoNormal>Hi all,<o:p></o:=
p></p><p class=3DMsoNormal><o:p>&nbsp;</o:p></p><p class=3DMsoNormal>A new =
version of the Trill Smart Endnodes draft has been posted.<o:p></o:p></p><p=
 class=3DMsoNormal><o:p>&nbsp;</o:p></p><p class=3DMsoNormal><a href=3D"htt=
p://datatracker.ietf.org/doc/draft-perlman-trill-smart-endnodes">http://dat=
atracker.ietf.org/doc/draft-perlman-trill-smart-endnodes</a><o:p></o:p></p>=
<p class=3DMsoNormal><o:p>&nbsp;</o:p></p><p class=3DMsoNormal>Kindly let u=
s know your comments.<o:p></o:p></p><p class=3DMsoNormal><o:p>&nbsp;</o:p><=
/p><p class=3DMsoNormal>Thanks,<o:p></o:p></p><p class=3DMsoNormal>kvk<o:p>=
</o:p></p></div></body></html>=

--_000_D5A6F3355F664C40AFB65BB1277D8D450186C7F7ADMAAX7MCDC101A_--

From pkoganti@Brocade.COM  Thu Jan  3 03:20:00 2013
Return-Path: <pkoganti@Brocade.COM>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4797F21F8431 for <trill@ietfa.amsl.com>; Thu,  3 Jan 2013 03:20:00 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.264
X-Spam-Level: 
X-Spam-Status: No, score=-3.264 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, IP_NOT_FRIENDLY=0.334, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0e5-R1XL7sFi for <trill@ietfa.amsl.com>; Thu,  3 Jan 2013 03:19:59 -0800 (PST)
Received: from mx0b-000f0801.pphosted.com (mx0b-000f0801.pphosted.com [67.231.152.113]) by ietfa.amsl.com (Postfix) with ESMTP id 0012521F8B19 for <trill@ietf.org>; Thu,  3 Jan 2013 03:19:53 -0800 (PST)
Received: from pps.filterd (m0000700 [127.0.0.1]) by mx0b-000f0801.pphosted.com (8.14.5/8.14.5) with SMTP id r03BHHID015656; Thu, 3 Jan 2013 03:19:50 -0800
Received: from hq1wp-exchub01.corp.brocade.com ([144.49.131.13]) by mx0b-000f0801.pphosted.com with ESMTP id 19kt2711qf-1 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Thu, 03 Jan 2013 03:19:50 -0800
Received: from HQ1WP-EXHUB01.corp.brocade.com (10.70.36.14) by HQ1WP-EXCHUB01.corp.brocade.com (10.70.36.99) with Microsoft SMTP Server (TLS) id 14.2.309.2; Thu, 3 Jan 2013 03:19:48 -0800
Received: from HQ1-EXCH02.corp.brocade.com ([fe80::c92a:772e:befa:c34c]) by HQ1WP-EXHUB01.corp.brocade.com ([::1]) with mapi; Thu, 3 Jan 2013 03:19:48 -0800
From: Phanidhar Koganti <pkoganti@Brocade.COM>
To: "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@Dell.com>, "trill@ietf.org" <trill@ietf.org>
Date: Thu, 3 Jan 2013 03:19:45 -0800
Thread-Topic: New version : TRILL Smart Endnodes draft
Thread-Index: Ac3pn3YBNeqnV+E/RhyPSQ5orD2kkAABB8iA
Message-ID: <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM>
In-Reply-To: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US
Content-Type: multipart/alternative; boundary="_000_6895EAE0CA8DEE40B92D7CA88BB521F32414007827HQ1EXCH02corp_"
MIME-Version: 1.0
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:5.9.8327, 1.0.431, 0.0.0000 definitions=2013-01-03_01:2013-01-03, 2013-01-03, 1970-01-01 signatures=0
X-Proofpoint-Spam-Details: rule=notspam policy=default score=0 spamscore=0 suspectscore=49 phishscore=0 adultscore=0 bulkscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=7.0.1-1211240000 definitions=main-1301030048
Cc: "d3e3e3@gmail.com" <d3e3e3@gmail.com>, "radiaperlman@gmail.com" <radiaperlman@gmail.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>
Subject: Re: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 03 Jan 2013 11:20:00 -0000

--_000_6895EAE0CA8DEE40B92D7CA88BB521F32414007827HQ1EXCH02corp_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

The draft defines the control and data path behavior from smart end-node to=
 edge-RB. When the packet reaches the egress RB I am presuming there will b=
e a inner-MAC lookup to deliver the packet to the egress end-node. Edge-RB =
to smart endnode cannot have TRILL encapsulation right? If so for my clarif=
ication we are saying the traffic towards the end-node will be non-TRILL wh=
ile towards the ingress edge-RB will be TRILL encapsulated?

Phani

From: trill-bounces@ietf.org [mailto:trill-bounces@ietf.org] On Behalf Of K=
esava_Vijaya_Krupak@Dell.com
Sent: Thursday, January 03, 2013 4:19 PM
To: trill@ietf.org
Cc: d3e3e3@gmail.com; radiaperlman@gmail.com; hu.fangwei@zte.com.cn
Subject: [trill] New version : TRILL Smart Endnodes draft

Hi all,

A new version of the Trill Smart Endnodes draft has been posted.

http://datatracker.ietf.org/doc/draft-perlman-trill-smart-endnodes

Kindly let us know your comments.

Thanks,
kvk

--_000_6895EAE0CA8DEE40B92D7CA88BB521F32414007827HQ1EXCH02corp_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micr=
osoft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" xmlns=3D"http:=
//www.w3.org/TR/REC-html40"><head><META HTTP-EQUIV=3D"Content-Type" CONTENT=
=3D"text/html; charset=3Dus-ascii"><meta name=3DGenerator content=3D"Micros=
oft Word 14 (filtered medium)"><style><!--
/* Font Definitions */
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:Tahoma;
	panose-1:2 11 6 4 3 5 4 4 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:11.0pt;
	font-family:"Calibri","sans-serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
span.EmailStyle17
	{mso-style-type:personal;
	font-family:"Calibri","sans-serif";
	color:windowtext;}
span.EmailStyle18
	{mso-style-type:personal-reply;
	font-family:"Calibri","sans-serif";
	color:#1F497D;}
.MsoChpDefault
	{mso-style-type:export-only;
	font-size:10.0pt;}
@page WordSection1
	{size:8.5in 11.0in;
	margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
	{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]--></head><body lang=3DEN-US link=3Dblue vli=
nk=3Dpurple><div class=3DWordSection1><p class=3DMsoNormal><span style=3D'c=
olor:#1F497D'>The draft defines the control and data path behavior from sma=
rt end-node to edge-RB. When the packet reaches the egress RB I am presumin=
g there will be a inner-MAC lookup to deliver the packet to the egress end-=
node. Edge-RB to smart endnode cannot have TRILL encapsulation right? If so=
 for my clarification we are saying the traffic towards the end-node will b=
e non-TRILL while towards the ingress edge-RB will be TRILL encapsulated?<o=
:p></o:p></span></p><p class=3DMsoNormal><span style=3D'color:#1F497D'><o:p=
>&nbsp;</o:p></span></p><p class=3DMsoNormal><span style=3D'color:#1F497D'>=
Phani<o:p></o:p></span></p><p class=3DMsoNormal><span style=3D'color:#1F497=
D'><o:p>&nbsp;</o:p></span></p><div><div style=3D'border:none;border-top:so=
lid #B5C4DF 1.0pt;padding:3.0pt 0in 0in 0in'><p class=3DMsoNormal><b><span =
style=3D'font-size:10.0pt;font-family:"Tahoma","sans-serif"'>From:</span></=
b><span style=3D'font-size:10.0pt;font-family:"Tahoma","sans-serif"'> trill=
-bounces@ietf.org [mailto:trill-bounces@ietf.org] <b>On Behalf Of </b>Kesav=
a_Vijaya_Krupak@Dell.com<br><b>Sent:</b> Thursday, January 03, 2013 4:19 PM=
<br><b>To:</b> trill@ietf.org<br><b>Cc:</b> d3e3e3@gmail.com; radiaperlman@=
gmail.com; hu.fangwei@zte.com.cn<br><b>Subject:</b> [trill] New version : T=
RILL Smart Endnodes draft<o:p></o:p></span></p></div></div><p class=3DMsoNo=
rmal><o:p>&nbsp;</o:p></p><p class=3DMsoNormal>Hi all,<o:p></o:p></p><p cla=
ss=3DMsoNormal><o:p>&nbsp;</o:p></p><p class=3DMsoNormal>A new version of t=
he Trill Smart Endnodes draft has been posted.<o:p></o:p></p><p class=3DMso=
Normal><o:p>&nbsp;</o:p></p><p class=3DMsoNormal><a href=3D"http://datatrac=
ker.ietf.org/doc/draft-perlman-trill-smart-endnodes">http://datatracker.iet=
f.org/doc/draft-perlman-trill-smart-endnodes</a><o:p></o:p></p><p class=3DM=
soNormal><o:p>&nbsp;</o:p></p><p class=3DMsoNormal>Kindly let us know your =
comments.<o:p></o:p></p><p class=3DMsoNormal><o:p>&nbsp;</o:p></p><p class=
=3DMsoNormal>Thanks,<o:p></o:p></p><p class=3DMsoNormal>kvk<o:p></o:p></p><=
/div></body></html>=

--_000_6895EAE0CA8DEE40B92D7CA88BB521F32414007827HQ1EXCH02corp_--

From ghanwani@gmail.com  Fri Jan  4 17:48:05 2013
Return-Path: <ghanwani@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5AE9721F875A for <trill@ietfa.amsl.com>; Fri,  4 Jan 2013 17:48:05 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.977
X-Spam-Level: 
X-Spam-Status: No, score=-2.977 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id UplOltVeD4Gl for <trill@ietfa.amsl.com>; Fri,  4 Jan 2013 17:48:04 -0800 (PST)
Received: from mail-ia0-f169.google.com (mail-ia0-f169.google.com [209.85.210.169]) by ietfa.amsl.com (Postfix) with ESMTP id 93B8121F872E for <trill@ietf.org>; Fri,  4 Jan 2013 17:48:04 -0800 (PST)
Received: by mail-ia0-f169.google.com with SMTP id u20so7805057iag.14 for <trill@ietf.org>; Fri, 04 Jan 2013 17:48:04 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:x-received:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=uqivG+kC2kwbZuXPC1DENt8iOOIsL82CysHEmKtPg/Q=; b=dvA75sqL6ZGPdwclf+UxvGOuNywiTZvsQVC5MhmzbOeSFA/cennxxpOagu3arUQyH/ FBR/WwN7+CfuiiePZKDcqjXo3Si6k3G31aygSsUI14Ruq+gmOAne8X9h1EYCnLt/G/PH o1Ud4EdhOfbJI7WcUy1N9+yq18/Jo5KZJgJjx5YgbFOlCeADN+dXolS50KafMPvg156n 09Fc1iAIRLS5cSXzZymc/xiDiHPTXtNBLukDfFW4MBzvZ7EU1HKGOFmDhHIRlbAweq9K arMlq6YWK89QCJDwh878KXlXQWYgJzFCk+bL2D7l8zKZuZ0ULTZC39s3S/6WygagTfTx eSZg==
MIME-Version: 1.0
X-Received: by 10.50.91.198 with SMTP id cg6mr277363igb.102.1357350483967; Fri, 04 Jan 2013 17:48:03 -0800 (PST)
Sender: ghanwani@gmail.com
Received: by 10.50.182.201 with HTTP; Fri, 4 Jan 2013 17:48:03 -0800 (PST)
In-Reply-To: <FBEA3E19AA24F847BA3AE74E2FE19356288937A0@xmb-rcd-x08.cisco.com>
References: <CAFOuuo4zvX5AtD-oGRRftuZaKmhY7C7-SvDjznMOdzUj+Q3fGQ@mail.gmail.com> <FBEA3E19AA24F847BA3AE74E2FE1935628892DF6@xmb-rcd-x08.cisco.com> <CAFOuuo5LP1EzajpeBri2KhTT-wf+vv=JwmTLma9_mxg7dM5PvQ@mail.gmail.com> <FBEA3E19AA24F847BA3AE74E2FE1935628892EAE@xmb-rcd-x08.cisco.com> <4552F0907735844E9204A62BBDD325E732138B02@SZXEML507-MBS.china.huawei.com> <EE1367B6-5498-492E-A57A-155312162CFC@gmail.com> <201212122025.qBCKPmM7013618@cichlid.raleigh.ibm.com> <0E430FEE-F2AD-4EA0-9E98-50762B563E9B@gmail.com> <201212122257.qBCMvdeW014734@cichlid.raleigh.ibm.com> <FBEA3E19AA24F847BA3AE74E2FE19356288937A0@xmb-rcd-x08.cisco.com>
Date: Fri, 4 Jan 2013 17:48:03 -0800
X-Google-Sender-Auth: pzlHapki4cz8X1juJu7Y1DkWpNY
Message-ID: <CA+-tSzz4oHsYuKus34hqHDUM1PC9v-WGnUWcwZSd5qdQBA=nDw@mail.gmail.com>
From: Anoop Ghanwani <anoop@alumni.duke.edu>
To: "Tissa Senevirathne (tsenevir)" <tsenevir@cisco.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Cc: Thomas Narten <narten@us.ibm.com>, Sam Aldrin <aldrin.ietf@gmail.com>, "trill@ietf.org" <trill@ietf.org>, Radia Perlman <radiaperlman@gmail.com>, Mingui Zhang <zhangmingui@huawei.com>
Subject: Re: [trill] Thoughts on active-active edge
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 05 Jan 2013 01:48:05 -0000

Chiming in late on an old thread as I took some time off.

I agree with Tissa.  It's not a good idea to trigger MAC moves
unnecessarily and poke the CPU, even if the CPU will eventually
ignore it.  Just not good design.  And it would be non-trivial
hardware change to address something like that.  (Otherwise,
we'd have IP RPF checks that worked with ECMP much
more widely available in hardware.)

Anoop

On Wed, Dec 12, 2012 at 3:14 PM, Tissa Senevirathne (tsenevir)
<tsenevir@cisco.com> wrote:
> I think we should answer the question in hand first.
>
> This thread is all about "saying it is ok to have MAC move in TRILL to so=
lve active-active edge"
>
> I am overwhelmingly against that statement and the argument for it is all=
 based on whole series of assumptions and ignoring the modern networking de=
vices and how they operate.
>
> Once that is resolved and behind we can have a separate discussion of CMT=
, hu draft and future approach etc.
>
> Thanks
> Tissa
>
> -----Original Message-----
> From: Thomas Narten [mailto:narten@us.ibm.com]
> Sent: Wednesday, December 12, 2012 2:58 PM
> To: Sam Aldrin
> Cc: Mingui Zhang; Tissa Senevirathne (tsenevir); Radia Perlman; trill@iet=
f.org
> Subject: Re: [trill] Thoughts on active-active edge
>
>
> Sam Aldrin <aldrin.ietf@gmail.com> writes:
>
>> On Dec 12, 2012, at 12:25 PM, Thomas Narten <narten@us.ibm.com> wrote:
>
>> >> I believe we have discussed these scenarios (different solution
>> >> types) previously.  WG should reach consensus and move forward with
>> >> the agreed solution.  Atleast from the IETF WG sessions, I presumed
>> >> there was a consensus on the solution.  Am I wrong in that
>> >> assumption?
>> >
>> > what "solution" are you referring above as maybe having consensus?
>
>> Cmt draft
>
> If you mean draft-ietf-trill-cmt-01.txt, I at least am not in agreement.
>
> I have several issues with the document.
>
> For starters, it's incomplete. It needs to be paired with another documen=
t to get an actual "standard" you can implement. We don't have such a compa=
nion document yet.  It's been suggested that draft-hu-trill-pseudonode-nick=
name-04.txt is one such document. But that document is not a WG document an=
d I'm not sure what its status is. (And it needs work...)
>
> As a general rule (not TRILL specific) I don't like WGs doing "partial wo=
rk" where they standardize one document, that is incomplete by itself, inde=
pendent of other documents that are needed to produce a complete implementa=
ble solution to some real problem.
>
> I've raised issues on the list about CMT and its unclarity wrt. LAG/MC-LA=
G, etc. I did not really get my questions answered.
>
> What I'd really like to start with on CMT is a clear problem statement. I=
.e., exactly what problem/use case it is trying to solve. I have not been a=
ble to get that out of the two  documents.
>
> As for draft-hu-trill-pseudonode-nickname-04.txt, I have other concerns. =
E.g., it talks about tunneling at one point, but that is completely unspeci=
fied.
>
> What I don't want to do is spend a lot of time on the *solution* details.=
 I first want a clear problem statement/use case. Without that, it is diffi=
cult to assess how well potential solutions address the stated problem.
>
> Thomas
>
>
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill

From d3e3e3@gmail.com  Mon Jan  7 19:25:29 2013
Return-Path: <d3e3e3@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 266A121F871D for <trill@ietfa.amsl.com>; Mon,  7 Jan 2013 19:25:29 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.599
X-Spam-Level: 
X-Spam-Status: No, score=-103.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id oYOCvAGo5Yq4 for <trill@ietfa.amsl.com>; Mon,  7 Jan 2013 19:25:28 -0800 (PST)
Received: from mail-ob0-f182.google.com (mail-ob0-f182.google.com [209.85.214.182]) by ietfa.amsl.com (Postfix) with ESMTP id 6E7ED21F872E for <trill@ietf.org>; Mon,  7 Jan 2013 19:25:28 -0800 (PST)
Received: by mail-ob0-f182.google.com with SMTP id 16so18406417obc.13 for <trill@ietf.org>; Mon, 07 Jan 2013 19:25:27 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:from:date:message-id:subject:to:content-type; bh=OrX0k1oUvTIHrZ9i0YRk5bNy/w/nP0q/Md0SvlDM7/8=; b=tKgMWxL1cV3/E+m0TidBJ1OPin0He/+Y7QttcCHIDFMcxp+iiYSTwUZFSuZx7EldPo IOVsCIT/brjWC2iQTxLPDkVez1XtOKGQdIK121MM+i/0ZZVwwdzD9Mp9oteuy9JhPdAs 9Q5FXqs1zNnexx66P8GcsbSDeUGvnGvDPwUDYs4YgHnjAlVuj+LeNMWlkKBleuIrU8eV YI8/cTW2O4iYREMCe9KjFufRSC6i21+ST2NGgSW9cAYokiBNOvEgiuRl/Pbhd4iPRE6j njR61BC+JbDaqy2+KaPR93TzSyRYwfaHwJHO39XHYNsrSd4Dvxbhi2yywArHJN+VeQ4p j6Zw==
Received: by 10.182.194.2 with SMTP id hs2mr43925353obc.97.1357615527708; Mon, 07 Jan 2013 19:25:27 -0800 (PST)
MIME-Version: 1.0
Received: by 10.76.144.105 with HTTP; Mon, 7 Jan 2013 19:25:07 -0800 (PST)
From: Donald Eastlake <d3e3e3@gmail.com>
Date: Mon, 7 Jan 2013 22:25:07 -0500
Message-ID: <CAF4+nEEaYi5E7-pFwfa1-pzFyCx94KXdcH5N2Zg+2V=KUnfBcA@mail.gmail.com>
To: trill@ietf.org
Content-Type: text/plain; charset=ISO-8859-1
Subject: [trill] Suggest improvements to draft-ietf-trill-esadi-01
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 08 Jan 2013 03:25:29 -0000

Hi,

Here are three suggested changes to draft-ietf-trill-esadi-01. I will
post a separate message on each so they can be discussed in separate
threads except for item 3 which I think is pretty straight forward.

1. Simplify the ESADI neighbor mechanism.

2. Expand the number of ESADI-LSP fragments available.

3. ESADI current provides only VLAN scoped flooding. This should be
expanded to also cover Fine Grained Label scoped flooding.

Thanks,
Donald
=============================
 Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
 155 Beaver Street, Milford, MA 01757 USA
 d3e3e3@gmail.com

From d3e3e3@gmail.com  Mon Jan  7 19:25:53 2013
Return-Path: <d3e3e3@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E6C2E21F872E for <trill@ietfa.amsl.com>; Mon,  7 Jan 2013 19:25:53 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.599
X-Spam-Level: 
X-Spam-Status: No, score=-103.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Cq4t+FN8gPoQ for <trill@ietfa.amsl.com>; Mon,  7 Jan 2013 19:25:53 -0800 (PST)
Received: from mail-oa0-f50.google.com (mail-oa0-f50.google.com [209.85.219.50]) by ietfa.amsl.com (Postfix) with ESMTP id 638CD21F871D for <trill@ietf.org>; Mon,  7 Jan 2013 19:25:53 -0800 (PST)
Received: by mail-oa0-f50.google.com with SMTP id n16so19014272oag.9 for <trill@ietf.org>; Mon, 07 Jan 2013 19:25:53 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:from:date:message-id:subject:to:content-type; bh=+HP9R7aNndrEIYJxj8L8VQvgQGvdz/AoMrTw7lnF0qw=; b=uG2epl2JbXbftK6BqrDKpgj0AexH3fZX0fKodKko0INFHzSx3CDlu+dTAG3qXwR08w 1MYoC6pyjQZ3OclbVqLgOG6JVwdrQ7QueLeOtIxmeHnY0L4zmOyWSspVCkVYsHh5QedN kdNHI4N4vma4UluCTOcSuvw4KhaZvPQurH7GHr9P2NJrUheHlVRXX7E2o5SowTq/ppw+ WPxTiPb3NzihVZnjxNvjmTPwRGmx38V7dCDmgcZVrfUj/sf04sxemcdjchjc6xZJ95U0 SNb8D0G2k12XyhuVzJHGA9IttXI7lr+x9xgnw4vRTpYUJVti6bGtrL/ggjfjYOmFcIsZ 8mBg==
Received: by 10.182.115.5 with SMTP id jk5mr2997329obb.7.1357615553020; Mon, 07 Jan 2013 19:25:53 -0800 (PST)
MIME-Version: 1.0
Received: by 10.76.144.105 with HTTP; Mon, 7 Jan 2013 19:25:32 -0800 (PST)
From: Donald Eastlake <d3e3e3@gmail.com>
Date: Mon, 7 Jan 2013 22:25:32 -0500
Message-ID: <CAF4+nEH8f_p_Fexo1d12CpaAdSD2wsa0_HK3dpO57KpTxgX55A@mail.gmail.com>
To: trill@ietf.org
Content-Type: text/plain; charset=ISO-8859-1
Subject: [trill] ESADI 1. Simplify ESADI Neighbor determination
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 08 Jan 2013 03:25:54 -0000

1. Simplify ESADI Neighbor determination

      The current draft requires that RBridge RB1 consider RB2 an ESADI
      neighbor for VLAN-X only if three conditions are met: (1) RB2 is
      data reachable from RB1, (2) RB2 has the ESADI participation bit
      on in its Interested VLANs sub-TLV for VLAN-X, and (3) RB1 is
      holding ESADI-LSP fragment zero for RB2. This last requirement (3)
      increases the state complexity of ESADI and the suggestion is
      that it be eliminated. If RB1 does not hold an ESADI-LSP fragment
      zero with an ESADI Parameters APPsub-TLV in it for RB2, then it
      treats RB2 as having default priority and CSNP time.

Thanks,
Donald
=============================
 Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
 155 Beaver Street, Milford, MA 01757 USA
 d3e3e3@gmail.com

From d3e3e3@gmail.com  Mon Jan  7 19:26:42 2013
Return-Path: <d3e3e3@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 440F121F872E for <trill@ietfa.amsl.com>; Mon,  7 Jan 2013 19:26:42 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.599
X-Spam-Level: 
X-Spam-Status: No, score=-103.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id iqpDPkky4trV for <trill@ietfa.amsl.com>; Mon,  7 Jan 2013 19:26:41 -0800 (PST)
Received: from mail-oa0-f42.google.com (mail-oa0-f42.google.com [209.85.219.42]) by ietfa.amsl.com (Postfix) with ESMTP id C0D4A21F871D for <trill@ietf.org>; Mon,  7 Jan 2013 19:26:41 -0800 (PST)
Received: by mail-oa0-f42.google.com with SMTP id j1so18920127oag.15 for <trill@ietf.org>; Mon, 07 Jan 2013 19:26:41 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:from:date:message-id:subject:to:content-type; bh=R/RDI/zv06hFCM7NB2CGjvNe3epi2DxdAI97cu5lliA=; b=IkPXQQeS4GRHdTQkk6R/1agRAFdmTnHctUm0/ljveQQRZPprsqK203WHnoHCxa/z4o CeGp2/cH4OuDhS8+nSBV4k467MtWyeFoR/2rhvHsOS/e7JgR48aFwHG515wH0LzOSgmP DQdXb6xMkXobmt9yGMx/jBW3ye+u/8gGIGrBRbGQLeceWvLLNWyaJDQhwWED/VzZMxRz k8YF70MJozZBkizTyBbi/T2C0XIVOVWAsn1nThtZFln/ORDxdmLKnpycxLDGnC/5EAux KnY26KFtymiGBAJV883WgTvcIeuqKLbExwcNLV43zaiU5RARagbLJyw9iZxLNIBhYo7C G0Dg==
Received: by 10.182.0.19 with SMTP id 19mr38636340oba.15.1357615601411; Mon, 07 Jan 2013 19:26:41 -0800 (PST)
MIME-Version: 1.0
Received: by 10.76.144.105 with HTTP; Mon, 7 Jan 2013 19:26:21 -0800 (PST)
From: Donald Eastlake <d3e3e3@gmail.com>
Date: Mon, 7 Jan 2013 22:26:21 -0500
Message-ID: <CAF4+nEE2R1Fc8mjuOUfX6m5nwbnbw38oJj=aDPTpkWefLjeHRA@mail.gmail.com>
To: trill@ietf.org
Content-Type: text/plain; charset=ISO-8859-1
Subject: [trill] ESADI 2. More ESADI-LSP fragments
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 08 Jan 2013 03:26:42 -0000

Expand ESADI data to 64K fragments.

      ESADI does no routing so there is never a pseudo-node on the
      ESADI pseudo-link. This allows re-purposing one byte of the LSP
      ID field to expand the number of possible fragments from 2**8 to
      2**16. In the IS-IS standard, the "LSP ID" field is defined as 8
      bytes. The top 6 are the router's System ID, the next byte is
      non-zero for pseudo-nodes, and the bottom byte is the fragment
      number. Change this so that in ESADI, there is just the System
      ID followed by two bytes of fragment number.  The same change is
      made to the LSP ID field of the Remaining Lifetime TLV which is
      used in EASDI-CSNP and ESADI-PSNP. The bottom byte of the ESADI
      CSNP/PSNP Header Source ID is always zero.

Thanks,
Donald
=============================
 Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
 155 Beaver Street, Milford, MA 01757 USA
 d3e3e3@gmail.com

From ayabaner@gmail.com  Tue Jan  8 12:51:25 2013
Return-Path: <ayabaner@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9F9161F0C6A for <trill@ietfa.amsl.com>; Tue,  8 Jan 2013 12:51:25 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.598
X-Spam-Level: 
X-Spam-Status: No, score=-3.598 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id WJT7D+DN1NQQ for <trill@ietfa.amsl.com>; Tue,  8 Jan 2013 12:51:25 -0800 (PST)
Received: from mail-vc0-f172.google.com (mail-vc0-f172.google.com [209.85.220.172]) by ietfa.amsl.com (Postfix) with ESMTP id E5C1D1F0C3E for <trill@ietf.org>; Tue,  8 Jan 2013 12:51:24 -0800 (PST)
Received: by mail-vc0-f172.google.com with SMTP id fw7so874067vcb.17 for <trill@ietf.org>; Tue, 08 Jan 2013 12:51:24 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=m9ECbcOiraCtA2IWWD4vmHBPZEp6q3ygasM97dEUclI=; b=PIkM6UOTOXqfy5sA9kKjNAAhgQcmAiz7dawmVRh+yB5TR3VHxYoURRfE1ZucOSTHhS ZiC+kn6pV3Tl8s43ElFDYs7Y9manWDObt2lxJ21NzmaZng8x8ByLqtMpQImFmLLq61gs z9jHBw5lUahgFS7Ye5AVCBNDJfk2mwMpQ/hLHOsR0/9y7qPV6/3lDc1skliW9t8jI0MT 0Zl/WkNQI5roKEGDQDWaS61bDL63R11o47jXwR93EWMVWdKenMqfMK8m3BHHankIbnta Bzjz0ejpFxgjYzNFKMOuurfQRL/jtndwdU4qvXgZ7OdYyU5qa4gEFem30QU047FzUXy9 NAJg==
MIME-Version: 1.0
Received: by 10.220.205.134 with SMTP id fq6mr2380384vcb.24.1357678284385; Tue, 08 Jan 2013 12:51:24 -0800 (PST)
Received: by 10.220.144.78 with HTTP; Tue, 8 Jan 2013 12:51:24 -0800 (PST)
In-Reply-To: <CAF4+nEH8f_p_Fexo1d12CpaAdSD2wsa0_HK3dpO57KpTxgX55A@mail.gmail.com>
References: <CAF4+nEH8f_p_Fexo1d12CpaAdSD2wsa0_HK3dpO57KpTxgX55A@mail.gmail.com>
Date: Tue, 8 Jan 2013 12:51:24 -0800
Message-ID: <CAHD03N_kFtW12cDSTbL1FYq1n8fWzrWho2t=LSHK1ESoPzWOuA@mail.gmail.com>
From: Ayan Banerjee <ayabaner@gmail.com>
To: Donald Eastlake <d3e3e3@gmail.com>
Content-Type: multipart/alternative; boundary=14dae9cdc3f3dc2e4604d2cd1b8f
Cc: "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] ESADI 1. Simplify ESADI Neighbor determination
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 08 Jan 2013 20:51:25 -0000

--14dae9cdc3f3dc2e4604d2cd1b8f
Content-Type: text/plain; charset=ISO-8859-1

This would be a nice change to have, since the intent to participate is
already present in the LSP on the core instance.

Thanks,
Ayan


On Mon, Jan 7, 2013 at 7:25 PM, Donald Eastlake <d3e3e3@gmail.com> wrote:

> 1. Simplify ESADI Neighbor determination
>
>       The current draft requires that RBridge RB1 consider RB2 an ESADI
>       neighbor for VLAN-X only if three conditions are met: (1) RB2 is
>       data reachable from RB1, (2) RB2 has the ESADI participation bit
>       on in its Interested VLANs sub-TLV for VLAN-X, and (3) RB1 is
>       holding ESADI-LSP fragment zero for RB2. This last requirement (3)
>       increases the state complexity of ESADI and the suggestion is
>       that it be eliminated. If RB1 does not hold an ESADI-LSP fragment
>       zero with an ESADI Parameters APPsub-TLV in it for RB2, then it
>       treats RB2 as having default priority and CSNP time.
>
> Thanks,
> Donald
> =============================
>  Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
>  155 Beaver Street, Milford, MA 01757 USA
>  d3e3e3@gmail.com
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill
>

--14dae9cdc3f3dc2e4604d2cd1b8f
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">This would be a nice change to have, since the intent to p=
articipate is already present in the LSP on the core instance. <br><br>Than=
ks,<br>Ayan<br></div><div class=3D"gmail_extra"><br><br><div class=3D"gmail=
_quote">
On Mon, Jan 7, 2013 at 7:25 PM, Donald Eastlake <span dir=3D"ltr">&lt;<a hr=
ef=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gmail.com</a>&gt;</=
span> wrote:<br><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8e=
x;border-left:1px #ccc solid;padding-left:1ex">
1. Simplify ESADI Neighbor determination<br>
<br>
=A0 =A0 =A0 The current draft requires that RBridge RB1 consider RB2 an ESA=
DI<br>
=A0 =A0 =A0 neighbor for VLAN-X only if three conditions are met: (1) RB2 i=
s<br>
=A0 =A0 =A0 data reachable from RB1, (2) RB2 has the ESADI participation bi=
t<br>
=A0 =A0 =A0 on in its Interested VLANs sub-TLV for VLAN-X, and (3) RB1 is<b=
r>
=A0 =A0 =A0 holding ESADI-LSP fragment zero for RB2. This last requirement =
(3)<br>
=A0 =A0 =A0 increases the state complexity of ESADI and the suggestion is<b=
r>
=A0 =A0 =A0 that it be eliminated. If RB1 does not hold an ESADI-LSP fragme=
nt<br>
=A0 =A0 =A0 zero with an ESADI Parameters APPsub-TLV in it for RB2, then it=
<br>
=A0 =A0 =A0 treats RB2 as having default priority and CSNP time.<br>
<br>
Thanks,<br>
Donald<br>
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D<br>
=A0Donald E. Eastlake 3rd =A0 +1-508-333-2270 (cell)<br>
=A0155 Beaver Street, Milford, MA 01757 USA<br>
=A0<a href=3D"mailto:d3e3e3@gmail.com">d3e3e3@gmail.com</a><br>
_______________________________________________<br>
trill mailing list<br>
<a href=3D"mailto:trill@ietf.org">trill@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/trill" target=3D"_blank">h=
ttps://www.ietf.org/mailman/listinfo/trill</a><br>
</blockquote></div><br></div>

--14dae9cdc3f3dc2e4604d2cd1b8f--

From radiaperlman@gmail.com  Tue Jan  8 13:11:17 2013
Return-Path: <radiaperlman@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0015821F853E for <trill@ietfa.amsl.com>; Tue,  8 Jan 2013 13:11:16 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.27
X-Spam-Level: 
X-Spam-Status: No, score=-3.27 tagged_above=-999 required=5 tests=[AWL=0.328,  BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id yZZGiO+iRWwd for <trill@ietfa.amsl.com>; Tue,  8 Jan 2013 13:11:16 -0800 (PST)
Received: from mail-lb0-f171.google.com (mail-lb0-f171.google.com [209.85.217.171]) by ietfa.amsl.com (Postfix) with ESMTP id AE51821F8526 for <trill@ietf.org>; Tue,  8 Jan 2013 13:11:15 -0800 (PST)
Received: by mail-lb0-f171.google.com with SMTP id gf7so730512lbb.30 for <trill@ietf.org>; Tue, 08 Jan 2013 13:11:14 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=dIQraDOhrfrwCbmwSv1c01+tcYK+dS+HpciZZDrrtjs=; b=0ayI1tKoqB1kNPctLlOnWm3n7LNrYUpCLma6FHCAsXVurF0qV1SwI38QZR7NQEac3g 0eDbPCFXctmVQkeryTMMR6Mdv4KfkLUMY2fvKXJx78COa/s5B9ehZUscvPPfDoorkQzY CU1VP5ljeRtjxnPmAyPVo3jXIfCbnOx9smeoKGI7vnGiazlwUyKi7eFVRfDauaWgtmbR GkOibhtyIVfyTElcvUg7qZxs/aUmsgCpWiDYZGUUybX+wQTFLNuMYptDXn85q6kEE1mn bs2+57lfybH0RNcpoWGJn4T+YaTNPw25/PhNaqGejiOighbTCZhc1LbV/Y1BcfpT7c40 0EOA==
MIME-Version: 1.0
Received: by 10.152.148.164 with SMTP id tt4mr22799977lab.28.1357679474537; Tue, 08 Jan 2013 13:11:14 -0800 (PST)
Received: by 10.112.64.17 with HTTP; Tue, 8 Jan 2013 13:11:14 -0800 (PST)
In-Reply-To: <CAHD03N_kFtW12cDSTbL1FYq1n8fWzrWho2t=LSHK1ESoPzWOuA@mail.gmail.com>
References: <CAF4+nEH8f_p_Fexo1d12CpaAdSD2wsa0_HK3dpO57KpTxgX55A@mail.gmail.com> <CAHD03N_kFtW12cDSTbL1FYq1n8fWzrWho2t=LSHK1ESoPzWOuA@mail.gmail.com>
Date: Tue, 8 Jan 2013 13:11:14 -0800
Message-ID: <CAFOuuo4NEWUTPLR3Xo4ei5ueizu05eGvLhTMpZsnUpLDzm9U2g@mail.gmail.com>
From: Radia Perlman <radiaperlman@gmail.com>
To: Ayan Banerjee <ayabaner@gmail.com>
Content-Type: multipart/alternative; boundary=e89a8f22c425cc72e604d2cd6241
Cc: Donald Eastlake <d3e3e3@gmail.com>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] ESADI 1. Simplify ESADI Neighbor determination
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 08 Jan 2013 21:11:17 -0000

--e89a8f22c425cc72e604d2cd6241
Content-Type: text/plain; charset=ISO-8859-1

I agree.

Radia

On Tue, Jan 8, 2013 at 12:51 PM, Ayan Banerjee <ayabaner@gmail.com> wrote:

> This would be a nice change to have, since the intent to participate is
> already present in the LSP on the core instance.
>
> Thanks,
> Ayan
>
>
> On Mon, Jan 7, 2013 at 7:25 PM, Donald Eastlake <d3e3e3@gmail.com> wrote:
>
>> 1. Simplify ESADI Neighbor determination
>>
>>       The current draft requires that RBridge RB1 consider RB2 an ESADI
>>       neighbor for VLAN-X only if three conditions are met: (1) RB2 is
>>       data reachable from RB1, (2) RB2 has the ESADI participation bit
>>       on in its Interested VLANs sub-TLV for VLAN-X, and (3) RB1 is
>>       holding ESADI-LSP fragment zero for RB2. This last requirement (3)
>>       increases the state complexity of ESADI and the suggestion is
>>       that it be eliminated. If RB1 does not hold an ESADI-LSP fragment
>>       zero with an ESADI Parameters APPsub-TLV in it for RB2, then it
>>       treats RB2 as having default priority and CSNP time.
>>
>> Thanks,
>> Donald
>> =============================
>>  Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
>>  155 Beaver Street, Milford, MA 01757 USA
>>  d3e3e3@gmail.com
>> _______________________________________________
>> trill mailing list
>> trill@ietf.org
>> https://www.ietf.org/mailman/listinfo/trill
>>
>
>
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill
>
>

--e89a8f22c425cc72e604d2cd6241
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

I agree.<div><br></div><div>Radia<br><br><div class=3D"gmail_quote">On Tue,=
 Jan 8, 2013 at 12:51 PM, Ayan Banerjee <span dir=3D"ltr">&lt;<a href=3D"ma=
ilto:ayabaner@gmail.com" target=3D"_blank">ayabaner@gmail.com</a>&gt;</span=
> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div dir=3D"ltr">This would be a nice change=
 to have, since the intent to participate is already present in the LSP on =
the core instance. <br>
<br>Thanks,<br>Ayan<br></div><div class=3D"HOEnZb"><div class=3D"h5"><div c=
lass=3D"gmail_extra"><br><br><div class=3D"gmail_quote">
On Mon, Jan 7, 2013 at 7:25 PM, Donald Eastlake <span dir=3D"ltr">&lt;<a hr=
ef=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gmail.com</a>&gt;</=
span> wrote:<br><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8e=
x;border-left:1px #ccc solid;padding-left:1ex">

1. Simplify ESADI Neighbor determination<br>
<br>
=A0 =A0 =A0 The current draft requires that RBridge RB1 consider RB2 an ESA=
DI<br>
=A0 =A0 =A0 neighbor for VLAN-X only if three conditions are met: (1) RB2 i=
s<br>
=A0 =A0 =A0 data reachable from RB1, (2) RB2 has the ESADI participation bi=
t<br>
=A0 =A0 =A0 on in its Interested VLANs sub-TLV for VLAN-X, and (3) RB1 is<b=
r>
=A0 =A0 =A0 holding ESADI-LSP fragment zero for RB2. This last requirement =
(3)<br>
=A0 =A0 =A0 increases the state complexity of ESADI and the suggestion is<b=
r>
=A0 =A0 =A0 that it be eliminated. If RB1 does not hold an ESADI-LSP fragme=
nt<br>
=A0 =A0 =A0 zero with an ESADI Parameters APPsub-TLV in it for RB2, then it=
<br>
=A0 =A0 =A0 treats RB2 as having default priority and CSNP time.<br>
<br>
Thanks,<br>
Donald<br>
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D<br>
=A0Donald E. Eastlake 3rd =A0 <a href=3D"tel:%2B1-508-333-2270" value=3D"+1=
5083332270" target=3D"_blank">+1-508-333-2270</a> (cell)<br>
=A0155 Beaver Street, Milford, MA 01757 USA<br>
=A0<a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gmail.com</=
a><br>
_______________________________________________<br>
trill mailing list<br>
<a href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/trill" target=3D"_blank">h=
ttps://www.ietf.org/mailman/listinfo/trill</a><br>
</blockquote></div><br></div>
</div></div><br>_______________________________________________<br>
trill mailing list<br>
<a href=3D"mailto:trill@ietf.org">trill@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/trill" target=3D"_blank">h=
ttps://www.ietf.org/mailman/listinfo/trill</a><br>
<br></blockquote></div><br></div>

--e89a8f22c425cc72e604d2cd6241--

From radiaperlman@gmail.com  Tue Jan  8 13:23:28 2013
Return-Path: <radiaperlman@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 46E3121F84F9 for <trill@ietfa.amsl.com>; Tue,  8 Jan 2013 13:23:28 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.052
X-Spam-Level: 
X-Spam-Status: No, score=-3.052 tagged_above=-999 required=5 tests=[AWL=-0.054, BAYES_00=-2.599, HTML_MESSAGE=0.001, J_CHICKENPOX_81=0.6, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id wyvbFqlmepCJ for <trill@ietfa.amsl.com>; Tue,  8 Jan 2013 13:23:27 -0800 (PST)
Received: from mail-lb0-f179.google.com (mail-lb0-f179.google.com [209.85.217.179]) by ietfa.amsl.com (Postfix) with ESMTP id 85C5A11E80AE for <trill@ietf.org>; Tue,  8 Jan 2013 13:23:26 -0800 (PST)
Received: by mail-lb0-f179.google.com with SMTP id gm13so739369lbb.38 for <trill@ietf.org>; Tue, 08 Jan 2013 13:23:25 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=sebINaI8i/NyBq2sddpQIAdX7m6GMXwANk4mxzwMKhw=; b=PJsvUUnA6kSzJwnwlMUDAPygfoReEz2s3TcBbB6u0QrS9OF35FFgwQPBssDc2Nb4dW BQuWP4uHu/h1EVQLCEsl2RipDpXphyl+MWYWfEy0L+DN8kmzKLnCzfpdvZfGmeoEmywK 2f0IBuhPXoJtAp8rBsErR8ZZesmxeQd3pF9GsnC6P99+zmZo5Ffe1xrymkuaABn0gRQI m0u5pTYX41X07IM/cgpSvKkvvytoSgnER0oWA5E1QHwT9Ot2UPBcraW/fcqQORJnF7GG JRx4FKCFhf9ub4tArxzkEVRw/nsEyxnGi5ZfrDJyNwtf7wtvfHbnN8i1TN1FtVhm312b JXFg==
MIME-Version: 1.0
Received: by 10.152.125.7 with SMTP id mm7mr63833094lab.2.1357680205422; Tue, 08 Jan 2013 13:23:25 -0800 (PST)
Received: by 10.112.64.17 with HTTP; Tue, 8 Jan 2013 13:23:25 -0800 (PST)
In-Reply-To: <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com>
Date: Tue, 8 Jan 2013 13:23:25 -0800
Message-ID: <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com>
From: Radia Perlman <radiaperlman@gmail.com>
To: Phanidhar Koganti <pkoganti@brocade.com>
Content-Type: multipart/alternative; boundary=f46d043743ef5ce1b604d2cd8e23
Cc: "d3e3e3@gmail.com" <d3e3e3@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 08 Jan 2013 21:23:28 -0000

--f46d043743ef5ce1b604d2cd8e23
Content-Type: text/plain; charset=ISO-8859-1

Most of what you said is correct.  One correction

a) When the packet reaches the egress RB I am presuming there will be a
inner-MAC lookup to deliver the packet to the egress end-node.

correct...when egress RB R has to look up the inner MAC address E, which it
has to do anyway without smart endnodes, in order to choose which port to
forward it on.  With smart endnodes, when R looks up MAC address E, it not
only discovers which port to send it on, but that E is "smart", in which
case R must leave the TRILL encapsulation

b) Edge-RB to smart endnode cannot have TRILL encapsulation right?

Not correct.  Traffic from egress RB to smart endnode will have TRILL
encapsulation.  So for example, let's say R is the egress RB, and R has a
link with two endnodes; E1 (which is smart) and E2 (which is not smart).

When R receives a TRILL packet from the campus, with TRILL destination
nickname=R, R looks at the MAC destination.  If it is E2, then R removes
the TRILL encapulation and forwards a native packet onto the link.  If it
is E1, then R forwards the packet with the TRILL header.

c) Edge-RB to smart endnode cannot have TRILL encapsulation right? If so
for my clarification we are saying the traffic towards the end-node will be
non-TRILL while towards the ingress edge-RB will be TRILL encapsulated?

Hopefully my answer to b) clarified this.  On the link between endnodes and
R, whether it's to or from endnode <--> edge RB, the packet will have TRILL
encapsulation if the endnode is smart, and not have TRILL encapsulation if
the endnode is not smart.

Radia


On Thu, Jan 3, 2013 at 3:19 AM, Phanidhar Koganti <pkoganti@brocade.com>wrote:

> The draft defines the control and data path behavior from smart end-node
> to edge-RB. When the packet reaches the egress RB I am presuming there will
> be a inner-MAC lookup to deliver the packet to the egress end-node. Edge-RB
> to smart endnode cannot have TRILL encapsulation right? If so for my
> clarification we are saying the traffic towards the end-node will be
> non-TRILL while towards the ingress edge-RB will be TRILL encapsulated?***
> *
>
> ** **
>
> Phani****
>
> ** **
>
> *From:* trill-bounces@ietf.org [mailto:trill-bounces@ietf.org] *On Behalf
> Of *Kesava_Vijaya_Krupak@Dell.com
> *Sent:* Thursday, January 03, 2013 4:19 PM
> *To:* trill@ietf.org
> *Cc:* d3e3e3@gmail.com; radiaperlman@gmail.com; hu.fangwei@zte.com.cn
> *Subject:* [trill] New version : TRILL Smart Endnodes draft****
>
> ** **
>
> Hi all,****
>
> ** **
>
> A new version of the Trill Smart Endnodes draft has been posted.****
>
> ** **
>
> http://datatracker.ietf.org/doc/draft-perlman-trill-smart-endnodes****
>
> ** **
>
> Kindly let us know your comments.****
>
> ** **
>
> Thanks,****
>
> kvk****
>

--f46d043743ef5ce1b604d2cd8e23
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Most of what you said is correct. =A0One correction<div><br></div><div>a)=
=A0<span style=3D"color:rgb(31,73,125)">When the packet reaches the egress =
RB I am presuming there will be a inner-MAC lookup to deliver the packet to=
 the egress end-node.</span></div>
<div><font color=3D"#1f497d"><br></font></div><div><font color=3D"#1f497d">=
correct...when egress RB R has to look up the inner MAC address E, which it=
 has to do anyway without smart endnodes, in order to choose which port to =
forward it on. =A0With smart endnodes, when R looks up MAC address E, it no=
t only discovers which port to send it on, but that E is &quot;smart&quot;,=
 in which case R must leave the TRILL encapsulation</font></div>
<div><font color=3D"#1f497d"><br></font></div><div><font color=3D"#1f497d">=
b)=A0</font><span style=3D"color:rgb(31,73,125)">Edge-RB to smart endnode c=
annot have TRILL encapsulation right?</span></div><div><font color=3D"#1f49=
7d"><br>
</font></div><div><font color=3D"#1f497d">Not correct. =A0Traffic from egre=
ss RB to smart endnode will have TRILL encapsulation. =A0So for example, le=
t&#39;s say R is the egress RB, and R has a link with two endnodes; E1 (whi=
ch is smart) and E2 (which is not smart).</font></div>
<div><font color=3D"#1f497d"><br></font></div><div><font color=3D"#1f497d">=
When R receives a TRILL packet from the campus, with TRILL destination nick=
name=3DR, R looks at the MAC destination. =A0If it is E2, then R removes th=
e TRILL encapulation and forwards a native packet onto the link. =A0If it i=
s E1, then R forwards the packet with the TRILL header.</font></div>
<div><font color=3D"#1f497d"><br></font></div><div><font color=3D"#1f497d">=
c)=A0</font><span style=3D"color:rgb(31,73,125)">Edge-RB to smart endnode c=
annot have TRILL encapsulation right? If so for my clarification we are say=
ing the traffic towards the end-node will be non-TRILL while towards the in=
gress edge-RB will be TRILL encapsulated?</span></div>
<div><span style=3D"color:rgb(31,73,125)"><br></span></div><div><font color=
=3D"#1f497d">Hopefully my answer to b) clarified this. =A0On the link betwe=
en endnodes and R, whether it&#39;s to or from endnode &lt;--&gt; edge RB, =
the packet will have TRILL encapsulation if the endnode is smart, and not h=
ave TRILL encapsulation if the endnode is not smart.</font></div>
<div><font color=3D"#1f497d"><br></font></div><div><font color=3D"#1f497d">=
Radia</font></div><div><font color=3D"#1f497d"><br></font></div><div><font =
color=3D"#1f497d"><br></font><div><div class=3D"gmail_quote">On Thu, Jan 3,=
 2013 at 3:19 AM, Phanidhar Koganti <span dir=3D"ltr">&lt;<a href=3D"mailto=
:pkoganti@brocade.com" target=3D"_blank">pkoganti@brocade.com</a>&gt;</span=
> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div lang=3D"EN-US" link=3D"blue" vlink=3D"p=
urple"><div><p class=3D"MsoNormal"><span style=3D"color:#1f497d">The draft =
defines the control and data path behavior from smart end-node to edge-RB. =
When the packet reaches the egress RB I am presuming there will be a inner-=
MAC lookup to deliver the packet to the egress end-node. Edge-RB to smart e=
ndnode cannot have TRILL encapsulation right? If so for my clarification we=
 are saying the traffic towards the end-node will be non-TRILL while toward=
s the ingress edge-RB will be TRILL encapsulated?<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"color:#1f497d"><u></u>=A0<u></u></spa=
n></p><p class=3D"MsoNormal"><span style=3D"color:#1f497d">Phani<u></u><u><=
/u></span></p><p class=3D"MsoNormal"><span style=3D"color:#1f497d"><u></u>=
=A0<u></u></span></p>
<div><div style=3D"border:none;border-top:solid #b5c4df 1.0pt;padding:3.0pt=
 0in 0in 0in"><p class=3D"MsoNormal"><b><span style=3D"font-size:10.0pt;fon=
t-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;">From:</span></b><span s=
tyle=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&qu=
ot;"> <a href=3D"mailto:trill-bounces@ietf.org" target=3D"_blank">trill-bou=
nces@ietf.org</a> [mailto:<a href=3D"mailto:trill-bounces@ietf.org" target=
=3D"_blank">trill-bounces@ietf.org</a>] <b>On Behalf Of </b>Kesava_Vijaya_K=
rupak@Dell.com<br>
<b>Sent:</b> Thursday, January 03, 2013 4:19 PM<br><b>To:</b> <a href=3D"ma=
ilto:trill@ietf.org" target=3D"_blank">trill@ietf.org</a><br><b>Cc:</b> <a =
href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gmail.com</a>; <a=
 href=3D"mailto:radiaperlman@gmail.com" target=3D"_blank">radiaperlman@gmai=
l.com</a>; <a href=3D"mailto:hu.fangwei@zte.com.cn" target=3D"_blank">hu.fa=
ngwei@zte.com.cn</a><br>
<b>Subject:</b> [trill] New version : TRILL Smart Endnodes draft<u></u><u><=
/u></span></p></div></div><div><div class=3D"h5"><p class=3D"MsoNormal"><u>=
</u>=A0<u></u></p><p class=3D"MsoNormal">Hi all,<u></u><u></u></p><p class=
=3D"MsoNormal">
<u></u>=A0<u></u></p><p class=3D"MsoNormal">A new version of the Trill Smar=
t Endnodes draft has been posted.<u></u><u></u></p><p class=3D"MsoNormal"><=
u></u>=A0<u></u></p><p class=3D"MsoNormal"><a href=3D"http://datatracker.ie=
tf.org/doc/draft-perlman-trill-smart-endnodes" target=3D"_blank">http://dat=
atracker.ietf.org/doc/draft-perlman-trill-smart-endnodes</a><u></u><u></u><=
/p>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p><p class=3D"MsoNormal">Kindly l=
et us know your comments.<u></u><u></u></p><p class=3D"MsoNormal"><u></u>=
=A0<u></u></p><p class=3D"MsoNormal">Thanks,<u></u><u></u></p><p class=3D"M=
soNormal">kvk<u></u><u></u></p>
</div></div></div></div></blockquote></div><br></div></div>

--f46d043743ef5ce1b604d2cd8e23--

From zhai.hongjun@zte.com.cn  Tue Jan  8 19:15:41 2013
Return-Path: <zhai.hongjun@zte.com.cn>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1184E1F0CAF; Tue,  8 Jan 2013 19:15:41 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -95.51
X-Spam-Level: 
X-Spam-Status: No, score=-95.51 tagged_above=-999 required=5 tests=[BAYES_05=-1.11, HTML_MESSAGE=0.001, MIME_BASE64_TEXT=1.753, MIME_CHARSET_FARAWAY=2.45, USER_IN_WHITELIST=-100, WEIRD_QUOTING=1.396]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id zfacdKik9qg7; Tue,  8 Jan 2013 19:15:40 -0800 (PST)
Received: from zte.com.cn (mx6.zte.com.cn [95.130.199.165]) by ietfa.amsl.com (Postfix) with ESMTP id C043C1F0C5F; Tue,  8 Jan 2013 19:15:39 -0800 (PST)
Received: from zte.com.cn (unknown [192.168.168.119]) by Websense Email Security Gateway with ESMTP id E511819387A2; Wed,  9 Jan 2013 11:15:27 +0800 (CST)
Received: from mse01.zte.com.cn (unknown [10.30.3.20]) by Websense Email Security Gateway with ESMTPS id 508507033D1; Wed,  9 Jan 2013 11:05:16 +0800 (CST)
Received: from notes_smtp.zte.com.cn ([10.30.1.239]) by mse01.zte.com.cn with ESMTP id r093FGbr055191; Wed, 9 Jan 2013 11:15:16 +0800 (GMT-8) (envelope-from zhai.hongjun@zte.com.cn)
In-Reply-To: <CAHD03N_kFtW12cDSTbL1FYq1n8fWzrWho2t=LSHK1ESoPzWOuA@mail.gmail.com>
To: Ayan Banerjee <ayabaner@gmail.com>
MIME-Version: 1.0
X-KeepSent: 8321E15A:57F0D3DF-48257AEE:0010C37F; type=4; name=$KeepSent
X-Mailer: Lotus Notes Release 6.5.6 March 06, 2007
Message-ID: <OF8321E15A.57F0D3DF-ON48257AEE.0010C37F-48257AEE.00122CC4@zte.com.cn>
From: zhai.hongjun@zte.com.cn
Date: Wed, 9 Jan 2013 11:15:17 +0800
X-MIMETrack: Serialize by Router on notes_smtp/zte_ltd(Release 8.5.3FP1 HF212|May 23, 2012) at 2013-01-09 11:15:13, Serialize complete at 2013-01-09 11:15:13
Content-Type: multipart/alternative; boundary="=_alternative 00122CC148257AEE_="
X-MAIL: mse01.zte.com.cn r093FGbr055191
Cc: Donald Eastlake <d3e3e3@gmail.com>, trill-bounces@ietf.org, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] ESADI 1. Simplify ESADI Neighbor determination
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 09 Jan 2013 03:15:41 -0000

This is a multipart message in MIME format.
--=_alternative 00122CC148257AEE_=
Content-Type: text/plain; charset="GB2312"
Content-Transfer-Encoding: base64
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--=_alternative 00122CC148257AEE_=
Content-Type: text/html; charset="GB2312"
Content-Transfer-Encoding: base64
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--=_alternative 00122CC148257AEE_=--

From pkoganti@Brocade.COM  Tue Jan  8 20:32:51 2013
Return-Path: <pkoganti@Brocade.COM>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E491221F84E8 for <trill@ietfa.amsl.com>; Tue,  8 Jan 2013 20:32:50 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.064
X-Spam-Level: 
X-Spam-Status: No, score=-2.064 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, IP_NOT_FRIENDLY=0.334, J_CHICKENPOX_33=0.6, J_CHICKENPOX_81=0.6, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id pG6P4ICIziD4 for <trill@ietfa.amsl.com>; Tue,  8 Jan 2013 20:32:48 -0800 (PST)
Received: from mx0a-000f0801.pphosted.com (mx0a-000f0801.pphosted.com [67.231.144.122]) by ietfa.amsl.com (Postfix) with ESMTP id CA6C51F0CB3 for <trill@ietf.org>; Tue,  8 Jan 2013 20:32:46 -0800 (PST)
Received: from pps.filterd (m0000542 [127.0.0.1]) by mx0a-000f0801.pphosted.com (8.14.5/8.14.5) with SMTP id r094WiMS011835; Tue, 8 Jan 2013 20:32:44 -0800
Received: from hq1wp-exchub01.corp.brocade.com ([144.49.131.13]) by mx0a-000f0801.pphosted.com with ESMTP id 19rta7073j-1 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Tue, 08 Jan 2013 20:32:43 -0800
Received: from HQ1WP-EXHUB01.corp.brocade.com (10.70.36.14) by HQ1WP-EXCHUB01.corp.brocade.com (10.70.36.99) with Microsoft SMTP Server (TLS) id 14.2.309.2; Tue, 8 Jan 2013 20:32:42 -0800
Received: from HQ1-EXCH02.corp.brocade.com ([fe80::c92a:772e:befa:c34c]) by HQ1WP-EXHUB01.corp.brocade.com ([::1]) with mapi; Tue, 8 Jan 2013 20:32:42 -0800
From: Phanidhar Koganti <pkoganti@Brocade.COM>
To: Radia Perlman <radiaperlman@gmail.com>
Date: Tue, 8 Jan 2013 20:32:40 -0800
Thread-Topic: New version : TRILL Smart Endnodes draft
Thread-Index: Ac3t5mdYwU/tWtyxTq6IpCYoVSBvWgAOxrdg
Message-ID: <6895EAE0CA8DEE40B92D7CA88BB521F32414007BC1@HQ1-EXCH02.corp.brocade.com>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com> <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com>
In-Reply-To: <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US
Content-Type: multipart/alternative; boundary="_000_6895EAE0CA8DEE40B92D7CA88BB521F32414007BC1HQ1EXCH02corp_"
MIME-Version: 1.0
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:5.9.8327, 1.0.431, 0.0.0000 definitions=2013-01-09_02:2013-01-08, 2013-01-09, 1970-01-01 signatures=0
X-Proofpoint-Spam-Details: rule=notspam policy=default score=0 spamscore=0 suspectscore=49 phishscore=0 adultscore=0 bulkscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=7.0.1-1211240000 definitions=main-1301080338
Cc: "d3e3e3@gmail.com" <d3e3e3@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 09 Jan 2013 04:32:51 -0000

--_000_6895EAE0CA8DEE40B92D7CA88BB521F32414007BC1HQ1EXCH02corp_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Thanks Radia. That clarifies for me. A more descriptive text in draft can s=
urely help vendors as they implement it in their hardware.

Another question on handling multi-homing
E can choose either R1 or R2's nickname, when encapsulating a
       frame, whether the encapsulated frame is sent via R1 or R2. If E
       wants to do active-active load splitting, and uses R1's nickname
       when forwarding through R1, and R2's nickname when forwarding
       through R2, this will cause distant RBridges (or smart endnodes)
       to keep changing their endnode table entry for D between (D, R1's
       nickname) and (D, R2's nickname). So it would be preferable for E
       to always encapsulate using the same nickname (R1 or R2) unless E
       detects a problem with connectivity using that nickname. And in
       this case, R1 and R2 need to be informed that the smart endnode
       might encapsulate with a different nickname, i.e., R1 might
       receive an encapsulated packet from smart endnode E using ingress
       nickname "R2".

If the smart endnode uses either R1 or R2 in the above example, would we ha=
ve TRILL RPF issues where upstream RBs see a packet with SRB=3Dsay R1 comin=
g from a wrong TRILL port? Let's say the RPF happens for multicast only the=
n we will atleast need to make sure the multicast (broadcast, unknown unica=
st and multcast) is sent via the edge-RB whose RBID the smart endnode is us=
ing.

This problem should not be there when using different RBID other than R1 an=
d R2 as described in approach-2

Phani
From: Radia Perlman [mailto:radiaperlman@gmail.com]
Sent: Wednesday, January 09, 2013 2:53 AM
To: Phanidhar Koganti
Cc: Kesava_Vijaya_Krupak@Dell.com; trill@ietf.org; d3e3e3@gmail.com; hu.fan=
gwei@zte.com.cn
Subject: Re: New version : TRILL Smart Endnodes draft

Most of what you said is correct.  One correction

a) When the packet reaches the egress RB I am presuming there will be a inn=
er-MAC lookup to deliver the packet to the egress end-node.

correct...when egress RB R has to look up the inner MAC address E, which it=
 has to do anyway without smart endnodes, in order to choose which port to =
forward it on.  With smart endnodes, when R looks up MAC address E, it not =
only discovers which port to send it on, but that E is "smart", in which ca=
se R must leave the TRILL encapsulation

b) Edge-RB to smart endnode cannot have TRILL encapsulation right?

Not correct.  Traffic from egress RB to smart endnode will have TRILL encap=
sulation.  So for example, let's say R is the egress RB, and R has a link w=
ith two endnodes; E1 (which is smart) and E2 (which is not smart).

When R receives a TRILL packet from the campus, with TRILL destination nick=
name=3DR, R looks at the MAC destination.  If it is E2, then R removes the =
TRILL encapulation and forwards a native packet onto the link.  If it is E1=
, then R forwards the packet with the TRILL header.

c) Edge-RB to smart endnode cannot have TRILL encapsulation right? If so fo=
r my clarification we are saying the traffic towards the end-node will be n=
on-TRILL while towards the ingress edge-RB will be TRILL encapsulated?

Hopefully my answer to b) clarified this.  On the link between endnodes and=
 R, whether it's to or from endnode <--> edge RB, the packet will have TRIL=
L encapsulation if the endnode is smart, and not have TRILL encapsulation i=
f the endnode is not smart.

Radia


On Thu, Jan 3, 2013 at 3:19 AM, Phanidhar Koganti <pkoganti@brocade.com<mai=
lto:pkoganti@brocade.com>> wrote:
The draft defines the control and data path behavior from smart end-node to=
 edge-RB. When the packet reaches the egress RB I am presuming there will b=
e a inner-MAC lookup to deliver the packet to the egress end-node. Edge-RB =
to smart endnode cannot have TRILL encapsulation right? If so for my clarif=
ication we are saying the traffic towards the end-node will be non-TRILL wh=
ile towards the ingress edge-RB will be TRILL encapsulated?

Phani

From: trill-bounces@ietf.org<mailto:trill-bounces@ietf.org> [mailto:trill-b=
ounces@ietf.org<mailto:trill-bounces@ietf.org>] On Behalf Of Kesava_Vijaya_=
Krupak@Dell.com<mailto:Kesava_Vijaya_Krupak@Dell.com>
Sent: Thursday, January 03, 2013 4:19 PM
To: trill@ietf.org<mailto:trill@ietf.org>
Cc: d3e3e3@gmail.com<mailto:d3e3e3@gmail.com>; radiaperlman@gmail.com<mailt=
o:radiaperlman@gmail.com>; hu.fangwei@zte.com.cn<mailto:hu.fangwei@zte.com.=
cn>
Subject: [trill] New version : TRILL Smart Endnodes draft

Hi all,

A new version of the Trill Smart Endnodes draft has been posted.

http://datatracker.ietf.org/doc/draft-perlman-trill-smart-endnodes

Kindly let us know your comments.

Thanks,
kvk


--_000_6895EAE0CA8DEE40B92D7CA88BB521F32414007BC1HQ1EXCH02corp_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micr=
osoft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" xmlns=3D"http:=
//www.w3.org/TR/REC-html40"><head><META HTTP-EQUIV=3D"Content-Type" CONTENT=
=3D"text/html; charset=3Dus-ascii"><meta name=3DGenerator content=3D"Micros=
oft Word 14 (filtered medium)"><style><!--
/* Font Definitions */
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:Tahoma;
	panose-1:2 11 6 4 3 5 4 4 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman","serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
pre
	{mso-style-priority:99;
	mso-style-link:"HTML Preformatted Char";
	margin:0in;
	margin-bottom:.0001pt;
	font-size:10.0pt;
	font-family:"Courier New";}
p.MsoAcetate, li.MsoAcetate, div.MsoAcetate
	{mso-style-priority:99;
	mso-style-link:"Balloon Text Char";
	margin:0in;
	margin-bottom:.0001pt;
	font-size:8.0pt;
	font-family:"Tahoma","sans-serif";}
span.BalloonTextChar
	{mso-style-name:"Balloon Text Char";
	mso-style-priority:99;
	mso-style-link:"Balloon Text";
	font-family:"Tahoma","sans-serif";}
span.EmailStyle19
	{mso-style-type:personal-reply;
	font-family:"Calibri","sans-serif";
	color:#1F497D;}
span.HTMLPreformattedChar
	{mso-style-name:"HTML Preformatted Char";
	mso-style-priority:99;
	mso-style-link:"HTML Preformatted";
	font-family:"Courier New";}
.MsoChpDefault
	{mso-style-type:export-only;
	font-family:"Calibri","sans-serif";}
@page WordSection1
	{size:8.5in 11.0in;
	margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
	{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]--></head><body lang=3DEN-US link=3Dblue vli=
nk=3Dpurple><div class=3DWordSection1><p class=3DMsoNormal><span style=3D'f=
ont-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>Thanks Ra=
dia. That clarifies for me. A more descriptive text in draft can surely hel=
p vendors as they implement it in their hardware.<o:p></o:p></span></p><p c=
lass=3DMsoNormal><span style=3D'font-size:11.0pt;font-family:"Calibri","san=
s-serif";color:#1F497D'><o:p>&nbsp;</o:p></span></p><p class=3DMsoNormal><s=
pan style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F4=
97D'>Another question on handling multi-homing<o:p></o:p></span></p><p clas=
s=3DMsoNormal style=3D'line-height:14.4pt'><span style=3D'font-size:10.0pt;=
font-family:"Courier New";color:black'>E can choose either R1 or R2's nickn=
ame, when encapsulating a<o:p></o:p></span></p><p class=3DMsoNormal style=
=3D'line-height:14.4pt'><span style=3D'font-size:10.0pt;font-family:"Courie=
r New";color:black'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; frame, whether the=
 encapsulated frame is sent via R1 or R2. If E<o:p></o:p></span></p><p clas=
s=3DMsoNormal style=3D'line-height:14.4pt'><span style=3D'font-size:10.0pt;=
font-family:"Courier New";color:black'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=
 wants to do active-active load splitting, and uses R1's nickname<o:p></o:p=
></span></p><p class=3DMsoNormal style=3D'line-height:14.4pt'><span style=
=3D'font-size:10.0pt;font-family:"Courier New";color:black'>&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;&nbsp; when forwarding through R1, and R2's nickname when fo=
rwarding<o:p></o:p></span></p><p class=3DMsoNormal style=3D'line-height:14.=
4pt'><span style=3D'font-size:10.0pt;font-family:"Courier New";color:black'=
>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; through R2, this will cause distant R=
Bridges (or smart endnodes)<o:p></o:p></span></p><p class=3DMsoNormal style=
=3D'line-height:14.4pt'><span style=3D'font-size:10.0pt;font-family:"Courie=
r New";color:black'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; to keep changing t=
heir endnode table entry for D between (D, R1's<o:p></o:p></span></p><p cla=
ss=3DMsoNormal style=3D'line-height:14.4pt'><span style=3D'font-size:10.0pt=
;font-family:"Courier New";color:black'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
; nickname) and (D, R2's nickname). So it would be preferable for E<o:p></o=
:p></span></p><p class=3DMsoNormal style=3D'line-height:14.4pt'><span style=
=3D'font-size:10.0pt;font-family:"Courier New";color:black'>&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;&nbsp; to always encapsulate using the same nickname (R1 or =
R2) unless E<o:p></o:p></span></p><p class=3DMsoNormal style=3D'line-height=
:14.4pt'><span style=3D'font-size:10.0pt;font-family:"Courier New";color:bl=
ack'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; detects a problem with connectivi=
ty using that nickname. And in<o:p></o:p></span></p><p class=3DMsoNormal st=
yle=3D'line-height:14.4pt'><span style=3D'font-size:10.0pt;font-family:"Cou=
rier New";color:black'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; this case, R1 a=
nd R2 need to be informed that the smart endnode<o:p></o:p></span></p><p cl=
ass=3DMsoNormal style=3D'line-height:14.4pt'><span style=3D'font-size:10.0p=
t;font-family:"Courier New";color:black'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p; might encapsulate with a different nickname, i.e., R1 might<o:p></o:p></=
span></p><p class=3DMsoNormal style=3D'line-height:14.4pt'><span style=3D'f=
ont-size:10.0pt;font-family:"Courier New";color:black'>&nbsp;&nbsp;&nbsp;&n=
bsp;&nbsp;&nbsp; receive an encapsulated packet from smart endnode E using =
ingress<o:p></o:p></span></p><p class=3DMsoNormal style=3D'line-height:14.4=
pt'><span style=3D'font-size:10.0pt;font-family:"Courier New";color:black'>=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; nickname &quot;R2&quot;.<o:p></o:p></s=
pan></p><p class=3DMsoNormal><span style=3D'font-size:11.0pt;font-family:"C=
alibri","sans-serif";color:#1F497D'><o:p>&nbsp;</o:p></span></p><p class=3D=
MsoNormal><span style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif=
";color:#1F497D'>If the smart endnode uses either R1 or R2 in the above exa=
mple, would we have TRILL RPF issues where upstream RBs see a packet with S=
RB=3Dsay R1 coming from a wrong TRILL port? Let&#8217;s say the RPF happens=
 for multicast only then we will atleast need to make sure the multicast (b=
roadcast, unknown unicast and multcast) is sent via the edge-RB whose RBID =
the smart endnode is using.<o:p></o:p></span></p><p class=3DMsoNormal><span=
 style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D=
'><o:p>&nbsp;</o:p></span></p><p class=3DMsoNormal><span style=3D'font-size=
:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>This problem shou=
ld not be there when using different RBID other than R1 and R2 as described=
 in approach-2<o:p></o:p></span></p><p class=3DMsoNormal><span style=3D'fon=
t-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'><o:p>&nbsp;=
</o:p></span></p><p class=3DMsoNormal><span style=3D'font-size:11.0pt;font-=
family:"Calibri","sans-serif";color:#1F497D'>Phani<o:p></o:p></span></p><p =
class=3DMsoNormal><b><span style=3D'font-size:10.0pt;font-family:"Tahoma","=
sans-serif"'>From:</span></b><span style=3D'font-size:10.0pt;font-family:"T=
ahoma","sans-serif"'> Radia Perlman [mailto:radiaperlman@gmail.com] <br><b>=
Sent:</b> Wednesday, January 09, 2013 2:53 AM<br><b>To:</b> Phanidhar Kogan=
ti<br><b>Cc:</b> Kesava_Vijaya_Krupak@Dell.com; trill@ietf.org; d3e3e3@gmai=
l.com; hu.fangwei@zte.com.cn<br><b>Subject:</b> Re: New version : TRILL Sma=
rt Endnodes draft<o:p></o:p></span></p><p class=3DMsoNormal><o:p>&nbsp;</o:=
p></p><p class=3DMsoNormal>Most of what you said is correct. &nbsp;One corr=
ection<o:p></o:p></p><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><=
div><p class=3DMsoNormal>a)&nbsp;<span style=3D'color:#1F497D'>When the pac=
ket reaches the egress RB I am presuming there will be a inner-MAC lookup t=
o deliver the packet to the egress end-node.</span><o:p></o:p></p></div><di=
v><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p class=3DMsoNormal=
><span style=3D'color:#1F497D'>correct...when egress RB R has to look up th=
e inner MAC address E, which it has to do anyway without smart endnodes, in=
 order to choose which port to forward it on. &nbsp;With smart endnodes, wh=
en R looks up MAC address E, it not only discovers which port to send it on=
, but that E is &quot;smart&quot;, in which case R must leave the TRILL enc=
apsulation</span><o:p></o:p></p></div><div><p class=3DMsoNormal><o:p>&nbsp;=
</o:p></p></div><div><p class=3DMsoNormal><span style=3D'color:#1F497D'>b)&=
nbsp;Edge-RB to smart endnode cannot have TRILL encapsulation right?</span>=
<o:p></o:p></p></div><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><=
div><p class=3DMsoNormal><span style=3D'color:#1F497D'>Not correct. &nbsp;T=
raffic from egress RB to smart endnode will have TRILL encapsulation. &nbsp=
;So for example, let's say R is the egress RB, and R has a link with two en=
dnodes; E1 (which is smart) and E2 (which is not smart).</span><o:p></o:p><=
/p></div><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p class=
=3DMsoNormal><span style=3D'color:#1F497D'>When R receives a TRILL packet f=
rom the campus, with TRILL destination nickname=3DR, R looks at the MAC des=
tination. &nbsp;If it is E2, then R removes the TRILL encapulation and forw=
ards a native packet onto the link. &nbsp;If it is E1, then R forwards the =
packet with the TRILL header.</span><o:p></o:p></p></div><div><p class=3DMs=
oNormal><o:p>&nbsp;</o:p></p></div><div><p class=3DMsoNormal><span style=3D=
'color:#1F497D'>c)&nbsp;Edge-RB to smart endnode cannot have TRILL encapsul=
ation right? If so for my clarification we are saying the traffic towards t=
he end-node will be non-TRILL while towards the ingress edge-RB will be TRI=
LL encapsulated?</span><o:p></o:p></p></div><div><p class=3DMsoNormal><o:p>=
&nbsp;</o:p></p></div><div><p class=3DMsoNormal><span style=3D'color:#1F497=
D'>Hopefully my answer to b) clarified this. &nbsp;On the link between endn=
odes and R, whether it's to or from endnode &lt;--&gt; edge RB, the packet =
will have TRILL encapsulation if the endnode is smart, and not have TRILL e=
ncapsulation if the endnode is not smart.</span><o:p></o:p></p></div><div><=
p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p class=3DMsoNormal><s=
pan style=3D'color:#1F497D'>Radia</span><o:p></o:p></p></div><div><p class=
=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p class=3DMsoNormal><o:p>&nbs=
p;</o:p></p><div><div><p class=3DMsoNormal>On Thu, Jan 3, 2013 at 3:19 AM, =
Phanidhar Koganti &lt;<a href=3D"mailto:pkoganti@brocade.com" target=3D"_bl=
ank">pkoganti@brocade.com</a>&gt; wrote:<o:p></o:p></p><div><div><p class=
=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><=
span style=3D'color:#1F497D'>The draft defines the control and data path be=
havior from smart end-node to edge-RB. When the packet reaches the egress R=
B I am presuming there will be a inner-MAC lookup to deliver the packet to =
the egress end-node. Edge-RB to smart endnode cannot have TRILL encapsulati=
on right? If so for my clarification we are saying the traffic towards the =
end-node will be non-TRILL while towards the ingress edge-RB will be TRILL =
encapsulated?</span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin=
-top-alt:auto;mso-margin-bottom-alt:auto'><span style=3D'color:#1F497D'>&nb=
sp;</span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:a=
uto;mso-margin-bottom-alt:auto'><span style=3D'color:#1F497D'>Phani</span><=
o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-mar=
gin-bottom-alt:auto'><span style=3D'color:#1F497D'>&nbsp;</span><o:p></o:p>=
</p><div><div style=3D'border:none;border-top:solid #B5C4DF 1.0pt;padding:3=
.0pt 0in 0in 0in'><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso=
-margin-bottom-alt:auto'><b><span style=3D'font-size:10.0pt;font-family:"Ta=
homa","sans-serif"'>From:</span></b><span style=3D'font-size:10.0pt;font-fa=
mily:"Tahoma","sans-serif"'> <a href=3D"mailto:trill-bounces@ietf.org" targ=
et=3D"_blank">trill-bounces@ietf.org</a> [mailto:<a href=3D"mailto:trill-bo=
unces@ietf.org" target=3D"_blank">trill-bounces@ietf.org</a>] <b>On Behalf =
Of </b><a href=3D"mailto:Kesava_Vijaya_Krupak@Dell.com">Kesava_Vijaya_Krupa=
k@Dell.com</a><br><b>Sent:</b> Thursday, January 03, 2013 4:19 PM<br><b>To:=
</b> <a href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ietf.org</a>=
<br><b>Cc:</b> <a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3=
@gmail.com</a>; <a href=3D"mailto:radiaperlman@gmail.com" target=3D"_blank"=
>radiaperlman@gmail.com</a>; <a href=3D"mailto:hu.fangwei@zte.com.cn" targe=
t=3D"_blank">hu.fangwei@zte.com.cn</a><br><b>Subject:</b> [trill] New versi=
on : TRILL Smart Endnodes draft</span><o:p></o:p></p></div></div><div><div>=
<p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt=
:auto'>&nbsp;<o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-al=
t:auto;mso-margin-bottom-alt:auto'>Hi all,<o:p></o:p></p><p class=3DMsoNorm=
al style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p>=
</o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-=
bottom-alt:auto'>A new version of the Trill Smart Endnodes draft has been p=
osted.<o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;=
mso-margin-bottom-alt:auto'>&nbsp;<o:p></o:p></p><p class=3DMsoNormal style=
=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><a href=3D"http://d=
atatracker.ietf.org/doc/draft-perlman-trill-smart-endnodes" target=3D"_blan=
k">http://datatracker.ietf.org/doc/draft-perlman-trill-smart-endnodes</a><o=
:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-marg=
in-bottom-alt:auto'>&nbsp;<o:p></o:p></p><p class=3DMsoNormal style=3D'mso-=
margin-top-alt:auto;mso-margin-bottom-alt:auto'>Kindly let us know your com=
ments.<o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;=
mso-margin-bottom-alt:auto'>&nbsp;<o:p></o:p></p><p class=3DMsoNormal style=
=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>Thanks,<o:p></o:p><=
/p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-=
alt:auto'>kvk<o:p></o:p></p></div></div></div></div></div><p class=3DMsoNor=
mal><o:p>&nbsp;</o:p></p></div></div></div></body></html>=

--_000_6895EAE0CA8DEE40B92D7CA88BB521F32414007BC1HQ1EXCH02corp_--

From radiaperlman@gmail.com  Tue Jan  8 22:21:36 2013
Return-Path: <radiaperlman@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id BCEE521F8793 for <trill@ietfa.amsl.com>; Tue,  8 Jan 2013 22:21:36 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.784
X-Spam-Level: 
X-Spam-Status: No, score=-2.784 tagged_above=-999 required=5 tests=[AWL=-0.386, BAYES_00=-2.599, HTML_MESSAGE=0.001, J_CHICKENPOX_33=0.6, J_CHICKENPOX_81=0.6, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ZTQGlxqCIUbg for <trill@ietfa.amsl.com>; Tue,  8 Jan 2013 22:21:35 -0800 (PST)
Received: from mail-la0-f47.google.com (mail-la0-f47.google.com [209.85.215.47]) by ietfa.amsl.com (Postfix) with ESMTP id ACF2F21F8792 for <trill@ietf.org>; Tue,  8 Jan 2013 22:21:34 -0800 (PST)
Received: by mail-la0-f47.google.com with SMTP id fh20so1430143lab.6 for <trill@ietf.org>; Tue, 08 Jan 2013 22:21:33 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=9wrG6DvVtT4RWQjvtU6bR5xEbhWmxoiFxIzbKOiH6Tk=; b=UsumKBJpGXvDc1L7OswUKJc4UiCMW1WsZ4eT31TfTKnsNjyRh9J8r5VMEyVixprqO1 psdU93rVbiK0VRHVa+0HH0s4bly47AGsKXd74skadRqKPimpirHesERB1sqLTDcMksH7 m/4ac7eBN/gMb03vkKHLbJ8NqCLdXDz/C3eLb6Sm9GTY9JMiWLvgZ0l4SqeCxSP+NGpM yUVuABfAPSFWNHW4YY1NUlT1FXhVrmLwjMvGSb3oseEmAiddjAGrimcklatS2SFR4PVF pT/Lu9hvPJlSJLpU36StBDBdAt6d5cjpIyvFipOc7KVotwgqy2PdT4MZeZ42F5X6BfwX RDOA==
MIME-Version: 1.0
Received: by 10.152.121.212 with SMTP id lm20mr63410579lab.42.1357712493414; Tue, 08 Jan 2013 22:21:33 -0800 (PST)
Received: by 10.112.64.17 with HTTP; Tue, 8 Jan 2013 22:21:33 -0800 (PST)
In-Reply-To: <6895EAE0CA8DEE40B92D7CA88BB521F32414007BC1@HQ1-EXCH02.corp.brocade.com>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com> <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com> <6895EAE0CA8DEE40B92D7CA88BB521F32414007BC1@HQ1-EXCH02.corp.brocade.com>
Date: Tue, 8 Jan 2013 22:21:33 -0800
Message-ID: <CAFOuuo626hw-c0kM4-1Fp-w2gByPenRnkFP3xNT+cTB4YcJ2kg@mail.gmail.com>
From: Radia Perlman <radiaperlman@gmail.com>
To: Phanidhar Koganti <pkoganti@brocade.com>
Content-Type: multipart/alternative; boundary=f46d04374ee1e0893004d2d512ec
Cc: "d3e3e3@gmail.com" <d3e3e3@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 09 Jan 2013 06:21:37 -0000

--f46d04374ee1e0893004d2d512ec
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

Good point (see Phani's comment below).  I'll summarize:

The current  text implies that it's OK for smart endnode E, multihomed to
R1 and R2, to choose one of the nicknames (say R1) for "ingress nickname"
in the TRILL header, and forward to either R1 or R2.  There are two
problems with E forwarding a TRILL-encapsulated frame using ingeress "R1",
through R2:

a) for multicast, likely to confuse the RPF check
b) for unicast, E would probably need to warn R2 that E is multihomed to R1
and will e using R1's nickname for ingress.

And actually, E won't get that much load splitting benefit from being
multihomed if it uses "R1" as ingress, since all traffic to E will go
through R1.  So, if E doesn't have a pseudonode nickname to use, then
perhaps E should just use one R1 or R2.

Or E could get its own nickname (I've been floating around the idea of a
"dumb RBridge", which is like a "smart endnode", except it gets its own
nickname, and does generate an LSP with "overloaded" flag.  It probably has
to receive LSPs and calculate Dijkstra, in order to know which port to send
multicast on (unless its attached true RBs are helpful enough to tell
it..."you can send multicast with trees X or Y through me")

So, it seems like there are several options which should be discussed, for
a smart endnode multihomed to R1 and R2
1) Always use a pseudnode nickname (but that's kind of complicated since
that nickname can only be used by smart endnodes that are multihomed to the
exact same set of RBridges)
2) Don't do active-active...choose one or R1 or R2 and always send
everything that way
3) Do active-active, but only for unicast...choose one of the nicknames,
say "R1", and it's OK to send unicast either through R1 or R2, but
multicast has to go through R1
4) Give E its own nickname (and somehow make sure it knows which port it
can multicast on)
5) Other possibilities?

Radia





On Tue, Jan 8, 2013 at 8:32 PM, Phanidhar Koganti <pkoganti@brocade.com>wro=
te:

> Thanks Radia. That clarifies for me. A more descriptive text in draft can
> surely help vendors as they implement it in their hardware.****
>
> ** **
>
> Another question on handling multi-homing****
>
> E can choose either R1 or R2's nickname, when encapsulating a****
>
>        frame, whether the encapsulated frame is sent via R1 or R2. If E**=
*
> *
>
>        wants to do active-active load splitting, and uses R1's nickname**=
*
> *
>
>        when forwarding through R1, and R2's nickname when forwarding****
>
>        through R2, this will cause distant RBridges (or smart endnodes)**=
*
> *
>
>        to keep changing their endnode table entry for D between (D, R1's*=
*
> **
>
>        nickname) and (D, R2's nickname). So it would be preferable for E*=
*
> **
>
>        to always encapsulate using the same nickname (R1 or R2) unless E*=
*
> **
>
>        detects a problem with connectivity using that nickname. And in***=
*
>
>        this case, R1 and R2 need to be informed that the smart endnode***=
*
>
>        might encapsulate with a different nickname, i.e., R1 might****
>
>        receive an encapsulated packet from smart endnode E using ingress*=
*
> **
>
>        nickname "R2".****
>
> ** **
>
> If the smart endnode uses either R1 or R2 in the above example, would we
> have TRILL RPF issues where upstream RBs see a packet with SRB=3Dsay R1
> coming from a wrong TRILL port? Let=92s say the RPF happens for multicast
> only then we will atleast need to make sure the multicast (broadcast,
> unknown unicast and multcast) is sent via the edge-RB whose RBID the smar=
t
> endnode is using.****
>
> ** **
>
> This problem should not be there when using different RBID other than R1
> and R2 as described in approach-2****
>
> ** **
>
> Phani****
>
> *From:* Radia Perlman [mailto:radiaperlman@gmail.com]
> *Sent:* Wednesday, January 09, 2013 2:53 AM
> *To:* Phanidhar Koganti
> *Cc:* Kesava_Vijaya_Krupak@Dell.com; trill@ietf.org; d3e3e3@gmail.com;
> hu.fangwei@zte.com.cn
> *Subject:* Re: New version : TRILL Smart Endnodes draft****
>
> ** **
>
> Most of what you said is correct.  One correction****
>
> ** **
>
> a) When the packet reaches the egress RB I am presuming there will be a
> inner-MAC lookup to deliver the packet to the egress end-node.****
>
> ** **
>
> correct...when egress RB R has to look up the inner MAC address E, which
> it has to do anyway without smart endnodes, in order to choose which port
> to forward it on.  With smart endnodes, when R looks up MAC address E, it
> not only discovers which port to send it on, but that E is "smart", in
> which case R must leave the TRILL encapsulation****
>
> ** **
>
> b) Edge-RB to smart endnode cannot have TRILL encapsulation right?****
>
> ** **
>
> Not correct.  Traffic from egress RB to smart endnode will have TRILL
> encapsulation.  So for example, let's say R is the egress RB, and R has a
> link with two endnodes; E1 (which is smart) and E2 (which is not smart).*=
*
> **
>
> ** **
>
> When R receives a TRILL packet from the campus, with TRILL destination
> nickname=3DR, R looks at the MAC destination.  If it is E2, then R remove=
s
> the TRILL encapulation and forwards a native packet onto the link.  If it
> is E1, then R forwards the packet with the TRILL header.****
>
> ** **
>
> c) Edge-RB to smart endnode cannot have TRILL encapsulation right? If so
> for my clarification we are saying the traffic towards the end-node will =
be
> non-TRILL while towards the ingress edge-RB will be TRILL encapsulated?**=
*
> *
>
> ** **
>
> Hopefully my answer to b) clarified this.  On the link between endnodes
> and R, whether it's to or from endnode <--> edge RB, the packet will have
> TRILL encapsulation if the endnode is smart, and not have TRILL
> encapsulation if the endnode is not smart.****
>
> ** **
>
> Radia****
>
> ** **
>
> ** **
>
> On Thu, Jan 3, 2013 at 3:19 AM, Phanidhar Koganti <pkoganti@brocade.com>
> wrote:****
>
> The draft defines the control and data path behavior from smart end-node
> to edge-RB. When the packet reaches the egress RB I am presuming there wi=
ll
> be a inner-MAC lookup to deliver the packet to the egress end-node. Edge-=
RB
> to smart endnode cannot have TRILL encapsulation right? If so for my
> clarification we are saying the traffic towards the end-node will be
> non-TRILL while towards the ingress edge-RB will be TRILL encapsulated?**=
*
> *
>
>  ****
>
> Phani****
>
>  ****
>
> *From:* trill-bounces@ietf.org [mailto:trill-bounces@ietf.org] *On Behalf
> Of *Kesava_Vijaya_Krupak@Dell.com
> *Sent:* Thursday, January 03, 2013 4:19 PM
> *To:* trill@ietf.org
> *Cc:* d3e3e3@gmail.com; radiaperlman@gmail.com; hu.fangwei@zte.com.cn
> *Subject:* [trill] New version : TRILL Smart Endnodes draft****
>
>  ****
>
> Hi all,****
>
>  ****
>
> A new version of the Trill Smart Endnodes draft has been posted.****
>
>  ****
>
> http://datatracker.ietf.org/doc/draft-perlman-trill-smart-endnodes****
>
>  ****
>
> Kindly let us know your comments.****
>
>  ****
>
> Thanks,****
>
> kvk****
>
> ** **
>

--f46d04374ee1e0893004d2d512ec
Content-Type: text/html; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

Good point (see Phani&#39;s comment below). =A0I&#39;ll summarize:<div><br>=
</div><div>The current =A0text implies that it&#39;s OK for smart endnode E=
, multihomed to R1 and R2, to choose one of the nicknames (say R1) for &quo=
t;ingress nickname&quot; in the TRILL header, and forward to either R1 or R=
2. =A0There are two problems with E forwarding a TRILL-encapsulated frame u=
sing ingeress &quot;R1&quot;, through R2:</div>
<div><br></div><div>a) for multicast, likely to confuse the RPF check</div>=
<div>b) for unicast, E would probably need to warn R2 that E is multihomed =
to R1 and will e using R1&#39;s nickname for ingress.</div><div><br></div>
<div>And actually, E won&#39;t get that much load splitting benefit from be=
ing multihomed if it uses &quot;R1&quot; as ingress, since all traffic to E=
 will go through R1. =A0So, if E doesn&#39;t have a pseudonode nickname to =
use, then perhaps E should just use one R1 or R2.</div>
<div><br></div><div>Or E could get its own nickname (I&#39;ve been floating=
 around the idea of a &quot;dumb RBridge&quot;, which is like a &quot;smart=
 endnode&quot;, except it gets its own nickname, and does generate an LSP w=
ith &quot;overloaded&quot; flag. =A0It probably has to receive LSPs and cal=
culate Dijkstra, in order to know which port to send multicast on (unless i=
ts attached true RBs are helpful enough to tell it...&quot;you can send mul=
ticast with trees X or Y through me&quot;)</div>
<div><br></div><div>So, it seems like there are several options which shoul=
d be discussed, for a smart endnode multihomed to R1 and R2</div><div>1) Al=
ways use a pseudnode nickname (but that&#39;s kind of complicated since tha=
t nickname can only be used by smart endnodes that are multihomed to the ex=
act same set of RBridges)</div>
<div>2) Don&#39;t do active-active...choose one or R1 or R2 and always send=
 everything that way</div><div>3) Do active-active, but only for unicast...=
choose one of the nicknames, say &quot;R1&quot;, and it&#39;s OK to send un=
icast either through R1 or R2, but multicast has to go through R1</div>
<div>4) Give E its own nickname (and somehow make sure it knows which port =
it can multicast on)</div><div>5) Other possibilities?</div><div><br></div>=
<div>Radia</div><div><br></div><div><br></div><div><br></div><div><br><br>
<div class=3D"gmail_quote">On Tue, Jan 8, 2013 at 8:32 PM, Phanidhar Kogant=
i <span dir=3D"ltr">&lt;<a href=3D"mailto:pkoganti@brocade.com" target=3D"_=
blank">pkoganti@brocade.com</a>&gt;</span> wrote:<br><blockquote class=3D"g=
mail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-l=
eft:1ex">
<div lang=3D"EN-US" link=3D"blue" vlink=3D"purple"><div><p class=3D"MsoNorm=
al"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&quot;s=
ans-serif&quot;;color:#1f497d">Thanks Radia. That clarifies for me. A more =
descriptive text in draft can surely help vendors as they implement it in t=
heir hardware.<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">Another question on ha=
ndling multi-homing<u></u><u></u></span></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt"><span style=3D"font-siz=
e:10.0pt;font-family:&quot;Courier New&quot;">E can choose either R1 or R2&=
#39;s nickname, when encapsulating a<u></u><u></u></span></p><p class=3D"Ms=
oNormal" style=3D"line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 frame, whether the encapsulated frame is sent via R1 or R2. If=
 E<u></u><u></u></span></p><p class=3D"MsoNormal" style=3D"line-height:14.4=
pt"><span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=
=A0=A0=A0=A0=A0=A0 wants to do active-active load splitting, and uses R1&#3=
9;s nickname<u></u><u></u></span></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt"><span style=3D"font-siz=
e:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=A0=A0=A0=A0 when forwa=
rding through R1, and R2&#39;s nickname when forwarding<u></u><u></u></span=
></p><p class=3D"MsoNormal" style=3D"line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 through R2, this will cause distant RBridges (or smart endnode=
s)<u></u><u></u></span></p><p class=3D"MsoNormal" style=3D"line-height:14.4=
pt"><span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=
=A0=A0=A0=A0=A0=A0 to keep changing their endnode table entry for D between=
 (D, R1&#39;s<u></u><u></u></span></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt"><span style=3D"font-siz=
e:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=A0=A0=A0=A0 nickname) =
and (D, R2&#39;s nickname). So it would be preferable for E<u></u><u></u></=
span></p><p class=3D"MsoNormal" style=3D"line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 to always encapsulate using the same nickname (R1 or R2) unles=
s E<u></u><u></u></span></p><p class=3D"MsoNormal" style=3D"line-height:14.=
4pt"><span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=
=A0=A0=A0=A0=A0=A0 detects a problem with connectivity using that nickname.=
 And in<u></u><u></u></span></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt"><span style=3D"font-siz=
e:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=A0=A0=A0=A0 this case,=
 R1 and R2 need to be informed that the smart endnode<u></u><u></u></span><=
/p><p class=3D"MsoNormal" style=3D"line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 might encapsulate with a different nickname, i.e., R1 might<u>=
</u><u></u></span></p><p class=3D"MsoNormal" style=3D"line-height:14.4pt"><=
span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 receive an encapsulated packet from smart endnode E using ingr=
ess<u></u><u></u></span></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt"><span style=3D"font-siz=
e:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=A0=A0=A0=A0 nickname &=
quot;R2&quot;.<u></u><u></u></span></p><p class=3D"MsoNormal"><span style=
=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&quot;sans-serif&quot;=
;color:#1f497d"><u></u>=A0<u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">If the smart endnode uses=
 either R1 or R2 in the above example, would we have TRILL RPF issues where=
 upstream RBs see a packet with SRB=3Dsay R1 coming from a wrong TRILL port=
? Let=92s say the RPF happens for multicast only then we will atleast need =
to make sure the multicast (broadcast, unknown unicast and multcast) is sen=
t via the edge-RB whose RBID the smart endnode is using.<u></u><u></u></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">This problem should no=
t be there when using different RBID other than R1 and R2 as described in a=
pproach-2<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">Phani<u></u><u></u></s=
pan></p>
<p class=3D"MsoNormal"><b><span style=3D"font-size:10.0pt;font-family:&quot=
;Tahoma&quot;,&quot;sans-serif&quot;">From:</span></b><span style=3D"font-s=
ize:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;"> Radia Pe=
rlman [mailto:<a href=3D"mailto:radiaperlman@gmail.com" target=3D"_blank">r=
adiaperlman@gmail.com</a>] <br>
<b>Sent:</b> Wednesday, January 09, 2013 2:53 AM<br><b>To:</b> Phanidhar Ko=
ganti<br><b>Cc:</b> Kesava_Vijaya_Krupak@Dell.com; <a href=3D"mailto:trill@=
ietf.org" target=3D"_blank">trill@ietf.org</a>; <a href=3D"mailto:d3e3e3@gm=
ail.com" target=3D"_blank">d3e3e3@gmail.com</a>; <a href=3D"mailto:hu.fangw=
ei@zte.com.cn" target=3D"_blank">hu.fangwei@zte.com.cn</a><br>
<b>Subject:</b> Re: New version : TRILL Smart Endnodes draft<u></u><u></u><=
/span></p><div><div class=3D"h5"><p class=3D"MsoNormal"><u></u>=A0<u></u></=
p><p class=3D"MsoNormal">Most of what you said is correct. =A0One correctio=
n<u></u><u></u></p>
<div><p class=3D"MsoNormal"><u></u>=A0<u></u></p></div><div><p class=3D"Mso=
Normal">a)=A0<span style=3D"color:#1f497d">When the packet reaches the egre=
ss RB I am presuming there will be a inner-MAC lookup to deliver the packet=
 to the egress end-node.</span><u></u><u></u></p>
</div><div><p class=3D"MsoNormal"><u></u>=A0<u></u></p></div><div><p class=
=3D"MsoNormal"><span style=3D"color:#1f497d">correct...when egress RB R has=
 to look up the inner MAC address E, which it has to do anyway without smar=
t endnodes, in order to choose which port to forward it on. =A0With smart e=
ndnodes, when R looks up MAC address E, it not only discovers which port to=
 send it on, but that E is &quot;smart&quot;, in which case R must leave th=
e TRILL encapsulation</span><u></u><u></u></p>
</div><div><p class=3D"MsoNormal"><u></u>=A0<u></u></p></div><div><p class=
=3D"MsoNormal"><span style=3D"color:#1f497d">b)=A0Edge-RB to smart endnode =
cannot have TRILL encapsulation right?</span><u></u><u></u></p></div><div><=
p class=3D"MsoNormal">
<u></u>=A0<u></u></p></div><div><p class=3D"MsoNormal"><span style=3D"color=
:#1f497d">Not correct. =A0Traffic from egress RB to smart endnode will have=
 TRILL encapsulation. =A0So for example, let&#39;s say R is the egress RB, =
and R has a link with two endnodes; E1 (which is smart) and E2 (which is no=
t smart).</span><u></u><u></u></p>
</div><div><p class=3D"MsoNormal"><u></u>=A0<u></u></p></div><div><p class=
=3D"MsoNormal"><span style=3D"color:#1f497d">When R receives a TRILL packet=
 from the campus, with TRILL destination nickname=3DR, R looks at the MAC d=
estination. =A0If it is E2, then R removes the TRILL encapulation and forwa=
rds a native packet onto the link. =A0If it is E1, then R forwards the pack=
et with the TRILL header.</span><u></u><u></u></p>
</div><div><p class=3D"MsoNormal"><u></u>=A0<u></u></p></div><div><p class=
=3D"MsoNormal"><span style=3D"color:#1f497d">c)=A0Edge-RB to smart endnode =
cannot have TRILL encapsulation right? If so for my clarification we are sa=
ying the traffic towards the end-node will be non-TRILL while towards the i=
ngress edge-RB will be TRILL encapsulated?</span><u></u><u></u></p>
</div><div><p class=3D"MsoNormal"><u></u>=A0<u></u></p></div><div><p class=
=3D"MsoNormal"><span style=3D"color:#1f497d">Hopefully my answer to b) clar=
ified this. =A0On the link between endnodes and R, whether it&#39;s to or f=
rom endnode &lt;--&gt; edge RB, the packet will have TRILL encapsulation if=
 the endnode is smart, and not have TRILL encapsulation if the endnode is n=
ot smart.</span><u></u><u></u></p>
</div><div><p class=3D"MsoNormal"><u></u>=A0<u></u></p></div><div><p class=
=3D"MsoNormal"><span style=3D"color:#1f497d">Radia</span><u></u><u></u></p>=
</div><div><p class=3D"MsoNormal"><u></u>=A0<u></u></p></div><div><p class=
=3D"MsoNormal">
<u></u>=A0<u></u></p><div><div><p class=3D"MsoNormal">On Thu, Jan 3, 2013 a=
t 3:19 AM, Phanidhar Koganti &lt;<a href=3D"mailto:pkoganti@brocade.com" ta=
rget=3D"_blank">pkoganti@brocade.com</a>&gt; wrote:<u></u><u></u></p><div><=
div>
<p class=3D"MsoNormal"><span style=3D"color:#1f497d">The draft defines the =
control and data path behavior from smart end-node to edge-RB. When the pac=
ket reaches the egress RB I am presuming there will be a inner-MAC lookup t=
o deliver the packet to the egress end-node. Edge-RB to smart endnode canno=
t have TRILL encapsulation right? If so for my clarification we are saying =
the traffic towards the end-node will be non-TRILL while towards the ingres=
s edge-RB will be TRILL encapsulated?</span><u></u><u></u></p>
<p class=3D"MsoNormal"><span style=3D"color:#1f497d">=A0</span><u></u><u></=
u></p><p class=3D"MsoNormal"><span style=3D"color:#1f497d">Phani</span><u><=
/u><u></u></p><p class=3D"MsoNormal"><span style=3D"color:#1f497d">=A0</spa=
n><u></u><u></u></p>
<div><div style=3D"border:none;border-top:solid #b5c4df 1.0pt;padding:3.0pt=
 0in 0in 0in"><p class=3D"MsoNormal"><b><span style=3D"font-size:10.0pt;fon=
t-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;">From:</span></b><span s=
tyle=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&qu=
ot;"> <a href=3D"mailto:trill-bounces@ietf.org" target=3D"_blank">trill-bou=
nces@ietf.org</a> [mailto:<a href=3D"mailto:trill-bounces@ietf.org" target=
=3D"_blank">trill-bounces@ietf.org</a>] <b>On Behalf Of </b><a href=3D"mail=
to:Kesava_Vijaya_Krupak@Dell.com" target=3D"_blank">Kesava_Vijaya_Krupak@De=
ll.com</a><br>
<b>Sent:</b> Thursday, January 03, 2013 4:19 PM<br><b>To:</b> <a href=3D"ma=
ilto:trill@ietf.org" target=3D"_blank">trill@ietf.org</a><br><b>Cc:</b> <a =
href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gmail.com</a>; <a=
 href=3D"mailto:radiaperlman@gmail.com" target=3D"_blank">radiaperlman@gmai=
l.com</a>; <a href=3D"mailto:hu.fangwei@zte.com.cn" target=3D"_blank">hu.fa=
ngwei@zte.com.cn</a><br>
<b>Subject:</b> [trill] New version : TRILL Smart Endnodes draft</span><u><=
/u><u></u></p></div></div><div><div><p class=3D"MsoNormal">=A0<u></u><u></u=
></p><p class=3D"MsoNormal">Hi all,<u></u><u></u></p><p class=3D"MsoNormal"=
>=A0<u></u><u></u></p>
<p class=3D"MsoNormal">A new version of the Trill Smart Endnodes draft has =
been posted.<u></u><u></u></p><p class=3D"MsoNormal">=A0<u></u><u></u></p><=
p class=3D"MsoNormal"><a href=3D"http://datatracker.ietf.org/doc/draft-perl=
man-trill-smart-endnodes" target=3D"_blank">http://datatracker.ietf.org/doc=
/draft-perlman-trill-smart-endnodes</a><u></u><u></u></p>
<p class=3D"MsoNormal">=A0<u></u><u></u></p><p class=3D"MsoNormal">Kindly l=
et us know your comments.<u></u><u></u></p><p class=3D"MsoNormal">=A0<u></u=
><u></u></p><p class=3D"MsoNormal">Thanks,<u></u><u></u></p><p class=3D"Mso=
Normal">kvk<u></u><u></u></p>
</div></div></div></div></div><p class=3D"MsoNormal"><u></u>=A0<u></u></p><=
/div></div></div></div></div></div></blockquote></div><br></div>

--f46d04374ee1e0893004d2d512ec--

From pkoganti@Brocade.COM  Wed Jan  9 00:24:58 2013
Return-Path: <pkoganti@Brocade.COM>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7A52621F84BB for <trill@ietfa.amsl.com>; Wed,  9 Jan 2013 00:24:58 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.064
X-Spam-Level: 
X-Spam-Status: No, score=-2.064 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, IP_NOT_FRIENDLY=0.334, J_CHICKENPOX_33=0.6, J_CHICKENPOX_81=0.6, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id u3FNSU3LnyuP for <trill@ietfa.amsl.com>; Wed,  9 Jan 2013 00:24:53 -0800 (PST)
Received: from mx0a-000f0801.pphosted.com (mx0a-000f0801.pphosted.com [67.231.144.122]) by ietfa.amsl.com (Postfix) with ESMTP id 642E021F8439 for <trill@ietf.org>; Wed,  9 Jan 2013 00:24:53 -0800 (PST)
Received: from pps.filterd (m0000542 [127.0.0.1]) by mx0a-000f0801.pphosted.com (8.14.5/8.14.5) with SMTP id r098OQTE029231; Wed, 9 Jan 2013 00:24:50 -0800
Received: from hq1wp-exchub01.corp.brocade.com ([144.49.131.13]) by mx0a-000f0801.pphosted.com with ESMTP id 19rta709js-1 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Wed, 09 Jan 2013 00:24:49 -0800
Received: from HQ1WP-EXHUB01.corp.brocade.com (10.70.36.14) by HQ1WP-EXCHUB01.corp.brocade.com (10.70.36.99) with Microsoft SMTP Server (TLS) id 14.2.309.2; Wed, 9 Jan 2013 00:24:49 -0800
Received: from HQ1-EXCH02.corp.brocade.com ([fe80::c92a:772e:befa:c34c]) by HQ1WP-EXHUB01.corp.brocade.com ([::1]) with mapi; Wed, 9 Jan 2013 00:24:48 -0800
From: Phanidhar Koganti <pkoganti@Brocade.COM>
To: Radia Perlman <radiaperlman@gmail.com>
Date: Wed, 9 Jan 2013 00:24:46 -0800
Thread-Topic: New version : TRILL Smart Endnodes draft
Thread-Index: Ac3uMZS6IOIGdQ6wTQubEUjijXkKlwACCo8g
Message-ID: <6895EAE0CA8DEE40B92D7CA88BB521F32414007BDC@HQ1-EXCH02.corp.brocade.com>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com> <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com> <6895EAE0CA8DEE40B92D7CA88BB521F32414007BC1@HQ1-EXCH02.corp.brocade.com> <CAFOuuo626hw-c0kM4-1Fp-w2gByPenRnkFP3xNT+cTB4YcJ2kg@mail.gmail.com>
In-Reply-To: <CAFOuuo626hw-c0kM4-1Fp-w2gByPenRnkFP3xNT+cTB4YcJ2kg@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US
Content-Type: multipart/alternative; boundary="_000_6895EAE0CA8DEE40B92D7CA88BB521F32414007BDCHQ1EXCH02corp_"
MIME-Version: 1.0
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:5.9.8327, 1.0.431, 0.0.0000 definitions=2013-01-09_02:2013-01-08, 2013-01-09, 1970-01-01 signatures=0
X-Proofpoint-Spam-Details: rule=notspam policy=default score=0 spamscore=0 suspectscore=49 phishscore=0 adultscore=0 bulkscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=7.0.1-1211240000 definitions=main-1301090006
Cc: "d3e3e3@gmail.com" <d3e3e3@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 09 Jan 2013 08:24:59 -0000

--_000_6895EAE0CA8DEE40B92D7CA88BB521F32414007BDCHQ1EXCH02corp_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Radia,

Valid point for unicast that the pair in the example needs to be aware that=
 it might get unicast traffic with TRILL Source RBID equal to its RBID. Thi=
s brings up an interesting scenario.

Let's say there are two smart end-nodes e1 & e2 which are both multi-homed =
behind R1 & R2. Also say both e1 & e2 pick R1 to use in TRILL source RBID, =
then when e1 is talking to e2 we end up in a encap where both TRILL SA/DA w=
ill have the same value =3D R1.

I am inclined towards a pseudo RB for the end-node with its own nickname, b=
ut am not clear on the impact on Dijkstra as the number of routable RBIDs w=
ill considerably go up.

Few additional concerns:

For multicast it will also trigger confusion on source suppression that nee=
ds to happen in the peer RB. Let's say there is a smart end-node e1 & e2 mu=
lti-homed behind R1 and R2. For multicast or flooded traffic we need source=
 suppression so that the traffic does not go back to the smart endnode. Or =
is the draft assuming that the smart end-node will perform the suppression =
if the packet comes back to it?

Also for the packet direction edge-RB to smart end-node, if we end up with =
unknown MAC then we should flood to all the locally hanging smart end-nodes=
 and regular end-nodes. When flooding towards smart end-nodes are we expect=
ed to continue the TRILL packet with M=3D0 or convert it to M=3D1 with T_DA=
=3DROOT. If we like to continue with M=3D0 then the edge-RB needs to suppor=
t unicast replication for multicast.

Phani

From: Radia Perlman [mailto:radiaperlman@gmail.com]
Sent: Wednesday, January 09, 2013 11:52 AM
To: Phanidhar Koganti
Cc: Kesava_Vijaya_Krupak@Dell.com; trill@ietf.org; d3e3e3@gmail.com; hu.fan=
gwei@zte.com.cn
Subject: Re: New version : TRILL Smart Endnodes draft

Good point (see Phani's comment below).  I'll summarize:

The current  text implies that it's OK for smart endnode E, multihomed to R=
1 and R2, to choose one of the nicknames (say R1) for "ingress nickname" in=
 the TRILL header, and forward to either R1 or R2.  There are two problems =
with E forwarding a TRILL-encapsulated frame using ingeress "R1", through R=
2:

a) for multicast, likely to confuse the RPF check
b) for unicast, E would probably need to warn R2 that E is multihomed to R1=
 and will e using R1's nickname for ingress.

And actually, E won't get that much load splitting benefit from being multi=
homed if it uses "R1" as ingress, since all traffic to E will go through R1=
.  So, if E doesn't have a pseudonode nickname to use, then perhaps E shoul=
d just use one R1 or R2.

Or E could get its own nickname (I've been floating around the idea of a "d=
umb RBridge", which is like a "smart endnode", except it gets its own nickn=
ame, and does generate an LSP with "overloaded" flag.  It probably has to r=
eceive LSPs and calculate Dijkstra, in order to know which port to send mul=
ticast on (unless its attached true RBs are helpful enough to tell it..."yo=
u can send multicast with trees X or Y through me")

So, it seems like there are several options which should be discussed, for =
a smart endnode multihomed to R1 and R2
1) Always use a pseudnode nickname (but that's kind of complicated since th=
at nickname can only be used by smart endnodes that are multihomed to the e=
xact same set of RBridges)
2) Don't do active-active...choose one or R1 or R2 and always send everythi=
ng that way
3) Do active-active, but only for unicast...choose one of the nicknames, sa=
y "R1", and it's OK to send unicast either through R1 or R2, but multicast =
has to go through R1
4) Give E its own nickname (and somehow make sure it knows which port it ca=
n multicast on)
5) Other possibilities?

Radia




On Tue, Jan 8, 2013 at 8:32 PM, Phanidhar Koganti <pkoganti@brocade.com<mai=
lto:pkoganti@brocade.com>> wrote:
Thanks Radia. That clarifies for me. A more descriptive text in draft can s=
urely help vendors as they implement it in their hardware.

Another question on handling multi-homing
E can choose either R1 or R2's nickname, when encapsulating a
       frame, whether the encapsulated frame is sent via R1 or R2. If E
       wants to do active-active load splitting, and uses R1's nickname
       when forwarding through R1, and R2's nickname when forwarding
       through R2, this will cause distant RBridges (or smart endnodes)
       to keep changing their endnode table entry for D between (D, R1's
       nickname) and (D, R2's nickname). So it would be preferable for E
       to always encapsulate using the same nickname (R1 or R2) unless E
       detects a problem with connectivity using that nickname. And in
       this case, R1 and R2 need to be informed that the smart endnode
       might encapsulate with a different nickname, i.e., R1 might
       receive an encapsulated packet from smart endnode E using ingress
       nickname "R2".

If the smart endnode uses either R1 or R2 in the above example, would we ha=
ve TRILL RPF issues where upstream RBs see a packet with SRB=3Dsay R1 comin=
g from a wrong TRILL port? Let's say the RPF happens for multicast only the=
n we will atleast need to make sure the multicast (broadcast, unknown unica=
st and multcast) is sent via the edge-RB whose RBID the smart endnode is us=
ing.

This problem should not be there when using different RBID other than R1 an=
d R2 as described in approach-2

Phani
From: Radia Perlman [mailto:radiaperlman@gmail.com<mailto:radiaperlman@gmai=
l.com>]
Sent: Wednesday, January 09, 2013 2:53 AM
To: Phanidhar Koganti
Cc: Kesava_Vijaya_Krupak@Dell.com<mailto:Kesava_Vijaya_Krupak@Dell.com>; tr=
ill@ietf.org<mailto:trill@ietf.org>; d3e3e3@gmail.com<mailto:d3e3e3@gmail.c=
om>; hu.fangwei@zte.com.cn<mailto:hu.fangwei@zte.com.cn>
Subject: Re: New version : TRILL Smart Endnodes draft

Most of what you said is correct.  One correction

a) When the packet reaches the egress RB I am presuming there will be a inn=
er-MAC lookup to deliver the packet to the egress end-node.

correct...when egress RB R has to look up the inner MAC address E, which it=
 has to do anyway without smart endnodes, in order to choose which port to =
forward it on.  With smart endnodes, when R looks up MAC address E, it not =
only discovers which port to send it on, but that E is "smart", in which ca=
se R must leave the TRILL encapsulation

b) Edge-RB to smart endnode cannot have TRILL encapsulation right?

Not correct.  Traffic from egress RB to smart endnode will have TRILL encap=
sulation.  So for example, let's say R is the egress RB, and R has a link w=
ith two endnodes; E1 (which is smart) and E2 (which is not smart).

When R receives a TRILL packet from the campus, with TRILL destination nick=
name=3DR, R looks at the MAC destination.  If it is E2, then R removes the =
TRILL encapulation and forwards a native packet onto the link.  If it is E1=
, then R forwards the packet with the TRILL header.

c) Edge-RB to smart endnode cannot have TRILL encapsulation right? If so fo=
r my clarification we are saying the traffic towards the end-node will be n=
on-TRILL while towards the ingress edge-RB will be TRILL encapsulated?

Hopefully my answer to b) clarified this.  On the link between endnodes and=
 R, whether it's to or from endnode <--> edge RB, the packet will have TRIL=
L encapsulation if the endnode is smart, and not have TRILL encapsulation i=
f the endnode is not smart.

Radia


On Thu, Jan 3, 2013 at 3:19 AM, Phanidhar Koganti <pkoganti@brocade.com<mai=
lto:pkoganti@brocade.com>> wrote:
The draft defines the control and data path behavior from smart end-node to=
 edge-RB. When the packet reaches the egress RB I am presuming there will b=
e a inner-MAC lookup to deliver the packet to the egress end-node. Edge-RB =
to smart endnode cannot have TRILL encapsulation right? If so for my clarif=
ication we are saying the traffic towards the end-node will be non-TRILL wh=
ile towards the ingress edge-RB will be TRILL encapsulated?

Phani

From: trill-bounces@ietf.org<mailto:trill-bounces@ietf.org> [mailto:trill-b=
ounces@ietf.org<mailto:trill-bounces@ietf.org>] On Behalf Of Kesava_Vijaya_=
Krupak@Dell.com<mailto:Kesava_Vijaya_Krupak@Dell.com>
Sent: Thursday, January 03, 2013 4:19 PM
To: trill@ietf.org<mailto:trill@ietf.org>
Cc: d3e3e3@gmail.com<mailto:d3e3e3@gmail.com>; radiaperlman@gmail.com<mailt=
o:radiaperlman@gmail.com>; hu.fangwei@zte.com.cn<mailto:hu.fangwei@zte.com.=
cn>
Subject: [trill] New version : TRILL Smart Endnodes draft

Hi all,

A new version of the Trill Smart Endnodes draft has been posted.

http://datatracker.ietf.org/doc/draft-perlman-trill-smart-endnodes

Kindly let us know your comments.

Thanks,
kvk



--_000_6895EAE0CA8DEE40B92D7CA88BB521F32414007BDCHQ1EXCH02corp_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micr=
osoft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" xmlns=3D"http:=
//www.w3.org/TR/REC-html40"><head><meta http-equiv=3DContent-Type content=
=3D"text/html; charset=3Dus-ascii"><meta name=3DGenerator content=3D"Micros=
oft Word 14 (filtered medium)"><style><!--
/* Font Definitions */
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:Tahoma;
	panose-1:2 11 6 4 3 5 4 4 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman","serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
p.MsoAcetate, li.MsoAcetate, div.MsoAcetate
	{mso-style-priority:99;
	mso-style-link:"Balloon Text Char";
	margin:0in;
	margin-bottom:.0001pt;
	font-size:8.0pt;
	font-family:"Tahoma","sans-serif";}
span.BalloonTextChar
	{mso-style-name:"Balloon Text Char";
	mso-style-priority:99;
	mso-style-link:"Balloon Text";
	font-family:"Tahoma","sans-serif";}
span.EmailStyle19
	{mso-style-type:personal-reply;
	font-family:"Calibri","sans-serif";
	color:#1F497D;}
.MsoChpDefault
	{mso-style-type:export-only;
	font-family:"Calibri","sans-serif";}
@page WordSection1
	{size:8.5in 11.0in;
	margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
	{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]--></head><body lang=3DEN-US link=3Dblue vli=
nk=3Dpurple><div class=3DWordSection1><p class=3DMsoNormal><span style=3D'f=
ont-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>Radia,<o:=
p></o:p></span></p><p class=3DMsoNormal><span style=3D'font-size:11.0pt;fon=
t-family:"Calibri","sans-serif";color:#1F497D'><o:p>&nbsp;</o:p></span></p>=
<p class=3DMsoNormal><span style=3D'font-size:11.0pt;font-family:"Calibri",=
"sans-serif";color:#1F497D'>Valid point for unicast that the pair in the ex=
ample needs to be aware that it might get unicast traffic with TRILL Source=
 RBID equal to its RBID. This brings up an interesting scenario. <o:p></o:p=
></span></p><p class=3DMsoNormal><span style=3D'font-size:11.0pt;font-famil=
y:"Calibri","sans-serif";color:#1F497D'><o:p>&nbsp;</o:p></span></p><p clas=
s=3DMsoNormal><span style=3D'font-size:11.0pt;font-family:"Calibri","sans-s=
erif";color:#1F497D'>Let&#8217;s say there are two smart end-nodes e1 &amp;=
 e2 which are both multi-homed behind R1 &amp; R2. Also say both e1 &amp; e=
2 pick R1 to use in TRILL source RBID, then when e1 is talking to e2 we end=
 up in a encap where both TRILL SA/DA will have the same value =3D R1.<o:p>=
</o:p></span></p><p class=3DMsoNormal><span style=3D'font-size:11.0pt;font-=
family:"Calibri","sans-serif";color:#1F497D'><o:p>&nbsp;</o:p></span></p><p=
 class=3DMsoNormal><span style=3D'font-size:11.0pt;font-family:"Calibri","s=
ans-serif";color:#1F497D'>I am inclined towards a pseudo RB for the end-nod=
e with its own nickname, but am not clear on the impact on Dijkstra as the =
number of routable RBIDs will considerably go up. <o:p></o:p></span></p><p =
class=3DMsoNormal><span style=3D'font-size:11.0pt;font-family:"Calibri","sa=
ns-serif";color:#1F497D'><o:p>&nbsp;</o:p></span></p><p class=3DMsoNormal><=
span style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F=
497D'>Few additional concerns:<o:p></o:p></span></p><p class=3DMsoNormal><s=
pan style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F4=
97D'><o:p>&nbsp;</o:p></span></p><p class=3DMsoNormal><span style=3D'font-s=
ize:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>For multicast =
it will also trigger confusion on source suppression that needs to happen i=
n the peer RB. Let&#8217;s say there is a smart end-node e1 &amp; e2 multi-=
homed behind R1 and R2. For multicast or flooded traffic we need source sup=
pression so that the traffic does not go back to the smart endnode. Or is t=
he draft assuming that the smart end-node will perform the suppression if t=
he packet comes back to it? <o:p></o:p></span></p><p class=3DMsoNormal><spa=
n style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'><o:p>&nbsp;</o:p></span></p><p class=3DMsoNormal><span style=3D'font-siz=
e:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>Also for the pac=
ket direction edge-RB to smart end-node, if we end up with unknown MAC then=
 we should flood to all the locally hanging smart end-nodes and regular end=
-nodes. When flooding towards smart end-nodes are we expected to continue t=
he TRILL packet with M=3D0 or convert it to M=3D1 with T_DA=3DROOT. If we l=
ike to continue with M=3D0 then the edge-RB needs to support unicast replic=
ation for multicast.<o:p></o:p></span></p><p class=3DMsoNormal><span style=
=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'><o:p=
>&nbsp;</o:p></span></p><p class=3DMsoNormal><span style=3D'font-size:11.0p=
t;font-family:"Calibri","sans-serif";color:#1F497D'>Phani<o:p></o:p></span>=
</p><p class=3DMsoNormal><span style=3D'font-size:11.0pt;font-family:"Calib=
ri","sans-serif";color:#1F497D'><o:p>&nbsp;</o:p></span></p><p class=3DMsoN=
ormal><b><span style=3D'font-size:10.0pt;font-family:"Tahoma","sans-serif"'=
>From:</span></b><span style=3D'font-size:10.0pt;font-family:"Tahoma","sans=
-serif"'> Radia Perlman [mailto:radiaperlman@gmail.com] <br><b>Sent:</b> We=
dnesday, January 09, 2013 11:52 AM<br><b>To:</b> Phanidhar Koganti<br><b>Cc=
:</b> Kesava_Vijaya_Krupak@Dell.com; trill@ietf.org; d3e3e3@gmail.com; hu.f=
angwei@zte.com.cn<br><b>Subject:</b> Re: New version : TRILL Smart Endnodes=
 draft<o:p></o:p></span></p><p class=3DMsoNormal><o:p>&nbsp;</o:p></p><p cl=
ass=3DMsoNormal>Good point (see Phani's comment below). &nbsp;I'll summariz=
e:<o:p></o:p></p><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div>=
<p class=3DMsoNormal>The current &nbsp;text implies that it's OK for smart =
endnode E, multihomed to R1 and R2, to choose one of the nicknames (say R1)=
 for &quot;ingress nickname&quot; in the TRILL header, and forward to eithe=
r R1 or R2. &nbsp;There are two problems with E forwarding a TRILL-encapsul=
ated frame using ingeress &quot;R1&quot;, through R2:<o:p></o:p></p></div><=
div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p class=3DMsoNorm=
al>a) for multicast, likely to confuse the RPF check<o:p></o:p></p></div><d=
iv><p class=3DMsoNormal>b) for unicast, E would probably need to warn R2 th=
at E is multihomed to R1 and will e using R1's nickname for ingress.<o:p></=
o:p></p></div><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p =
class=3DMsoNormal>And actually, E won't get that much load splitting benefi=
t from being multihomed if it uses &quot;R1&quot; as ingress, since all tra=
ffic to E will go through R1. &nbsp;So, if E doesn't have a pseudonode nick=
name to use, then perhaps E should just use one R1 or R2.<o:p></o:p></p></d=
iv><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p class=3DMso=
Normal>Or E could get its own nickname (I've been floating around the idea =
of a &quot;dumb RBridge&quot;, which is like a &quot;smart endnode&quot;, e=
xcept it gets its own nickname, and does generate an LSP with &quot;overloa=
ded&quot; flag. &nbsp;It probably has to receive LSPs and calculate Dijkstr=
a, in order to know which port to send multicast on (unless its attached tr=
ue RBs are helpful enough to tell it...&quot;you can send multicast with tr=
ees X or Y through me&quot;)<o:p></o:p></p></div><div><p class=3DMsoNormal>=
<o:p>&nbsp;</o:p></p></div><div><p class=3DMsoNormal>So, it seems like ther=
e are several options which should be discussed, for a smart endnode multih=
omed to R1 and R2<o:p></o:p></p></div><div><p class=3DMsoNormal>1) Always u=
se a pseudnode nickname (but that's kind of complicated since that nickname=
 can only be used by smart endnodes that are multihomed to the exact same s=
et of RBridges)<o:p></o:p></p></div><div><p class=3DMsoNormal>2) Don't do a=
ctive-active...choose one or R1 or R2 and always send everything that way<o=
:p></o:p></p></div><div><p class=3DMsoNormal>3) Do active-active, but only =
for unicast...choose one of the nicknames, say &quot;R1&quot;, and it's OK =
to send unicast either through R1 or R2, but multicast has to go through R1=
<o:p></o:p></p></div><div><p class=3DMsoNormal>4) Give E its own nickname (=
and somehow make sure it knows which port it can multicast on)<o:p></o:p></=
p></div><div><p class=3DMsoNormal>5) Other possibilities?<o:p></o:p></p></d=
iv><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p class=3DMso=
Normal>Radia<o:p></o:p></p></div><div><p class=3DMsoNormal><o:p>&nbsp;</o:p=
></p></div><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p cla=
ss=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p class=3DMsoNormal style=
=3D'margin-bottom:12.0pt'><o:p>&nbsp;</o:p></p><div><p class=3DMsoNormal>On=
 Tue, Jan 8, 2013 at 8:32 PM, Phanidhar Koganti &lt;<a href=3D"mailto:pkoga=
nti@brocade.com" target=3D"_blank">pkoganti@brocade.com</a>&gt; wrote:<o:p>=
</o:p></p><div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;m=
so-margin-bottom-alt:auto'><span style=3D'font-size:11.0pt;font-family:"Cal=
ibri","sans-serif";color:#1F497D'>Thanks Radia. That clarifies for me. A mo=
re descriptive text in draft can surely help vendors as they implement it i=
n their hardware.</span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-ma=
rgin-top-alt:auto;mso-margin-bottom-alt:auto'><span style=3D'font-size:11.0=
pt;font-family:"Calibri","sans-serif";color:#1F497D'>&nbsp;</span><o:p></o:=
p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bott=
om-alt:auto'><span style=3D'font-size:11.0pt;font-family:"Calibri","sans-se=
rif";color:#1F497D'>Another question on handling multi-homing</span><o:p></=
o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bo=
ttom-alt:auto;line-height:14.4pt'><span style=3D'font-size:10.0pt;font-fami=
ly:"Courier New"'>E can choose either R1 or R2's nickname, when encapsulati=
ng a</span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:=
auto;mso-margin-bottom-alt:auto;line-height:14.4pt'><span style=3D'font-siz=
e:10.0pt;font-family:"Courier New"'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; fr=
ame, whether the encapsulated frame is sent via R1 or R2. If E</span><o:p><=
/o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-b=
ottom-alt:auto;line-height:14.4pt'><span style=3D'font-size:10.0pt;font-fam=
ily:"Courier New"'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; wants to do active-=
active load splitting, and uses R1's nickname</span><o:p></o:p></p><p class=
=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto;li=
ne-height:14.4pt'><span style=3D'font-size:10.0pt;font-family:"Courier New"=
'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; when forwarding through R1, and R2's=
 nickname when forwarding</span><o:p></o:p></p><p class=3DMsoNormal style=
=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto;line-height:14.4pt'>=
<span style=3D'font-size:10.0pt;font-family:"Courier New"'>&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp; through R2, this will cause distant RBridges (or smart=
 endnodes)</span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-to=
p-alt:auto;mso-margin-bottom-alt:auto;line-height:14.4pt'><span style=3D'fo=
nt-size:10.0pt;font-family:"Courier New"'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp; to keep changing their endnode table entry for D between (D, R1's</span=
><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-m=
argin-bottom-alt:auto;line-height:14.4pt'><span style=3D'font-size:10.0pt;f=
ont-family:"Courier New"'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; nickname) an=
d (D, R2's nickname). So it would be preferable for E</span><o:p></o:p></p>=
<p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt=
:auto;line-height:14.4pt'><span style=3D'font-size:10.0pt;font-family:"Cour=
ier New"'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; to always encapsulate using =
the same nickname (R1 or R2) unless E</span><o:p></o:p></p><p class=3DMsoNo=
rmal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto;line-heigh=
t:14.4pt'><span style=3D'font-size:10.0pt;font-family:"Courier New"'>&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; detects a problem with connectivity using th=
at nickname. And in</span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-=
margin-top-alt:auto;mso-margin-bottom-alt:auto;line-height:14.4pt'><span st=
yle=3D'font-size:10.0pt;font-family:"Courier New"'>&nbsp;&nbsp;&nbsp;&nbsp;=
&nbsp;&nbsp; this case, R1 and R2 need to be informed that the smart endnod=
e</span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:aut=
o;mso-margin-bottom-alt:auto;line-height:14.4pt'><span style=3D'font-size:1=
0.0pt;font-family:"Courier New"'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; might=
 encapsulate with a different nickname, i.e., R1 might</span><o:p></o:p></p=
><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-al=
t:auto;line-height:14.4pt'><span style=3D'font-size:10.0pt;font-family:"Cou=
rier New"'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; receive an encapsulated pac=
ket from smart endnode E using ingress</span><o:p></o:p></p><p class=3DMsoN=
ormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto;line-heig=
ht:14.4pt'><span style=3D'font-size:10.0pt;font-family:"Courier New"'>&nbsp=
;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; nickname &quot;R2&quot;.</span><o:p></o:p><=
/p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-=
alt:auto'><span style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif=
";color:#1F497D'>&nbsp;</span><o:p></o:p></p><p class=3DMsoNormal style=3D'=
mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span style=3D'font-siz=
e:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>If the smart end=
node uses either R1 or R2 in the above example, would we have TRILL RPF iss=
ues where upstream RBs see a packet with SRB=3Dsay R1 coming from a wrong T=
RILL port? Let&#8217;s say the RPF happens for multicast only then we will =
atleast need to make sure the multicast (broadcast, unknown unicast and mul=
tcast) is sent via the edge-RB whose RBID the smart endnode is using.</span=
><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-m=
argin-bottom-alt:auto'><span style=3D'font-size:11.0pt;font-family:"Calibri=
","sans-serif";color:#1F497D'>&nbsp;</span><o:p></o:p></p><p class=3DMsoNor=
mal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span styl=
e=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>Thi=
s problem should not be there when using different RBID other than R1 and R=
2 as described in approach-2</span><o:p></o:p></p><p class=3DMsoNormal styl=
e=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span style=3D'fon=
t-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>&nbsp;</spa=
n><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-=
margin-bottom-alt:auto'><span style=3D'font-size:11.0pt;font-family:"Calibr=
i","sans-serif";color:#1F497D'>Phani</span><o:p></o:p></p><p class=3DMsoNor=
mal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><b><span s=
tyle=3D'font-size:10.0pt;font-family:"Tahoma","sans-serif"'>From:</span></b=
><span style=3D'font-size:10.0pt;font-family:"Tahoma","sans-serif"'> Radia =
Perlman [mailto:<a href=3D"mailto:radiaperlman@gmail.com" target=3D"_blank"=
>radiaperlman@gmail.com</a>] <br><b>Sent:</b> Wednesday, January 09, 2013 2=
:53 AM<br><b>To:</b> Phanidhar Koganti<br><b>Cc:</b> <a href=3D"mailto:Kesa=
va_Vijaya_Krupak@Dell.com">Kesava_Vijaya_Krupak@Dell.com</a>; <a href=3D"ma=
ilto:trill@ietf.org" target=3D"_blank">trill@ietf.org</a>; <a href=3D"mailt=
o:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gmail.com</a>; <a href=3D"mail=
to:hu.fangwei@zte.com.cn" target=3D"_blank">hu.fangwei@zte.com.cn</a><br><b=
>Subject:</b> Re: New version : TRILL Smart Endnodes draft</span><o:p></o:p=
></p><div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-ma=
rgin-bottom-alt:auto'>&nbsp;<o:p></o:p></p><p class=3DMsoNormal style=3D'ms=
o-margin-top-alt:auto;mso-margin-bottom-alt:auto'>Most of what you said is =
correct. &nbsp;One correction<o:p></o:p></p><div><p class=3DMsoNormal style=
=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p></o:p></=
p></div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-marg=
in-bottom-alt:auto'>a)&nbsp;<span style=3D'color:#1F497D'>When the packet r=
eaches the egress RB I am presuming there will be a inner-MAC lookup to del=
iver the packet to the egress end-node.</span><o:p></o:p></p></div><div><p =
class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:au=
to'>&nbsp;<o:p></o:p></p></div><div><p class=3DMsoNormal style=3D'mso-margi=
n-top-alt:auto;mso-margin-bottom-alt:auto'><span style=3D'color:#1F497D'>co=
rrect...when egress RB R has to look up the inner MAC address E, which it h=
as to do anyway without smart endnodes, in order to choose which port to fo=
rward it on. &nbsp;With smart endnodes, when R looks up MAC address E, it n=
ot only discovers which port to send it on, but that E is &quot;smart&quot;=
, in which case R must leave the TRILL encapsulation</span><o:p></o:p></p><=
/div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-=
bottom-alt:auto'>&nbsp;<o:p></o:p></p></div><div><p class=3DMsoNormal style=
=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span style=3D'colo=
r:#1F497D'>b)&nbsp;Edge-RB to smart endnode cannot have TRILL encapsulation=
 right?</span><o:p></o:p></p></div><div><p class=3DMsoNormal style=3D'mso-m=
argin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p></o:p></p></div><=
div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom=
-alt:auto'><span style=3D'color:#1F497D'>Not correct. &nbsp;Traffic from eg=
ress RB to smart endnode will have TRILL encapsulation. &nbsp;So for exampl=
e, let's say R is the egress RB, and R has a link with two endnodes; E1 (wh=
ich is smart) and E2 (which is not smart).</span><o:p></o:p></p></div><div>=
<p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt=
:auto'>&nbsp;<o:p></o:p></p></div><div><p class=3DMsoNormal style=3D'mso-ma=
rgin-top-alt:auto;mso-margin-bottom-alt:auto'><span style=3D'color:#1F497D'=
>When R receives a TRILL packet from the campus, with TRILL destination nic=
kname=3DR, R looks at the MAC destination. &nbsp;If it is E2, then R remove=
s the TRILL encapulation and forwards a native packet onto the link. &nbsp;=
If it is E1, then R forwards the packet with the TRILL header.</span><o:p><=
/o:p></p></div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;m=
so-margin-bottom-alt:auto'>&nbsp;<o:p></o:p></p></div><div><p class=3DMsoNo=
rmal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span sty=
le=3D'color:#1F497D'>c)&nbsp;Edge-RB to smart endnode cannot have TRILL enc=
apsulation right? If so for my clarification we are saying the traffic towa=
rds the end-node will be non-TRILL while towards the ingress edge-RB will b=
e TRILL encapsulated?</span><o:p></o:p></p></div><div><p class=3DMsoNormal =
style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p></o=
:p></p></div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso=
-margin-bottom-alt:auto'><span style=3D'color:#1F497D'>Hopefully my answer =
to b) clarified this. &nbsp;On the link between endnodes and R, whether it'=
s to or from endnode &lt;--&gt; edge RB, the packet will have TRILL encapsu=
lation if the endnode is smart, and not have TRILL encapsulation if the end=
node is not smart.</span><o:p></o:p></p></div><div><p class=3DMsoNormal sty=
le=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p></o:p>=
</p></div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-ma=
rgin-bottom-alt:auto'><span style=3D'color:#1F497D'>Radia</span><o:p></o:p>=
</p></div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-ma=
rgin-bottom-alt:auto'>&nbsp;<o:p></o:p></p></div><div><p class=3DMsoNormal =
style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p></o=
:p></p><div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-=
margin-bottom-alt:auto'>On Thu, Jan 3, 2013 at 3:19 AM, Phanidhar Koganti &=
lt;<a href=3D"mailto:pkoganti@brocade.com" target=3D"_blank">pkoganti@broca=
de.com</a>&gt; wrote:<o:p></o:p></p><div><div><p class=3DMsoNormal style=3D=
'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span style=3D'color:#=
1F497D'>The draft defines the control and data path behavior from smart end=
-node to edge-RB. When the packet reaches the egress RB I am presuming ther=
e will be a inner-MAC lookup to deliver the packet to the egress end-node. =
Edge-RB to smart endnode cannot have TRILL encapsulation right? If so for m=
y clarification we are saying the traffic towards the end-node will be non-=
TRILL while towards the ingress edge-RB will be TRILL encapsulated?</span><=
o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-mar=
gin-bottom-alt:auto'><span style=3D'color:#1F497D'>&nbsp;</span><o:p></o:p>=
</p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom=
-alt:auto'><span style=3D'color:#1F497D'>Phani</span><o:p></o:p></p><p clas=
s=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>=
<span style=3D'color:#1F497D'>&nbsp;</span><o:p></o:p></p><div><div style=
=3D'border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0in 0in 0in'><=
p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:=
auto'><b><span style=3D'font-size:10.0pt;font-family:"Tahoma","sans-serif"'=
>From:</span></b><span style=3D'font-size:10.0pt;font-family:"Tahoma","sans=
-serif"'> <a href=3D"mailto:trill-bounces@ietf.org" target=3D"_blank">trill=
-bounces@ietf.org</a> [mailto:<a href=3D"mailto:trill-bounces@ietf.org" tar=
get=3D"_blank">trill-bounces@ietf.org</a>] <b>On Behalf Of </b><a href=3D"m=
ailto:Kesava_Vijaya_Krupak@Dell.com" target=3D"_blank">Kesava_Vijaya_Krupak=
@Dell.com</a><br><b>Sent:</b> Thursday, January 03, 2013 4:19 PM<br><b>To:<=
/b> <a href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ietf.org</a><=
br><b>Cc:</b> <a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@=
gmail.com</a>; <a href=3D"mailto:radiaperlman@gmail.com" target=3D"_blank">=
radiaperlman@gmail.com</a>; <a href=3D"mailto:hu.fangwei@zte.com.cn" target=
=3D"_blank">hu.fangwei@zte.com.cn</a><br><b>Subject:</b> [trill] New versio=
n : TRILL Smart Endnodes draft</span><o:p></o:p></p></div></div><div><div><=
p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:=
auto'>&nbsp;<o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt=
:auto;mso-margin-bottom-alt:auto'>Hi all,<o:p></o:p></p><p class=3DMsoNorma=
l style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p><=
/o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-b=
ottom-alt:auto'>A new version of the Trill Smart Endnodes draft has been po=
sted.<o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;m=
so-margin-bottom-alt:auto'>&nbsp;<o:p></o:p></p><p class=3DMsoNormal style=
=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><a href=3D"http://d=
atatracker.ietf.org/doc/draft-perlman-trill-smart-endnodes" target=3D"_blan=
k">http://datatracker.ietf.org/doc/draft-perlman-trill-smart-endnodes</a><o=
:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-marg=
in-bottom-alt:auto'>&nbsp;<o:p></o:p></p><p class=3DMsoNormal style=3D'mso-=
margin-top-alt:auto;mso-margin-bottom-alt:auto'>Kindly let us know your com=
ments.<o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;=
mso-margin-bottom-alt:auto'>&nbsp;<o:p></o:p></p><p class=3DMsoNormal style=
=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>Thanks,<o:p></o:p><=
/p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-=
alt:auto'>kvk<o:p></o:p></p></div></div></div></div></div><p class=3DMsoNor=
mal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p=
></o:p></p></div></div></div></div></div></div></div><p class=3DMsoNormal><=
o:p>&nbsp;</o:p></p></div></div></body></html>=

--_000_6895EAE0CA8DEE40B92D7CA88BB521F32414007BDCHQ1EXCH02corp_--

From radiaperlman@gmail.com  Wed Jan  9 00:36:24 2013
Return-Path: <radiaperlman@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3291421F85BF for <trill@ietfa.amsl.com>; Wed,  9 Jan 2013 00:36:24 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.729
X-Spam-Level: 
X-Spam-Status: No, score=-2.729 tagged_above=-999 required=5 tests=[AWL=-0.331, BAYES_00=-2.599, HTML_MESSAGE=0.001, J_CHICKENPOX_33=0.6, J_CHICKENPOX_81=0.6, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Wy60l6O5Bnqc for <trill@ietfa.amsl.com>; Wed,  9 Jan 2013 00:36:22 -0800 (PST)
Received: from mail-ea0-f181.google.com (mail-ea0-f181.google.com [209.85.215.181]) by ietfa.amsl.com (Postfix) with ESMTP id 1884521F8589 for <trill@ietf.org>; Wed,  9 Jan 2013 00:36:21 -0800 (PST)
Received: by mail-ea0-f181.google.com with SMTP id k14so530372eaa.26 for <trill@ietf.org>; Wed, 09 Jan 2013 00:36:21 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=UHZQLCppKj2t+mqNrN2dbRBkSWtmIendDE3Nxjn4jrQ=; b=zmCM9qvXmlkuZ5tAR7zcsx+0jnnPGgrjlFMUCYIpRtKsL9H+iTvsBFhHSGmUO52lUb 4Gj5yf3A4iFVRR7Htzur0BLPRByPYd93HXAR8RrfZR/nMGjdC45ZsZAWB94ntfJ4xGvk nNQwT8PhGREZSgNZhYBARb/yFXezQskNivbL6hpZK63YeWKOuECErL05F2dqx7/eE2I/ ai7hFTLpEuCkIiZYFwx5l/37jv4+PPTxbnr9nlQketL3HQzlM2e+vrEB4WTR9C8jDrD0 vNPx6RusHYrTepACDDkHaLinr8gCabpfC5uAuyU7atDhVmUJqSDaULoVYhmEgu7c9JT7 lN+g==
MIME-Version: 1.0
Received: by 10.14.174.132 with SMTP id x4mr177222262eel.39.1357720581158; Wed, 09 Jan 2013 00:36:21 -0800 (PST)
Received: by 10.14.130.66 with HTTP; Wed, 9 Jan 2013 00:36:20 -0800 (PST)
In-Reply-To: <6895EAE0CA8DEE40B92D7CA88BB521F32414007BDC@HQ1-EXCH02.corp.brocade.com>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com> <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com> <6895EAE0CA8DEE40B92D7CA88BB521F32414007BC1@HQ1-EXCH02.corp.brocade.com> <CAFOuuo626hw-c0kM4-1Fp-w2gByPenRnkFP3xNT+cTB4YcJ2kg@mail.gmail.com> <6895EAE0CA8DEE40B92D7CA88BB521F32414007BDC@HQ1-EXCH02.corp.brocade.com>
Date: Wed, 9 Jan 2013 00:36:20 -0800
Message-ID: <CAFOuuo6CK7=1uKU3j6WCSVEjd6PqYRxpsvC55Y66mVCPN_zTzg@mail.gmail.com>
From: Radia Perlman <radiaperlman@gmail.com>
To: Phanidhar Koganti <pkoganti@brocade.com>
Content-Type: multipart/alternative; boundary=047d7b621de2f1b72004d2d6f420
Cc: "d3e3e3@gmail.com" <d3e3e3@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 09 Jan 2013 08:36:24 -0000

--047d7b621de2f1b72004d2d6f420
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

Hmm.  Interesting example...

Let's first ignore multihoming, and assuming that two smart endnodes, E1
and E2, are both connected to R1.  I think the case works where E1 and E2
are on the same shared link...they will learn that the other can be reached
directly, and talk to each other without TRILL encapsulation.

But let's consider the case where E1 and E2 are both singly homed to R, but
using different ports.  Then they will have to talk to each other with
TRILL encapsulation with both ingress and egress nickname=3DR1.  That's a b=
it
weird, but I think it works.  Does it?

-------
As for multicast showing up twice...it shouldn't...only one RB should be
attached to a link with an endnode via any particular tree, so I think only
one copy of a multicast packet should be delivered.

--------
As for having each smart endnode use its own nickname...the real problem is
nickname exhaustion (running out of nicknames).  There's only 64K
nicknames, unless we do the multilevel thing where nicknames only need to
be unique within an area, but the RBs that forward between level 1 and
level 2 have to rewrite the nickname fields.

----
As for unknown destinations/learning...I thought we considered that.  But
smart endnodes will not be unknown to the attached RB...they explicitly
announce themselves.

Radia



On Wed, Jan 9, 2013 at 12:24 AM, Phanidhar Koganti <pkoganti@brocade.com>wr=
ote:

> Radia,****
>
> ** **
>
> Valid point for unicast that the pair in the example needs to be aware
> that it might get unicast traffic with TRILL Source RBID equal to its RBI=
D.
> This brings up an interesting scenario. ****
>
> ** **
>
> Let=92s say there are two smart end-nodes e1 & e2 which are both multi-ho=
med
> behind R1 & R2. Also say both e1 & e2 pick R1 to use in TRILL source RBID=
,
> then when e1 is talking to e2 we end up in a encap where both TRILL SA/DA
> will have the same value =3D R1.****
>
> ** **
>
> I am inclined towards a pseudo RB for the end-node with its own nickname,
> but am not clear on the impact on Dijkstra as the number of routable RBID=
s
> will considerably go up. ****
>
> ** **
>
> Few additional concerns:****
>
> ** **
>
> For multicast it will also trigger confusion on source suppression that
> needs to happen in the peer RB. Let=92s say there is a smart end-node e1 =
& e2
> multi-homed behind R1 and R2. For multicast or flooded traffic we need
> source suppression so that the traffic does not go back to the smart
> endnode. Or is the draft assuming that the smart end-node will perform th=
e
> suppression if the packet comes back to it? ****
>
> ** **
>
> Also for the packet direction edge-RB to smart end-node, if we end up wit=
h
> unknown MAC then we should flood to all the locally hanging smart end-nod=
es
> and regular end-nodes. When flooding towards smart end-nodes are we
> expected to continue the TRILL packet with M=3D0 or convert it to M=3D1 w=
ith
> T_DA=3DROOT. If we like to continue with M=3D0 then the edge-RB needs to
> support unicast replication for multicast.****
>
> ** **
>
> Phani****
>
> ** **
>
> *From:* Radia Perlman [mailto:radiaperlman@gmail.com]
> *Sent:* Wednesday, January 09, 2013 11:52 AM
>
> *To:* Phanidhar Koganti
> *Cc:* Kesava_Vijaya_Krupak@Dell.com; trill@ietf.org; d3e3e3@gmail.com;
> hu.fangwei@zte.com.cn
> *Subject:* Re: New version : TRILL Smart Endnodes draft****
>
> ** **
>
> Good point (see Phani's comment below).  I'll summarize:****
>
> ** **
>
> The current  text implies that it's OK for smart endnode E, multihomed to
> R1 and R2, to choose one of the nicknames (say R1) for "ingress nickname"
> in the TRILL header, and forward to either R1 or R2.  There are two
> problems with E forwarding a TRILL-encapsulated frame using ingeress "R1"=
,
> through R2:****
>
> ** **
>
> a) for multicast, likely to confuse the RPF check****
>
> b) for unicast, E would probably need to warn R2 that E is multihomed to
> R1 and will e using R1's nickname for ingress.****
>
> ** **
>
> And actually, E won't get that much load splitting benefit from being
> multihomed if it uses "R1" as ingress, since all traffic to E will go
> through R1.  So, if E doesn't have a pseudonode nickname to use, then
> perhaps E should just use one R1 or R2.****
>
> ** **
>
> Or E could get its own nickname (I've been floating around the idea of a
> "dumb RBridge", which is like a "smart endnode", except it gets its own
> nickname, and does generate an LSP with "overloaded" flag.  It probably h=
as
> to receive LSPs and calculate Dijkstra, in order to know which port to se=
nd
> multicast on (unless its attached true RBs are helpful enough to tell
> it..."you can send multicast with trees X or Y through me")****
>
> ** **
>
> So, it seems like there are several options which should be discussed, fo=
r
> a smart endnode multihomed to R1 and R2****
>
> 1) Always use a pseudnode nickname (but that's kind of complicated since
> that nickname can only be used by smart endnodes that are multihomed to t=
he
> exact same set of RBridges)****
>
> 2) Don't do active-active...choose one or R1 or R2 and always send
> everything that way****
>
> 3) Do active-active, but only for unicast...choose one of the nicknames,
> say "R1", and it's OK to send unicast either through R1 or R2, but
> multicast has to go through R1****
>
> 4) Give E its own nickname (and somehow make sure it knows which port it
> can multicast on)****
>
> 5) Other possibilities?****
>
> ** **
>
> Radia****
>
> ** **
>
> ** **
>
> ** **
>
> ** **
>
> On Tue, Jan 8, 2013 at 8:32 PM, Phanidhar Koganti <pkoganti@brocade.com>
> wrote:****
>
> Thanks Radia. That clarifies for me. A more descriptive text in draft can
> surely help vendors as they implement it in their hardware.****
>
>  ****
>
> Another question on handling multi-homing****
>
> E can choose either R1 or R2's nickname, when encapsulating a****
>
>        frame, whether the encapsulated frame is sent via R1 or R2. If E**=
*
> *
>
>        wants to do active-active load splitting, and uses R1's nickname**=
*
> *
>
>        when forwarding through R1, and R2's nickname when forwarding****
>
>        through R2, this will cause distant RBridges (or smart endnodes)**=
*
> *
>
>        to keep changing their endnode table entry for D between (D, R1's*=
*
> **
>
>        nickname) and (D, R2's nickname). So it would be preferable for E*=
*
> **
>
>        to always encapsulate using the same nickname (R1 or R2) unless E*=
*
> **
>
>        detects a problem with connectivity using that nickname. And in***=
*
>
>        this case, R1 and R2 need to be informed that the smart endnode***=
*
>
>        might encapsulate with a different nickname, i.e., R1 might****
>
>        receive an encapsulated packet from smart endnode E using ingress*=
*
> **
>
>        nickname "R2".****
>
>  ****
>
> If the smart endnode uses either R1 or R2 in the above example, would we
> have TRILL RPF issues where upstream RBs see a packet with SRB=3Dsay R1
> coming from a wrong TRILL port? Let=92s say the RPF happens for multicast
> only then we will atleast need to make sure the multicast (broadcast,
> unknown unicast and multcast) is sent via the edge-RB whose RBID the smar=
t
> endnode is using.****
>
>  ****
>
> This problem should not be there when using different RBID other than R1
> and R2 as described in approach-2****
>
>  ****
>
> Phani****
>
> *From:* Radia Perlman [mailto:radiaperlman@gmail.com]
> *Sent:* Wednesday, January 09, 2013 2:53 AM
> *To:* Phanidhar Koganti
> *Cc:* Kesava_Vijaya_Krupak@Dell.com; trill@ietf.org; d3e3e3@gmail.com;
> hu.fangwei@zte.com.cn
> *Subject:* Re: New version : TRILL Smart Endnodes draft****
>
>  ****
>
> Most of what you said is correct.  One correction****
>
>  ****
>
> a) When the packet reaches the egress RB I am presuming there will be a
> inner-MAC lookup to deliver the packet to the egress end-node.****
>
>  ****
>
> correct...when egress RB R has to look up the inner MAC address E, which
> it has to do anyway without smart endnodes, in order to choose which port
> to forward it on.  With smart endnodes, when R looks up MAC address E, it
> not only discovers which port to send it on, but that E is "smart", in
> which case R must leave the TRILL encapsulation****
>
>  ****
>
> b) Edge-RB to smart endnode cannot have TRILL encapsulation right?****
>
>  ****
>
> Not correct.  Traffic from egress RB to smart endnode will have TRILL
> encapsulation.  So for example, let's say R is the egress RB, and R has a
> link with two endnodes; E1 (which is smart) and E2 (which is not smart).*=
*
> **
>
>  ****
>
> When R receives a TRILL packet from the campus, with TRILL destination
> nickname=3DR, R looks at the MAC destination.  If it is E2, then R remove=
s
> the TRILL encapulation and forwards a native packet onto the link.  If it
> is E1, then R forwards the packet with the TRILL header.****
>
>  ****
>
> c) Edge-RB to smart endnode cannot have TRILL encapsulation right? If so
> for my clarification we are saying the traffic towards the end-node will =
be
> non-TRILL while towards the ingress edge-RB will be TRILL encapsulated?**=
*
> *
>
>  ****
>
> Hopefully my answer to b) clarified this.  On the link between endnodes
> and R, whether it's to or from endnode <--> edge RB, the packet will have
> TRILL encapsulation if the endnode is smart, and not have TRILL
> encapsulation if the endnode is not smart.****
>
>  ****
>
> Radia****
>
>  ****
>
>  ****
>
> On Thu, Jan 3, 2013 at 3:19 AM, Phanidhar Koganti <pkoganti@brocade.com>
> wrote:****
>
> The draft defines the control and data path behavior from smart end-node
> to edge-RB. When the packet reaches the egress RB I am presuming there wi=
ll
> be a inner-MAC lookup to deliver the packet to the egress end-node. Edge-=
RB
> to smart endnode cannot have TRILL encapsulation right? If so for my
> clarification we are saying the traffic towards the end-node will be
> non-TRILL while towards the ingress edge-RB will be TRILL encapsulated?**=
*
> *
>
>  ****
>
> Phani****
>
>  ****
>
> *From:* trill-bounces@ietf.org [mailto:trill-bounces@ietf.org] *On Behalf
> Of *Kesava_Vijaya_Krupak@Dell.com
> *Sent:* Thursday, January 03, 2013 4:19 PM
> *To:* trill@ietf.org
> *Cc:* d3e3e3@gmail.com; radiaperlman@gmail.com; hu.fangwei@zte.com.cn
> *Subject:* [trill] New version : TRILL Smart Endnodes draft****
>
>  ****
>
> Hi all,****
>
>  ****
>
> A new version of the Trill Smart Endnodes draft has been posted.****
>
>  ****
>
> http://datatracker.ietf.org/doc/draft-perlman-trill-smart-endnodes****
>
>  ****
>
> Kindly let us know your comments.****
>
>  ****
>
> Thanks,****
>
> kvk****
>
>  ****
>
> ** **
>

--047d7b621de2f1b72004d2d6f420
Content-Type: text/html; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

Hmm. =A0Interesting example...<div><br></div><div>Let&#39;s first ignore mu=
ltihoming, and assuming that two smart endnodes, E1 and E2, are both connec=
ted to R1. =A0I think the case works where E1 and E2 are on the same shared=
 link...they will learn that the other can be reached directly, and talk to=
 each other without TRILL encapsulation.</div>
<div><br></div><div>But let&#39;s consider the case where E1 and E2 are bot=
h singly homed to R, but using different ports. =A0Then they will have to t=
alk to each other with TRILL encapsulation with both ingress and egress nic=
kname=3DR1. =A0That&#39;s a bit weird, but I think it works. =A0Does it?</d=
iv>
<div><br></div><div>-------</div><div>As for multicast showing up twice...i=
t shouldn&#39;t...only one RB should be attached to a link with an endnode =
via any particular tree, so I think only one copy of a multicast packet sho=
uld be delivered.</div>
<div><br></div><div>--------</div><div>As for having each smart endnode use=
 its own nickname...the real problem is nickname exhaustion (running out of=
 nicknames). =A0There&#39;s only 64K nicknames, unless we do the multilevel=
 thing where nicknames only need to be unique within an area, but the RBs t=
hat forward between level 1 and level 2 have to rewrite the nickname fields=
.</div>
<div><br></div><div>----</div><div>As for unknown destinations/learning...I=
 thought we considered that. =A0But smart endnodes will not be unknown to t=
he attached RB...they explicitly announce themselves.</div><div><br></div>
<div>Radia</div><div><br></div><div><br></div><div><br><div class=3D"gmail_=
quote">On Wed, Jan 9, 2013 at 12:24 AM, Phanidhar Koganti <span dir=3D"ltr"=
>&lt;<a href=3D"mailto:pkoganti@brocade.com" target=3D"_blank">pkoganti@bro=
cade.com</a>&gt;</span> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div lang=3D"EN-US" link=3D"blue" vlink=3D"p=
urple"><div><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-fam=
ily:&quot;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">Radia,<u></u>=
<u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">Valid point for unicas=
t that the pair in the example needs to be aware that it might get unicast =
traffic with TRILL Source RBID equal to its RBID. This brings up an interes=
ting scenario. <u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">Let=92s say there are =
two smart end-nodes e1 &amp; e2 which are both multi-homed behind R1 &amp; =
R2. Also say both e1 &amp; e2 pick R1 to use in TRILL source RBID, then whe=
n e1 is talking to e2 we end up in a encap where both TRILL SA/DA will have=
 the same value =3D R1.<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">I am inclined towards =
a pseudo RB for the end-node with its own nickname, but am not clear on the=
 impact on Dijkstra as the number of routable RBIDs will considerably go up=
. <u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">Few additional concern=
s:<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">For multicast it will =
also trigger confusion on source suppression that needs to happen in the pe=
er RB. Let=92s say there is a smart end-node e1 &amp; e2 multi-homed behind=
 R1 and R2. For multicast or flooded traffic we need source suppression so =
that the traffic does not go back to the smart endnode. Or is the draft ass=
uming that the smart end-node will perform the suppression if the packet co=
mes back to it? <u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">Also for the packet di=
rection edge-RB to smart end-node, if we end up with unknown MAC then we sh=
ould flood to all the locally hanging smart end-nodes and regular end-nodes=
. When flooding towards smart end-nodes are we expected to continue the TRI=
LL packet with M=3D0 or convert it to M=3D1 with T_DA=3DROOT. If we like to=
 continue with M=3D0 then the edge-RB needs to support unicast replication =
for multicast.<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">Phani<u></u><u></u></s=
pan></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p><p class=3D"MsoNormal"><b><span style=3D"font-size:10.0pt;font-family:&q=
uot;Tahoma&quot;,&quot;sans-serif&quot;">From:</span></b><span style=3D"fon=
t-size:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;"> Radia=
 Perlman [mailto:<a href=3D"mailto:radiaperlman@gmail.com" target=3D"_blank=
">radiaperlman@gmail.com</a>] <br>
<b>Sent:</b> Wednesday, January 09, 2013 11:52 AM</span></p><div><div class=
=3D"h5"><br><b>To:</b> Phanidhar Koganti<br><b>Cc:</b> Kesava_Vijaya_Krupak=
@Dell.com; <a href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ietf.o=
rg</a>; <a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gmail.=
com</a>; <a href=3D"mailto:hu.fangwei@zte.com.cn" target=3D"_blank">hu.fang=
wei@zte.com.cn</a><br>
<b>Subject:</b> Re: New version : TRILL Smart Endnodes draft<u></u><u></u><=
/div></div><p></p><div><div class=3D"h5"><p class=3D"MsoNormal"><u></u>=A0<=
u></u></p><p class=3D"MsoNormal">Good point (see Phani&#39;s comment below)=
. =A0I&#39;ll summarize:<u></u><u></u></p>
<div><p class=3D"MsoNormal"><u></u>=A0<u></u></p></div><div><p class=3D"Mso=
Normal">The current =A0text implies that it&#39;s OK for smart endnode E, m=
ultihomed to R1 and R2, to choose one of the nicknames (say R1) for &quot;i=
ngress nickname&quot; in the TRILL header, and forward to either R1 or R2. =
=A0There are two problems with E forwarding a TRILL-encapsulated frame usin=
g ingeress &quot;R1&quot;, through R2:<u></u><u></u></p>
</div><div><p class=3D"MsoNormal"><u></u>=A0<u></u></p></div><div><p class=
=3D"MsoNormal">a) for multicast, likely to confuse the RPF check<u></u><u><=
/u></p></div><div><p class=3D"MsoNormal">b) for unicast, E would probably n=
eed to warn R2 that E is multihomed to R1 and will e using R1&#39;s nicknam=
e for ingress.<u></u><u></u></p>
</div><div><p class=3D"MsoNormal"><u></u>=A0<u></u></p></div><div><p class=
=3D"MsoNormal">And actually, E won&#39;t get that much load splitting benef=
it from being multihomed if it uses &quot;R1&quot; as ingress, since all tr=
affic to E will go through R1. =A0So, if E doesn&#39;t have a pseudonode ni=
ckname to use, then perhaps E should just use one R1 or R2.<u></u><u></u></=
p>
</div><div><p class=3D"MsoNormal"><u></u>=A0<u></u></p></div><div><p class=
=3D"MsoNormal">Or E could get its own nickname (I&#39;ve been floating arou=
nd the idea of a &quot;dumb RBridge&quot;, which is like a &quot;smart endn=
ode&quot;, except it gets its own nickname, and does generate an LSP with &=
quot;overloaded&quot; flag. =A0It probably has to receive LSPs and calculat=
e Dijkstra, in order to know which port to send multicast on (unless its at=
tached true RBs are helpful enough to tell it...&quot;you can send multicas=
t with trees X or Y through me&quot;)<u></u><u></u></p>
</div><div><p class=3D"MsoNormal"><u></u>=A0<u></u></p></div><div><p class=
=3D"MsoNormal">So, it seems like there are several options which should be =
discussed, for a smart endnode multihomed to R1 and R2<u></u><u></u></p></d=
iv>
<div><p class=3D"MsoNormal">1) Always use a pseudnode nickname (but that&#3=
9;s kind of complicated since that nickname can only be used by smart endno=
des that are multihomed to the exact same set of RBridges)<u></u><u></u></p=
>
</div><div><p class=3D"MsoNormal">2) Don&#39;t do active-active...choose on=
e or R1 or R2 and always send everything that way<u></u><u></u></p></div><d=
iv><p class=3D"MsoNormal">3) Do active-active, but only for unicast...choos=
e one of the nicknames, say &quot;R1&quot;, and it&#39;s OK to send unicast=
 either through R1 or R2, but multicast has to go through R1<u></u><u></u><=
/p>
</div><div><p class=3D"MsoNormal">4) Give E its own nickname (and somehow m=
ake sure it knows which port it can multicast on)<u></u><u></u></p></div><d=
iv><p class=3D"MsoNormal">5) Other possibilities?<u></u><u></u></p></div><d=
iv>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p></div><div><p class=3D"MsoNorma=
l">Radia<u></u><u></u></p></div><div><p class=3D"MsoNormal"><u></u>=A0<u></=
u></p></div><div><p class=3D"MsoNormal"><u></u>=A0<u></u></p></div><div><p =
class=3D"MsoNormal">
<u></u>=A0<u></u></p></div><div><p class=3D"MsoNormal" style=3D"margin-bott=
om:12.0pt"><u></u>=A0<u></u></p><div><p class=3D"MsoNormal">On Tue, Jan 8, =
2013 at 8:32 PM, Phanidhar Koganti &lt;<a href=3D"mailto:pkoganti@brocade.c=
om" target=3D"_blank">pkoganti@brocade.com</a>&gt; wrote:<u></u><u></u></p>
<div><div><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-famil=
y:&quot;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">Thanks Radia. T=
hat clarifies for me. A more descriptive text in draft can surely help vend=
ors as they implement it in their hardware.</span><u></u><u></u></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">=A0</span><u></u><u></u><=
/p><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">Another question on ha=
ndling multi-homing</span><u></u><u></u></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt"><span style=3D"font-siz=
e:10.0pt;font-family:&quot;Courier New&quot;">E can choose either R1 or R2&=
#39;s nickname, when encapsulating a</span><u></u><u></u></p><p class=3D"Ms=
oNormal" style=3D"line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 frame, whether the encapsulated frame is sent via R1 or R2. If=
 E</span><u></u><u></u></p><p class=3D"MsoNormal" style=3D"line-height:14.4=
pt"><span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=
=A0=A0=A0=A0=A0=A0 wants to do active-active load splitting, and uses R1&#3=
9;s nickname</span><u></u><u></u></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt"><span style=3D"font-siz=
e:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=A0=A0=A0=A0 when forwa=
rding through R1, and R2&#39;s nickname when forwarding</span><u></u><u></u=
></p><p class=3D"MsoNormal" style=3D"line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 through R2, this will cause distant RBridges (or smart endnode=
s)</span><u></u><u></u></p><p class=3D"MsoNormal" style=3D"line-height:14.4=
pt"><span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=
=A0=A0=A0=A0=A0=A0 to keep changing their endnode table entry for D between=
 (D, R1&#39;s</span><u></u><u></u></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt"><span style=3D"font-siz=
e:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=A0=A0=A0=A0 nickname) =
and (D, R2&#39;s nickname). So it would be preferable for E</span><u></u><u=
></u></p><p class=3D"MsoNormal" style=3D"line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 to always encapsulate using the same nickname (R1 or R2) unles=
s E</span><u></u><u></u></p><p class=3D"MsoNormal" style=3D"line-height:14.=
4pt"><span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=
=A0=A0=A0=A0=A0=A0 detects a problem with connectivity using that nickname.=
 And in</span><u></u><u></u></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt"><span style=3D"font-siz=
e:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=A0=A0=A0=A0 this case,=
 R1 and R2 need to be informed that the smart endnode</span><u></u><u></u><=
/p><p class=3D"MsoNormal" style=3D"line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 might encapsulate with a different nickname, i.e., R1 might</s=
pan><u></u><u></u></p><p class=3D"MsoNormal" style=3D"line-height:14.4pt"><=
span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 receive an encapsulated packet from smart endnode E using ingr=
ess</span><u></u><u></u></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt"><span style=3D"font-siz=
e:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=A0=A0=A0=A0 nickname &=
quot;R2&quot;.</span><u></u><u></u></p><p class=3D"MsoNormal"><span style=
=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&quot;sans-serif&quot;=
;color:#1f497d">=A0</span><u></u><u></u></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">If the smart endnode uses=
 either R1 or R2 in the above example, would we have TRILL RPF issues where=
 upstream RBs see a packet with SRB=3Dsay R1 coming from a wrong TRILL port=
? Let=92s say the RPF happens for multicast only then we will atleast need =
to make sure the multicast (broadcast, unknown unicast and multcast) is sen=
t via the edge-RB whose RBID the smart endnode is using.</span><u></u><u></=
u></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">=A0</span><u></u><u></u><=
/p><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">This problem should no=
t be there when using different RBID other than R1 and R2 as described in a=
pproach-2</span><u></u><u></u></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">=A0</span><u></u><u></u><=
/p><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">Phani</span><u></u><u>=
</u></p>
<p class=3D"MsoNormal"><b><span style=3D"font-size:10.0pt;font-family:&quot=
;Tahoma&quot;,&quot;sans-serif&quot;">From:</span></b><span style=3D"font-s=
ize:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;"> Radia Pe=
rlman [mailto:<a href=3D"mailto:radiaperlman@gmail.com" target=3D"_blank">r=
adiaperlman@gmail.com</a>] <br>
<b>Sent:</b> Wednesday, January 09, 2013 2:53 AM<br><b>To:</b> Phanidhar Ko=
ganti<br><b>Cc:</b> <a href=3D"mailto:Kesava_Vijaya_Krupak@Dell.com" target=
=3D"_blank">Kesava_Vijaya_Krupak@Dell.com</a>; <a href=3D"mailto:trill@ietf=
.org" target=3D"_blank">trill@ietf.org</a>; <a href=3D"mailto:d3e3e3@gmail.=
com" target=3D"_blank">d3e3e3@gmail.com</a>; <a href=3D"mailto:hu.fangwei@z=
te.com.cn" target=3D"_blank">hu.fangwei@zte.com.cn</a><br>
<b>Subject:</b> Re: New version : TRILL Smart Endnodes draft</span><u></u><=
u></u></p><div><div><p class=3D"MsoNormal">=A0<u></u><u></u></p><p class=3D=
"MsoNormal">Most of what you said is correct. =A0One correction<u></u><u></=
u></p>
<div><p class=3D"MsoNormal">=A0<u></u><u></u></p></div><div><p class=3D"Mso=
Normal">a)=A0<span style=3D"color:#1f497d">When the packet reaches the egre=
ss RB I am presuming there will be a inner-MAC lookup to deliver the packet=
 to the egress end-node.</span><u></u><u></u></p>
</div><div><p class=3D"MsoNormal">=A0<u></u><u></u></p></div><div><p class=
=3D"MsoNormal"><span style=3D"color:#1f497d">correct...when egress RB R has=
 to look up the inner MAC address E, which it has to do anyway without smar=
t endnodes, in order to choose which port to forward it on. =A0With smart e=
ndnodes, when R looks up MAC address E, it not only discovers which port to=
 send it on, but that E is &quot;smart&quot;, in which case R must leave th=
e TRILL encapsulation</span><u></u><u></u></p>
</div><div><p class=3D"MsoNormal">=A0<u></u><u></u></p></div><div><p class=
=3D"MsoNormal"><span style=3D"color:#1f497d">b)=A0Edge-RB to smart endnode =
cannot have TRILL encapsulation right?</span><u></u><u></u></p></div><div><=
p class=3D"MsoNormal">
=A0<u></u><u></u></p></div><div><p class=3D"MsoNormal"><span style=3D"color=
:#1f497d">Not correct. =A0Traffic from egress RB to smart endnode will have=
 TRILL encapsulation. =A0So for example, let&#39;s say R is the egress RB, =
and R has a link with two endnodes; E1 (which is smart) and E2 (which is no=
t smart).</span><u></u><u></u></p>
</div><div><p class=3D"MsoNormal">=A0<u></u><u></u></p></div><div><p class=
=3D"MsoNormal"><span style=3D"color:#1f497d">When R receives a TRILL packet=
 from the campus, with TRILL destination nickname=3DR, R looks at the MAC d=
estination. =A0If it is E2, then R removes the TRILL encapulation and forwa=
rds a native packet onto the link. =A0If it is E1, then R forwards the pack=
et with the TRILL header.</span><u></u><u></u></p>
</div><div><p class=3D"MsoNormal">=A0<u></u><u></u></p></div><div><p class=
=3D"MsoNormal"><span style=3D"color:#1f497d">c)=A0Edge-RB to smart endnode =
cannot have TRILL encapsulation right? If so for my clarification we are sa=
ying the traffic towards the end-node will be non-TRILL while towards the i=
ngress edge-RB will be TRILL encapsulated?</span><u></u><u></u></p>
</div><div><p class=3D"MsoNormal">=A0<u></u><u></u></p></div><div><p class=
=3D"MsoNormal"><span style=3D"color:#1f497d">Hopefully my answer to b) clar=
ified this. =A0On the link between endnodes and R, whether it&#39;s to or f=
rom endnode &lt;--&gt; edge RB, the packet will have TRILL encapsulation if=
 the endnode is smart, and not have TRILL encapsulation if the endnode is n=
ot smart.</span><u></u><u></u></p>
</div><div><p class=3D"MsoNormal">=A0<u></u><u></u></p></div><div><p class=
=3D"MsoNormal"><span style=3D"color:#1f497d">Radia</span><u></u><u></u></p>=
</div><div><p class=3D"MsoNormal">=A0<u></u><u></u></p></div><div><p class=
=3D"MsoNormal">
=A0<u></u><u></u></p><div><div><p class=3D"MsoNormal">On Thu, Jan 3, 2013 a=
t 3:19 AM, Phanidhar Koganti &lt;<a href=3D"mailto:pkoganti@brocade.com" ta=
rget=3D"_blank">pkoganti@brocade.com</a>&gt; wrote:<u></u><u></u></p><div><=
div>
<p class=3D"MsoNormal"><span style=3D"color:#1f497d">The draft defines the =
control and data path behavior from smart end-node to edge-RB. When the pac=
ket reaches the egress RB I am presuming there will be a inner-MAC lookup t=
o deliver the packet to the egress end-node. Edge-RB to smart endnode canno=
t have TRILL encapsulation right? If so for my clarification we are saying =
the traffic towards the end-node will be non-TRILL while towards the ingres=
s edge-RB will be TRILL encapsulated?</span><u></u><u></u></p>
<p class=3D"MsoNormal"><span style=3D"color:#1f497d">=A0</span><u></u><u></=
u></p><p class=3D"MsoNormal"><span style=3D"color:#1f497d">Phani</span><u><=
/u><u></u></p><p class=3D"MsoNormal"><span style=3D"color:#1f497d">=A0</spa=
n><u></u><u></u></p>
<div><div style=3D"border:none;border-top:solid #b5c4df 1.0pt;padding:3.0pt=
 0in 0in 0in"><p class=3D"MsoNormal"><b><span style=3D"font-size:10.0pt;fon=
t-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;">From:</span></b><span s=
tyle=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&qu=
ot;"> <a href=3D"mailto:trill-bounces@ietf.org" target=3D"_blank">trill-bou=
nces@ietf.org</a> [mailto:<a href=3D"mailto:trill-bounces@ietf.org" target=
=3D"_blank">trill-bounces@ietf.org</a>] <b>On Behalf Of </b><a href=3D"mail=
to:Kesava_Vijaya_Krupak@Dell.com" target=3D"_blank">Kesava_Vijaya_Krupak@De=
ll.com</a><br>
<b>Sent:</b> Thursday, January 03, 2013 4:19 PM<br><b>To:</b> <a href=3D"ma=
ilto:trill@ietf.org" target=3D"_blank">trill@ietf.org</a><br><b>Cc:</b> <a =
href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gmail.com</a>; <a=
 href=3D"mailto:radiaperlman@gmail.com" target=3D"_blank">radiaperlman@gmai=
l.com</a>; <a href=3D"mailto:hu.fangwei@zte.com.cn" target=3D"_blank">hu.fa=
ngwei@zte.com.cn</a><br>
<b>Subject:</b> [trill] New version : TRILL Smart Endnodes draft</span><u><=
/u><u></u></p></div></div><div><div><p class=3D"MsoNormal">=A0<u></u><u></u=
></p><p class=3D"MsoNormal">Hi all,<u></u><u></u></p><p class=3D"MsoNormal"=
>=A0<u></u><u></u></p>
<p class=3D"MsoNormal">A new version of the Trill Smart Endnodes draft has =
been posted.<u></u><u></u></p><p class=3D"MsoNormal">=A0<u></u><u></u></p><=
p class=3D"MsoNormal"><a href=3D"http://datatracker.ietf.org/doc/draft-perl=
man-trill-smart-endnodes" target=3D"_blank">http://datatracker.ietf.org/doc=
/draft-perlman-trill-smart-endnodes</a><u></u><u></u></p>
<p class=3D"MsoNormal">=A0<u></u><u></u></p><p class=3D"MsoNormal">Kindly l=
et us know your comments.<u></u><u></u></p><p class=3D"MsoNormal">=A0<u></u=
><u></u></p><p class=3D"MsoNormal">Thanks,<u></u><u></u></p><p class=3D"Mso=
Normal">kvk<u></u><u></u></p>
</div></div></div></div></div><p class=3D"MsoNormal">=A0<u></u><u></u></p><=
/div></div></div></div></div></div></div><p class=3D"MsoNormal"><u></u>=A0<=
u></u></p></div></div></div></div></div></blockquote></div><br></div>

--047d7b621de2f1b72004d2d6f420--

From pkoganti@Brocade.COM  Wed Jan  9 00:47:25 2013
Return-Path: <pkoganti@Brocade.COM>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 00B7321F8501 for <trill@ietfa.amsl.com>; Wed,  9 Jan 2013 00:47:25 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.064
X-Spam-Level: 
X-Spam-Status: No, score=-2.064 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, IP_NOT_FRIENDLY=0.334, J_CHICKENPOX_33=0.6, J_CHICKENPOX_81=0.6, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id lo2u-7aQdIo8 for <trill@ietfa.amsl.com>; Wed,  9 Jan 2013 00:47:18 -0800 (PST)
Received: from mx0a-000f0801.pphosted.com (mx0a-000f0801.pphosted.com [67.231.144.122]) by ietfa.amsl.com (Postfix) with ESMTP id 636F321F8507 for <trill@ietf.org>; Wed,  9 Jan 2013 00:47:18 -0800 (PST)
Received: from pps.filterd (m0000542 [127.0.0.1]) by mx0a-000f0801.pphosted.com (8.14.5/8.14.5) with SMTP id r098lExo018374; Wed, 9 Jan 2013 00:47:14 -0800
Received: from hq1wp-exchub01.corp.brocade.com ([144.49.131.13]) by mx0a-000f0801.pphosted.com with ESMTP id 19rta709u8-1 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Wed, 09 Jan 2013 00:47:13 -0800
Received: from HQ1WP-EXHUB01.corp.brocade.com (10.70.36.14) by HQ1WP-EXCHUB01.corp.brocade.com (10.70.36.99) with Microsoft SMTP Server (TLS) id 14.2.309.2; Wed, 9 Jan 2013 00:47:04 -0800
Received: from HQ1-EXCH02.corp.brocade.com ([fe80::c92a:772e:befa:c34c]) by HQ1WP-EXHUB01.corp.brocade.com ([::1]) with mapi; Wed, 9 Jan 2013 00:47:04 -0800
From: Phanidhar Koganti <pkoganti@Brocade.COM>
To: Radia Perlman <radiaperlman@gmail.com>
Date: Wed, 9 Jan 2013 00:47:02 -0800
Thread-Topic: New version : TRILL Smart Endnodes draft
Thread-Index: Ac3uRGvGCbwcv0mnTG6DWkgeGFy53gAAKqtw
Message-ID: <6895EAE0CA8DEE40B92D7CA88BB521F32414007BE1@HQ1-EXCH02.corp.brocade.com>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com> <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com> <6895EAE0CA8DEE40B92D7CA88BB521F32414007BC1@HQ1-EXCH02.corp.brocade.com> <CAFOuuo626hw-c0kM4-1Fp-w2gByPenRnkFP3xNT+cTB4YcJ2kg@mail.gmail.com> <6895EAE0CA8DEE40B92D7CA88BB521F32414007BDC@HQ1-EXCH02.corp.brocade.com> <CAFOuuo6CK7=1uKU3j6WCSVEjd6PqYRxpsvC55Y66mVCPN_zTzg@mail.gmail.com>
In-Reply-To: <CAFOuuo6CK7=1uKU3j6WCSVEjd6PqYRxpsvC55Y66mVCPN_zTzg@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US
Content-Type: multipart/alternative; boundary="_000_6895EAE0CA8DEE40B92D7CA88BB521F32414007BE1HQ1EXCH02corp_"
MIME-Version: 1.0
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:5.9.8327, 1.0.431, 0.0.0000 definitions=2013-01-09_02:2013-01-08, 2013-01-09, 1970-01-01 signatures=0
X-Proofpoint-Spam-Details: rule=notspam policy=default score=0 spamscore=0 suspectscore=49 phishscore=0 adultscore=0 bulkscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=7.0.1-1211240000 definitions=main-1301090014
Cc: "d3e3e3@gmail.com" <d3e3e3@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 09 Jan 2013 08:47:25 -0000

--_000_6895EAE0CA8DEE40B92D7CA88BB521F32414007BE1HQ1EXCH02corp_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Inline

From: Radia Perlman [mailto:radiaperlman@gmail.com]
Sent: Wednesday, January 09, 2013 2:06 PM
To: Phanidhar Koganti
Cc: Kesava_Vijaya_Krupak@Dell.com; trill@ietf.org; d3e3e3@gmail.com; hu.fan=
gwei@zte.com.cn
Subject: Re: New version : TRILL Smart Endnodes draft

Hmm.  Interesting example...

Let's first ignore multihoming, and assuming that two smart endnodes, E1 an=
d E2, are both connected to R1.  I think the case works where E1 and E2 are=
 on the same shared link...they will learn that the other can be reached di=
rectly, and talk to each other without TRILL encapsulation.

But let's consider the case where E1 and E2 are both singly homed to R, but=
 using different ports.  Then they will have to talk to each other with TRI=
LL encapsulation with both ingress and egress nickname=3DR1.  That's a bit =
weird, but I think it works.  Does it?
[Phanidhar Koganti] We can make it work :). It sounds weird so will need to=
 put more thought....

-------
As for multicast showing up twice...it shouldn't...only one RB should be at=
tached to a link with an endnode via any particular tree, so I think only o=
ne copy of a multicast packet should be delivered.
[Phanidhar Koganti] Got you, so the link is chosen on a tree basis and not =
like a LAG where each side picks the link for multicast without coordinatio=
n with remote.

--------
As for having each smart endnode use its own nickname...the real problem is=
 nickname exhaustion (running out of nicknames).  There's only 64K nickname=
s, unless we do the multilevel thing where nicknames only need to be unique=
 within an area, but the RBs that forward between level 1 and level 2 have =
to rewrite the nickname fields.
[Phanidhar Koganti] Yes the nickname space can be localized to an area and =
the need for nickname rewrites will come into picture for very large scale =
deployments and can potentially be justified.

----
As for unknown destinations/learning...I thought we considered that.  But s=
mart endnodes will not be unknown to the attached RB...they explicitly anno=
unce themselves.
[Phanidhar Koganti] Agree if smart endnodes are always known we are good.

Radia



On Wed, Jan 9, 2013 at 12:24 AM, Phanidhar Koganti <pkoganti@brocade.com<ma=
ilto:pkoganti@brocade.com>> wrote:
Radia,

Valid point for unicast that the pair in the example needs to be aware that=
 it might get unicast traffic with TRILL Source RBID equal to its RBID. Thi=
s brings up an interesting scenario.

Let's say there are two smart end-nodes e1 & e2 which are both multi-homed =
behind R1 & R2. Also say both e1 & e2 pick R1 to use in TRILL source RBID, =
then when e1 is talking to e2 we end up in a encap where both TRILL SA/DA w=
ill have the same value =3D R1.

I am inclined towards a pseudo RB for the end-node with its own nickname, b=
ut am not clear on the impact on Dijkstra as the number of routable RBIDs w=
ill considerably go up.

Few additional concerns:

For multicast it will also trigger confusion on source suppression that nee=
ds to happen in the peer RB. Let's say there is a smart end-node e1 & e2 mu=
lti-homed behind R1 and R2. For multicast or flooded traffic we need source=
 suppression so that the traffic does not go back to the smart endnode. Or =
is the draft assuming that the smart end-node will perform the suppression =
if the packet comes back to it?

Also for the packet direction edge-RB to smart end-node, if we end up with =
unknown MAC then we should flood to all the locally hanging smart end-nodes=
 and regular end-nodes. When flooding towards smart end-nodes are we expect=
ed to continue the TRILL packet with M=3D0 or convert it to M=3D1 with T_DA=
=3DROOT. If we like to continue with M=3D0 then the edge-RB needs to suppor=
t unicast replication for multicast.

Phani

From: Radia Perlman [mailto:radiaperlman@gmail.com<mailto:radiaperlman@gmai=
l.com>]
Sent: Wednesday, January 09, 2013 11:52 AM

To: Phanidhar Koganti
Cc: Kesava_Vijaya_Krupak@Dell.com<mailto:Kesava_Vijaya_Krupak@Dell.com>; tr=
ill@ietf.org<mailto:trill@ietf.org>; d3e3e3@gmail.com<mailto:d3e3e3@gmail.c=
om>; hu.fangwei@zte.com.cn<mailto:hu.fangwei@zte.com.cn>
Subject: Re: New version : TRILL Smart Endnodes draft

Good point (see Phani's comment below).  I'll summarize:

The current  text implies that it's OK for smart endnode E, multihomed to R=
1 and R2, to choose one of the nicknames (say R1) for "ingress nickname" in=
 the TRILL header, and forward to either R1 or R2.  There are two problems =
with E forwarding a TRILL-encapsulated frame using ingeress "R1", through R=
2:

a) for multicast, likely to confuse the RPF check
b) for unicast, E would probably need to warn R2 that E is multihomed to R1=
 and will e using R1's nickname for ingress.

And actually, E won't get that much load splitting benefit from being multi=
homed if it uses "R1" as ingress, since all traffic to E will go through R1=
.  So, if E doesn't have a pseudonode nickname to use, then perhaps E shoul=
d just use one R1 or R2.

Or E could get its own nickname (I've been floating around the idea of a "d=
umb RBridge", which is like a "smart endnode", except it gets its own nickn=
ame, and does generate an LSP with "overloaded" flag.  It probably has to r=
eceive LSPs and calculate Dijkstra, in order to know which port to send mul=
ticast on (unless its attached true RBs are helpful enough to tell it..."yo=
u can send multicast with trees X or Y through me")

So, it seems like there are several options which should be discussed, for =
a smart endnode multihomed to R1 and R2
1) Always use a pseudnode nickname (but that's kind of complicated since th=
at nickname can only be used by smart endnodes that are multihomed to the e=
xact same set of RBridges)
2) Don't do active-active...choose one or R1 or R2 and always send everythi=
ng that way
3) Do active-active, but only for unicast...choose one of the nicknames, sa=
y "R1", and it's OK to send unicast either through R1 or R2, but multicast =
has to go through R1
4) Give E its own nickname (and somehow make sure it knows which port it ca=
n multicast on)
5) Other possibilities?

Radia




On Tue, Jan 8, 2013 at 8:32 PM, Phanidhar Koganti <pkoganti@brocade.com<mai=
lto:pkoganti@brocade.com>> wrote:
Thanks Radia. That clarifies for me. A more descriptive text in draft can s=
urely help vendors as they implement it in their hardware.

Another question on handling multi-homing
E can choose either R1 or R2's nickname, when encapsulating a
       frame, whether the encapsulated frame is sent via R1 or R2. If E
       wants to do active-active load splitting, and uses R1's nickname
       when forwarding through R1, and R2's nickname when forwarding
       through R2, this will cause distant RBridges (or smart endnodes)
       to keep changing their endnode table entry for D between (D, R1's
       nickname) and (D, R2's nickname). So it would be preferable for E
       to always encapsulate using the same nickname (R1 or R2) unless E
       detects a problem with connectivity using that nickname. And in
       this case, R1 and R2 need to be informed that the smart endnode
       might encapsulate with a different nickname, i.e., R1 might
       receive an encapsulated packet from smart endnode E using ingress
       nickname "R2".

If the smart endnode uses either R1 or R2 in the above example, would we ha=
ve TRILL RPF issues where upstream RBs see a packet with SRB=3Dsay R1 comin=
g from a wrong TRILL port? Let's say the RPF happens for multicast only the=
n we will atleast need to make sure the multicast (broadcast, unknown unica=
st and multcast) is sent via the edge-RB whose RBID the smart endnode is us=
ing.

This problem should not be there when using different RBID other than R1 an=
d R2 as described in approach-2

Phani
From: Radia Perlman [mailto:radiaperlman@gmail.com<mailto:radiaperlman@gmai=
l.com>]
Sent: Wednesday, January 09, 2013 2:53 AM
To: Phanidhar Koganti
Cc: Kesava_Vijaya_Krupak@Dell.com<mailto:Kesava_Vijaya_Krupak@Dell.com>; tr=
ill@ietf.org<mailto:trill@ietf.org>; d3e3e3@gmail.com<mailto:d3e3e3@gmail.c=
om>; hu.fangwei@zte.com.cn<mailto:hu.fangwei@zte.com.cn>
Subject: Re: New version : TRILL Smart Endnodes draft

Most of what you said is correct.  One correction

a) When the packet reaches the egress RB I am presuming there will be a inn=
er-MAC lookup to deliver the packet to the egress end-node.

correct...when egress RB R has to look up the inner MAC address E, which it=
 has to do anyway without smart endnodes, in order to choose which port to =
forward it on.  With smart endnodes, when R looks up MAC address E, it not =
only discovers which port to send it on, but that E is "smart", in which ca=
se R must leave the TRILL encapsulation

b) Edge-RB to smart endnode cannot have TRILL encapsulation right?

Not correct.  Traffic from egress RB to smart endnode will have TRILL encap=
sulation.  So for example, let's say R is the egress RB, and R has a link w=
ith two endnodes; E1 (which is smart) and E2 (which is not smart).

When R receives a TRILL packet from the campus, with TRILL destination nick=
name=3DR, R looks at the MAC destination.  If it is E2, then R removes the =
TRILL encapulation and forwards a native packet onto the link.  If it is E1=
, then R forwards the packet with the TRILL header.

c) Edge-RB to smart endnode cannot have TRILL encapsulation right? If so fo=
r my clarification we are saying the traffic towards the end-node will be n=
on-TRILL while towards the ingress edge-RB will be TRILL encapsulated?

Hopefully my answer to b) clarified this.  On the link between endnodes and=
 R, whether it's to or from endnode <--> edge RB, the packet will have TRIL=
L encapsulation if the endnode is smart, and not have TRILL encapsulation i=
f the endnode is not smart.

Radia


On Thu, Jan 3, 2013 at 3:19 AM, Phanidhar Koganti <pkoganti@brocade.com<mai=
lto:pkoganti@brocade.com>> wrote:
The draft defines the control and data path behavior from smart end-node to=
 edge-RB. When the packet reaches the egress RB I am presuming there will b=
e a inner-MAC lookup to deliver the packet to the egress end-node. Edge-RB =
to smart endnode cannot have TRILL encapsulation right? If so for my clarif=
ication we are saying the traffic towards the end-node will be non-TRILL wh=
ile towards the ingress edge-RB will be TRILL encapsulated?

Phani

From: trill-bounces@ietf.org<mailto:trill-bounces@ietf.org> [mailto:trill-b=
ounces@ietf.org<mailto:trill-bounces@ietf.org>] On Behalf Of Kesava_Vijaya_=
Krupak@Dell.com<mailto:Kesava_Vijaya_Krupak@Dell.com>
Sent: Thursday, January 03, 2013 4:19 PM
To: trill@ietf.org<mailto:trill@ietf.org>
Cc: d3e3e3@gmail.com<mailto:d3e3e3@gmail.com>; radiaperlman@gmail.com<mailt=
o:radiaperlman@gmail.com>; hu.fangwei@zte.com.cn<mailto:hu.fangwei@zte.com.=
cn>
Subject: [trill] New version : TRILL Smart Endnodes draft

Hi all,

A new version of the Trill Smart Endnodes draft has been posted.

http://datatracker.ietf.org/doc/draft-perlman-trill-smart-endnodes

Kindly let us know your comments.

Thanks,
kvk




--_000_6895EAE0CA8DEE40B92D7CA88BB521F32414007BE1HQ1EXCH02corp_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micr=
osoft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" xmlns=3D"http:=
//www.w3.org/TR/REC-html40"><head><META HTTP-EQUIV=3D"Content-Type" CONTENT=
=3D"text/html; charset=3Dus-ascii"><meta name=3DGenerator content=3D"Micros=
oft Word 14 (filtered medium)"><style><!--
/* Font Definitions */
@font-face
	{font-family:Wingdings;
	panose-1:5 0 0 0 0 0 0 0 0 0;}
@font-face
	{font-family:Wingdings;
	panose-1:5 0 0 0 0 0 0 0 0 0;}
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:Tahoma;
	panose-1:2 11 6 4 3 5 4 4 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman","serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
p
	{mso-style-priority:99;
	mso-margin-top-alt:auto;
	margin-right:0in;
	mso-margin-bottom-alt:auto;
	margin-left:0in;
	font-size:12.0pt;
	font-family:"Times New Roman","serif";}
p.MsoAcetate, li.MsoAcetate, div.MsoAcetate
	{mso-style-priority:99;
	mso-style-link:"Balloon Text Char";
	margin:0in;
	margin-bottom:.0001pt;
	font-size:8.0pt;
	font-family:"Tahoma","sans-serif";}
span.BalloonTextChar
	{mso-style-name:"Balloon Text Char";
	mso-style-priority:99;
	mso-style-link:"Balloon Text";
	font-family:"Tahoma","sans-serif";}
span.EmailStyle20
	{mso-style-type:personal-reply;
	font-family:"Calibri","sans-serif";
	color:#1F497D;}
.MsoChpDefault
	{mso-style-type:export-only;
	font-family:"Calibri","sans-serif";}
@page WordSection1
	{size:8.5in 11.0in;
	margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
	{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]--></head><body lang=3DEN-US link=3Dblue vli=
nk=3Dpurple><div class=3DWordSection1><p class=3DMsoNormal><span style=3D'f=
ont-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>Inline<o:=
p></o:p></span></p><p class=3DMsoNormal><span style=3D'font-size:11.0pt;fon=
t-family:"Calibri","sans-serif";color:#1F497D'><o:p>&nbsp;</o:p></span></p>=
<p class=3DMsoNormal><b><span style=3D'font-size:10.0pt;font-family:"Tahoma=
","sans-serif"'>From:</span></b><span style=3D'font-size:10.0pt;font-family=
:"Tahoma","sans-serif"'> Radia Perlman [mailto:radiaperlman@gmail.com] <br>=
<b>Sent:</b> Wednesday, January 09, 2013 2:06 PM<br><b>To:</b> Phanidhar Ko=
ganti<br><b>Cc:</b> Kesava_Vijaya_Krupak@Dell.com; trill@ietf.org; d3e3e3@g=
mail.com; hu.fangwei@zte.com.cn<br><b>Subject:</b> Re: New version : TRILL =
Smart Endnodes draft<o:p></o:p></span></p><p class=3DMsoNormal><o:p>&nbsp;<=
/o:p></p><p class=3DMsoNormal>Hmm. &nbsp;Interesting example...<o:p></o:p><=
/p><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p class=3DMso=
Normal>Let's first ignore multihoming, and assuming that two smart endnodes=
, E1 and E2, are both connected to R1. &nbsp;I think the case works where E=
1 and E2 are on the same shared link...they will learn that the other can b=
e reached directly, and talk to each other without TRILL encapsulation.<o:p=
></o:p></p></div><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div>=
<p class=3DMsoNormal>But let's consider the case where E1 and E2 are both s=
ingly homed to R, but using different ports. &nbsp;Then they will have to t=
alk to each other with TRILL encapsulation with both ingress and egress nic=
kname=3DR1. &nbsp;That's a bit weird, but I think it works. &nbsp;Does it?<=
o:p></o:p></p><p class=3DMsoNormal><b><i><span style=3D'font-size:11.0pt;fo=
nt-family:"Calibri","sans-serif";color:#1F497D'>[Phanidhar Koganti] We can =
make it work </span></i></b><b><i><span style=3D'font-size:11.0pt;font-fami=
ly:Wingdings;color:#1F497D'>J</span></i></b><b><i><span style=3D'font-size:=
11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>. It sounds weird =
so will need to put more thought&#8230;.</span></i></b><span style=3D'font-=
size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'><o:p></o:p></=
span></p></div><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p=
 class=3DMsoNormal>-------<o:p></o:p></p></div><div><p class=3DMsoNormal>As=
 for multicast showing up twice...it shouldn't...only one RB should be atta=
ched to a link with an endnode via any particular tree, so I think only one=
 copy of a multicast packet should be delivered.<o:p></o:p></p><p class=3DM=
soNormal><b><i><span style=3D'font-size:11.0pt;font-family:"Calibri","sans-=
serif";color:#1F497D'>[Phanidhar Koganti] Got you, so the link is chosen on=
 a tree basis and not like a LAG where each side picks the link for multica=
st without coordination with remote.</span></i></b><span style=3D'font-size=
:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'><o:p></o:p></span=
></p></div><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p cla=
ss=3DMsoNormal>--------<o:p></o:p></p></div><div><p class=3DMsoNormal>As fo=
r having each smart endnode use its own nickname...the real problem is nick=
name exhaustion (running out of nicknames). &nbsp;There's only 64K nickname=
s, unless we do the multilevel thing where nicknames only need to be unique=
 within an area, but the RBs that forward between level 1 and level 2 have =
to rewrite the nickname fields.<span style=3D'color:#1F497D'><o:p></o:p></s=
pan></p><p class=3DMsoNormal><b><i><span style=3D'font-size:11.0pt;font-fam=
ily:"Calibri","sans-serif";color:#1F497D'>[Phanidhar Koganti] Yes the nickn=
ame space can be localized to an area and the need for nickname rewrites wi=
ll come into picture for very large scale deployments and can potentially b=
e justified.</span></i></b><span style=3D'font-size:11.0pt;font-family:"Cal=
ibri","sans-serif";color:#1F497D'><o:p></o:p></span></p></div><div><p class=
=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p class=3DMsoNormal>----<o:p>=
</o:p></p></div><div><p class=3DMsoNormal>As for unknown destinations/learn=
ing...I thought we considered that. &nbsp;But smart endnodes will not be un=
known to the attached RB...they explicitly announce themselves.<o:p></o:p><=
/p><p class=3DMsoNormal><b><i><span style=3D'font-size:11.0pt;font-family:"=
Calibri","sans-serif";color:#1F497D'>[Phanidhar Koganti] Agree if smart end=
nodes are always known we are good.</span></i></b><span style=3D'font-size:=
11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'><o:p></o:p></span>=
</p></div><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p clas=
s=3DMsoNormal>Radia<o:p></o:p></p></div><div><p class=3DMsoNormal><o:p>&nbs=
p;</o:p></p></div><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div=
><p class=3DMsoNormal><o:p>&nbsp;</o:p></p><div><p class=3DMsoNormal>On Wed=
, Jan 9, 2013 at 12:24 AM, Phanidhar Koganti &lt;<a href=3D"mailto:pkoganti=
@brocade.com" target=3D"_blank">pkoganti@brocade.com</a>&gt; wrote:<o:p></o=
:p></p><div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-=
margin-bottom-alt:auto'><span style=3D'font-size:11.0pt;font-family:"Calibr=
i","sans-serif";color:#1F497D'>Radia,</span><o:p></o:p></p><p class=3DMsoNo=
rmal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span sty=
le=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>&n=
bsp;</span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:=
auto;mso-margin-bottom-alt:auto'><span style=3D'font-size:11.0pt;font-famil=
y:"Calibri","sans-serif";color:#1F497D'>Valid point for unicast that the pa=
ir in the example needs to be aware that it might get unicast traffic with =
TRILL Source RBID equal to its RBID. This brings up an interesting scenario=
. </span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:au=
to;mso-margin-bottom-alt:auto'><span style=3D'font-size:11.0pt;font-family:=
"Calibri","sans-serif";color:#1F497D'>&nbsp;</span><o:p></o:p></p><p class=
=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><=
span style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F=
497D'>Let&#8217;s say there are two smart end-nodes e1 &amp; e2 which are b=
oth multi-homed behind R1 &amp; R2. Also say both e1 &amp; e2 pick R1 to us=
e in TRILL source RBID, then when e1 is talking to e2 we end up in a encap =
where both TRILL SA/DA will have the same value =3D R1.</span><o:p></o:p></=
p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto'><span style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif"=
;color:#1F497D'>&nbsp;</span><o:p></o:p></p><p class=3DMsoNormal style=3D'm=
so-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span style=3D'font-size=
:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>I am inclined tow=
ards a pseudo RB for the end-node with its own nickname, but am not clear o=
n the impact on Dijkstra as the number of routable RBIDs will considerably =
go up. </span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-a=
lt:auto;mso-margin-bottom-alt:auto'><span style=3D'font-size:11.0pt;font-fa=
mily:"Calibri","sans-serif";color:#1F497D'>&nbsp;</span><o:p></o:p></p><p c=
lass=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:aut=
o'><span style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color=
:#1F497D'>Few additional concerns:</span><o:p></o:p></p><p class=3DMsoNorma=
l style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span style=
=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>&nbs=
p;</span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:au=
to;mso-margin-bottom-alt:auto'><span style=3D'font-size:11.0pt;font-family:=
"Calibri","sans-serif";color:#1F497D'>For multicast it will also trigger co=
nfusion on source suppression that needs to happen in the peer RB. Let&#821=
7;s say there is a smart end-node e1 &amp; e2 multi-homed behind R1 and R2.=
 For multicast or flooded traffic we need source suppression so that the tr=
affic does not go back to the smart endnode. Or is the draft assuming that =
the smart end-node will perform the suppression if the packet comes back to=
 it? </span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt=
:auto;mso-margin-bottom-alt:auto'><span style=3D'font-size:11.0pt;font-fami=
ly:"Calibri","sans-serif";color:#1F497D'>&nbsp;</span><o:p></o:p></p><p cla=
ss=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'=
><span style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#=
1F497D'>Also for the packet direction edge-RB to smart end-node, if we end =
up with unknown MAC then we should flood to all the locally hanging smart e=
nd-nodes and regular end-nodes. When flooding towards smart end-nodes are w=
e expected to continue the TRILL packet with M=3D0 or convert it to M=3D1 w=
ith T_DA=3DROOT. If we like to continue with M=3D0 then the edge-RB needs t=
o support unicast replication for multicast.</span><o:p></o:p></p><p class=
=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><=
span style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F=
497D'>&nbsp;</span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-=
top-alt:auto;mso-margin-bottom-alt:auto'><span style=3D'font-size:11.0pt;fo=
nt-family:"Calibri","sans-serif";color:#1F497D'>Phani</span><o:p></o:p></p>=
<p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt=
:auto'><span style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";c=
olor:#1F497D'>&nbsp;</span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso=
-margin-top-alt:auto;mso-margin-bottom-alt:auto'><b><span style=3D'font-siz=
e:10.0pt;font-family:"Tahoma","sans-serif"'>From:</span></b><span style=3D'=
font-size:10.0pt;font-family:"Tahoma","sans-serif"'> Radia Perlman [mailto:=
<a href=3D"mailto:radiaperlman@gmail.com" target=3D"_blank">radiaperlman@gm=
ail.com</a>] <br><b>Sent:</b> Wednesday, January 09, 2013 11:52 AM</span><o=
:p></o:p></p><div><div><p class=3DMsoNormal><br><b>To:</b> Phanidhar Kogant=
i<br><b>Cc:</b> <a href=3D"mailto:Kesava_Vijaya_Krupak@Dell.com">Kesava_Vij=
aya_Krupak@Dell.com</a>; <a href=3D"mailto:trill@ietf.org" target=3D"_blank=
">trill@ietf.org</a>; <a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank"=
>d3e3e3@gmail.com</a>; <a href=3D"mailto:hu.fangwei@zte.com.cn" target=3D"_=
blank">hu.fangwei@zte.com.cn</a><br><b>Subject:</b> Re: New version : TRILL=
 Smart Endnodes draft<o:p></o:p></p></div></div><div><div><p class=3DMsoNor=
mal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p=
></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin=
-bottom-alt:auto'>Good point (see Phani's comment below). &nbsp;I'll summar=
ize:<o:p></o:p></p><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:au=
to;mso-margin-bottom-alt:auto'>&nbsp;<o:p></o:p></p></div><div><p class=3DM=
soNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>The c=
urrent &nbsp;text implies that it's OK for smart endnode E, multihomed to R=
1 and R2, to choose one of the nicknames (say R1) for &quot;ingress nicknam=
e&quot; in the TRILL header, and forward to either R1 or R2. &nbsp;There ar=
e two problems with E forwarding a TRILL-encapsulated frame using ingeress =
&quot;R1&quot;, through R2:<o:p></o:p></p></div><div><p class=3DMsoNormal s=
tyle=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p></o:=
p></p></div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-=
margin-bottom-alt:auto'>a) for multicast, likely to confuse the RPF check<o=
:p></o:p></p></div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:au=
to;mso-margin-bottom-alt:auto'>b) for unicast, E would probably need to war=
n R2 that E is multihomed to R1 and will e using R1's nickname for ingress.=
<o:p></o:p></p></div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:=
auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p></o:p></p></div><div><p class=
=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>A=
nd actually, E won't get that much load splitting benefit from being multih=
omed if it uses &quot;R1&quot; as ingress, since all traffic to E will go t=
hrough R1. &nbsp;So, if E doesn't have a pseudonode nickname to use, then p=
erhaps E should just use one R1 or R2.<o:p></o:p></p></div><div><p class=3D=
MsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbs=
p;<o:p></o:p></p></div><div><p class=3DMsoNormal style=3D'mso-margin-top-al=
t:auto;mso-margin-bottom-alt:auto'>Or E could get its own nickname (I've be=
en floating around the idea of a &quot;dumb RBridge&quot;, which is like a =
&quot;smart endnode&quot;, except it gets its own nickname, and does genera=
te an LSP with &quot;overloaded&quot; flag. &nbsp;It probably has to receiv=
e LSPs and calculate Dijkstra, in order to know which port to send multicas=
t on (unless its attached true RBs are helpful enough to tell it...&quot;yo=
u can send multicast with trees X or Y through me&quot;)<o:p></o:p></p></di=
v><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bot=
tom-alt:auto'>&nbsp;<o:p></o:p></p></div><div><p class=3DMsoNormal style=3D=
'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>So, it seems like ther=
e are several options which should be discussed, for a smart endnode multih=
omed to R1 and R2<o:p></o:p></p></div><div><p class=3DMsoNormal style=3D'ms=
o-margin-top-alt:auto;mso-margin-bottom-alt:auto'>1) Always use a pseudnode=
 nickname (but that's kind of complicated since that nickname can only be u=
sed by smart endnodes that are multihomed to the exact same set of RBridges=
)<o:p></o:p></p></div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt=
:auto;mso-margin-bottom-alt:auto'>2) Don't do active-active...choose one or=
 R1 or R2 and always send everything that way<o:p></o:p></p></div><div><p c=
lass=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:aut=
o'>3) Do active-active, but only for unicast...choose one of the nicknames,=
 say &quot;R1&quot;, and it's OK to send unicast either through R1 or R2, b=
ut multicast has to go through R1<o:p></o:p></p></div><div><p class=3DMsoNo=
rmal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>4) Give E=
 its own nickname (and somehow make sure it knows which port it can multica=
st on)<o:p></o:p></p></div><div><p class=3DMsoNormal style=3D'mso-margin-to=
p-alt:auto;mso-margin-bottom-alt:auto'>5) Other possibilities?<o:p></o:p></=
p></div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-marg=
in-bottom-alt:auto'>&nbsp;<o:p></o:p></p></div><div><p class=3DMsoNormal st=
yle=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>Radia<o:p></o:p>=
</p></div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-ma=
rgin-bottom-alt:auto'>&nbsp;<o:p></o:p></p></div><div><p class=3DMsoNormal =
style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p></o=
:p></p></div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso=
-margin-bottom-alt:auto'>&nbsp;<o:p></o:p></p></div><div><p class=3DMsoNorm=
al style=3D'mso-margin-top-alt:auto;margin-bottom:12.0pt'>&nbsp;<o:p></o:p>=
</p><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-b=
ottom-alt:auto'>On Tue, Jan 8, 2013 at 8:32 PM, Phanidhar Koganti &lt;<a hr=
ef=3D"mailto:pkoganti@brocade.com" target=3D"_blank">pkoganti@brocade.com</=
a>&gt; wrote:<o:p></o:p></p><div><div><p class=3DMsoNormal style=3D'mso-mar=
gin-top-alt:auto;mso-margin-bottom-alt:auto'><span style=3D'font-size:11.0p=
t;font-family:"Calibri","sans-serif";color:#1F497D'>Thanks Radia. That clar=
ifies for me. A more descriptive text in draft can surely help vendors as t=
hey implement it in their hardware.</span><o:p></o:p></p><p class=3DMsoNorm=
al style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span style=
=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>&nbs=
p;</span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:au=
to;mso-margin-bottom-alt:auto'><span style=3D'font-size:11.0pt;font-family:=
"Calibri","sans-serif";color:#1F497D'>Another question on handling multi-ho=
ming</span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:=
auto;mso-margin-bottom-alt:auto;line-height:14.4pt'><span style=3D'font-siz=
e:10.0pt;font-family:"Courier New"'>E can choose either R1 or R2's nickname=
, when encapsulating a</span><o:p></o:p></p><p class=3DMsoNormal style=3D'm=
so-margin-top-alt:auto;mso-margin-bottom-alt:auto;line-height:14.4pt'><span=
 style=3D'font-size:10.0pt;font-family:"Courier New"'>&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp; frame, whether the encapsulated frame is sent via R1 or R2.=
 If E</span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt=
:auto;mso-margin-bottom-alt:auto;line-height:14.4pt'><span style=3D'font-si=
ze:10.0pt;font-family:"Courier New"'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; w=
ants to do active-active load splitting, and uses R1's nickname</span><o:p>=
</o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-=
bottom-alt:auto;line-height:14.4pt'><span style=3D'font-size:10.0pt;font-fa=
mily:"Courier New"'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; when forwarding th=
rough R1, and R2's nickname when forwarding</span><o:p></o:p></p><p class=
=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto;li=
ne-height:14.4pt'><span style=3D'font-size:10.0pt;font-family:"Courier New"=
'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; through R2, this will cause distant =
RBridges (or smart endnodes)</span><o:p></o:p></p><p class=3DMsoNormal styl=
e=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto;line-height:14.4pt'=
><span style=3D'font-size:10.0pt;font-family:"Courier New"'>&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;&nbsp; to keep changing their endnode table entry for D betw=
een (D, R1's</span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-=
top-alt:auto;mso-margin-bottom-alt:auto;line-height:14.4pt'><span style=3D'=
font-size:10.0pt;font-family:"Courier New"'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp; nickname) and (D, R2's nickname). So it would be preferable for E</sp=
an><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso=
-margin-bottom-alt:auto;line-height:14.4pt'><span style=3D'font-size:10.0pt=
;font-family:"Courier New"'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; to always =
encapsulate using the same nickname (R1 or R2) unless E</span><o:p></o:p></=
p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto;line-height:14.4pt'><span style=3D'font-size:10.0pt;font-family:"Co=
urier New"'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; detects a problem with con=
nectivity using that nickname. And in</span><o:p></o:p></p><p class=3DMsoNo=
rmal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto;line-heigh=
t:14.4pt'><span style=3D'font-size:10.0pt;font-family:"Courier New"'>&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; this case, R1 and R2 need to be informed tha=
t the smart endnode</span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-=
margin-top-alt:auto;mso-margin-bottom-alt:auto;line-height:14.4pt'><span st=
yle=3D'font-size:10.0pt;font-family:"Courier New"'>&nbsp;&nbsp;&nbsp;&nbsp;=
&nbsp;&nbsp; might encapsulate with a different nickname, i.e., R1 might</s=
pan><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;ms=
o-margin-bottom-alt:auto;line-height:14.4pt'><span style=3D'font-size:10.0p=
t;font-family:"Courier New"'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; receive a=
n encapsulated packet from smart endnode E using ingress</span><o:p></o:p><=
/p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-=
alt:auto;line-height:14.4pt'><span style=3D'font-size:10.0pt;font-family:"C=
ourier New"'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; nickname &quot;R2&quot;.<=
/span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;=
mso-margin-bottom-alt:auto'><span style=3D'font-size:11.0pt;font-family:"Ca=
libri","sans-serif";color:#1F497D'>&nbsp;</span><o:p></o:p></p><p class=3DM=
soNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span=
 style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D=
'>If the smart endnode uses either R1 or R2 in the above example, would we =
have TRILL RPF issues where upstream RBs see a packet with SRB=3Dsay R1 com=
ing from a wrong TRILL port? Let&#8217;s say the RPF happens for multicast =
only then we will atleast need to make sure the multicast (broadcast, unkno=
wn unicast and multcast) is sent via the edge-RB whose RBID the smart endno=
de is using.</span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-=
top-alt:auto;mso-margin-bottom-alt:auto'><span style=3D'font-size:11.0pt;fo=
nt-family:"Calibri","sans-serif";color:#1F497D'>&nbsp;</span><o:p></o:p></p=
><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-al=
t:auto'><span style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";=
color:#1F497D'>This problem should not be there when using different RBID o=
ther than R1 and R2 as described in approach-2</span><o:p></o:p></p><p clas=
s=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>=
<span style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1=
F497D'>&nbsp;</span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin=
-top-alt:auto;mso-margin-bottom-alt:auto'><span style=3D'font-size:11.0pt;f=
ont-family:"Calibri","sans-serif";color:#1F497D'>Phani</span><o:p></o:p></p=
><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-al=
t:auto'><b><span style=3D'font-size:10.0pt;font-family:"Tahoma","sans-serif=
"'>From:</span></b><span style=3D'font-size:10.0pt;font-family:"Tahoma","sa=
ns-serif"'> Radia Perlman [mailto:<a href=3D"mailto:radiaperlman@gmail.com"=
 target=3D"_blank">radiaperlman@gmail.com</a>] <br><b>Sent:</b> Wednesday, =
January 09, 2013 2:53 AM<br><b>To:</b> Phanidhar Koganti<br><b>Cc:</b> <a h=
ref=3D"mailto:Kesava_Vijaya_Krupak@Dell.com" target=3D"_blank">Kesava_Vijay=
a_Krupak@Dell.com</a>; <a href=3D"mailto:trill@ietf.org" target=3D"_blank">=
trill@ietf.org</a>; <a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d=
3e3e3@gmail.com</a>; <a href=3D"mailto:hu.fangwei@zte.com.cn" target=3D"_bl=
ank">hu.fangwei@zte.com.cn</a><br><b>Subject:</b> Re: New version : TRILL S=
mart Endnodes draft</span><o:p></o:p></p><div><div><p class=3DMsoNormal sty=
le=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p></o:p>=
</p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom=
-alt:auto'>Most of what you said is correct. &nbsp;One correction<o:p></o:p=
></p><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-=
bottom-alt:auto'>&nbsp;<o:p></o:p></p></div><div><p class=3DMsoNormal style=
=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>a)&nbsp;<span style=
=3D'color:#1F497D'>When the packet reaches the egress RB I am presuming the=
re will be a inner-MAC lookup to deliver the packet to the egress end-node.=
</span><o:p></o:p></p></div><div><p class=3DMsoNormal style=3D'mso-margin-t=
op-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p></o:p></p></div><div><p =
class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:au=
to'><span style=3D'color:#1F497D'>correct...when egress RB R has to look up=
 the inner MAC address E, which it has to do anyway without smart endnodes,=
 in order to choose which port to forward it on. &nbsp;With smart endnodes,=
 when R looks up MAC address E, it not only discovers which port to send it=
 on, but that E is &quot;smart&quot;, in which case R must leave the TRILL =
encapsulation</span><o:p></o:p></p></div><div><p class=3DMsoNormal style=3D=
'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p></o:p></p><=
/div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-=
bottom-alt:auto'><span style=3D'color:#1F497D'>b)&nbsp;Edge-RB to smart end=
node cannot have TRILL encapsulation right?</span><o:p></o:p></p></div><div=
><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-al=
t:auto'>&nbsp;<o:p></o:p></p></div><div><p class=3DMsoNormal style=3D'mso-m=
argin-top-alt:auto;mso-margin-bottom-alt:auto'><span style=3D'color:#1F497D=
'>Not correct. &nbsp;Traffic from egress RB to smart endnode will have TRIL=
L encapsulation. &nbsp;So for example, let's say R is the egress RB, and R =
has a link with two endnodes; E1 (which is smart) and E2 (which is not smar=
t).</span><o:p></o:p></p></div><div><p class=3DMsoNormal style=3D'mso-margi=
n-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p></o:p></p></div><div>=
<p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt=
:auto'><span style=3D'color:#1F497D'>When R receives a TRILL packet from th=
e campus, with TRILL destination nickname=3DR, R looks at the MAC destinati=
on. &nbsp;If it is E2, then R removes the TRILL encapulation and forwards a=
 native packet onto the link. &nbsp;If it is E1, then R forwards the packet=
 with the TRILL header.</span><o:p></o:p></p></div><div><p class=3DMsoNorma=
l style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p><=
/o:p></p></div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;m=
so-margin-bottom-alt:auto'><span style=3D'color:#1F497D'>c)&nbsp;Edge-RB to=
 smart endnode cannot have TRILL encapsulation right? If so for my clarific=
ation we are saying the traffic towards the end-node will be non-TRILL whil=
e towards the ingress edge-RB will be TRILL encapsulated?</span><o:p></o:p>=
</p></div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-ma=
rgin-bottom-alt:auto'>&nbsp;<o:p></o:p></p></div><div><p class=3DMsoNormal =
style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span style=3D=
'color:#1F497D'>Hopefully my answer to b) clarified this. &nbsp;On the link=
 between endnodes and R, whether it's to or from endnode &lt;--&gt; edge RB=
, the packet will have TRILL encapsulation if the endnode is smart, and not=
 have TRILL encapsulation if the endnode is not smart.</span><o:p></o:p></p=
></div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margi=
n-bottom-alt:auto'>&nbsp;<o:p></o:p></p></div><div><p class=3DMsoNormal sty=
le=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span style=3D'co=
lor:#1F497D'>Radia</span><o:p></o:p></p></div><div><p class=3DMsoNormal sty=
le=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p></o:p>=
</p></div><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-ma=
rgin-bottom-alt:auto'>&nbsp;<o:p></o:p></p><div><div><p class=3DMsoNormal s=
tyle=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>On Thu, Jan 3, =
2013 at 3:19 AM, Phanidhar Koganti &lt;<a href=3D"mailto:pkoganti@brocade.c=
om" target=3D"_blank">pkoganti@brocade.com</a>&gt; wrote:<o:p></o:p></p><di=
v><div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bot=
tom-alt:auto'><span style=3D'color:#1F497D'>The draft defines the control a=
nd data path behavior from smart end-node to edge-RB. When the packet reach=
es the egress RB I am presuming there will be a inner-MAC lookup to deliver=
 the packet to the egress end-node. Edge-RB to smart endnode cannot have TR=
ILL encapsulation right? If so for my clarification we are saying the traff=
ic towards the end-node will be non-TRILL while towards the ingress edge-RB=
 will be TRILL encapsulated?</span><o:p></o:p></p><p class=3DMsoNormal styl=
e=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span style=3D'col=
or:#1F497D'>&nbsp;</span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-m=
argin-top-alt:auto;mso-margin-bottom-alt:auto'><span style=3D'color:#1F497D=
'>Phani</span><o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-a=
lt:auto;mso-margin-bottom-alt:auto'><span style=3D'color:#1F497D'>&nbsp;</s=
pan><o:p></o:p></p><div><div style=3D'border:none;border-top:solid #B5C4DF =
1.0pt;padding:3.0pt 0in 0in 0in'><p class=3DMsoNormal style=3D'mso-margin-t=
op-alt:auto;mso-margin-bottom-alt:auto'><b><span style=3D'font-size:10.0pt;=
font-family:"Tahoma","sans-serif"'>From:</span></b><span style=3D'font-size=
:10.0pt;font-family:"Tahoma","sans-serif"'> <a href=3D"mailto:trill-bounces=
@ietf.org" target=3D"_blank">trill-bounces@ietf.org</a> [mailto:<a href=3D"=
mailto:trill-bounces@ietf.org" target=3D"_blank">trill-bounces@ietf.org</a>=
] <b>On Behalf Of </b><a href=3D"mailto:Kesava_Vijaya_Krupak@Dell.com" targ=
et=3D"_blank">Kesava_Vijaya_Krupak@Dell.com</a><br><b>Sent:</b> Thursday, J=
anuary 03, 2013 4:19 PM<br><b>To:</b> <a href=3D"mailto:trill@ietf.org" tar=
get=3D"_blank">trill@ietf.org</a><br><b>Cc:</b> <a href=3D"mailto:d3e3e3@gm=
ail.com" target=3D"_blank">d3e3e3@gmail.com</a>; <a href=3D"mailto:radiaper=
lman@gmail.com" target=3D"_blank">radiaperlman@gmail.com</a>; <a href=3D"ma=
ilto:hu.fangwei@zte.com.cn" target=3D"_blank">hu.fangwei@zte.com.cn</a><br>=
<b>Subject:</b> [trill] New version : TRILL Smart Endnodes draft</span><o:p=
></o:p></p></div></div><div><div><p class=3DMsoNormal style=3D'mso-margin-t=
op-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p></o:p></p><p class=3DMso=
Normal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>Hi all,=
<o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-ma=
rgin-bottom-alt:auto'>&nbsp;<o:p></o:p></p><p class=3DMsoNormal style=3D'ms=
o-margin-top-alt:auto;mso-margin-bottom-alt:auto'>A new version of the Tril=
l Smart Endnodes draft has been posted.<o:p></o:p></p><p class=3DMsoNormal =
style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p></o=
:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bot=
tom-alt:auto'><a href=3D"http://datatracker.ietf.org/doc/draft-perlman-tril=
l-smart-endnodes" target=3D"_blank">http://datatracker.ietf.org/doc/draft-p=
erlman-trill-smart-endnodes</a><o:p></o:p></p><p class=3DMsoNormal style=3D=
'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p></o:p></p><=
p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:=
auto'>Kindly let us know your comments.<o:p></o:p></p><p class=3DMsoNormal =
style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p></o=
:p></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bot=
tom-alt:auto'>Thanks,<o:p></o:p></p><p class=3DMsoNormal style=3D'mso-margi=
n-top-alt:auto;mso-margin-bottom-alt:auto'>kvk<o:p></o:p></p></div></div></=
div></div></div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-m=
argin-bottom-alt:auto'>&nbsp;<o:p></o:p></p></div></div></div></div></div><=
/div></div><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin=
-bottom-alt:auto'>&nbsp;<o:p></o:p></p></div></div></div></div></div></div>=
<p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div></div></body></html>=

--_000_6895EAE0CA8DEE40B92D7CA88BB521F32414007BE1HQ1EXCH02corp_--

From linda.dunbar@huawei.com  Wed Jan  9 12:55:48 2013
Return-Path: <linda.dunbar@huawei.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 89E161F0CFD for <trill@ietfa.amsl.com>; Wed,  9 Jan 2013 12:55:48 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -5.998
X-Spam-Level: 
X-Spam-Status: No, score=-5.998 tagged_above=-999 required=5 tests=[AWL=-0.601, BAYES_00=-2.599, HTML_MESSAGE=0.001, J_CHICKENPOX_33=0.6, J_CHICKENPOX_81=0.6, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ZQYNH3sLln8g for <trill@ietfa.amsl.com>; Wed,  9 Jan 2013 12:55:43 -0800 (PST)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) by ietfa.amsl.com (Postfix) with ESMTP id E804E1F0CFC for <trill@ietf.org>; Wed,  9 Jan 2013 12:55:41 -0800 (PST)
Received: from 172.18.7.190 (EHLO lhreml204-edg.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.5-GA FastPath queued) with ESMTP id AOO95512; Wed, 09 Jan 2013 20:55:39 +0000 (GMT)
Received: from LHREML405-HUB.china.huawei.com (10.201.5.242) by lhreml204-edg.china.huawei.com (172.18.7.223) with Microsoft SMTP Server (TLS) id 14.1.323.3; Wed, 9 Jan 2013 20:54:40 +0000
Received: from DFWEML405-HUB.china.huawei.com (10.193.5.102) by lhreml405-hub.china.huawei.com (10.201.5.242) with Microsoft SMTP Server (TLS) id 14.1.323.3; Wed, 9 Jan 2013 20:55:38 +0000
Received: from DFWEML505-MBX.china.huawei.com ([10.124.31.100]) by dfweml405-hub.china.huawei.com ([10.193.5.102]) with mapi id 14.01.0323.003; Wed, 9 Jan 2013 12:55:33 -0800
From: Linda Dunbar <linda.dunbar@huawei.com>
To: Radia Perlman <radiaperlman@gmail.com>, Phanidhar Koganti <pkoganti@brocade.com>
Thread-Topic: Questions to  TRILL Smart Endnodes draft
Thread-Index: AQHN7quqE8c/67/j9UujxW5Rg87JVg==
Date: Wed, 9 Jan 2013 20:55:32 +0000
Message-ID: <4A95BA014132FF49AE685FAB4B9F17F6450E01E4@dfweml505-mbx>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com> <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com> <6895EAE0CA8DEE40B92D7CA88BB521F32414007BC1@HQ1-EXCH02.corp.brocade.com> <CAFOuuo626hw-c0kM4-1Fp-w2gByPenRnkFP3xNT+cTB4YcJ2kg@mail.gmail.com>
In-Reply-To: <CAFOuuo626hw-c0kM4-1Fp-w2gByPenRnkFP3xNT+cTB4YcJ2kg@mail.gmail.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.192.11.236]
Content-Type: multipart/alternative; boundary="_000_4A95BA014132FF49AE685FAB4B9F17F6450E01E4dfweml505mbx_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Cc: "d3e3e3@gmail.com" <d3e3e3@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: [trill] Questions to  TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 09 Jan 2013 20:55:48 -0000

--_000_4A95BA014132FF49AE685FAB4B9F17F6450E01E4dfweml505mbx_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Radia, et al,

The draft states that Smart End Node does both TRILL encapsulation and Deca=
psulation.
When there are multiple end nodes attached to one RBridge, this one RBridge=
 is the egress node for data frames  to all those end nodes.
If the RBridge doesn't decapsulate the TRILL header, how does it send the d=
ata frame to the correct end node?

https://datatracker.ietf.org/doc/draft-dunbar-trill-directory-assisted-enca=
p/ describes another type of smart node which only encapsulates TRILL heade=
r when sending data frames to remote nodes (i.e. attached to different RBri=
dges). All RBridges decapsulate TRILL header for data frames received from =
TRILL domain and towards directly attached end nodes, so that the data fram=
es can be forwarded by Ethernet destination address. Do you think this appr=
oach will work?

Thanks, Linda


From: trill-bounces@ietf.org [mailto:trill-bounces@ietf.org] On Behalf Of R=
adia Perlman
Sent: Wednesday, January 09, 2013 12:22 AM
To: Phanidhar Koganti
Cc: d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com; hu.fangwei@zte.com.cn;=
 trill@ietf.org
Subject: Re: [trill] New version : TRILL Smart Endnodes draft

Good point (see Phani's comment below).  I'll summarize:

The current  text implies that it's OK for smart endnode E, multihomed to R=
1 and R2, to choose one of the nicknames (say R1) for "ingress nickname" in=
 the TRILL header, and forward to either R1 or R2.  There are two problems =
with E forwarding a TRILL-encapsulated frame using ingeress "R1", through R=
2:

a) for multicast, likely to confuse the RPF check
b) for unicast, E would probably need to warn R2 that E is multihomed to R1=
 and will e using R1's nickname for ingress.

And actually, E won't get that much load splitting benefit from being multi=
homed if it uses "R1" as ingress, since all traffic to E will go through R1=
.  So, if E doesn't have a pseudonode nickname to use, then perhaps E shoul=
d just use one R1 or R2.

Or E could get its own nickname (I've been floating around the idea of a "d=
umb RBridge", which is like a "smart endnode", except it gets its own nickn=
ame, and does generate an LSP with "overloaded" flag.  It probably has to r=
eceive LSPs and calculate Dijkstra, in order to know which port to send mul=
ticast on (unless its attached true RBs are helpful enough to tell it..."yo=
u can send multicast with trees X or Y through me")

So, it seems like there are several options which should be discussed, for =
a smart endnode multihomed to R1 and R2
1) Always use a pseudnode nickname (but that's kind of complicated since th=
at nickname can only be used by smart endnodes that are multihomed to the e=
xact same set of RBridges)
2) Don't do active-active...choose one or R1 or R2 and always send everythi=
ng that way
3) Do active-active, but only for unicast...choose one of the nicknames, sa=
y "R1", and it's OK to send unicast either through R1 or R2, but multicast =
has to go through R1
4) Give E its own nickname (and somehow make sure it knows which port it ca=
n multicast on)
5) Other possibilities?

Radia




On Tue, Jan 8, 2013 at 8:32 PM, Phanidhar Koganti <pkoganti@brocade.com<mai=
lto:pkoganti@brocade.com>> wrote:
Thanks Radia. That clarifies for me. A more descriptive text in draft can s=
urely help vendors as they implement it in their hardware.

Another question on handling multi-homing
E can choose either R1 or R2's nickname, when encapsulating a
       frame, whether the encapsulated frame is sent via R1 or R2. If E
       wants to do active-active load splitting, and uses R1's nickname
       when forwarding through R1, and R2's nickname when forwarding
       through R2, this will cause distant RBridges (or smart endnodes)
       to keep changing their endnode table entry for D between (D, R1's
       nickname) and (D, R2's nickname). So it would be preferable for E
       to always encapsulate using the same nickname (R1 or R2) unless E
       detects a problem with connectivity using that nickname. And in
       this case, R1 and R2 need to be informed that the smart endnode
       might encapsulate with a different nickname, i.e., R1 might
       receive an encapsulated packet from smart endnode E using ingress
       nickname "R2".

If the smart endnode uses either R1 or R2 in the above example, would we ha=
ve TRILL RPF issues where upstream RBs see a packet with SRB=3Dsay R1 comin=
g from a wrong TRILL port? Let's say the RPF happens for multicast only the=
n we will atleast need to make sure the multicast (broadcast, unknown unica=
st and multcast) is sent via the edge-RB whose RBID the smart endnode is us=
ing.

This problem should not be there when using different RBID other than R1 an=
d R2 as described in approach-2

Phani
From: Radia Perlman [mailto:radiaperlman@gmail.com<mailto:radiaperlman@gmai=
l.com>]
Sent: Wednesday, January 09, 2013 2:53 AM
To: Phanidhar Koganti
Cc: Kesava_Vijaya_Krupak@Dell.com; trill@ietf.org<mailto:trill@ietf.org>; d=
3e3e3@gmail.com<mailto:d3e3e3@gmail.com>; hu.fangwei@zte.com.cn<mailto:hu.f=
angwei@zte.com.cn>
Subject: Re: New version : TRILL Smart Endnodes draft

Most of what you said is correct.  One correction

a) When the packet reaches the egress RB I am presuming there will be a inn=
er-MAC lookup to deliver the packet to the egress end-node.

correct...when egress RB R has to look up the inner MAC address E, which it=
 has to do anyway without smart endnodes, in order to choose which port to =
forward it on.  With smart endnodes, when R looks up MAC address E, it not =
only discovers which port to send it on, but that E is "smart", in which ca=
se R must leave the TRILL encapsulation

b) Edge-RB to smart endnode cannot have TRILL encapsulation right?

Not correct.  Traffic from egress RB to smart endnode will have TRILL encap=
sulation.  So for example, let's say R is the egress RB, and R has a link w=
ith two endnodes; E1 (which is smart) and E2 (which is not smart).

When R receives a TRILL packet from the campus, with TRILL destination nick=
name=3DR, R looks at the MAC destination.  If it is E2, then R removes the =
TRILL encapulation and forwards a native packet onto the link.  If it is E1=
, then R forwards the packet with the TRILL header.

c) Edge-RB to smart endnode cannot have TRILL encapsulation right? If so fo=
r my clarification we are saying the traffic towards the end-node will be n=
on-TRILL while towards the ingress edge-RB will be TRILL encapsulated?

Hopefully my answer to b) clarified this.  On the link between endnodes and=
 R, whether it's to or from endnode <--> edge RB, the packet will have TRIL=
L encapsulation if the endnode is smart, and not have TRILL encapsulation i=
f the endnode is not smart.

Radia


On Thu, Jan 3, 2013 at 3:19 AM, Phanidhar Koganti <pkoganti@brocade.com<mai=
lto:pkoganti@brocade.com>> wrote:
The draft defines the control and data path behavior from smart end-node to=
 edge-RB. When the packet reaches the egress RB I am presuming there will b=
e a inner-MAC lookup to deliver the packet to the egress end-node. Edge-RB =
to smart endnode cannot have TRILL encapsulation right? If so for my clarif=
ication we are saying the traffic towards the end-node will be non-TRILL wh=
ile towards the ingress edge-RB will be TRILL encapsulated?

Phani

From: trill-bounces@ietf.org<mailto:trill-bounces@ietf.org> [mailto:trill-b=
ounces@ietf.org<mailto:trill-bounces@ietf.org>] On Behalf Of Kesava_Vijaya_=
Krupak@Dell.com<mailto:Kesava_Vijaya_Krupak@Dell.com>
Sent: Thursday, January 03, 2013 4:19 PM
To: trill@ietf.org<mailto:trill@ietf.org>
Cc: d3e3e3@gmail.com<mailto:d3e3e3@gmail.com>; radiaperlman@gmail.com<mailt=
o:radiaperlman@gmail.com>; hu.fangwei@zte.com.cn<mailto:hu.fangwei@zte.com.=
cn>
Subject: [trill] New version : TRILL Smart Endnodes draft

Hi all,

A new version of the Trill Smart Endnodes draft has been posted.

http://datatracker.ietf.org/doc/draft-perlman-trill-smart-endnodes

Kindly let us know your comments.

Thanks,
kvk



--_000_4A95BA014132FF49AE685FAB4B9F17F6450E01E4dfweml505mbx_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micr=
osoft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:x=3D"urn:schemas-microsoft-com:office:excel" xmlns:p=3D"urn:schemas-m=
icrosoft-com:office:powerpoint" xmlns:a=3D"urn:schemas-microsoft-com:office=
:access" xmlns:dt=3D"uuid:C2F41010-65B3-11d1-A29F-00AA00C14882" xmlns:s=3D"=
uuid:BDC6E3F0-6DA3-11d1-A2A3-00AA00C14882" xmlns:rs=3D"urn:schemas-microsof=
t-com:rowset" xmlns:z=3D"#RowsetSchema" xmlns:b=3D"urn:schemas-microsoft-co=
m:office:publisher" xmlns:ss=3D"urn:schemas-microsoft-com:office:spreadshee=
t" xmlns:c=3D"urn:schemas-microsoft-com:office:component:spreadsheet" xmlns=
:odc=3D"urn:schemas-microsoft-com:office:odc" xmlns:oa=3D"urn:schemas-micro=
soft-com:office:activation" xmlns:html=3D"http://www.w3.org/TR/REC-html40" =
xmlns:q=3D"http://schemas.xmlsoap.org/soap/envelope/" xmlns:rtc=3D"http://m=
icrosoft.com/officenet/conferencing" xmlns:D=3D"DAV:" xmlns:Repl=3D"http://=
schemas.microsoft.com/repl/" xmlns:mt=3D"http://schemas.microsoft.com/share=
point/soap/meetings/" xmlns:x2=3D"http://schemas.microsoft.com/office/excel=
/2003/xml" xmlns:ppda=3D"http://www.passport.com/NameSpace.xsd" xmlns:ois=
=3D"http://schemas.microsoft.com/sharepoint/soap/ois/" xmlns:dir=3D"http://=
schemas.microsoft.com/sharepoint/soap/directory/" xmlns:ds=3D"http://www.w3=
.org/2000/09/xmldsig#" xmlns:dsp=3D"http://schemas.microsoft.com/sharepoint=
/dsp" xmlns:udc=3D"http://schemas.microsoft.com/data/udc" xmlns:xsd=3D"http=
://www.w3.org/2001/XMLSchema" xmlns:sub=3D"http://schemas.microsoft.com/sha=
repoint/soap/2002/1/alerts/" xmlns:ec=3D"http://www.w3.org/2001/04/xmlenc#"=
 xmlns:sp=3D"http://schemas.microsoft.com/sharepoint/" xmlns:sps=3D"http://=
schemas.microsoft.com/sharepoint/soap/" xmlns:xsi=3D"http://www.w3.org/2001=
/XMLSchema-instance" xmlns:udcs=3D"http://schemas.microsoft.com/data/udc/so=
ap" xmlns:udcxf=3D"http://schemas.microsoft.com/data/udc/xmlfile" xmlns:udc=
p2p=3D"http://schemas.microsoft.com/data/udc/parttopart" xmlns:wf=3D"http:/=
/schemas.microsoft.com/sharepoint/soap/workflow/" xmlns:dsss=3D"http://sche=
mas.microsoft.com/office/2006/digsig-setup" xmlns:dssi=3D"http://schemas.mi=
crosoft.com/office/2006/digsig" xmlns:mdssi=3D"http://schemas.openxmlformat=
s.org/package/2006/digital-signature" xmlns:mver=3D"http://schemas.openxmlf=
ormats.org/markup-compatibility/2006" xmlns:m=3D"http://schemas.microsoft.c=
om/office/2004/12/omml" xmlns:mrels=3D"http://schemas.openxmlformats.org/pa=
ckage/2006/relationships" xmlns:spwp=3D"http://microsoft.com/sharepoint/web=
partpages" xmlns:ex12t=3D"http://schemas.microsoft.com/exchange/services/20=
06/types" xmlns:ex12m=3D"http://schemas.microsoft.com/exchange/services/200=
6/messages" xmlns:pptsl=3D"http://schemas.microsoft.com/sharepoint/soap/Sli=
deLibrary/" xmlns:spsl=3D"http://microsoft.com/webservices/SharePointPortal=
Server/PublishedLinksService" xmlns:Z=3D"urn:schemas-microsoft-com:" xmlns:=
st=3D"&#1;" xmlns=3D"http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Dus-ascii"=
>
<meta name=3D"Generator" content=3D"Microsoft Word 12 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
	{font-family:SimSun;
	panose-1:2 1 6 0 3 1 1 1 1 1;}
@font-face
	{font-family:"Cambria Math";
	panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:Tahoma;
	panose-1:2 11 6 4 3 5 4 4 2 4;}
@font-face
	{font-family:SimSun;
	panose-1:2 1 6 0 3 1 1 1 1 1;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman","serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
span.EmailStyle17
	{mso-style-type:personal-reply;
	font-family:"Calibri","sans-serif";
	color:#1F497D;}
.MsoChpDefault
	{mso-style-type:export-only;}
@page WordSection1
	{size:8.5in 11.0in;
	margin:1.0in 1.25in 1.0in 1.25in;}
div.WordSection1
	{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang=3D"EN-US" link=3D"blue" vlink=3D"purple">
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">Radia, et al,
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">The draft states that Sma=
rt End Node does both TRILL encapsulation and Decapsulation.
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">When there are multiple e=
nd nodes attached to one RBridge, this one RBridge is the egress node for d=
ata frames &nbsp;to all those end nodes. &nbsp;<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">If the RBridge doesn&#821=
7;t decapsulate the TRILL header, how does it send the data frame to the co=
rrect end node?
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><a href=3D"https://datatr=
acker.ietf.org/doc/draft-dunbar-trill-directory-assisted-encap/">https://da=
tatracker.ietf.org/doc/draft-dunbar-trill-directory-assisted-encap/</a>
</span><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&quo=
t;sans-serif&quot;;color:#1F497D">describes another type of smart node whic=
h only encapsulates TRILL header when sending data frames to remote nodes (=
i.e. attached to different RBridges). All RBridges decapsulate
 TRILL header for data frames received from TRILL domain and towards direct=
ly attached end nodes, so that the data frames can be forwarded by Ethernet=
 destination address. Do you think this approach will work?<o:p></o:p></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">Thanks, Linda<o:p></o:p><=
/span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<div style=3D"border:none;border-left:solid blue 1.5pt;padding:0in 0in 0in =
4.0pt">
<div>
<div style=3D"border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:10.0pt;font-family:&quot=
;Tahoma&quot;,&quot;sans-serif&quot;">From:</span></b><span style=3D"font-s=
ize:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;"> trill-bo=
unces@ietf.org [mailto:trill-bounces@ietf.org]
<b>On Behalf Of </b>Radia Perlman<br>
<b>Sent:</b> Wednesday, January 09, 2013 12:22 AM<br>
<b>To:</b> Phanidhar Koganti<br>
<b>Cc:</b> d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com; hu.fangwei@zte.=
com.cn; trill@ietf.org<br>
<b>Subject:</b> Re: [trill] New version : TRILL Smart Endnodes draft<o:p></=
o:p></span></p>
</div>
</div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Good point (see Phani's comment below). &nbsp;I'll s=
ummarize:<o:p></o:p></p>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal">The current &nbsp;text implies that it's OK for smar=
t endnode E, multihomed to R1 and R2, to choose one of the nicknames (say R=
1) for &quot;ingress nickname&quot; in the TRILL header, and forward to eit=
her R1 or R2. &nbsp;There are two problems with E forwarding
 a TRILL-encapsulated frame using ingeress &quot;R1&quot;, through R2:<o:p>=
</o:p></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal">a) for multicast, likely to confuse the RPF check<o:=
p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal">b) for unicast, E would probably need to warn R2 tha=
t E is multihomed to R1 and will e using R1's nickname for ingress.<o:p></o=
:p></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal">And actually, E won't get that much load splitting b=
enefit from being multihomed if it uses &quot;R1&quot; as ingress, since al=
l traffic to E will go through R1. &nbsp;So, if E doesn't have a pseudonode=
 nickname to use, then perhaps E should just use
 one R1 or R2.<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal">Or E could get its own nickname (I've been floating =
around the idea of a &quot;dumb RBridge&quot;, which is like a &quot;smart =
endnode&quot;, except it gets its own nickname, and does generate an LSP wi=
th &quot;overloaded&quot; flag. &nbsp;It probably has to receive LSPs
 and calculate Dijkstra, in order to know which port to send multicast on (=
unless its attached true RBs are helpful enough to tell it...&quot;you can =
send multicast with trees X or Y through me&quot;)<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal">So, it seems like there are several options which sh=
ould be discussed, for a smart endnode multihomed to R1 and R2<o:p></o:p></=
p>
</div>
<div>
<p class=3D"MsoNormal">1) Always use a pseudnode nickname (but that's kind =
of complicated since that nickname can only be used by smart endnodes that =
are multihomed to the exact same set of RBridges)<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal">2) Don't do active-active...choose one or R1 or R2 a=
nd always send everything that way<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal">3) Do active-active, but only for unicast...choose o=
ne of the nicknames, say &quot;R1&quot;, and it's OK to send unicast either=
 through R1 or R2, but multicast has to go through R1<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal">4) Give E its own nickname (and somehow make sure it=
 knows which port it can multicast on)<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal">5) Other possibilities?<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal">Radia<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><o:p>&nbsp;</o:p></p>
<div>
<p class=3D"MsoNormal">On Tue, Jan 8, 2013 at 8:32 PM, Phanidhar Koganti &l=
t;<a href=3D"mailto:pkoganti@brocade.com" target=3D"_blank">pkoganti@brocad=
e.com</a>&gt; wrote:<o:p></o:p></p>
<div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">Thanks Radia. That clarifies for me. A =
more descriptive text in draft can surely help vendors as
 they implement it in their hardware.</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">&nbsp;</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">Another question on handling multi-homi=
ng</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto;line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">E can =
choose either R1 or R2's nickname, when encapsulating a</span><o:p></o:p></=
p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto;line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; frame, whether the encapsulated frame is sen=
t via R1 or R2. If E</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto;line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; wants to do active-active load splitting, an=
d uses R1's nickname</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto;line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; when forwarding through R1, and R2's nicknam=
e when forwarding</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto;line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; through R2, this will cause distant RBridges=
 (or smart endnodes)</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto;line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; to keep changing their endnode table entry f=
or D between (D, R1's</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto;line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; nickname) and (D, R2's nickname). So it woul=
d be preferable for E</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto;line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; to always encapsulate using the same nicknam=
e (R1 or R2) unless E</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto;line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; detects a problem with connectivity using th=
at nickname. And in</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto;line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; this case, R1 and R2 need to be informed tha=
t the smart endnode</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto;line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; might encapsulate with a different nickname,=
 i.e., R1 might</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto;line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; receive an encapsulated packet from smart en=
dnode E using ingress</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto;line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">&nbsp;=
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; nickname &quot;R2&quot;.</span><o:p></o:p></=
p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">&nbsp;</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">If the smart endnode uses either R1 or =
R2 in the above example, would we have TRILL RPF issues where
 upstream RBs see a packet with SRB=3Dsay R1 coming from a wrong TRILL port=
? Let&#8217;s say the RPF happens for multicast only then we will atleast n=
eed to make sure the multicast (broadcast, unknown unicast and multcast) is=
 sent via the edge-RB whose RBID the smart
 endnode is using.</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">&nbsp;</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">This problem should not be there when u=
sing different RBID other than R1 and R2 as described in approach-2</span><=
o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">&nbsp;</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">Phani</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><b><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,=
&quot;sans-serif&quot;">From:</span></b><span style=3D"font-size:10.0pt;fon=
t-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;"> Radia Perlman [mailto:=
<a href=3D"mailto:radiaperlman@gmail.com" target=3D"_blank">radiaperlman@gm=
ail.com</a>]
<br>
<b>Sent:</b> Wednesday, January 09, 2013 2:53 AM<br>
<b>To:</b> Phanidhar Koganti<br>
<b>Cc:</b> Kesava_Vijaya_Krupak@Dell.com; <a href=3D"mailto:trill@ietf.org"=
 target=3D"_blank">
trill@ietf.org</a>; <a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d=
3e3e3@gmail.com</a>;
<a href=3D"mailto:hu.fangwei@zte.com.cn" target=3D"_blank">hu.fangwei@zte.c=
om.cn</a><br>
<b>Subject:</b> Re: New version : TRILL Smart Endnodes draft</span><o:p></o=
:p></p>
<div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">Most of what you said is correct. &nbsp;One correction<o:p></o:p><=
/p>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">a)&nbsp;<span style=3D"color:#1F497D">When the packet reaches the =
egress RB I am presuming there will be a inner-MAC lookup to deliver the pa=
cket to the egress end-node.</span><o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"color:#1F497D">correct...when egress RB R has to lo=
ok up the inner MAC address E, which it has to do anyway without smart endn=
odes, in order to choose which port to
 forward it on. &nbsp;With smart endnodes, when R looks up MAC address E, i=
t not only discovers which port to send it on, but that E is &quot;smart&qu=
ot;, in which case R must leave the TRILL encapsulation</span><o:p></o:p></=
p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"color:#1F497D">b)&nbsp;Edge-RB to smart endnode can=
not have TRILL encapsulation right?</span><o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"color:#1F497D">Not correct. &nbsp;Traffic from egre=
ss RB to smart endnode will have TRILL encapsulation. &nbsp;So for example,=
 let's say R is the egress RB, and R has a link
 with two endnodes; E1 (which is smart) and E2 (which is not smart).</span>=
<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"color:#1F497D">When R receives a TRILL packet from =
the campus, with TRILL destination nickname=3DR, R looks at the MAC destina=
tion. &nbsp;If it is E2, then R removes the TRILL
 encapulation and forwards a native packet onto the link. &nbsp;If it is E1=
, then R forwards the packet with the TRILL header.</span><o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"color:#1F497D">c)&nbsp;Edge-RB to smart endnode can=
not have TRILL encapsulation right? If so for my clarification we are sayin=
g the traffic towards the end-node will be
 non-TRILL while towards the ingress edge-RB will be TRILL encapsulated?</s=
pan><o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"color:#1F497D">Hopefully my answer to b) clarified =
this. &nbsp;On the link between endnodes and R, whether it's to or from end=
node &lt;--&gt; edge RB, the packet will have TRILL
 encapsulation if the endnode is smart, and not have TRILL encapsulation if=
 the endnode is not smart.</span><o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"color:#1F497D">Radia</span><o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
<div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">On Thu, Jan 3, 2013 at 3:19 AM, Phanidhar Koganti &lt;<a href=3D"m=
ailto:pkoganti@brocade.com" target=3D"_blank">pkoganti@brocade.com</a>&gt; =
wrote:<o:p></o:p></p>
<div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"color:#1F497D">The draft defines the control and da=
ta path behavior from smart end-node to edge-RB. When the packet reaches th=
e egress RB I am presuming there will
 be a inner-MAC lookup to deliver the packet to the egress end-node. Edge-R=
B to smart endnode cannot have TRILL encapsulation right? If so for my clar=
ification we are saying the traffic towards the end-node will be non-TRILL =
while towards the ingress edge-RB
 will be TRILL encapsulated?</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"color:#1F497D">&nbsp;</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"color:#1F497D">Phani</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"color:#1F497D">&nbsp;</span><o:p></o:p></p>
<div>
<div style=3D"border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><b><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,=
&quot;sans-serif&quot;">From:</span></b><span style=3D"font-size:10.0pt;fon=
t-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;">
<a href=3D"mailto:trill-bounces@ietf.org" target=3D"_blank">trill-bounces@i=
etf.org</a> [mailto:<a href=3D"mailto:trill-bounces@ietf.org" target=3D"_bl=
ank">trill-bounces@ietf.org</a>]
<b>On Behalf Of </b><a href=3D"mailto:Kesava_Vijaya_Krupak@Dell.com" target=
=3D"_blank">Kesava_Vijaya_Krupak@Dell.com</a><br>
<b>Sent:</b> Thursday, January 03, 2013 4:19 PM<br>
<b>To:</b> <a href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ietf.o=
rg</a><br>
<b>Cc:</b> <a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gma=
il.com</a>;
<a href=3D"mailto:radiaperlman@gmail.com" target=3D"_blank">radiaperlman@gm=
ail.com</a>;
<a href=3D"mailto:hu.fangwei@zte.com.cn" target=3D"_blank">hu.fangwei@zte.c=
om.cn</a><br>
<b>Subject:</b> [trill] New version : TRILL Smart Endnodes draft</span><o:p=
></o:p></p>
</div>
</div>
<div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">Hi all,<o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">A new version of the Trill Smart Endnodes draft has been posted.<o=
:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><a href=3D"http://datatracker.ietf.org/doc/draft-perlman-trill-sma=
rt-endnodes" target=3D"_blank">http://datatracker.ietf.org/doc/draft-perlma=
n-trill-smart-endnodes</a><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">Kindly let us know your comments.<o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">Thanks,<o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">kvk<o:p></o:p></p>
</div>
</div>
</div>
</div>
</div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
</div>
</div>
</body>
</html>

--_000_4A95BA014132FF49AE685FAB4B9F17F6450E01E4dfweml505mbx_--

From radiaperlman@gmail.com  Wed Jan  9 14:32:10 2013
Return-Path: <radiaperlman@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5635921F843B for <trill@ietfa.amsl.com>; Wed,  9 Jan 2013 14:32:10 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.772
X-Spam-Level: 
X-Spam-Status: No, score=-2.772 tagged_above=-999 required=5 tests=[AWL=-0.374, BAYES_00=-2.599, HTML_MESSAGE=0.001, J_CHICKENPOX_33=0.6, J_CHICKENPOX_81=0.6, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id d1ZpFXLJjONr for <trill@ietfa.amsl.com>; Wed,  9 Jan 2013 14:32:08 -0800 (PST)
Received: from mail-lb0-f169.google.com (mail-lb0-f169.google.com [209.85.217.169]) by ietfa.amsl.com (Postfix) with ESMTP id 2B1C221F8436 for <trill@ietf.org>; Wed,  9 Jan 2013 14:32:00 -0800 (PST)
Received: by mail-lb0-f169.google.com with SMTP id gk1so408lbb.0 for <trill@ietf.org>; Wed, 09 Jan 2013 14:31:59 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=8hRSANNXNWVief+98FJFl/vqZDwwUnK383S/TDsnFRU=; b=Y1oTGNifjhd8Lr7scQsAHpix1TnOa6e9YaX077JfHdwRBxXa8Fo74smLE5AkWhEriE hHGT+1LJ1oLaUG36tFx3ZtHVTKtgakuZ4Ejob5Yxchq3JH8m05DSQNt70tgwSwbdq+QO 8J5Hz2GWgznRiQ5puuF+8G3EzKqXCK7SNeoc3P18lULZKAQEaWLA8zf2BhCCcXqftfCO 6Tph/PRhLANs9jBpopLLpH5T/MwY64nTrqqNrjfjlW/manXWowVyWUf6mzYlP2y7f+9n vCoOyXBA9saoHxPV90jZIvfA4GMx4mDucAIu2/m3VJKtA4WwXEO72XZjfvIcbZx4GwoG arpg==
MIME-Version: 1.0
Received: by 10.112.86.232 with SMTP id s8mr28606339lbz.86.1357770719775; Wed, 09 Jan 2013 14:31:59 -0800 (PST)
Received: by 10.112.64.17 with HTTP; Wed, 9 Jan 2013 14:31:59 -0800 (PST)
In-Reply-To: <4A95BA014132FF49AE685FAB4B9F17F6450E01E4@dfweml505-mbx>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com> <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com> <6895EAE0CA8DEE40B92D7CA88BB521F32414007BC1@HQ1-EXCH02.corp.brocade.com> <CAFOuuo626hw-c0kM4-1Fp-w2gByPenRnkFP3xNT+cTB4YcJ2kg@mail.gmail.com> <4A95BA014132FF49AE685FAB4B9F17F6450E01E4@dfweml505-mbx>
Date: Wed, 9 Jan 2013 14:31:59 -0800
Message-ID: <CAFOuuo6=wDeKiG-zjSnkcO3B2iTbazpsGHNYJLSfOxYZsh9kFA@mail.gmail.com>
From: Radia Perlman <radiaperlman@gmail.com>
To: Linda Dunbar <linda.dunbar@huawei.com>
Content-Type: multipart/alternative; boundary=f46d0401f95b704b3504d2e2a13e
Cc: "d3e3e3@gmail.com" <d3e3e3@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Questions to TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 09 Jan 2013 22:32:10 -0000

--f46d0401f95b704b3504d2e2a13e
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

I think that's in the smart endnode draft. I hope it is (I'll have to
reread it).  But it was certainly discussed in email.

If there is a smart endnode on a shared link with other endnodes, it learns
about its endnode neighbors, and sends directly to them (without
encapsulation).  If smart E wants to talk to unknown destination D, E has
to send two copies; one (without encapsulation), multicast on the link, and
the other (with encapsulation), forwarded to the attached RBridge R.

R has been warned by E of all the MAC addresses E owns.  R must drop all
packets it receives on the link if either the source or destination MAC is
one of the ones owned by a smart endnode on the link.

Radia

On Wed, Jan 9, 2013 at 12:55 PM, Linda Dunbar <linda.dunbar@huawei.com>wrot=
e:

>  Radia, et al, ****
>
> ** **
>
> The draft states that Smart End Node does both TRILL encapsulation and
> Decapsulation. ****
>
> When there are multiple end nodes attached to one RBridge, this one
> RBridge is the egress node for data frames  to all those end nodes.  ****
>
> If the RBridge doesn=92t decapsulate the TRILL header, how does it send t=
he
> data frame to the correct end node? ****
>
> ** **
>
>
> https://datatracker.ietf.org/doc/draft-dunbar-trill-directory-assisted-en=
cap/ describes
> another type of smart node which only encapsulates TRILL header when
> sending data frames to remote nodes (i.e. attached to different RBridges)=
.
> All RBridges decapsulate TRILL header for data frames received from TRILL
> domain and towards directly attached end nodes, so that the data frames c=
an
> be forwarded by Ethernet destination address. Do you think this approach
> will work?****
>
> ** **
>
> Thanks, Linda****
>
> ** **
>
> ** **
>
> *From:* trill-bounces@ietf.org [mailto:trill-bounces@ietf.org] *On Behalf
> Of *Radia Perlman
> *Sent:* Wednesday, January 09, 2013 12:22 AM
> *To:* Phanidhar Koganti
> *Cc:* d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com;
> hu.fangwei@zte.com.cn; trill@ietf.org
> *Subject:* Re: [trill] New version : TRILL Smart Endnodes draft****
>
> ** **
>
> Good point (see Phani's comment below).  I'll summarize:****
>
> ** **
>
> The current  text implies that it's OK for smart endnode E, multihomed to
> R1 and R2, to choose one of the nicknames (say R1) for "ingress nickname"
> in the TRILL header, and forward to either R1 or R2.  There are two
> problems with E forwarding a TRILL-encapsulated frame using ingeress "R1"=
,
> through R2:****
>
> ** **
>
> a) for multicast, likely to confuse the RPF check****
>
> b) for unicast, E would probably need to warn R2 that E is multihomed to
> R1 and will e using R1's nickname for ingress.****
>
> ** **
>
> And actually, E won't get that much load splitting benefit from being
> multihomed if it uses "R1" as ingress, since all traffic to E will go
> through R1.  So, if E doesn't have a pseudonode nickname to use, then
> perhaps E should just use one R1 or R2.****
>
> ** **
>
> Or E could get its own nickname (I've been floating around the idea of a
> "dumb RBridge", which is like a "smart endnode", except it gets its own
> nickname, and does generate an LSP with "overloaded" flag.  It probably h=
as
> to receive LSPs and calculate Dijkstra, in order to know which port to se=
nd
> multicast on (unless its attached true RBs are helpful enough to tell
> it..."you can send multicast with trees X or Y through me")****
>
> ** **
>
> So, it seems like there are several options which should be discussed, fo=
r
> a smart endnode multihomed to R1 and R2****
>
> 1) Always use a pseudnode nickname (but that's kind of complicated since
> that nickname can only be used by smart endnodes that are multihomed to t=
he
> exact same set of RBridges)****
>
> 2) Don't do active-active...choose one or R1 or R2 and always send
> everything that way****
>
> 3) Do active-active, but only for unicast...choose one of the nicknames,
> say "R1", and it's OK to send unicast either through R1 or R2, but
> multicast has to go through R1****
>
> 4) Give E its own nickname (and somehow make sure it knows which port it
> can multicast on)****
>
> 5) Other possibilities?****
>
> ** **
>
> Radia****
>
> ** **
>
> ** **
>
> ** **
>
> ** **
>
> On Tue, Jan 8, 2013 at 8:32 PM, Phanidhar Koganti <pkoganti@brocade.com>
> wrote:****
>
> Thanks Radia. That clarifies for me. A more descriptive text in draft can
> surely help vendors as they implement it in their hardware.****
>
>  ****
>
> Another question on handling multi-homing****
>
> E can choose either R1 or R2's nickname, when encapsulating a****
>
>        frame, whether the encapsulated frame is sent via R1 or R2. If E**=
*
> *
>
>        wants to do active-active load splitting, and uses R1's nickname**=
*
> *
>
>        when forwarding through R1, and R2's nickname when forwarding****
>
>        through R2, this will cause distant RBridges (or smart endnodes)**=
*
> *
>
>        to keep changing their endnode table entry for D between (D, R1's*=
*
> **
>
>        nickname) and (D, R2's nickname). So it would be preferable for E*=
*
> **
>
>        to always encapsulate using the same nickname (R1 or R2) unless E*=
*
> **
>
>        detects a problem with connectivity using that nickname. And in***=
*
>
>        this case, R1 and R2 need to be informed that the smart endnode***=
*
>
>        might encapsulate with a different nickname, i.e., R1 might****
>
>        receive an encapsulated packet from smart endnode E using ingress*=
*
> **
>
>        nickname "R2".****
>
>  ****
>
> If the smart endnode uses either R1 or R2 in the above example, would we
> have TRILL RPF issues where upstream RBs see a packet with SRB=3Dsay R1
> coming from a wrong TRILL port? Let=92s say the RPF happens for multicast
> only then we will atleast need to make sure the multicast (broadcast,
> unknown unicast and multcast) is sent via the edge-RB whose RBID the smar=
t
> endnode is using.****
>
>  ****
>
> This problem should not be there when using different RBID other than R1
> and R2 as described in approach-2****
>
>  ****
>
> Phani****
>
> *From:* Radia Perlman [mailto:radiaperlman@gmail.com]
> *Sent:* Wednesday, January 09, 2013 2:53 AM
> *To:* Phanidhar Koganti
> *Cc:* Kesava_Vijaya_Krupak@Dell.com; trill@ietf.org; d3e3e3@gmail.com;
> hu.fangwei@zte.com.cn
> *Subject:* Re: New version : TRILL Smart Endnodes draft****
>
>  ****
>
> Most of what you said is correct.  One correction****
>
>  ****
>
> a) When the packet reaches the egress RB I am presuming there will be a
> inner-MAC lookup to deliver the packet to the egress end-node.****
>
>  ****
>
> correct...when egress RB R has to look up the inner MAC address E, which
> it has to do anyway without smart endnodes, in order to choose which port
> to forward it on.  With smart endnodes, when R looks up MAC address E, it
> not only discovers which port to send it on, but that E is "smart", in
> which case R must leave the TRILL encapsulation****
>
>  ****
>
> b) Edge-RB to smart endnode cannot have TRILL encapsulation right?****
>
>  ****
>
> Not correct.  Traffic from egress RB to smart endnode will have TRILL
> encapsulation.  So for example, let's say R is the egress RB, and R has a
> link with two endnodes; E1 (which is smart) and E2 (which is not smart).*=
*
> **
>
>  ****
>
> When R receives a TRILL packet from the campus, with TRILL destination
> nickname=3DR, R looks at the MAC destination.  If it is E2, then R remove=
s
> the TRILL encapulation and forwards a native packet onto the link.  If it
> is E1, then R forwards the packet with the TRILL header.****
>
>  ****
>
> c) Edge-RB to smart endnode cannot have TRILL encapsulation right? If so
> for my clarification we are saying the traffic towards the end-node will =
be
> non-TRILL while towards the ingress edge-RB will be TRILL encapsulated?**=
*
> *
>
>  ****
>
> Hopefully my answer to b) clarified this.  On the link between endnodes
> and R, whether it's to or from endnode <--> edge RB, the packet will have
> TRILL encapsulation if the endnode is smart, and not have TRILL
> encapsulation if the endnode is not smart.****
>
>  ****
>
> Radia****
>
>  ****
>
>  ****
>
> On Thu, Jan 3, 2013 at 3:19 AM, Phanidhar Koganti <pkoganti@brocade.com>
> wrote:****
>
> The draft defines the control and data path behavior from smart end-node
> to edge-RB. When the packet reaches the egress RB I am presuming there wi=
ll
> be a inner-MAC lookup to deliver the packet to the egress end-node. Edge-=
RB
> to smart endnode cannot have TRILL encapsulation right? If so for my
> clarification we are saying the traffic towards the end-node will be
> non-TRILL while towards the ingress edge-RB will be TRILL encapsulated?**=
*
> *
>
>  ****
>
> Phani****
>
>  ****
>
> *From:* trill-bounces@ietf.org [mailto:trill-bounces@ietf.org] *On Behalf
> Of *Kesava_Vijaya_Krupak@Dell.com
> *Sent:* Thursday, January 03, 2013 4:19 PM
> *To:* trill@ietf.org
> *Cc:* d3e3e3@gmail.com; radiaperlman@gmail.com; hu.fangwei@zte.com.cn
> *Subject:* [trill] New version : TRILL Smart Endnodes draft****
>
>  ****
>
> Hi all,****
>
>  ****
>
> A new version of the Trill Smart Endnodes draft has been posted.****
>
>  ****
>
> http://datatracker.ietf.org/doc/draft-perlman-trill-smart-endnodes****
>
>  ****
>
> Kindly let us know your comments.****
>
>  ****
>
> Thanks,****
>
> kvk****
>
>  ****
>
> ** **
>
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill
>
>

--f46d0401f95b704b3504d2e2a13e
Content-Type: text/html; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

I think that&#39;s in the smart endnode draft. I hope it is (I&#39;ll have =
to reread it). =A0But it was certainly discussed in email.<div><br></div><d=
iv>If there is a smart endnode on a shared link with other endnodes, it lea=
rns about its endnode neighbors, and sends directly to them (without encaps=
ulation). =A0If smart E wants to talk to unknown destination D, E has to se=
nd two copies; one (without encapsulation), multicast on the link, and the =
other (with encapsulation), forwarded to the attached RBridge R.</div>
<div><br></div><div>R has been warned by E of all the MAC addresses E owns.=
 =A0R must drop all packets it receives on the link if either the source or=
 destination MAC is one of the ones owned by a smart endnode on the link.</=
div>
<div><br></div><div>Radia</div><div><br><div class=3D"gmail_quote">On Wed, =
Jan 9, 2013 at 12:55 PM, Linda Dunbar <span dir=3D"ltr">&lt;<a href=3D"mail=
to:linda.dunbar@huawei.com" target=3D"_blank">linda.dunbar@huawei.com</a>&g=
t;</span> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">





<div lang=3D"EN-US" link=3D"blue" vlink=3D"purple">
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">Radia, et al,
<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">The draft states that Sma=
rt End Node does both TRILL encapsulation and Decapsulation.
<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">When there are multiple e=
nd nodes attached to one RBridge, this one RBridge is the egress node for d=
ata frames =A0to all those end nodes. =A0<u></u><u></u></span></p>

<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">If the RBridge doesn=92t =
decapsulate the TRILL header, how does it send the data frame to the correc=
t end node?
<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><a href=3D"https://datatr=
acker.ietf.org/doc/draft-dunbar-trill-directory-assisted-encap/" target=3D"=
_blank">https://datatracker.ietf.org/doc/draft-dunbar-trill-directory-assis=
ted-encap/</a>
</span><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&quo=
t;sans-serif&quot;;color:#1f497d">describes another type of smart node whic=
h only encapsulates TRILL header when sending data frames to remote nodes (=
i.e. attached to different RBridges). All RBridges decapsulate
 TRILL header for data frames received from TRILL domain and towards direct=
ly attached end nodes, so that the data frames can be forwarded by Ethernet=
 destination address. Do you think this approach will work?<u></u><u></u></=
span></p>

<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">Thanks, Linda<u></u><u></=
u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
<div style=3D"border:none;border-left:solid blue 1.5pt;padding:0in 0in 0in =
4.0pt">
<div>
<div style=3D"border:none;border-top:solid #b5c4df 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:10.0pt;font-family:&quot=
;Tahoma&quot;,&quot;sans-serif&quot;">From:</span></b><span style=3D"font-s=
ize:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;"> <a href=
=3D"mailto:trill-bounces@ietf.org" target=3D"_blank">trill-bounces@ietf.org=
</a> [mailto:<a href=3D"mailto:trill-bounces@ietf.org" target=3D"_blank">tr=
ill-bounces@ietf.org</a>]
<b>On Behalf Of </b>Radia Perlman<br>
<b>Sent:</b> Wednesday, January 09, 2013 12:22 AM<br>
<b>To:</b> Phanidhar Koganti<br>
<b>Cc:</b> <a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gma=
il.com</a>; Kesava_Vijaya_Krupak@Dell.com; <a href=3D"mailto:hu.fangwei@zte=
.com.cn" target=3D"_blank">hu.fangwei@zte.com.cn</a>; <a href=3D"mailto:tri=
ll@ietf.org" target=3D"_blank">trill@ietf.org</a><br>

<b>Subject:</b> Re: [trill] New version : TRILL Smart Endnodes draft<u></u>=
<u></u></span></p>
</div>
</div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
<p class=3D"MsoNormal">Good point (see Phani&#39;s comment below). =A0I&#39=
;ll summarize:<u></u><u></u></p>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
<div>
<p class=3D"MsoNormal">The current =A0text implies that it&#39;s OK for sma=
rt endnode E, multihomed to R1 and R2, to choose one of the nicknames (say =
R1) for &quot;ingress nickname&quot; in the TRILL header, and forward to ei=
ther R1 or R2. =A0There are two problems with E forwarding
 a TRILL-encapsulated frame using ingeress &quot;R1&quot;, through R2:<u></=
u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
<div>
<p class=3D"MsoNormal">a) for multicast, likely to confuse the RPF check<u>=
</u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal">b) for unicast, E would probably need to warn R2 tha=
t E is multihomed to R1 and will e using R1&#39;s nickname for ingress.<u><=
/u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
<div>
<p class=3D"MsoNormal">And actually, E won&#39;t get that much load splitti=
ng benefit from being multihomed if it uses &quot;R1&quot; as ingress, sinc=
e all traffic to E will go through R1. =A0So, if E doesn&#39;t have a pseud=
onode nickname to use, then perhaps E should just use
 one R1 or R2.<u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
<div>
<p class=3D"MsoNormal">Or E could get its own nickname (I&#39;ve been float=
ing around the idea of a &quot;dumb RBridge&quot;, which is like a &quot;sm=
art endnode&quot;, except it gets its own nickname, and does generate an LS=
P with &quot;overloaded&quot; flag. =A0It probably has to receive LSPs
 and calculate Dijkstra, in order to know which port to send multicast on (=
unless its attached true RBs are helpful enough to tell it...&quot;you can =
send multicast with trees X or Y through me&quot;)<u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
<div>
<p class=3D"MsoNormal">So, it seems like there are several options which sh=
ould be discussed, for a smart endnode multihomed to R1 and R2<u></u><u></u=
></p>
</div>
<div>
<p class=3D"MsoNormal">1) Always use a pseudnode nickname (but that&#39;s k=
ind of complicated since that nickname can only be used by smart endnodes t=
hat are multihomed to the exact same set of RBridges)<u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal">2) Don&#39;t do active-active...choose one or R1 or =
R2 and always send everything that way<u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal">3) Do active-active, but only for unicast...choose o=
ne of the nicknames, say &quot;R1&quot;, and it&#39;s OK to send unicast ei=
ther through R1 or R2, but multicast has to go through R1<u></u><u></u></p>

</div>
<div>
<p class=3D"MsoNormal">4) Give E its own nickname (and somehow make sure it=
 knows which port it can multicast on)<u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal">5) Other possibilities?<u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
<div>
<p class=3D"MsoNormal">Radia<u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><u></u>=A0<u></u></p>
<div>
<p class=3D"MsoNormal">On Tue, Jan 8, 2013 at 8:32 PM, Phanidhar Koganti &l=
t;<a href=3D"mailto:pkoganti@brocade.com" target=3D"_blank">pkoganti@brocad=
e.com</a>&gt; wrote:<u></u><u></u></p>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">Thanks Radia. That clarif=
ies for me. A more descriptive text in draft can surely help vendors as
 they implement it in their hardware.</span><u></u><u></u></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">=A0</span><u></u><u></u><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">Another question on handl=
ing multi-homing</span><u></u><u></u></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">E can =
choose either R1 or R2&#39;s nickname, when encapsulating a</span><u></u><u=
></u></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 frame, whether the encapsulated frame is sent via R1 or R2. If=
 E</span><u></u><u></u></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 wants to do active-active load splitting, and uses R1&#39;s ni=
ckname</span><u></u><u></u></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 when forwarding through R1, and R2&#39;s nickname when forward=
ing</span><u></u><u></u></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 through R2, this will cause distant RBridges (or smart endnode=
s)</span><u></u><u></u></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 to keep changing their endnode table entry for D between (D, R=
1&#39;s</span><u></u><u></u></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 nickname) and (D, R2&#39;s nickname). So it would be preferabl=
e for E</span><u></u><u></u></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 to always encapsulate using the same nickname (R1 or R2) unles=
s E</span><u></u><u></u></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 detects a problem with connectivity using that nickname. And i=
n</span><u></u><u></u></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 this case, R1 and R2 need to be informed that the smart endnod=
e</span><u></u><u></u></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 might encapsulate with a different nickname, i.e., R1 might</s=
pan><u></u><u></u></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 receive an encapsulated packet from smart endnode E using ingr=
ess</span><u></u><u></u></p>
<p class=3D"MsoNormal" style=3D"line-height:14.4pt">
<span style=3D"font-size:10.0pt;font-family:&quot;Courier New&quot;">=A0=A0=
=A0=A0=A0=A0 nickname &quot;R2&quot;.</span><u></u><u></u></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">=A0</span><u></u><u></u><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">If the smart endnode uses=
 either R1 or R2 in the above example, would we have TRILL RPF issues where
 upstream RBs see a packet with SRB=3Dsay R1 coming from a wrong TRILL port=
? Let=92s say the RPF happens for multicast only then we will atleast need =
to make sure the multicast (broadcast, unknown unicast and multcast) is sen=
t via the edge-RB whose RBID the smart
 endnode is using.</span><u></u><u></u></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">=A0</span><u></u><u></u><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">This problem should not b=
e there when using different RBID other than R1 and R2 as described in appr=
oach-2</span><u></u><u></u></p>

<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">=A0</span><u></u><u></u><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">Phani</span><u></u><u></u=
></p>
<p class=3D"MsoNormal"><b><span style=3D"font-size:10.0pt;font-family:&quot=
;Tahoma&quot;,&quot;sans-serif&quot;">From:</span></b><span style=3D"font-s=
ize:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;"> Radia Pe=
rlman [mailto:<a href=3D"mailto:radiaperlman@gmail.com" target=3D"_blank">r=
adiaperlman@gmail.com</a>]
<br>
<b>Sent:</b> Wednesday, January 09, 2013 2:53 AM<br>
<b>To:</b> Phanidhar Koganti<br>
<b>Cc:</b> Kesava_Vijaya_Krupak@Dell.com; <a href=3D"mailto:trill@ietf.org"=
 target=3D"_blank">
trill@ietf.org</a>; <a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d=
3e3e3@gmail.com</a>;
<a href=3D"mailto:hu.fangwei@zte.com.cn" target=3D"_blank">hu.fangwei@zte.c=
om.cn</a><br>
<b>Subject:</b> Re: New version : TRILL Smart Endnodes draft</span><u></u><=
u></u></p>
<div>
<div>
<p class=3D"MsoNormal">=A0<u></u><u></u></p>
<p class=3D"MsoNormal">Most of what you said is correct. =A0One correction<=
u></u><u></u></p>
<div>
<p class=3D"MsoNormal">=A0<u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal">a)=A0<span style=3D"color:#1f497d">When the packet r=
eaches the egress RB I am presuming there will be a inner-MAC lookup to del=
iver the packet to the egress end-node.</span><u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal">=A0<u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:#1f497d">correct...when egress =
RB R has to look up the inner MAC address E, which it has to do anyway with=
out smart endnodes, in order to choose which port to
 forward it on. =A0With smart endnodes, when R looks up MAC address E, it n=
ot only discovers which port to send it on, but that E is &quot;smart&quot;=
, in which case R must leave the TRILL encapsulation</span><u></u><u></u></=
p>

</div>
<div>
<p class=3D"MsoNormal">=A0<u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:#1f497d">b)=A0Edge-RB to smart =
endnode cannot have TRILL encapsulation right?</span><u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal">=A0<u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:#1f497d">Not correct. =A0Traffi=
c from egress RB to smart endnode will have TRILL encapsulation. =A0So for =
example, let&#39;s say R is the egress RB, and R has a link
 with two endnodes; E1 (which is smart) and E2 (which is not smart).</span>=
<u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal">=A0<u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:#1f497d">When R receives a TRIL=
L packet from the campus, with TRILL destination nickname=3DR, R looks at t=
he MAC destination. =A0If it is E2, then R removes the TRILL
 encapulation and forwards a native packet onto the link. =A0If it is E1, t=
hen R forwards the packet with the TRILL header.</span><u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal">=A0<u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:#1f497d">c)=A0Edge-RB to smart =
endnode cannot have TRILL encapsulation right? If so for my clarification w=
e are saying the traffic towards the end-node will be
 non-TRILL while towards the ingress edge-RB will be TRILL encapsulated?</s=
pan><u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal">=A0<u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:#1f497d">Hopefully my answer to=
 b) clarified this. =A0On the link between endnodes and R, whether it&#39;s=
 to or from endnode &lt;--&gt; edge RB, the packet will have TRILL
 encapsulation if the endnode is smart, and not have TRILL encapsulation if=
 the endnode is not smart.</span><u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal">=A0<u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:#1f497d">Radia</span><u></u><u>=
</u></p>
</div>
<div>
<p class=3D"MsoNormal">=A0<u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal">=A0<u></u><u></u></p>
<div>
<div>
<p class=3D"MsoNormal">On Thu, Jan 3, 2013 at 3:19 AM, Phanidhar Koganti &l=
t;<a href=3D"mailto:pkoganti@brocade.com" target=3D"_blank">pkoganti@brocad=
e.com</a>&gt; wrote:<u></u><u></u></p>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:#1f497d">The draft defines the =
control and data path behavior from smart end-node to edge-RB. When the pac=
ket reaches the egress RB I am presuming there will
 be a inner-MAC lookup to deliver the packet to the egress end-node. Edge-R=
B to smart endnode cannot have TRILL encapsulation right? If so for my clar=
ification we are saying the traffic towards the end-node will be non-TRILL =
while towards the ingress edge-RB
 will be TRILL encapsulated?</span><u></u><u></u></p>
<p class=3D"MsoNormal"><span style=3D"color:#1f497d">=A0</span><u></u><u></=
u></p>
<p class=3D"MsoNormal"><span style=3D"color:#1f497d">Phani</span><u></u><u>=
</u></p>
<p class=3D"MsoNormal"><span style=3D"color:#1f497d">=A0</span><u></u><u></=
u></p>
<div>
<div style=3D"border:none;border-top:solid #b5c4df 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:10.0pt;font-family:&quot=
;Tahoma&quot;,&quot;sans-serif&quot;">From:</span></b><span style=3D"font-s=
ize:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;">
<a href=3D"mailto:trill-bounces@ietf.org" target=3D"_blank">trill-bounces@i=
etf.org</a> [mailto:<a href=3D"mailto:trill-bounces@ietf.org" target=3D"_bl=
ank">trill-bounces@ietf.org</a>]
<b>On Behalf Of </b><a href=3D"mailto:Kesava_Vijaya_Krupak@Dell.com" target=
=3D"_blank">Kesava_Vijaya_Krupak@Dell.com</a><br>
<b>Sent:</b> Thursday, January 03, 2013 4:19 PM<br>
<b>To:</b> <a href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ietf.o=
rg</a><br>
<b>Cc:</b> <a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gma=
il.com</a>;
<a href=3D"mailto:radiaperlman@gmail.com" target=3D"_blank">radiaperlman@gm=
ail.com</a>;
<a href=3D"mailto:hu.fangwei@zte.com.cn" target=3D"_blank">hu.fangwei@zte.c=
om.cn</a><br>
<b>Subject:</b> [trill] New version : TRILL Smart Endnodes draft</span><u><=
/u><u></u></p>
</div>
</div>
<div>
<div>
<p class=3D"MsoNormal">=A0<u></u><u></u></p>
<p class=3D"MsoNormal">Hi all,<u></u><u></u></p>
<p class=3D"MsoNormal">=A0<u></u><u></u></p>
<p class=3D"MsoNormal">A new version of the Trill Smart Endnodes draft has =
been posted.<u></u><u></u></p>
<p class=3D"MsoNormal">=A0<u></u><u></u></p>
<p class=3D"MsoNormal"><a href=3D"http://datatracker.ietf.org/doc/draft-per=
lman-trill-smart-endnodes" target=3D"_blank">http://datatracker.ietf.org/do=
c/draft-perlman-trill-smart-endnodes</a><u></u><u></u></p>
<p class=3D"MsoNormal">=A0<u></u><u></u></p>
<p class=3D"MsoNormal">Kindly let us know your comments.<u></u><u></u></p>
<p class=3D"MsoNormal">=A0<u></u><u></u></p>
<p class=3D"MsoNormal">Thanks,<u></u><u></u></p>
<p class=3D"MsoNormal">kvk<u></u><u></u></p>
</div>
</div>
</div>
</div>
</div>
<p class=3D"MsoNormal">=A0<u></u><u></u></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
</div>
</div>
</div>

<br>_______________________________________________<br>
trill mailing list<br>
<a href=3D"mailto:trill@ietf.org">trill@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/trill" target=3D"_blank">h=
ttps://www.ietf.org/mailman/listinfo/trill</a><br>
<br></blockquote></div><br></div>

--f46d0401f95b704b3504d2e2a13e--

From linda.dunbar@huawei.com  Thu Jan 10 09:42:51 2013
Return-Path: <linda.dunbar@huawei.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7BB6921F8A27 for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 09:42:51 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.298
X-Spam-Level: 
X-Spam-Status: No, score=-6.298 tagged_above=-999 required=5 tests=[AWL=0.300,  BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id pkS8EF6sEk8S for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 09:42:50 -0800 (PST)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) by ietfa.amsl.com (Postfix) with ESMTP id 126C821F89FC for <trill@ietf.org>; Thu, 10 Jan 2013 09:42:48 -0800 (PST)
Received: from 172.18.7.190 (EHLO lhreml203-edg.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.5-GA FastPath queued) with ESMTP id ANJ25784; Thu, 10 Jan 2013 17:42:47 +0000 (GMT)
Received: from LHREML403-HUB.china.huawei.com (10.201.5.217) by lhreml203-edg.huawei.com (172.18.7.221) with Microsoft SMTP Server (TLS) id 14.1.323.3; Thu, 10 Jan 2013 17:41:31 +0000
Received: from DFWEML404-HUB.china.huawei.com (10.193.5.203) by lhreml403-hub.china.huawei.com (10.201.5.217) with Microsoft SMTP Server (TLS) id 14.1.323.3; Thu, 10 Jan 2013 17:42:45 +0000
Received: from DFWEML505-MBX.china.huawei.com ([10.124.31.100]) by dfweml404-hub.china.huawei.com ([10.193.5.203]) with mapi id 14.01.0323.003; Thu, 10 Jan 2013 09:42:39 -0800
From: Linda Dunbar <linda.dunbar@huawei.com>
To: Radia Perlman <radiaperlman@gmail.com>, Phanidhar Koganti <pkoganti@brocade.com>
Thread-Topic: [trill] New version : TRILL Smart Endnodes draft
Thread-Index: AQHN7eZvBDHzFA5/eke6FKO9FjPC95hC0tRg
Date: Thu, 10 Jan 2013 17:42:38 +0000
Message-ID: <4A95BA014132FF49AE685FAB4B9F17F6450E0647@dfweml505-mbx>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com> <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com>
In-Reply-To: <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.150.48]
Content-Type: multipart/alternative; boundary="_000_4A95BA014132FF49AE685FAB4B9F17F6450E0647dfweml505mbx_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Cc: "d3e3e3@gmail.com" <d3e3e3@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 10 Jan 2013 17:42:51 -0000

--_000_4A95BA014132FF49AE685FAB4B9F17F6450E0647dfweml505mbx_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Radia,

Questions inserted below:


a) When the packet reaches the egress RB I am presuming there will be a inn=
er-MAC lookup to deliver the packet to the egress end-node.

correct...when egress RB R has to look up the inner MAC address E, which it=
 has to do anyway without smart endnodes, in order to choose which port to =
forward it on.  With smart endnodes, when R looks up MAC address E, it not =
only discovers which port to send it on, but that E is "smart", in which ca=
se R must leave the TRILL encapsulation

[Linda] Why can't the egress RB R simply decapsulates the TRILL header and =
forward the data frame towards the attached end nodes? Regardless if the DA=
 is smart node or not?
So RB processing is much simpler. It doesn't have to attach different Ether=
net header based on the inner DA. The smart node can receive Ethernet frame=
s anyway (for communication within the directly attached LAN).

Thanks, Linda


--_000_4A95BA014132FF49AE685FAB4B9F17F6450E0647dfweml505mbx_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micr=
osoft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:x=3D"urn:schemas-microsoft-com:office:excel" xmlns:p=3D"urn:schemas-m=
icrosoft-com:office:powerpoint" xmlns:a=3D"urn:schemas-microsoft-com:office=
:access" xmlns:dt=3D"uuid:C2F41010-65B3-11d1-A29F-00AA00C14882" xmlns:s=3D"=
uuid:BDC6E3F0-6DA3-11d1-A2A3-00AA00C14882" xmlns:rs=3D"urn:schemas-microsof=
t-com:rowset" xmlns:z=3D"#RowsetSchema" xmlns:b=3D"urn:schemas-microsoft-co=
m:office:publisher" xmlns:ss=3D"urn:schemas-microsoft-com:office:spreadshee=
t" xmlns:c=3D"urn:schemas-microsoft-com:office:component:spreadsheet" xmlns=
:odc=3D"urn:schemas-microsoft-com:office:odc" xmlns:oa=3D"urn:schemas-micro=
soft-com:office:activation" xmlns:html=3D"http://www.w3.org/TR/REC-html40" =
xmlns:q=3D"http://schemas.xmlsoap.org/soap/envelope/" xmlns:rtc=3D"http://m=
icrosoft.com/officenet/conferencing" xmlns:D=3D"DAV:" xmlns:Repl=3D"http://=
schemas.microsoft.com/repl/" xmlns:mt=3D"http://schemas.microsoft.com/share=
point/soap/meetings/" xmlns:x2=3D"http://schemas.microsoft.com/office/excel=
/2003/xml" xmlns:ppda=3D"http://www.passport.com/NameSpace.xsd" xmlns:ois=
=3D"http://schemas.microsoft.com/sharepoint/soap/ois/" xmlns:dir=3D"http://=
schemas.microsoft.com/sharepoint/soap/directory/" xmlns:ds=3D"http://www.w3=
.org/2000/09/xmldsig#" xmlns:dsp=3D"http://schemas.microsoft.com/sharepoint=
/dsp" xmlns:udc=3D"http://schemas.microsoft.com/data/udc" xmlns:xsd=3D"http=
://www.w3.org/2001/XMLSchema" xmlns:sub=3D"http://schemas.microsoft.com/sha=
repoint/soap/2002/1/alerts/" xmlns:ec=3D"http://www.w3.org/2001/04/xmlenc#"=
 xmlns:sp=3D"http://schemas.microsoft.com/sharepoint/" xmlns:sps=3D"http://=
schemas.microsoft.com/sharepoint/soap/" xmlns:xsi=3D"http://www.w3.org/2001=
/XMLSchema-instance" xmlns:udcs=3D"http://schemas.microsoft.com/data/udc/so=
ap" xmlns:udcxf=3D"http://schemas.microsoft.com/data/udc/xmlfile" xmlns:udc=
p2p=3D"http://schemas.microsoft.com/data/udc/parttopart" xmlns:wf=3D"http:/=
/schemas.microsoft.com/sharepoint/soap/workflow/" xmlns:dsss=3D"http://sche=
mas.microsoft.com/office/2006/digsig-setup" xmlns:dssi=3D"http://schemas.mi=
crosoft.com/office/2006/digsig" xmlns:mdssi=3D"http://schemas.openxmlformat=
s.org/package/2006/digital-signature" xmlns:mver=3D"http://schemas.openxmlf=
ormats.org/markup-compatibility/2006" xmlns:m=3D"http://schemas.microsoft.c=
om/office/2004/12/omml" xmlns:mrels=3D"http://schemas.openxmlformats.org/pa=
ckage/2006/relationships" xmlns:spwp=3D"http://microsoft.com/sharepoint/web=
partpages" xmlns:ex12t=3D"http://schemas.microsoft.com/exchange/services/20=
06/types" xmlns:ex12m=3D"http://schemas.microsoft.com/exchange/services/200=
6/messages" xmlns:pptsl=3D"http://schemas.microsoft.com/sharepoint/soap/Sli=
deLibrary/" xmlns:spsl=3D"http://microsoft.com/webservices/SharePointPortal=
Server/PublishedLinksService" xmlns:Z=3D"urn:schemas-microsoft-com:" xmlns:=
st=3D"&#1;" xmlns=3D"http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Dus-ascii"=
>
<meta name=3D"Generator" content=3D"Microsoft Word 12 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
	{font-family:SimSun;
	panose-1:2 1 6 0 3 1 1 1 1 1;}
@font-face
	{font-family:"Cambria Math";
	panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:SimSun;
	panose-1:2 1 6 0 3 1 1 1 1 1;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman","serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
span.EmailStyle17
	{mso-style-type:personal-reply;
	font-family:"Calibri","sans-serif";
	color:#1F497D;}
.MsoChpDefault
	{mso-style-type:export-only;}
@page WordSection1
	{size:8.5in 11.0in;
	margin:1.0in 1.25in 1.0in 1.25in;}
div.WordSection1
	{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang=3D"EN-US" link=3D"blue" vlink=3D"purple">
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">Radia,
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">Questions inserted below:=
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<div style=3D"border:none;border-left:solid blue 1.5pt;padding:0in 0in 0in =
4.0pt">
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal">a)&nbsp;<span style=3D"color:#1F497D">When the packe=
t reaches the egress RB I am presuming there will be a inner-MAC lookup to =
deliver the packet to the egress end-node.</span><o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">correct...when egress =
RB R has to look up the inner MAC address E, which it has to do anyway with=
out smart endnodes, in order to choose which port to forward it on. &nbsp;W=
ith smart endnodes, when R looks up MAC address
 E, it not only discovers which port to send it on, but that E is &quot;sma=
rt&quot;, in which case R must leave the TRILL encapsulation</span><o:p></o=
:p></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#943634">[Linda] Why can&#8217;t t=
he egress RB R simply decapsulates the TRILL header and forward the data fr=
ame towards the attached end nodes? Regardless if the DA is smart
 node or not?<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#943634">So RB processing is much =
simpler. It doesn&#8217;t have to attach different Ethernet header based on=
 the inner DA. The smart node can receive Ethernet frames anyway
 (for communication within the directly attached LAN). <o:p></o:p></span></=
p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
</div>
<div>
<div>
<div>
<div>
<div>
<div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"color:#1F497D">Thanks, Linda</span><o:p></o:p></p>
</div>
</div>
</div>
</div>
</div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
</div>
</div>
</div>
</body>
</html>

--_000_4A95BA014132FF49AE685FAB4B9F17F6450E0647dfweml505mbx_--

From radiaperlman@gmail.com  Thu Jan 10 09:54:10 2013
Return-Path: <radiaperlman@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id DFF8E21F87B2 for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 09:54:09 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.341
X-Spam-Level: 
X-Spam-Status: No, score=-3.341 tagged_above=-999 required=5 tests=[AWL=0.257,  BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 8rOiufAAM5YK for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 09:54:09 -0800 (PST)
Received: from mail-la0-f51.google.com (mail-la0-f51.google.com [209.85.215.51]) by ietfa.amsl.com (Postfix) with ESMTP id A3AC521F86CE for <trill@ietf.org>; Thu, 10 Jan 2013 09:54:08 -0800 (PST)
Received: by mail-la0-f51.google.com with SMTP id fj20so909939lab.24 for <trill@ietf.org>; Thu, 10 Jan 2013 09:54:07 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=jZCxTidKm1eFL0unDmKiHWSzxbdok17Ov1cnQnZ8JSA=; b=gueEUYsH6xeXiJNe6SrfUEchNox4e6Q9vbfYqOupbiVFQO4CSSwcH7siMORLO3sfM2 WgoaJIbCuHCZf4DRm4qNL4BCuc2Oc3FumVx3UuvIhYXZr2xuSVwM3g+m4BvHiTbhDu49 4eDey1eWuPMeEqaje429Po1d2ZWpZP6kf48YBR+Ixgd7LI/907B5vOtla4bxkosQNZqq Aqxf6/PomwcmvSD8v5IrJ+646v6PNZBRNq7YKfI49nEDwgQOM9p+P56DlId9GQqEiPHO mK5zM1E/535LPCGkwTCiQmv6kprEFeyrnA5Q7OpoMlaxjUKm1FCULkyKaPnZFIBVr/nc ej2g==
MIME-Version: 1.0
Received: by 10.112.24.226 with SMTP id x2mr29716470lbf.97.1357840447350; Thu, 10 Jan 2013 09:54:07 -0800 (PST)
Received: by 10.112.64.17 with HTTP; Thu, 10 Jan 2013 09:54:07 -0800 (PST)
In-Reply-To: <4A95BA014132FF49AE685FAB4B9F17F6450E0647@dfweml505-mbx>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com> <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com> <4A95BA014132FF49AE685FAB4B9F17F6450E0647@dfweml505-mbx>
Date: Thu, 10 Jan 2013 09:54:07 -0800
Message-ID: <CAFOuuo7B74Y4iptName9fcXzYfarhV79t2_V7zPsDoBncsLFig@mail.gmail.com>
From: Radia Perlman <radiaperlman@gmail.com>
To: Linda Dunbar <linda.dunbar@huawei.com>
Content-Type: multipart/alternative; boundary=e0cb4efe35b486aa5604d2f2dd5a
Cc: "d3e3e3@gmail.com" <d3e3e3@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 10 Jan 2013 17:54:10 -0000

--e0cb4efe35b486aa5604d2f2dd5a
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

The reason RB has to leave it encapsulated when forwarding to a smart
endnode is for the case where the smart endnode will be learning
(destination MAC, egress nickname) from received traffic.

Radia

On Thu, Jan 10, 2013 at 9:42 AM, Linda Dunbar <linda.dunbar@huawei.com>wrot=
e:

>  Radia, ****
>
> ** **
>
> Questions inserted below:****
>
> ** **
>
> ** **
>
> a) When the packet reaches the egress RB I am presuming there will be a
> inner-MAC lookup to deliver the packet to the egress end-node.****
>
> ** **
>
> correct...when egress RB R has to look up the inner MAC address E, which
> it has to do anyway without smart endnodes, in order to choose which port
> to forward it on.  With smart endnodes, when R looks up MAC address E, it
> not only discovers which port to send it on, but that E is "smart", in
> which case R must leave the TRILL encapsulation****
>
> ** **
>
> [Linda] Why can=92t the egress RB R simply decapsulates the TRILL header =
and
> forward the data frame towards the attached end nodes? Regardless if the =
DA
> is smart node or not?****
>
> So RB processing is much simpler. It doesn=92t have to attach different
> Ethernet header based on the inner DA. The smart node can receive Etherne=
t
> frames anyway (for communication within the directly attached LAN). ****
>
> ** **
>
> Thanks, Linda****
>
> ** **
>
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill
>
>

--e0cb4efe35b486aa5604d2f2dd5a
Content-Type: text/html; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

The reason RB has to leave it encapsulated when forwarding to a smart endno=
de is for the case where the smart endnode will be learning (destination MA=
C, egress nickname) from received traffic.<div><br></div><div>Radia<br><br>
<div class=3D"gmail_quote">On Thu, Jan 10, 2013 at 9:42 AM, Linda Dunbar <s=
pan dir=3D"ltr">&lt;<a href=3D"mailto:linda.dunbar@huawei.com" target=3D"_b=
lank">linda.dunbar@huawei.com</a>&gt;</span> wrote:<br><blockquote class=3D=
"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding=
-left:1ex">






<div lang=3D"EN-US" link=3D"blue" vlink=3D"purple">
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">Radia,
<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">Questions inserted below:=
<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
<div style=3D"border:none;border-left:solid blue 1.5pt;padding:0in 0in 0in =
4.0pt"><div class=3D"im">
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
<div>
<p class=3D"MsoNormal">a)=A0<span style=3D"color:#1f497d">When the packet r=
eaches the egress RB I am presuming there will be a inner-MAC lookup to del=
iver the packet to the egress end-node.</span><u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
</div><div><div class=3D"im">
<p class=3D"MsoNormal"><span style=3D"color:#1f497d">correct...when egress =
RB R has to look up the inner MAC address E, which it has to do anyway with=
out smart endnodes, in order to choose which port to forward it on. =A0With=
 smart endnodes, when R looks up MAC address
 E, it not only discovers which port to send it on, but that E is &quot;sma=
rt&quot;, in which case R must leave the TRILL encapsulation</span><u></u><=
u></u></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
</div><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&q=
uot;Calibri&quot;,&quot;sans-serif&quot;;color:#943634">[Linda] Why can=92t=
 the egress RB R simply decapsulates the TRILL header and forward the data =
frame towards the attached end nodes? Regardless if the DA is smart
 node or not?<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#943634">So RB processing is much =
simpler. It doesn=92t have to attach different Ethernet header based on the=
 inner DA. The smart node can receive Ethernet frames anyway
 (for communication within the directly attached LAN). <u></u><u></u></span=
></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
</div>
<div>
<div>
<div>
<div>
<div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:#1f497d">Thanks, Linda</span><u=
></u><u></u></p>
</div>
</div>
</div>
</div>
</div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
</div>
</div>
</div>
</div>

<br>_______________________________________________<br>
trill mailing list<br>
<a href=3D"mailto:trill@ietf.org">trill@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/trill" target=3D"_blank">h=
ttps://www.ietf.org/mailman/listinfo/trill</a><br>
<br></blockquote></div><br></div>

--e0cb4efe35b486aa5604d2f2dd5a--

From linda.dunbar@huawei.com  Thu Jan 10 10:03:28 2013
Return-Path: <linda.dunbar@huawei.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7E64321F886B for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 10:03:28 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.398
X-Spam-Level: 
X-Spam-Status: No, score=-6.398 tagged_above=-999 required=5 tests=[AWL=0.200,  BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id W0dB6GAMJu1y for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 10:03:27 -0800 (PST)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) by ietfa.amsl.com (Postfix) with ESMTP id CB8D621F8930 for <trill@ietf.org>; Thu, 10 Jan 2013 10:03:25 -0800 (PST)
Received: from 172.18.7.190 (EHLO lhreml203-edg.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.5-GA FastPath queued) with ESMTP id ANJ26583; Thu, 10 Jan 2013 18:03:20 +0000 (GMT)
Received: from LHREML403-HUB.china.huawei.com (10.201.5.217) by lhreml203-edg.huawei.com (172.18.7.221) with Microsoft SMTP Server (TLS) id 14.1.323.3; Thu, 10 Jan 2013 18:02:05 +0000
Received: from DFWEML408-HUB.china.huawei.com (10.193.5.134) by lhreml403-hub.china.huawei.com (10.201.5.217) with Microsoft SMTP Server (TLS) id 14.1.323.3; Thu, 10 Jan 2013 18:03:19 +0000
Received: from DFWEML505-MBX.china.huawei.com ([10.124.31.100]) by dfweml408-hub.china.huawei.com ([10.193.5.134]) with mapi id 14.01.0323.003; Thu, 10 Jan 2013 10:03:08 -0800
From: Linda Dunbar <linda.dunbar@huawei.com>
To: Linda Dunbar <linda.dunbar@huawei.com>, Radia Perlman <radiaperlman@gmail.com>, Phanidhar Koganti <pkoganti@brocade.com>
Thread-Topic: [trill] New version : TRILL Smart Endnodes draft
Thread-Index: AQHN7eZvBDHzFA5/eke6FKO9FjPC95hC0tRggAAJ9mA=
Date: Thu, 10 Jan 2013 18:03:07 +0000
Message-ID: <4A95BA014132FF49AE685FAB4B9F17F6450E06AB@dfweml505-mbx>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com> <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com> 
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.192.11.236]
Content-Type: multipart/alternative; boundary="_000_4A95BA014132FF49AE685FAB4B9F17F6450E06ABdfweml505mbx_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Cc: "d3e3e3@gmail.com" <d3e3e3@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 10 Jan 2013 18:03:28 -0000

--_000_4A95BA014132FF49AE685FAB4B9F17F6450E06ABdfweml505mbx_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

It is definitely simpler for Egress RBridge to decapsulate all received dat=
a frames:

*        So it doesn't need determine if the target end node is smart  or n=
ot

*        It doesn't have to attached different Ethernet Destination based o=
n the inner header

*        It is consistent for all data frames and all end nodes.

*        Smart End node can receive native Ethernet frames anyway. It doesn=
't cause any extra processing on the Smart End node either.

Linda

From: Linda Dunbar
Sent: Thursday, January 10, 2013 11:43 AM
To: 'Radia Perlman'; Phanidhar Koganti
Cc: d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com; hu.fangwei@zte.com.cn;=
 trill@ietf.org
Subject: RE: [trill] New version : TRILL Smart Endnodes draft

Radia,

Questions inserted below:


a) When the packet reaches the egress RB I am presuming there will be a inn=
er-MAC lookup to deliver the packet to the egress end-node.

correct...when egress RB R has to look up the inner MAC address E, which it=
 has to do anyway without smart endnodes, in order to choose which port to =
forward it on.  With smart endnodes, when R looks up MAC address E, it not =
only discovers which port to send it on, but that E is "smart", in which ca=
se R must leave the TRILL encapsulation

[Linda] Why can't the egress RB R simply decapsulates the TRILL header and =
forward the data frame towards the attached end nodes? Regardless if the DA=
 is smart node or not?
So RB processing is much simpler. It doesn't have to attach different Ether=
net header based on the inner DA. The smart node can receive Ethernet frame=
s anyway (for communication within the directly attached LAN).

Thanks, Linda


--_000_4A95BA014132FF49AE685FAB4B9F17F6450E06ABdfweml505mbx_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micr=
osoft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:x=3D"urn:schemas-microsoft-com:office:excel" xmlns:p=3D"urn:schemas-m=
icrosoft-com:office:powerpoint" xmlns:a=3D"urn:schemas-microsoft-com:office=
:access" xmlns:dt=3D"uuid:C2F41010-65B3-11d1-A29F-00AA00C14882" xmlns:s=3D"=
uuid:BDC6E3F0-6DA3-11d1-A2A3-00AA00C14882" xmlns:rs=3D"urn:schemas-microsof=
t-com:rowset" xmlns:z=3D"#RowsetSchema" xmlns:b=3D"urn:schemas-microsoft-co=
m:office:publisher" xmlns:ss=3D"urn:schemas-microsoft-com:office:spreadshee=
t" xmlns:c=3D"urn:schemas-microsoft-com:office:component:spreadsheet" xmlns=
:odc=3D"urn:schemas-microsoft-com:office:odc" xmlns:oa=3D"urn:schemas-micro=
soft-com:office:activation" xmlns:html=3D"http://www.w3.org/TR/REC-html40" =
xmlns:q=3D"http://schemas.xmlsoap.org/soap/envelope/" xmlns:rtc=3D"http://m=
icrosoft.com/officenet/conferencing" xmlns:D=3D"DAV:" xmlns:Repl=3D"http://=
schemas.microsoft.com/repl/" xmlns:mt=3D"http://schemas.microsoft.com/share=
point/soap/meetings/" xmlns:x2=3D"http://schemas.microsoft.com/office/excel=
/2003/xml" xmlns:ppda=3D"http://www.passport.com/NameSpace.xsd" xmlns:ois=
=3D"http://schemas.microsoft.com/sharepoint/soap/ois/" xmlns:dir=3D"http://=
schemas.microsoft.com/sharepoint/soap/directory/" xmlns:ds=3D"http://www.w3=
.org/2000/09/xmldsig#" xmlns:dsp=3D"http://schemas.microsoft.com/sharepoint=
/dsp" xmlns:udc=3D"http://schemas.microsoft.com/data/udc" xmlns:xsd=3D"http=
://www.w3.org/2001/XMLSchema" xmlns:sub=3D"http://schemas.microsoft.com/sha=
repoint/soap/2002/1/alerts/" xmlns:ec=3D"http://www.w3.org/2001/04/xmlenc#"=
 xmlns:sp=3D"http://schemas.microsoft.com/sharepoint/" xmlns:sps=3D"http://=
schemas.microsoft.com/sharepoint/soap/" xmlns:xsi=3D"http://www.w3.org/2001=
/XMLSchema-instance" xmlns:udcs=3D"http://schemas.microsoft.com/data/udc/so=
ap" xmlns:udcxf=3D"http://schemas.microsoft.com/data/udc/xmlfile" xmlns:udc=
p2p=3D"http://schemas.microsoft.com/data/udc/parttopart" xmlns:wf=3D"http:/=
/schemas.microsoft.com/sharepoint/soap/workflow/" xmlns:dsss=3D"http://sche=
mas.microsoft.com/office/2006/digsig-setup" xmlns:dssi=3D"http://schemas.mi=
crosoft.com/office/2006/digsig" xmlns:mdssi=3D"http://schemas.openxmlformat=
s.org/package/2006/digital-signature" xmlns:mver=3D"http://schemas.openxmlf=
ormats.org/markup-compatibility/2006" xmlns:m=3D"http://schemas.microsoft.c=
om/office/2004/12/omml" xmlns:mrels=3D"http://schemas.openxmlformats.org/pa=
ckage/2006/relationships" xmlns:spwp=3D"http://microsoft.com/sharepoint/web=
partpages" xmlns:ex12t=3D"http://schemas.microsoft.com/exchange/services/20=
06/types" xmlns:ex12m=3D"http://schemas.microsoft.com/exchange/services/200=
6/messages" xmlns:pptsl=3D"http://schemas.microsoft.com/sharepoint/soap/Sli=
deLibrary/" xmlns:spsl=3D"http://microsoft.com/webservices/SharePointPortal=
Server/PublishedLinksService" xmlns:Z=3D"urn:schemas-microsoft-com:" xmlns:=
st=3D"&#1;" xmlns=3D"http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Dus-ascii"=
>
<meta name=3D"Generator" content=3D"Microsoft Word 12 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
	{font-family:Wingdings;
	panose-1:5 0 0 0 0 0 0 0 0 0;}
@font-face
	{font-family:SimSun;
	panose-1:2 1 6 0 3 1 1 1 1 1;}
@font-face
	{font-family:"Cambria Math";
	panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:Tahoma;
	panose-1:2 11 6 4 3 5 4 4 2 4;}
@font-face
	{font-family:SimSun;
	panose-1:2 1 6 0 3 1 1 1 1 1;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman","serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
p.MsoListParagraph, li.MsoListParagraph, div.MsoListParagraph
	{mso-style-priority:34;
	margin-top:0in;
	margin-right:0in;
	margin-bottom:0in;
	margin-left:.5in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman","serif";}
span.EmailStyle17
	{mso-style-type:personal;
	font-family:"Calibri","sans-serif";
	color:#1F497D;}
span.EmailStyle18
	{mso-style-type:personal-reply;
	font-family:"Calibri","sans-serif";
	color:#1F497D;}
.MsoChpDefault
	{mso-style-type:export-only;
	font-size:10.0pt;}
@page WordSection1
	{size:8.5in 11.0in;
	margin:1.0in 1.25in 1.0in 1.25in;}
div.WordSection1
	{page:WordSection1;}
/* List Definitions */
@list l0
	{mso-list-id:103697583;
	mso-list-type:hybrid;
	mso-list-template-ids:1061299432 67698689 67698691 67698693 67698689 67698=
691 67698693 67698689 67698691 67698693;}
@list l0:level1
	{mso-level-number-format:bullet;
	mso-level-text:\F0B7;
	mso-level-tab-stop:none;
	mso-level-number-position:left;
	text-indent:-.25in;
	font-family:Symbol;}
ol
	{margin-bottom:0in;}
ul
	{margin-bottom:0in;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang=3D"EN-US" link=3D"blue" vlink=3D"purple">
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">It is definitely simpler =
for Egress RBridge to decapsulate all received data frames:<o:p></o:p></spa=
n></p>
<p class=3D"MsoListParagraph" style=3D"text-indent:-.25in;mso-list:l0 level=
1 lfo1"><![if !supportLists]><span style=3D"font-size:11.0pt;font-family:Sy=
mbol;color:#1F497D"><span style=3D"mso-list:Ignore">&middot;<span style=3D"=
font:7.0pt &quot;Times New Roman&quot;">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;
</span></span></span><![endif]><span style=3D"font-size:11.0pt;font-family:=
&quot;Calibri&quot;,&quot;sans-serif&quot;;color:#1F497D">So it doesn&#8217=
;t need determine if the target end node is smart &nbsp;or not<o:p></o:p></=
span></p>
<p class=3D"MsoListParagraph" style=3D"text-indent:-.25in;mso-list:l0 level=
1 lfo1"><![if !supportLists]><span style=3D"font-size:11.0pt;font-family:Sy=
mbol;color:#1F497D"><span style=3D"mso-list:Ignore">&middot;<span style=3D"=
font:7.0pt &quot;Times New Roman&quot;">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;
</span></span></span><![endif]><span style=3D"font-size:11.0pt;font-family:=
&quot;Calibri&quot;,&quot;sans-serif&quot;;color:#1F497D">It doesn&#8217;t =
have to attached different Ethernet Destination based on the inner header<o=
:p></o:p></span></p>
<p class=3D"MsoListParagraph" style=3D"text-indent:-.25in;mso-list:l0 level=
1 lfo1"><![if !supportLists]><span style=3D"font-size:11.0pt;font-family:Sy=
mbol;color:#1F497D"><span style=3D"mso-list:Ignore">&middot;<span style=3D"=
font:7.0pt &quot;Times New Roman&quot;">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;
</span></span></span><![endif]><span style=3D"font-size:11.0pt;font-family:=
&quot;Calibri&quot;,&quot;sans-serif&quot;;color:#1F497D">It is consistent =
for all data frames and all end nodes.
<o:p></o:p></span></p>
<p class=3D"MsoListParagraph" style=3D"text-indent:-.25in;mso-list:l0 level=
1 lfo1"><![if !supportLists]><span style=3D"font-size:11.0pt;font-family:Sy=
mbol;color:#1F497D"><span style=3D"mso-list:Ignore">&middot;<span style=3D"=
font:7.0pt &quot;Times New Roman&quot;">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;
</span></span></span><![endif]><span style=3D"font-size:11.0pt;font-family:=
&quot;Calibri&quot;,&quot;sans-serif&quot;;color:#1F497D">Smart End node ca=
n receive native Ethernet frames anyway. It doesn&#8217;t cause any extra p=
rocessing on the Smart End node either.
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">Linda<o:p></o:p></span></=
p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<div style=3D"border:none;border-left:solid blue 1.5pt;padding:0in 0in 0in =
4.0pt">
<div>
<div style=3D"border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:10.0pt;font-family:&quot=
;Tahoma&quot;,&quot;sans-serif&quot;">From:</span></b><span style=3D"font-s=
ize:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;"> Linda Du=
nbar
<br>
<b>Sent:</b> Thursday, January 10, 2013 11:43 AM<br>
<b>To:</b> 'Radia Perlman'; Phanidhar Koganti<br>
<b>Cc:</b> d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com; hu.fangwei@zte.=
com.cn; trill@ietf.org<br>
<b>Subject:</b> RE: [trill] New version : TRILL Smart Endnodes draft<o:p></=
o:p></span></p>
</div>
</div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">Radia,
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">Questions inserted below:=
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<div style=3D"border:none;border-left:solid blue 1.5pt;padding:0in 0in 0in =
4.0pt">
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal">a)&nbsp;<span style=3D"color:#1F497D">When the packe=
t reaches the egress RB I am presuming there will be a inner-MAC lookup to =
deliver the packet to the egress end-node.</span><o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">correct...when egress =
RB R has to look up the inner MAC address E, which it has to do anyway with=
out smart endnodes, in order to choose which port to forward it on. &nbsp;W=
ith smart endnodes, when R looks up MAC address
 E, it not only discovers which port to send it on, but that E is &quot;sma=
rt&quot;, in which case R must leave the TRILL encapsulation</span><o:p></o=
:p></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#943634">[Linda] Why can&#8217;t t=
he egress RB R simply decapsulates the TRILL header and forward the data fr=
ame towards the attached end nodes? Regardless if the DA is smart
 node or not?<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#943634">So RB processing is much =
simpler. It doesn&#8217;t have to attach different Ethernet header based on=
 the inner DA. The smart node can receive Ethernet frames anyway
 (for communication within the directly attached LAN). <o:p></o:p></span></=
p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
</div>
<div>
<div>
<div>
<div>
<div>
<div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"color:#1F497D">Thanks, Linda</span><o:p></o:p></p>
</div>
</div>
</div>
</div>
</div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
</div>
</div>
</div>
</div>
</body>
</html>

--_000_4A95BA014132FF49AE685FAB4B9F17F6450E06ABdfweml505mbx_--

From linda.dunbar@huawei.com  Thu Jan 10 10:18:53 2013
Return-Path: <linda.dunbar@huawei.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C168721F8A11 for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 10:18:53 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.448
X-Spam-Level: 
X-Spam-Status: No, score=-6.448 tagged_above=-999 required=5 tests=[AWL=0.150,  BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id mgDvJwj+ypRF for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 10:18:51 -0800 (PST)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) by ietfa.amsl.com (Postfix) with ESMTP id AFE9221F8A3E for <trill@ietf.org>; Thu, 10 Jan 2013 10:18:49 -0800 (PST)
Received: from 172.18.7.190 (EHLO lhreml204-edg.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.5-GA FastPath queued) with ESMTP id ANJ27181; Thu, 10 Jan 2013 18:18:47 +0000 (GMT)
Received: from LHREML405-HUB.china.huawei.com (10.201.5.242) by lhreml204-edg.china.huawei.com (172.18.7.223) with Microsoft SMTP Server (TLS) id 14.1.323.3; Thu, 10 Jan 2013 18:17:46 +0000
Received: from DFWEML405-HUB.china.huawei.com (10.193.5.102) by lhreml405-hub.china.huawei.com (10.201.5.242) with Microsoft SMTP Server (TLS) id 14.1.323.3; Thu, 10 Jan 2013 18:18:47 +0000
Received: from DFWEML505-MBX.china.huawei.com ([10.124.31.100]) by dfweml405-hub.china.huawei.com ([10.193.5.102]) with mapi id 14.01.0323.003; Thu, 10 Jan 2013 10:18:42 -0800
From: Linda Dunbar <linda.dunbar@huawei.com>
To: Radia Perlman <radiaperlman@gmail.com>
Thread-Topic: [trill] New version : TRILL Smart Endnodes draft
Thread-Index: AQHN7eZvBDHzFA5/eke6FKO9FjPC95hC0tRggACOc4D//3yQEA==
Date: Thu, 10 Jan 2013 18:18:41 +0000
Message-ID: <4A95BA014132FF49AE685FAB4B9F17F6450E06CD@dfweml505-mbx>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com> <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com> <4A95BA014132FF49AE685FAB4B9F17F6450E0647@dfweml505-mbx> <CAFOuuo7B74Y4iptName9fcXzYfarhV79t2_V7zPsDoBncsLFig@mail.gmail.com>
In-Reply-To: <CAFOuuo7B74Y4iptName9fcXzYfarhV79t2_V7zPsDoBncsLFig@mail.gmail.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.192.11.236]
Content-Type: multipart/alternative; boundary="_000_4A95BA014132FF49AE685FAB4B9F17F6450E06CDdfweml505mbx_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Cc: "d3e3e3@gmail.com" <d3e3e3@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 10 Jan 2013 18:18:53 -0000

--_000_4A95BA014132FF49AE685FAB4B9F17F6450E06CDdfweml505mbx_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Radia,

It would be much simpler to let the egress RBridge send a freshly learned m=
apping to the Smart node when decapsulating a TRILL frame. By doing so, the=
 (egress) RBridge doesn't have make any change to its current learning proc=
essing
If the RBridge can't keep up all the learned mapping due to massive number =
of end nodes, the egress RBridge doesn't  have to cache the mapping (or set=
 the timer for the cache very small).

You can use the Hello message between RBridge and smart node to carry this =
mapping information.

Alternatively, let the Smart Node be assisted by Directory Servers, which m=
akes lot of sense in Data Center environment.

Linda

From: Radia Perlman [mailto:radiaperlman@gmail.com]
Sent: Thursday, January 10, 2013 11:54 AM
To: Linda Dunbar
Cc: Phanidhar Koganti; d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com; hu.=
fangwei@zte.com.cn; trill@ietf.org
Subject: Re: [trill] New version : TRILL Smart Endnodes draft

The reason RB has to leave it encapsulated when forwarding to a smart endno=
de is for the case where the smart endnode will be learning (destination MA=
C, egress nickname) from received traffic.

Radia
On Thu, Jan 10, 2013 at 9:42 AM, Linda Dunbar <linda.dunbar@huawei.com<mail=
to:linda.dunbar@huawei.com>> wrote:
Radia,

Questions inserted below:


a) When the packet reaches the egress RB I am presuming there will be a inn=
er-MAC lookup to deliver the packet to the egress end-node.

correct...when egress RB R has to look up the inner MAC address E, which it=
 has to do anyway without smart endnodes, in order to choose which port to =
forward it on.  With smart endnodes, when R looks up MAC address E, it not =
only discovers which port to send it on, but that E is "smart", in which ca=
se R must leave the TRILL encapsulation

[Linda] Why can't the egress RB R simply decapsulates the TRILL header and =
forward the data frame towards the attached end nodes? Regardless if the DA=
 is smart node or not?
So RB processing is much simpler. It doesn't have to attach different Ether=
net header based on the inner DA. The smart node can receive Ethernet frame=
s anyway (for communication within the directly attached LAN).

Thanks, Linda


_______________________________________________
trill mailing list
trill@ietf.org<mailto:trill@ietf.org>
https://www.ietf.org/mailman/listinfo/trill


--_000_4A95BA014132FF49AE685FAB4B9F17F6450E06CDdfweml505mbx_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micr=
osoft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:x=3D"urn:schemas-microsoft-com:office:excel" xmlns:p=3D"urn:schemas-m=
icrosoft-com:office:powerpoint" xmlns:a=3D"urn:schemas-microsoft-com:office=
:access" xmlns:dt=3D"uuid:C2F41010-65B3-11d1-A29F-00AA00C14882" xmlns:s=3D"=
uuid:BDC6E3F0-6DA3-11d1-A2A3-00AA00C14882" xmlns:rs=3D"urn:schemas-microsof=
t-com:rowset" xmlns:z=3D"#RowsetSchema" xmlns:b=3D"urn:schemas-microsoft-co=
m:office:publisher" xmlns:ss=3D"urn:schemas-microsoft-com:office:spreadshee=
t" xmlns:c=3D"urn:schemas-microsoft-com:office:component:spreadsheet" xmlns=
:odc=3D"urn:schemas-microsoft-com:office:odc" xmlns:oa=3D"urn:schemas-micro=
soft-com:office:activation" xmlns:html=3D"http://www.w3.org/TR/REC-html40" =
xmlns:q=3D"http://schemas.xmlsoap.org/soap/envelope/" xmlns:rtc=3D"http://m=
icrosoft.com/officenet/conferencing" xmlns:D=3D"DAV:" xmlns:Repl=3D"http://=
schemas.microsoft.com/repl/" xmlns:mt=3D"http://schemas.microsoft.com/share=
point/soap/meetings/" xmlns:x2=3D"http://schemas.microsoft.com/office/excel=
/2003/xml" xmlns:ppda=3D"http://www.passport.com/NameSpace.xsd" xmlns:ois=
=3D"http://schemas.microsoft.com/sharepoint/soap/ois/" xmlns:dir=3D"http://=
schemas.microsoft.com/sharepoint/soap/directory/" xmlns:ds=3D"http://www.w3=
.org/2000/09/xmldsig#" xmlns:dsp=3D"http://schemas.microsoft.com/sharepoint=
/dsp" xmlns:udc=3D"http://schemas.microsoft.com/data/udc" xmlns:xsd=3D"http=
://www.w3.org/2001/XMLSchema" xmlns:sub=3D"http://schemas.microsoft.com/sha=
repoint/soap/2002/1/alerts/" xmlns:ec=3D"http://www.w3.org/2001/04/xmlenc#"=
 xmlns:sp=3D"http://schemas.microsoft.com/sharepoint/" xmlns:sps=3D"http://=
schemas.microsoft.com/sharepoint/soap/" xmlns:xsi=3D"http://www.w3.org/2001=
/XMLSchema-instance" xmlns:udcs=3D"http://schemas.microsoft.com/data/udc/so=
ap" xmlns:udcxf=3D"http://schemas.microsoft.com/data/udc/xmlfile" xmlns:udc=
p2p=3D"http://schemas.microsoft.com/data/udc/parttopart" xmlns:wf=3D"http:/=
/schemas.microsoft.com/sharepoint/soap/workflow/" xmlns:dsss=3D"http://sche=
mas.microsoft.com/office/2006/digsig-setup" xmlns:dssi=3D"http://schemas.mi=
crosoft.com/office/2006/digsig" xmlns:mdssi=3D"http://schemas.openxmlformat=
s.org/package/2006/digital-signature" xmlns:mver=3D"http://schemas.openxmlf=
ormats.org/markup-compatibility/2006" xmlns:m=3D"http://schemas.microsoft.c=
om/office/2004/12/omml" xmlns:mrels=3D"http://schemas.openxmlformats.org/pa=
ckage/2006/relationships" xmlns:spwp=3D"http://microsoft.com/sharepoint/web=
partpages" xmlns:ex12t=3D"http://schemas.microsoft.com/exchange/services/20=
06/types" xmlns:ex12m=3D"http://schemas.microsoft.com/exchange/services/200=
6/messages" xmlns:pptsl=3D"http://schemas.microsoft.com/sharepoint/soap/Sli=
deLibrary/" xmlns:spsl=3D"http://microsoft.com/webservices/SharePointPortal=
Server/PublishedLinksService" xmlns:Z=3D"urn:schemas-microsoft-com:" xmlns:=
st=3D"&#1;" xmlns=3D"http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Dus-ascii"=
>
<meta name=3D"Generator" content=3D"Microsoft Word 12 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
	{font-family:Wingdings;
	panose-1:5 0 0 0 0 0 0 0 0 0;}
@font-face
	{font-family:SimSun;
	panose-1:2 1 6 0 3 1 1 1 1 1;}
@font-face
	{font-family:"Cambria Math";
	panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:Tahoma;
	panose-1:2 11 6 4 3 5 4 4 2 4;}
@font-face
	{font-family:SimSun;
	panose-1:2 1 6 0 3 1 1 1 1 1;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman","serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
p.MsoListParagraph, li.MsoListParagraph, div.MsoListParagraph
	{mso-style-priority:34;
	margin-top:0in;
	margin-right:0in;
	margin-bottom:0in;
	margin-left:.5in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman","serif";}
span.EmailStyle17
	{mso-style-type:personal-reply;
	font-family:"Calibri","sans-serif";
	color:#1F497D;}
.MsoChpDefault
	{mso-style-type:export-only;}
@page WordSection1
	{size:8.5in 11.0in;
	margin:1.0in 1.25in 1.0in 1.25in;}
div.WordSection1
	{page:WordSection1;}
/* List Definitions */
@list l0
	{mso-list-id:201287769;
	mso-list-type:hybrid;
	mso-list-template-ids:-2057913154 2085796794 67698691 67698693 67698689 67=
698691 67698693 67698689 67698691 67698693;}
@list l0:level1
	{mso-level-start-at:0;
	mso-level-number-format:bullet;
	mso-level-text:-;
	mso-level-tab-stop:none;
	mso-level-number-position:left;
	text-indent:-.25in;
	font-family:"Calibri","sans-serif";
	mso-fareast-font-family:SimSun;
	mso-bidi-font-family:"Times New Roman";}
ol
	{margin-bottom:0in;}
ul
	{margin-bottom:0in;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang=3D"EN-US" link=3D"blue" vlink=3D"purple">
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">Radia,
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">It would be much simpler =
to let the egress RBridge send a freshly learned mapping to the Smart node =
when decapsulating a TRILL frame. By doing so, the (egress)
 RBridge doesn&#8217;t have make any change to its current learning process=
ing <o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">If the RBridge can&#8217;=
t keep up all the learned mapping due to massive number of end nodes, the e=
gress RBridge doesn&#8217;t &nbsp;have to cache the mapping (or set the tim=
er
 for the cache very small). <o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">You can use the Hello mes=
sage between RBridge and smart node to carry this mapping information.
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">Alternatively, let the Sm=
art Node be assisted by Directory Servers, which makes lot of sense in Data=
 Center environment.
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">Linda<o:p></o:p></span></=
p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<div style=3D"border:none;border-left:solid blue 1.5pt;padding:0in 0in 0in =
4.0pt">
<div>
<div style=3D"border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:10.0pt;font-family:&quot=
;Tahoma&quot;,&quot;sans-serif&quot;">From:</span></b><span style=3D"font-s=
ize:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;"> Radia Pe=
rlman [mailto:radiaperlman@gmail.com]
<br>
<b>Sent:</b> Thursday, January 10, 2013 11:54 AM<br>
<b>To:</b> Linda Dunbar<br>
<b>Cc:</b> Phanidhar Koganti; d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.c=
om; hu.fangwei@zte.com.cn; trill@ietf.org<br>
<b>Subject:</b> Re: [trill] New version : TRILL Smart Endnodes draft<o:p></=
o:p></span></p>
</div>
</div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">The reason RB has to leave it encapsulated when forw=
arding to a smart endnode is for the case where the smart endnode will be l=
earning (destination MAC, egress nickname) from received traffic.<o:p></o:p=
></p>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt">Radia<o:p></o:p></p>
<div>
<p class=3D"MsoNormal">On Thu, Jan 10, 2013 at 9:42 AM, Linda Dunbar &lt;<a=
 href=3D"mailto:linda.dunbar@huawei.com" target=3D"_blank">linda.dunbar@hua=
wei.com</a>&gt; wrote:<o:p></o:p></p>
<div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">Radia,
</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">&nbsp;</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">Questions inserted below:</span><o:p></=
o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">&nbsp;</span><o:p></o:p></p>
<div style=3D"border:none;border-left:solid blue 1.5pt;padding:0in 0in 0in =
4.0pt">
<div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">a)&nbsp;<span style=3D"color:#1F497D">When the packet reaches the =
egress RB I am presuming there will be a inner-MAC lookup to deliver the pa=
cket to the egress end-node.</span><o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"color:#1F497D">correct...when egress RB R has to lo=
ok up the inner MAC address E, which it has to do anyway without smart endn=
odes, in order to choose which port to
 forward it on. &nbsp;With smart endnodes, when R looks up MAC address E, i=
t not only discovers which port to send it on, but that E is &quot;smart&qu=
ot;, in which case R must leave the TRILL encapsulation</span><o:p></o:p></=
p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">&nbsp;</span><o:p></o:p></p>
</div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#943634">[Linda] Why can&#8217;t the egress RB R=
 simply decapsulates the TRILL header and forward the data frame
 towards the attached end nodes? Regardless if the DA is smart node or not?=
</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#943634">So RB processing is much simpler. It do=
esn&#8217;t have to attach different Ethernet header based on the
 inner DA. The smart node can receive Ethernet frames anyway (for communica=
tion within the directly attached LAN).
</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">&nbsp;</span><o:p></o:p></p>
</div>
<div>
<div>
<div>
<div>
<div>
<div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"color:#1F497D">Thanks, Linda</span><o:p></o:p></p>
</div>
</div>
</div>
</div>
</div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
</div>
</div>
</div>
</div>
</div>
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><br>
_______________________________________________<br>
trill mailing list<br>
<a href=3D"mailto:trill@ietf.org">trill@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/trill" target=3D"_blank">h=
ttps://www.ietf.org/mailman/listinfo/trill</a><o:p></o:p></p>
</div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
</div>
</div>
</body>
</html>

--_000_4A95BA014132FF49AE685FAB4B9F17F6450E06CDdfweml505mbx_--

From radiaperlman@gmail.com  Thu Jan 10 10:24:07 2013
Return-Path: <radiaperlman@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CC2EA21F8A4A for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 10:24:05 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.361
X-Spam-Level: 
X-Spam-Status: No, score=-3.361 tagged_above=-999 required=5 tests=[AWL=0.237,  BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id geUHGOPw7o28 for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 10:24:04 -0800 (PST)
Received: from mail-la0-f41.google.com (mail-la0-f41.google.com [209.85.215.41]) by ietfa.amsl.com (Postfix) with ESMTP id DF29021F886B for <trill@ietf.org>; Thu, 10 Jan 2013 10:24:03 -0800 (PST)
Received: by mail-la0-f41.google.com with SMTP id em20so970149lab.0 for <trill@ietf.org>; Thu, 10 Jan 2013 10:24:02 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=20OcfyK1NnU+q9VN74SpDfCaUMO08Kw6idYLuQ0fkDI=; b=EVPQp09MMkLvsum0dDDNLq/xdorTUdalaPMD5a+WFVDRl1DW8oih3tL0rDrRHZmzcO rApMm+mQwB+5syyCdDe+QAEadc3fjHH2gfXd+uYiEtjN2vxGlQJJetCOszo/r2ucWcpu lfBGJ8lu6NF/N/TC9oYzJ5Zjd3DUJVXUt0LEX9gha2QVM0z5EZImDmZYqaDEvAXmCitP 1S4HAuUBggI6S3XQWt6nCaqYYRJ5620r3GBAxwaQACQO6O6sIwhTE0gLXMa9pRqjMdvt XhRZfyUGyOrMzvCatmNetyyWw152J+bJ0ddXD6IK6Pu4y6DZaotmS9T6VzIexIHCdl3h V1CA==
MIME-Version: 1.0
Received: by 10.152.148.164 with SMTP id tt4mr29689020lab.28.1357842242617; Thu, 10 Jan 2013 10:24:02 -0800 (PST)
Received: by 10.112.64.17 with HTTP; Thu, 10 Jan 2013 10:24:02 -0800 (PST)
In-Reply-To: <4A95BA014132FF49AE685FAB4B9F17F6450E06AB@dfweml505-mbx>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com> <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com> <4A95BA014132FF49AE685FAB4B9F17F6450E06AB@dfweml505-mbx>
Date: Thu, 10 Jan 2013 10:24:02 -0800
Message-ID: <CAFOuuo5z=LepKF1FSD0PSS0zGCC7eBdBEX4eQ+WQXaOa1+ZArA@mail.gmail.com>
From: Radia Perlman <radiaperlman@gmail.com>
To: Linda Dunbar <linda.dunbar@huawei.com>
Content-Type: multipart/alternative; boundary=e89a8f22c42588409204d2f3485e
Cc: "d3e3e3@gmail.com" <d3e3e3@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 10 Jan 2013 18:24:07 -0000

--e89a8f22c42588409204d2f3485e
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

Yes, it's simpler, but it doesn't work, unless we want to say that smart
endnodes can never learn from data traffic...they can only work if they are
configured with (MAC, egress nickname) of all destinations they might want
to talk to, or must have a directory and request from the directory.

I think we need to allow smart endnodes to learn from data traffic, in
which, unfortunately (because as you point out, it's more complicated) the
RB has to consciously know whether it's forwarding to a smart endnode or
not, and then leave it encapsulated in that case.

Radia

On Thu, Jan 10, 2013 at 10:03 AM, Linda Dunbar <linda.dunbar@huawei.com>wro=
te:

>  It is definitely simpler for Egress RBridge to decapsulate all received
> data frames:****
>
> **=B7        **So it doesn=92t need determine if the target end node is s=
mart
>  or not****
>
> **=B7        **It doesn=92t have to attached different Ethernet Destinati=
on
> based on the inner header****
>
> **=B7        **It is consistent for all data frames and all end nodes. **=
**
>
> **=B7        **Smart End node can receive native Ethernet frames anyway. =
It
> doesn=92t cause any extra processing on the Smart End node either. ****
>
> ** **
>
> Linda****
>
> ** **
>
> *From:* Linda Dunbar
> *Sent:* Thursday, January 10, 2013 11:43 AM
> *To:* 'Radia Perlman'; Phanidhar Koganti
> *Cc:* d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com;
> hu.fangwei@zte.com.cn; trill@ietf.org
> *Subject:* RE: [trill] New version : TRILL Smart Endnodes draft****
>
> ** **
>
> Radia, ****
>
> ** **
>
> Questions inserted below:****
>
> ** **
>
> ** **
>
> a) When the packet reaches the egress RB I am presuming there will be a
> inner-MAC lookup to deliver the packet to the egress end-node.****
>
> ** **
>
> correct...when egress RB R has to look up the inner MAC address E, which
> it has to do anyway without smart endnodes, in order to choose which port
> to forward it on.  With smart endnodes, when R looks up MAC address E, it
> not only discovers which port to send it on, but that E is "smart", in
> which case R must leave the TRILL encapsulation****
>
> ** **
>
> [Linda] Why can=92t the egress RB R simply decapsulates the TRILL header =
and
> forward the data frame towards the attached end nodes? Regardless if the =
DA
> is smart node or not?****
>
> So RB processing is much simpler. It doesn=92t have to attach different
> Ethernet header based on the inner DA. The smart node can receive Etherne=
t
> frames anyway (for communication within the directly attached LAN). ****
>
> ** **
>
> Thanks, Linda****
>
> ** **
>
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill
>
>

--e89a8f22c42588409204d2f3485e
Content-Type: text/html; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

Yes, it&#39;s simpler, but it doesn&#39;t work, unless we want to say that =
smart endnodes can never learn from data traffic...they can only work if th=
ey are configured with (MAC, egress nickname) of all destinations they migh=
t want to talk to, or must have a directory and request from the directory.=
<div>
<br></div><div>I think we need to allow smart endnodes to learn from data t=
raffic, in which, unfortunately (because as you point out, it&#39;s more co=
mplicated) the RB has to consciously know whether it&#39;s forwarding to a =
smart endnode or not, and then leave it encapsulated in that case.</div>
<div><br></div><div>Radia<br><br><div class=3D"gmail_quote">On Thu, Jan 10,=
 2013 at 10:03 AM, Linda Dunbar <span dir=3D"ltr">&lt;<a href=3D"mailto:lin=
da.dunbar@huawei.com" target=3D"_blank">linda.dunbar@huawei.com</a>&gt;</sp=
an> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">





<div lang=3D"EN-US" link=3D"blue" vlink=3D"purple">
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">It is definitely simpler =
for Egress RBridge to decapsulate all received data frames:<u></u><u></u></=
span></p>

<p><u></u><span style=3D"font-size:11.0pt;font-family:Symbol;color:#1f497d"=
><span>=B7<span style=3D"font:7.0pt &quot;Times New Roman&quot;">=A0=A0=A0=
=A0=A0=A0=A0
</span></span></span><u></u><span style=3D"font-size:11.0pt;font-family:&qu=
ot;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">So it doesn=92t need=
 determine if the target end node is smart =A0or not<u></u><u></u></span></=
p>

<p><u></u><span style=3D"font-size:11.0pt;font-family:Symbol;color:#1f497d"=
><span>=B7<span style=3D"font:7.0pt &quot;Times New Roman&quot;">=A0=A0=A0=
=A0=A0=A0=A0
</span></span></span><u></u><span style=3D"font-size:11.0pt;font-family:&qu=
ot;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">It doesn=92t have to=
 attached different Ethernet Destination based on the inner header<u></u><u=
></u></span></p>

<p><u></u><span style=3D"font-size:11.0pt;font-family:Symbol;color:#1f497d"=
><span>=B7<span style=3D"font:7.0pt &quot;Times New Roman&quot;">=A0=A0=A0=
=A0=A0=A0=A0
</span></span></span><u></u><span style=3D"font-size:11.0pt;font-family:&qu=
ot;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">It is consistent for=
 all data frames and all end nodes.
<u></u><u></u></span></p>
<p><u></u><span style=3D"font-size:11.0pt;font-family:Symbol;color:#1f497d"=
><span>=B7<span style=3D"font:7.0pt &quot;Times New Roman&quot;">=A0=A0=A0=
=A0=A0=A0=A0
</span></span></span><u></u><span style=3D"font-size:11.0pt;font-family:&qu=
ot;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">Smart End node can r=
eceive native Ethernet frames anyway. It doesn=92t cause any extra processi=
ng on the Smart End node either.
<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">Linda<u></u><u></u></span=
></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
<div style=3D"border:none;border-left:solid blue 1.5pt;padding:0in 0in 0in =
4.0pt">
<div>
<div style=3D"border:none;border-top:solid #b5c4df 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:10.0pt;font-family:&quot=
;Tahoma&quot;,&quot;sans-serif&quot;">From:</span></b><span style=3D"font-s=
ize:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;"> Linda Du=
nbar
<br>
<b>Sent:</b> Thursday, January 10, 2013 11:43 AM<br>
<b>To:</b> &#39;Radia Perlman&#39;; Phanidhar Koganti<br>
<b>Cc:</b> <a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gma=
il.com</a>; Kesava_Vijaya_Krupak@Dell.com; <a href=3D"mailto:hu.fangwei@zte=
.com.cn" target=3D"_blank">hu.fangwei@zte.com.cn</a>; <a href=3D"mailto:tri=
ll@ietf.org" target=3D"_blank">trill@ietf.org</a><br>

<b>Subject:</b> RE: [trill] New version : TRILL Smart Endnodes draft<u></u>=
<u></u></span></p>
</div>
</div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">Radia,
<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">Questions inserted below:=
<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
<div style=3D"border:none;border-left:solid blue 1.5pt;padding:0in 0in 0in =
4.0pt"><div class=3D"im">
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
<div>
<p class=3D"MsoNormal">a)=A0<span style=3D"color:#1f497d">When the packet r=
eaches the egress RB I am presuming there will be a inner-MAC lookup to del=
iver the packet to the egress end-node.</span><u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
</div><div><div class=3D"im">
<p class=3D"MsoNormal"><span style=3D"color:#1f497d">correct...when egress =
RB R has to look up the inner MAC address E, which it has to do anyway with=
out smart endnodes, in order to choose which port to forward it on. =A0With=
 smart endnodes, when R looks up MAC address
 E, it not only discovers which port to send it on, but that E is &quot;sma=
rt&quot;, in which case R must leave the TRILL encapsulation</span><u></u><=
u></u></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
</div><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&q=
uot;Calibri&quot;,&quot;sans-serif&quot;;color:#943634">[Linda] Why can=92t=
 the egress RB R simply decapsulates the TRILL header and forward the data =
frame towards the attached end nodes? Regardless if the DA is smart
 node or not?<u></u><u></u></span></p><div class=3D"im">
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#943634">So RB processing is much =
simpler. It doesn=92t have to attach different Ethernet header based on the=
 inner DA. The smart node can receive Ethernet frames anyway
 (for communication within the directly attached LAN). <u></u><u></u></span=
></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
</div></div>
<div>
<div>
<div>
<div>
<div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:#1f497d">Thanks, Linda</span><u=
></u><u></u></p>
</div>
</div>
</div>
</div>
</div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
</div>
</div>
</div>
</div>
</div>

<br>_______________________________________________<br>
trill mailing list<br>
<a href=3D"mailto:trill@ietf.org">trill@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/trill" target=3D"_blank">h=
ttps://www.ietf.org/mailman/listinfo/trill</a><br>
<br></blockquote></div><br></div>

--e89a8f22c42588409204d2f3485e--

From d3e3e3@gmail.com  Thu Jan 10 12:16:07 2013
Return-Path: <d3e3e3@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 66B9A21F8836 for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 12:16:07 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.599
X-Spam-Level: 
X-Spam-Status: No, score=-103.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id m9quhMFmRlmp for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 12:16:06 -0800 (PST)
Received: from mail-ob0-f182.google.com (mail-ob0-f182.google.com [209.85.214.182]) by ietfa.amsl.com (Postfix) with ESMTP id 9F79F21F8833 for <trill@ietf.org>; Thu, 10 Jan 2013 12:16:06 -0800 (PST)
Received: by mail-ob0-f182.google.com with SMTP id 16so998739obc.41 for <trill@ietf.org>; Thu, 10 Jan 2013 12:16:06 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type:content-transfer-encoding; bh=bQWy2L4KQWuEJAn5wdIhIyd+MT+9ITGWLkcbbaQf7tc=; b=Tvs822jjx+e1LMGghR7/56dTmghxt5KGFA//o/mVQyZPfjwqbiA+sGQu+6Q7G2hTh4 RGXT5tqrhuddXVNsVCe3HN76rQGUBItm7saNyvo/J1BLVGSdvTwQ/3xSz+1K2CaHVvvP +Ec3NfCFnKC4YglgT/E7RIy/Wi01N1zbiAIkQ1KgugxEOSpWQKJLDdR7kOia7jsCziCp fCwUzIe9PQG7tfQ1nJRqTbxFczW/VEYTvjZDtdvYmx8XNHZwW7ljtau/EzwTxtbCbfmi aVCPJ8PyqzdpX9IsW/QPKawF3d+XaeuzKAKTMgzn1ZQGgyRm+o7PPWVVK40vxPG2zh22 PPfQ==
Received: by 10.60.7.167 with SMTP id k7mr44042707oea.20.1357848966068; Thu, 10 Jan 2013 12:16:06 -0800 (PST)
MIME-Version: 1.0
Received: by 10.76.110.239 with HTTP; Thu, 10 Jan 2013 12:15:46 -0800 (PST)
In-Reply-To: <CAFOuuo5z=LepKF1FSD0PSS0zGCC7eBdBEX4eQ+WQXaOa1+ZArA@mail.gmail.com>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com> <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com> <4A95BA014132FF49AE685FAB4B9F17F6450E06AB@dfweml505-mbx> <CAFOuuo5z=LepKF1FSD0PSS0zGCC7eBdBEX4eQ+WQXaOa1+ZArA@mail.gmail.com>
From: Donald Eastlake <d3e3e3@gmail.com>
Date: Thu, 10 Jan 2013 15:15:46 -0500
Message-ID: <CAF4+nEHdTX+g-A3kMykr+eaDP3KXf+zwh5fGvQOX7w=4xz31kw@mail.gmail.com>
To: Linda Dunbar <linda.dunbar@huawei.com>
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable
Cc: Radia Perlman <radiaperlman@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 10 Jan 2013 20:16:07 -0000

There may also be reasons to communicate other TRILL Data packet
information with the smart end node, besides the ingress nickname
needed for learning. For example, a smart end node might want to be
able to send and receive TRILL Data packets with fine grained labels.

Thanks,
Donald
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D
 Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
 155 Beaver Street, Milford, MA 01757 USA
 d3e3e3@gmail.com


On Thu, Jan 10, 2013 at 1:24 PM, Radia Perlman <radiaperlman@gmail.com> wro=
te:
> Yes, it's simpler, but it doesn't work, unless we want to say that smart
> endnodes can never learn from data traffic...they can only work if they a=
re
> configured with (MAC, egress nickname) of all destinations they might wan=
t
> to talk to, or must have a directory and request from the directory.
>
> I think we need to allow smart endnodes to learn from data traffic, in
> which, unfortunately (because as you point out, it's more complicated) th=
e
> RB has to consciously know whether it's forwarding to a smart endnode or
> not, and then leave it encapsulated in that case.
>
> Radia
>
> On Thu, Jan 10, 2013 at 10:03 AM, Linda Dunbar <linda.dunbar@huawei.com>
> wrote:
>>
>> It is definitely simpler for Egress RBridge to decapsulate all received
>> data frames:
>>
>> =B7        So it doesn=92t need determine if the target end node is smar=
t  or
>> not
>>
>> =B7        It doesn=92t have to attached different Ethernet Destination =
based
>> on the inner header
>>
>> =B7        It is consistent for all data frames and all end nodes.
>>
>> =B7        Smart End node can receive native Ethernet frames anyway. It
>> doesn=92t cause any extra processing on the Smart End node either.
>>
>>
>>
>> Linda
>>
>>
>>
>> From: Linda Dunbar
>> Sent: Thursday, January 10, 2013 11:43 AM
>> To: 'Radia Perlman'; Phanidhar Koganti
>> Cc: d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com;
>> hu.fangwei@zte.com.cn; trill@ietf.org
>> Subject: RE: [trill] New version : TRILL Smart Endnodes draft
>>
>>
>>
>> Radia,
>>
>>
>>
>> Questions inserted below:
>>
>>
>>
>>
>>
>> a) When the packet reaches the egress RB I am presuming there will be a
>> inner-MAC lookup to deliver the packet to the egress end-node.
>>
>>
>>
>> correct...when egress RB R has to look up the inner MAC address E, which
>> it has to do anyway without smart endnodes, in order to choose which por=
t to
>> forward it on.  With smart endnodes, when R looks up MAC address E, it n=
ot
>> only discovers which port to send it on, but that E is "smart", in which
>> case R must leave the TRILL encapsulation
>>
>>
>>
>> [Linda] Why can=92t the egress RB R simply decapsulates the TRILL header=
 and
>> forward the data frame towards the attached end nodes? Regardless if the=
 DA
>> is smart node or not?
>>
>> So RB processing is much simpler. It doesn=92t have to attach different
>> Ethernet header based on the inner DA. The smart node can receive Ethern=
et
>> frames anyway (for communication within the directly attached LAN).
>>
>>
>>
>> Thanks, Linda
>>
>>
>>
>>
>> _______________________________________________
>> trill mailing list
>> trill@ietf.org
>> https://www.ietf.org/mailman/listinfo/trill
>>
>

From linda.dunbar@huawei.com  Thu Jan 10 12:56:47 2013
Return-Path: <linda.dunbar@huawei.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CEDAD21F88E1 for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 12:56:47 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.479
X-Spam-Level: 
X-Spam-Status: No, score=-6.479 tagged_above=-999 required=5 tests=[AWL=0.120,  BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 2joKJqzH34GM for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 12:56:47 -0800 (PST)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) by ietfa.amsl.com (Postfix) with ESMTP id A64F121F88DD for <trill@ietf.org>; Thu, 10 Jan 2013 12:56:43 -0800 (PST)
Received: from 172.18.7.190 (EHLO lhreml203-edg.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.5-GA FastPath queued) with ESMTP id AOQ20220; Thu, 10 Jan 2013 20:56:42 +0000 (GMT)
Received: from LHREML404-HUB.china.huawei.com (10.201.5.218) by lhreml203-edg.huawei.com (172.18.7.221) with Microsoft SMTP Server (TLS) id 14.1.323.3; Thu, 10 Jan 2013 20:55:27 +0000
Received: from DFWEML404-HUB.china.huawei.com (10.193.5.203) by lhreml404-hub.china.huawei.com (10.201.5.218) with Microsoft SMTP Server (TLS) id 14.1.323.3; Fri, 11 Jan 2013 04:56:41 +0800
Received: from DFWEML505-MBX.china.huawei.com ([10.124.31.100]) by dfweml404-hub.china.huawei.com ([10.193.5.203]) with mapi id 14.01.0323.003; Thu, 10 Jan 2013 12:56:35 -0800
From: Linda Dunbar <linda.dunbar@huawei.com>
To: Donald Eastlake <d3e3e3@gmail.com>
Thread-Topic: [trill] New version : TRILL Smart Endnodes draft
Thread-Index: AQHN7eZvBDHzFA5/eke6FKO9FjPC95hC0tRggAAJ9mCAAIzZAIAAHzgA//+E1EA=
Date: Thu, 10 Jan 2013 20:56:34 +0000
Message-ID: <4A95BA014132FF49AE685FAB4B9F17F6450E0780@dfweml505-mbx>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com> <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com> <4A95BA014132FF49AE685FAB4B9F17F6450E06AB@dfweml505-mbx> <CAFOuuo5z=LepKF1FSD0PSS0zGCC7eBdBEX4eQ+WQXaOa1+ZArA@mail.gmail.com> <CAF4+nEHdTX+g-A3kMykr+eaDP3KXf+zwh5fGvQOX7w=4xz31kw@mail.gmail.com>
In-Reply-To: <CAF4+nEHdTX+g-A3kMykr+eaDP3KXf+zwh5fGvQOX7w=4xz31kw@mail.gmail.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.192.11.236]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Cc: "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, Radia Perlman <radiaperlman@gmail.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 10 Jan 2013 20:56:47 -0000

Donald,=20

Can those "other TRILL data packet information" to be communicated to Smart=
 node being carried by the special "Hello" message between RBridge and Smar=
t nodes?

Linda =20

> -----Original Message-----
> From: trill-bounces@ietf.org [mailto:trill-bounces@ietf.org] On Behalf
> Of Donald Eastlake
> Sent: Thursday, January 10, 2013 2:16 PM
> To: Linda Dunbar
> Cc: Radia Perlman; Kesava_Vijaya_Krupak@Dell.com; hu.fangwei@zte.com.cn;
> trill@ietf.org
> Subject: Re: [trill] New version : TRILL Smart Endnodes draft
>=20
> There may also be reasons to communicate other TRILL Data packet
> information with the smart end node, besides the ingress nickname
> needed for learning. For example, a smart end node might want to be
> able to send and receive TRILL Data packets with fine grained labels.
>=20
> Thanks,
> Donald
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D
>  Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
>  155 Beaver Street, Milford, MA 01757 USA
>  d3e3e3@gmail.com
>=20
>=20
> On Thu, Jan 10, 2013 at 1:24 PM, Radia Perlman <radiaperlman@gmail.com>
> wrote:
> > Yes, it's simpler, but it doesn't work, unless we want to say that
> smart
> > endnodes can never learn from data traffic...they can only work if
> they are
> > configured with (MAC, egress nickname) of all destinations they might
> want
> > to talk to, or must have a directory and request from the directory.
> >
> > I think we need to allow smart endnodes to learn from data traffic,
> in
> > which, unfortunately (because as you point out, it's more complicated)
> the
> > RB has to consciously know whether it's forwarding to a smart endnode
> or
> > not, and then leave it encapsulated in that case.
> >
> > Radia
> >
> > On Thu, Jan 10, 2013 at 10:03 AM, Linda Dunbar
> <linda.dunbar@huawei.com>
> > wrote:
> >>
> >> It is definitely simpler for Egress RBridge to decapsulate all
> received
> >> data frames:
> >>
> >> *        So it doesn't need determine if the target end node is
> smart  or
> >> not
> >>
> >> *        It doesn't have to attached different Ethernet Destination
> based
> >> on the inner header
> >>
> >> *        It is consistent for all data frames and all end nodes.
> >>
> >> *        Smart End node can receive native Ethernet frames anyway.
> It
> >> doesn't cause any extra processing on the Smart End node either.
> >>
> >>
> >>
> >> Linda
> >>
> >>
> >>
> >> From: Linda Dunbar
> >> Sent: Thursday, January 10, 2013 11:43 AM
> >> To: 'Radia Perlman'; Phanidhar Koganti
> >> Cc: d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com;
> >> hu.fangwei@zte.com.cn; trill@ietf.org
> >> Subject: RE: [trill] New version : TRILL Smart Endnodes draft
> >>
> >>
> >>
> >> Radia,
> >>
> >>
> >>
> >> Questions inserted below:
> >>
> >>
> >>
> >>
> >>
> >> a) When the packet reaches the egress RB I am presuming there will
> be a
> >> inner-MAC lookup to deliver the packet to the egress end-node.
> >>
> >>
> >>
> >> correct...when egress RB R has to look up the inner MAC address E,
> which
> >> it has to do anyway without smart endnodes, in order to choose which
> port to
> >> forward it on.  With smart endnodes, when R looks up MAC address E,
> it not
> >> only discovers which port to send it on, but that E is "smart", in
> which
> >> case R must leave the TRILL encapsulation
> >>
> >>
> >>
> >> [Linda] Why can't the egress RB R simply decapsulates the TRILL
> header and
> >> forward the data frame towards the attached end nodes? Regardless if
> the DA
> >> is smart node or not?
> >>
> >> So RB processing is much simpler. It doesn't have to attach
> different
> >> Ethernet header based on the inner DA. The smart node can receive
> Ethernet
> >> frames anyway (for communication within the directly attached LAN).
> >>
> >>
> >>
> >> Thanks, Linda
> >>
> >>
> >>
> >>
> >> _______________________________________________
> >> trill mailing list
> >> trill@ietf.org
> >> https://www.ietf.org/mailman/listinfo/trill
> >>
> >
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill

From d3e3e3@gmail.com  Thu Jan 10 13:22:15 2013
Return-Path: <d3e3e3@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CA88921F862D for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 13:22:15 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.599
X-Spam-Level: 
X-Spam-Status: No, score=-103.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id iLvGJ0TdQ+jr for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 13:22:11 -0800 (PST)
Received: from mail-ob0-f182.google.com (mail-ob0-f182.google.com [209.85.214.182]) by ietfa.amsl.com (Postfix) with ESMTP id 7D70E21F857A for <trill@ietf.org>; Thu, 10 Jan 2013 13:22:11 -0800 (PST)
Received: by mail-ob0-f182.google.com with SMTP id 16so1085168obc.13 for <trill@ietf.org>; Thu, 10 Jan 2013 13:22:11 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=bzOnyiAizGK80d+zPgXwdw0gFt4sheN4LlfmQOou9ss=; b=nFVJCSKojEwtDSntvAlGYWUQxiGgJNaNL6W+JMNu5VxF8hESLm6gVKqZ96OV1BStMy +qpw0MPtT5vY0FYRwlqCiDWXx9VLQ0bdy+PehSEgtpSVHwYcp3e4+ez7g1fTaf2lhedH 9gbF7UVs5wAOWWdO0q0Vi2UfUUmN7QvjnZg5RDLNQd9T4iXd6nBEyCd//2N+uehj3uby SmNHuxlTvf/VUFuTa+jkqzTtZXD/zARGNixb2aqmwrBNsbA3PBY0mB2rFBOR4FdPeRkH 7Zx3wwv5DlZJE0eCRP4p4XKxzMJ4WMR5BxtgvSHSS/0xuBvzPHtYT4MyR0cco/vq1eaP /Ugg==
Received: by 10.60.27.166 with SMTP id u6mr41960364oeg.80.1357852930950; Thu, 10 Jan 2013 13:22:10 -0800 (PST)
MIME-Version: 1.0
Received: by 10.76.110.239 with HTTP; Thu, 10 Jan 2013 13:21:50 -0800 (PST)
In-Reply-To: <4A95BA014132FF49AE685FAB4B9F17F6450E0780@dfweml505-mbx>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com> <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com> <4A95BA014132FF49AE685FAB4B9F17F6450E06AB@dfweml505-mbx> <CAFOuuo5z=LepKF1FSD0PSS0zGCC7eBdBEX4eQ+WQXaOa1+ZArA@mail.gmail.com> <CAF4+nEHdTX+g-A3kMykr+eaDP3KXf+zwh5fGvQOX7w=4xz31kw@mail.gmail.com> <4A95BA014132FF49AE685FAB4B9F17F6450E0780@dfweml505-mbx>
From: Donald Eastlake <d3e3e3@gmail.com>
Date: Thu, 10 Jan 2013 16:21:50 -0500
Message-ID: <CAF4+nEH37TG7J7nCD-iV=PfuT59vX2NP3h1T+zi+mqb0BwFiKQ@mail.gmail.com>
To: Linda Dunbar <linda.dunbar@huawei.com>
Content-Type: text/plain; charset=ISO-8859-1
Cc: "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, Radia Perlman <radiaperlman@gmail.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 10 Jan 2013 21:22:15 -0000

Linda,

On Thu, Jan 10, 2013 at 3:56 PM, Linda Dunbar <linda.dunbar@huawei.com> wrote:
> Donald,
>
> Can those "other TRILL data packet information" to be communicated to Smart node being carried by the special "Hello" message between RBridge and Smart nodes?

No. By "TRILL data packet information" I mean information tied to the
payload in the packet and which might vary from data packet to data
packet. Stuff in any "special Hello" or whatever other sort of control
packets there are seems appropriate for information related to the
state of the RBridge to Smart End node but not for information tied to
a data payload.

Thanks,
Donald
=============================
 Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
 155 Beaver Street, Milford, MA 01757 USA
 d3e3e3@gmail.com

> Linda
>
>> -----Original Message-----
>> From: trill-bounces@ietf.org [mailto:trill-bounces@ietf.org] On Behalf
>> Of Donald Eastlake
>> Sent: Thursday, January 10, 2013 2:16 PM
>> To: Linda Dunbar
>> Cc: Radia Perlman; Kesava_Vijaya_Krupak@Dell.com; hu.fangwei@zte.com.cn;
>> trill@ietf.org
>> Subject: Re: [trill] New version : TRILL Smart Endnodes draft
>>
>> There may also be reasons to communicate other TRILL Data packet
>> information with the smart end node, besides the ingress nickname
>> needed for learning. For example, a smart end node might want to be
>> able to send and receive TRILL Data packets with fine grained labels.
>>
>> Thanks,
>> Donald
>> =============================
>>  Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
>>  155 Beaver Street, Milford, MA 01757 USA
>>  d3e3e3@gmail.com
>>
>>
>> On Thu, Jan 10, 2013 at 1:24 PM, Radia Perlman <radiaperlman@gmail.com>
>> wrote:
>> > Yes, it's simpler, but it doesn't work, unless we want to say that
>> smart
>> > endnodes can never learn from data traffic...they can only work if
>> they are
>> > configured with (MAC, egress nickname) of all destinations they might
>> want
>> > to talk to, or must have a directory and request from the directory.
>> >
>> > I think we need to allow smart endnodes to learn from data traffic,
>> in
>> > which, unfortunately (because as you point out, it's more complicated)
>> the
>> > RB has to consciously know whether it's forwarding to a smart endnode
>> or
>> > not, and then leave it encapsulated in that case.
>> >
>> > Radia
>> >
>> > On Thu, Jan 10, 2013 at 10:03 AM, Linda Dunbar
>> <linda.dunbar@huawei.com>
>> > wrote:
>> >>
>> >> It is definitely simpler for Egress RBridge to decapsulate all
>> received
>> >> data frames:
>> >>
>> >> *        So it doesn't need determine if the target end node is
>> smart  or
>> >> not
>> >>
>> >> *        It doesn't have to attached different Ethernet Destination
>> based
>> >> on the inner header
>> >>
>> >> *        It is consistent for all data frames and all end nodes.
>> >>
>> >> *        Smart End node can receive native Ethernet frames anyway.
>> It
>> >> doesn't cause any extra processing on the Smart End node either.
>> >>
>> >>
>> >>
>> >> Linda
>> >>
>> >>
>> >>
>> >> From: Linda Dunbar
>> >> Sent: Thursday, January 10, 2013 11:43 AM
>> >> To: 'Radia Perlman'; Phanidhar Koganti
>> >> Cc: d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com;
>> >> hu.fangwei@zte.com.cn; trill@ietf.org
>> >> Subject: RE: [trill] New version : TRILL Smart Endnodes draft
>> >>
>> >>
>> >>
>> >> Radia,
>> >>
>> >>
>> >>
>> >> Questions inserted below:
>> >>
>> >>
>> >>
>> >>
>> >>
>> >> a) When the packet reaches the egress RB I am presuming there will
>> be a
>> >> inner-MAC lookup to deliver the packet to the egress end-node.
>> >>
>> >>
>> >>
>> >> correct...when egress RB R has to look up the inner MAC address E,
>> which
>> >> it has to do anyway without smart endnodes, in order to choose which
>> port to
>> >> forward it on.  With smart endnodes, when R looks up MAC address E,
>> it not
>> >> only discovers which port to send it on, but that E is "smart", in
>> which
>> >> case R must leave the TRILL encapsulation
>> >>
>> >>
>> >>
>> >> [Linda] Why can't the egress RB R simply decapsulates the TRILL
>> header and
>> >> forward the data frame towards the attached end nodes? Regardless if
>> the DA
>> >> is smart node or not?
>> >>
>> >> So RB processing is much simpler. It doesn't have to attach
>> different
>> >> Ethernet header based on the inner DA. The smart node can receive
>> Ethernet
>> >> frames anyway (for communication within the directly attached LAN).
>> >>
>> >>
>> >>
>> >> Thanks, Linda
>> >>
>> >>
>> >>
>> >>
>> >> _______________________________________________
>> >> trill mailing list
>> >> trill@ietf.org
>> >> https://www.ietf.org/mailman/listinfo/trill
>> >>
>> >
>> _______________________________________________
>> trill mailing list
>> trill@ietf.org
>> https://www.ietf.org/mailman/listinfo/trill
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill

From d3e3e3@gmail.com  Thu Jan 10 14:36:17 2013
Return-Path: <d3e3e3@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3DD0321F8696 for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 14:36:17 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.999
X-Spam-Level: 
X-Spam-Status: No, score=-102.999 tagged_above=-999 required=5 tests=[AWL=-0.600, BAYES_00=-2.599, J_CHICKENPOX_34=0.6, J_CHICKENPOX_55=0.6, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ku5cxNfuys1t for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 14:36:16 -0800 (PST)
Received: from mail-oa0-f47.google.com (mail-oa0-f47.google.com [209.85.219.47]) by ietfa.amsl.com (Postfix) with ESMTP id 9AEE121F8694 for <trill@ietf.org>; Thu, 10 Jan 2013 14:36:16 -0800 (PST)
Received: by mail-oa0-f47.google.com with SMTP id h1so1172634oag.6 for <trill@ietf.org>; Thu, 10 Jan 2013 14:36:16 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:from:date:message-id:subject:to:content-type; bh=4YRselCipSKMJF9WwDKLeAzyNkPld93Ztc35qNkZaqI=; b=ipLcvRbcGU6lallv5jP+N+u+WkW6DEf/8ZNWclVdA+jC7Sf5Ev+p+vPjxfjU3MjxXh MQdEdcGh79NfiuOleiJRGYuQqBJ7pnj6CDlBTIV6R1ILis9iJCKvOt+9f8hALsvnUg6a wl4lPr3s4OGiEwfukVF/kE6o7IAvLL/ITNrTUURO43ppYKxS8KQJRGIJdmj1iXwjx3cv MHLzcK3i5HoHPjxh30wAn38bp6uvOUwrIUwTwQyXEnb60CUmtVKEHvadClnRvLu4twX1 i7Lv5IQh2mGEdzqIdojC7+sfAR+O5C9FWccG7LlFmGoTGWD3IGK07dhz+w5EIxJCk75U Rghg==
Received: by 10.182.131.3 with SMTP id oi3mr48849047obb.13.1357857376034; Thu, 10 Jan 2013 14:36:16 -0800 (PST)
MIME-Version: 1.0
Received: by 10.76.110.239 with HTTP; Thu, 10 Jan 2013 14:35:55 -0800 (PST)
From: Donald Eastlake <d3e3e3@gmail.com>
Date: Thu, 10 Jan 2013 17:35:55 -0500
Message-ID: <CAF4+nEHkw3=F3XWUqEYWm1fu8GvYn+Jm_a4buqGs2sAwrVerfg@mail.gmail.com>
To: trill@ietf.org
Content-Type: text/plain; charset=ISO-8859-1
Subject: [trill] ESADI 4. Multiple attachment of the same MAC
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 10 Jan 2013 22:36:17 -0000

Notwithstanding the claim that there were three suggested changes to
ESADI below is a forth :-)

Thanks,
Donald
=============================
 Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
 155 Beaver Street, Milford, MA 01757 USA
 d3e3e3@gmail.com


Update RFC 6325 by updating the last sentence of the last paragraph of
Section 4.2 as shown below to provide better traffic spreading while
avoiding possible address flip-flopping. (The preceding two sentences
are also shown for context.)

   Explanation: Assume some end station has two or more ports with the
   same MAC&VLAN that are each connected to different RBridges (RB1,
   RB2, ...).  With ESADI, some other RBridge, RB0, can persistently
   see that MAC&VLAN connected to multile RBridges. When RB0 ingresses
   a frame for that MAC&VLAN, the current RFC 6325 text permits a wide
   range of behavior. In particular, it would permit RB0 to use some
   rule such as always send to the egress with the lowest System ID,
   which would put all of this traffic through one of the egress RBs
   and one of the end station ports.  There would be no load spreading
   even if there were multiple different ingress RBridges and/or
   different MAC addresses. It also would also permit RB0 to send
   different traffic to different egresses by doing ECMP at a flow
   levee which would likely result in return traffic to RB0 from RB1,
   RB2, ... for the same MAC&VLAN. The resulting address flip-flopping
   could cause problems. The change below fixes this by requiring RB0
   to use only one egress for a particular MAC&VLAN but to select that
   egress pseudo-randomly based on the topology including MAC
   reachabilty.  Assuming multiple ingress RBridges and/or multiple
   MAC addresses, this should result in load spreading without without
   address flip-floping.

OLD: ...  However, duplicates of the same MAC within a VLAN can appear
      in ESADI data and between ESADI data and addresses learned from
      the observation of received and decapsulated frames, entered by
      manual configuration, or learned through Layer 2 registration
      protocols.  If duplicate MAC addresses occur within a VLAN, RB2
      sends frames to the MAC with the highest confidence.  If
      confidences are also tied between the duplicates, for
      consistency it is suggested that RB2 direct all such frames (or
      all such frames in the same ECMP flow) toward the same egress
      RBridge; however, the use of other policies will not cause a
      network problem since transit RBridges do not examine the
      Inner.MacDA for known unicast frames.

NEW: ...  However, duplicates of the same MAC within a VLAN can appear
      in ESADI data and between ESADI data and addresses learned from
      the observation of received and decapsulated frames, entered by
      manual configuration, or learned through Layer 2 registration
      protocols.  If duplicate MAC addresses occur within a VLAN, RB2
      sends frames to the MAC with the highest confidence.

      If confidences are also tied between the duplicates then RB2
      MUST, in a pseudo-random way [RFC4086], select one of the egress
      RBridges to which the destination MAC appears to be attached and
      send all traffic for the destination MAC and VLAN to that
      egress. Such pseudo-random selection will, over a population of
      ingress RBridges, probabilistically spread traffic over the
      possible egress RBridges. Reasonable inputs to the pseudo-random
      selection are the ingress RBridge System ID and/or nickname, the
      VLAN, the destination MAC address, and a vector of the RBridges
      with connectivity to that MAC and VLAN. There is no need for
      different RBridges to use the same pseudo-random function.

From zhai.hongjun@zte.com.cn  Thu Jan 10 16:45:07 2013
Return-Path: <zhai.hongjun@zte.com.cn>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 426CA21F878F; Thu, 10 Jan 2013 16:45:07 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -101.202
X-Spam-Level: 
X-Spam-Status: No, score=-101.202 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, USER_IN_WHITELIST=-100, WEIRD_QUOTING=1.396]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id uX1M5uBHn8X4; Thu, 10 Jan 2013 16:45:06 -0800 (PST)
Received: from zte.com.cn (mx5.zte.com.cn [63.217.80.70]) by ietfa.amsl.com (Postfix) with ESMTP id B884321F844A; Thu, 10 Jan 2013 16:45:05 -0800 (PST)
Received: from mse01.zte.com.cn (unknown [10.30.3.20]) by Websense Email Security Gateway with ESMTPS id 547C012718CC; Fri, 11 Jan 2013 08:47:24 +0800 (CST)
Received: from notes_smtp.zte.com.cn ([10.30.1.239]) by mse01.zte.com.cn with ESMTP id r0B0ihPi067303; Fri, 11 Jan 2013 08:44:43 +0800 (GMT-8) (envelope-from zhai.hongjun@zte.com.cn)
In-Reply-To: <CAF4+nEHdTX+g-A3kMykr+eaDP3KXf+zwh5fGvQOX7w=4xz31kw@mail.gmail.com>
To: Donald Eastlake <d3e3e3@gmail.com>
MIME-Version: 1.0
X-KeepSent: 537A9CFF:886515DC-48257AF0:00028D26; type=4; name=$KeepSent
X-Mailer: Lotus Notes Release 6.5.6 March 06, 2007
Message-ID: <OF537A9CFF.886515DC-ON48257AF0.00028D26-48257AF0.0004650E@zte.com.cn>
From: zhai.hongjun@zte.com.cn
Date: Fri, 11 Jan 2013 08:44:46 +0800
X-MIMETrack: Serialize by Router on notes_smtp/zte_ltd(Release 8.5.3FP1 HF212|May 23, 2012) at 2013-01-11 08:44:36, Serialize complete at 2013-01-11 08:44:36
Content-Type: multipart/alternative; boundary="=_alternative 0004650748257AF0_="
X-MAIL: mse01.zte.com.cn r0B0ihPi067303
Cc: trill-bounces@ietf.org, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, Radia Perlman <radiaperlman@gmail.com>, Linda Dunbar <linda.dunbar@huawei.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 11 Jan 2013 00:45:07 -0000

This is a multipart message in MIME format.
--=_alternative 0004650748257AF0_=
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: base64
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--=_alternative 0004650748257AF0_=
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: base64
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=
--=_alternative 0004650748257AF0_=--

From d3e3e3@gmail.com  Thu Jan 10 17:23:49 2013
Return-Path: <d3e3e3@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9213E21F88C8 for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 17:23:49 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -101.609
X-Spam-Level: 
X-Spam-Status: No, score=-101.609 tagged_above=-999 required=5 tests=[AWL=-1.857, BAYES_00=-2.599, HTML_MESSAGE=0.001, MIME_CHARSET_FARAWAY=2.45, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100, WEIRD_QUOTING=1.396]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id U9+TupB5DSAh for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 17:23:46 -0800 (PST)
Received: from mail-ob0-f178.google.com (mail-ob0-f178.google.com [209.85.214.178]) by ietfa.amsl.com (Postfix) with ESMTP id D154E21F8930 for <trill@ietf.org>; Thu, 10 Jan 2013 17:23:45 -0800 (PST)
Received: by mail-ob0-f178.google.com with SMTP id eh20so1236060obb.23 for <trill@ietf.org>; Thu, 10 Jan 2013 17:23:45 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=I87SftScxlnPLNfny6LRZN0Bv0s8K0LloSkS9laEcdY=; b=ckJKciZYmUsEJbPpDpLrzRVGbtGAyON5mC7eXXDxP5vtr9DnJ8Omujl7OSpaInXbGY GKlk9a8jMgDlU92am1ewgNZjASYRuvX0QyogN4H0diapW9ZP5Nm9QP0A21/EMyWgDE/E ySM7IHEFmnEcyQ4txQG0BBQNIhn8GgcKvasaflFp65jXFk/Qz11zsQQoHKQ3n1u5DwhI w67lIQ3xocR8AX3UFDgT6Oh5lzxQ2x+tUxUH5teIBD1KmehhOCpL3t9p7uXzxeTyQSDv MB/xJ+keBCPvKoHc2TLjL+JoU0tRO9q5Y/8JV53qHD2OAMucWgqCNG9Fx7YtX5xQYiwg gDVQ==
Received: by 10.182.150.72 with SMTP id ug8mr53209795obb.1.1357867425056; Thu, 10 Jan 2013 17:23:45 -0800 (PST)
MIME-Version: 1.0
Received: by 10.76.110.239 with HTTP; Thu, 10 Jan 2013 17:23:24 -0800 (PST)
In-Reply-To: <OF537A9CFF.886515DC-ON48257AF0.00028D26-48257AF0.0004650E@zte.com.cn>
References: <CAF4+nEHdTX+g-A3kMykr+eaDP3KXf+zwh5fGvQOX7w=4xz31kw@mail.gmail.com> <OF537A9CFF.886515DC-ON48257AF0.00028D26-48257AF0.0004650E@zte.com.cn>
From: Donald Eastlake <d3e3e3@gmail.com>
Date: Thu, 10 Jan 2013 20:23:24 -0500
Message-ID: <CAF4+nEF890TCnGiXV34eHrNE6kano76QcrR4Q8JJ5X5xxwz+xw@mail.gmail.com>
To: zhai.hongjun@zte.com.cn
Content-Type: multipart/alternative; boundary=f46d044469ad85c94c04d2f925c8
Cc: "trill@ietf.org" <trill@ietf.org>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, Radia Perlman <radiaperlman@gmail.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, Linda Dunbar <linda.dunbar@huawei.com>
Subject: Re: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 11 Jan 2013 01:23:49 -0000

--f46d044469ad85c94c04d2f925c8
Content-Type: text/plain; charset=GB2312
Content-Transfer-Encoding: quoted-printable

I believe FGL support should be announced in Hellos.

I think the main reason for an end station to want to receive/send FGL
TRILL Data packets is because it wants to handle data packets with a very
large number of different labels. This strikes me as a bit unusual,
probably a large gateway or server, so the network would normally be
configured so that the TRILL switches it is connected with support FGL.

Thanks,
Donald
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D
 Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
 155 Beaver Street, Milford, MA 01757 USA
 d3e3e3@gmail.com


On Thu, Jan 10, 2013 at 7:44 PM, <zhai.hongjun@zte.com.cn> wrote:

>
> Hi Donald,
>
> > There may also be reasons to communicate other TRILL Data packet
> > information with the smart end node, besides the ingress nickname
> > needed for learning. For example, a smart end node might want to be
> > able to send and receive TRILL Data packets with fine grained labels.
>
> As smartnodes have no Link State Database information for TRILL campus,
> they do not know whether all the RBridges in TRILL campus support FGL
> or not, or which RBridges support FGL and which does not support FGL.
> Therefore, if a smartnode sends TRILL multi-destination data packets
> with FGL, the packets might be discarded by some RBridgs that do not
> support FGL. In this case, how such a sartnode makes sure those FGL
> packets are able to arrive at their destination(s)?
>
>
>
> Best Regards,
> Zhai Hongjun
> """"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
> Protocol Development Dept.VI, Central R&D Institute, ZTE Corporation
> No. 68, Zijinghua Road, Yuhuatai District, Nanjing, P.R.China, 210012
>
> Zhai Hongjun
>
> Tel: +86-25-52877345
> Email: zhai.hongjun@zte.com.cn
> """""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
>
>
>
>
>  *Donald Eastlake <d3e3e3@gmail.com>*
> =B7=A2=BC=FE=C8=CB:  trill-bounces@ietf.org
>
> 2013-01-11 04:15
>   =CA=D5=BC=FE=C8=CB
> Linda Dunbar <linda.dunbar@huawei.com>
> =B3=AD=CB=CD
> Radia Perlman <radiaperlman@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com" <
> Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <
> hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
> =D6=F7=CC=E2
> Re: [trill] New version : TRILL Smart Endnodes draft
>
>
>
>
> There may also be reasons to communicate other TRILL Data packet
> information with the smart end node, besides the ingress nickname
> needed for learning. For example, a smart end node might want to be
> able to send and receive TRILL Data packets with fine grained labels.
>
> Thanks,
> Donald
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D
> Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
> 155 Beaver Street, Milford, MA 01757 USA
> d3e3e3@gmail.com
>
>
> On Thu, Jan 10, 2013 at 1:24 PM, Radia Perlman <radiaperlman@gmail.com>
> wrote:
> > Yes, it's simpler, but it doesn't work, unless we want to say that smar=
t
> > endnodes can never learn from data traffic...they can only work if they
> are
> > configured with (MAC, egress nickname) of all destinations they might
> want
> > to talk to, or must have a directory and request from the directory.
> >
> > I think we need to allow smart endnodes to learn from data traffic, in
> > which, unfortunately (because as you point out, it's more complicated)
> the
> > RB has to consciously know whether it's forwarding to a smart endnode o=
r
> > not, and then leave it encapsulated in that case.
> >
> > Radia
> >
> > On Thu, Jan 10, 2013 at 10:03 AM, Linda Dunbar <linda.dunbar@huawei.com=
>
> > wrote:
> >>
> >> It is definitely simpler for Egress RBridge to decapsulate all receive=
d
> >> data frames:
> >>
> >> =A1=A4        So it doesn=A1=AFt need determine if the target end node=
 is smart
>  or
> >> not
> >>
> >> =A1=A4        It doesn=A1=AFt have to attached different Ethernet Dest=
ination
> based
> >> on the inner header
> >>
> >> =A1=A4        It is consistent for all data frames and all end nodes.
> >>
> >> =A1=A4        Smart End node can receive native Ethernet frames anyway=
. It
> >> doesn=A1=AFt cause any extra processing on the Smart End node either.
> >>
> >>
> >>
> >> Linda
> >>
> >>
> >>
> >> From: Linda Dunbar
> >> Sent: Thursday, January 10, 2013 11:43 AM
> >> To: 'Radia Perlman'; Phanidhar Koganti
> >> Cc: d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com;
> >> hu.fangwei@zte.com.cn; trill@ietf.org
> >> Subject: RE: [trill] New version : TRILL Smart Endnodes draft
> >>
> >>
> >>
> >> Radia,
> >>
> >>
> >>
> >> Questions inserted below:
> >>
> >>
> >>
> >>
> >>
> >> a) When the packet reaches the egress RB I am presuming there will be =
a
> >> inner-MAC lookup to deliver the packet to the egress end-node.
> >>
> >>
> >>
> >> correct...when egress RB R has to look up the inner MAC address E, whi=
ch
> >> it has to do anyway without smart endnodes, in order to choose which
> port to
> >> forward it on.  With smart endnodes, when R looks up MAC address E, it
> not
> >> only discovers which port to send it on, but that E is "smart", in whi=
ch
> >> case R must leave the TRILL encapsulation
> >>
> >>
> >>
> >> [Linda] Why can=A1=AFt the egress RB R simply decapsulates the TRILL h=
eader
> and
> >> forward the data frame towards the attached end nodes? Regardless if
> the DA
> >> is smart node or not?
> >>
> >> So RB processing is much simpler. It doesn=A1=AFt have to attach diffe=
rent
> >> Ethernet header based on the inner DA. The smart node can receive
> Ethernet
> >> frames anyway (for communication within the directly attached LAN).
> >>
> >>
> >>
> >> Thanks, Linda
> >>
> >>
> >>
> >>
> >> _______________________________________________
> >> trill mailing list
> >> trill@ietf.org
> >> https://www.ietf.org/mailman/listinfo/trill
> >>
> >
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill
>
>

--f46d044469ad85c94c04d2f925c8
Content-Type: text/html; charset=GB2312
Content-Transfer-Encoding: quoted-printable

I believe FGL support should be announced in Hellos.<div><br></div><div>I t=
hink the main reason for an end station to want to receive/send FGL TRILL D=
ata packets is because it wants to handle data packets with a very large nu=
mber of different labels. This strikes me as a bit unusual, probably a larg=
e gateway or server, so the network would normally be configured so that th=
e TRILL switches it is connected with support FGL.</div>

<div><br clear=3D"all"><div>Thanks,<br>Donald<br>=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D<br>&nbsp;Do=
nald E. Eastlake 3rd&nbsp;&nbsp; +1-508-333-2270 (cell)<br>&nbsp;155 Beaver=
 Street,&nbsp;Milford, MA 01757 USA<br>&nbsp;<a href=3D"mailto:d3e3e3@gmail=
.com" target=3D"_blank">d3e3e3@gmail.com</a></div>


<br><br><div class=3D"gmail_quote">On Thu, Jan 10, 2013 at 7:44 PM,  <span =
dir=3D"ltr">&lt;<a href=3D"mailto:zhai.hongjun@zte.com.cn" target=3D"_blank=
">zhai.hongjun@zte.com.cn</a>&gt;</span> wrote:<br><blockquote class=3D"gma=
il_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-lef=
t:1ex">


<br><font face=3D"sans-serif">Hi Donald,</font>
<br><div class=3D"im">
<br><tt><font>&gt; There may also be reasons to communicate other
TRILL Data packet<br>
&gt; information with the smart end node, besides the ingress nickname<br>
&gt; needed for learning. For example, a smart end node might want to be<br=
>
&gt; able to send and receive TRILL Data packets with fine grained labels.<=
/font></tt>
<br>
<br></div><tt><font>As smartnodes have no Link State Database information
for TRILL campus, </font></tt>
<br><tt><font>they do not know whether all the RBridges in TRILL
campus support FGL </font></tt>
<br><tt><font>or not, or which RBridges support FGL and which does
not support FGL.</font></tt>
<br><tt><font>Therefore, if a smartnode sends TRILL multi-destination
data packets </font></tt>
<br><tt><font>with FGL, the packets might be discarded by some RBridgs
that do not </font></tt>
<br><tt><font>support FGL. In this case, how such a sartnode makes
sure those FGL </font></tt>
<br><tt><font>packets are able to arrive at their destination(s)?</font></t=
t>
<br><tt><font>&nbsp;<br>
</font></tt><font face=3D"sans-serif"><br>
<br>
Best Regards,<br>
Zhai Hongjun<br>
&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&qu=
ot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;=
&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&qu=
ot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;=
&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&qu=
ot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;<br>


 Protocol Development Dept.VI, Central R&amp;D Institute, ZTE Corporation<b=
r>
 No. 68, Zijinghua Road, Yuhuatai District, Nanjing, P.R.China, 210012<br>
 <br>
 Zhai Hongjun<br>
 <br>
 Tel: <a href=3D"tel:%2B86-25-52877345" value=3D"+862552877345" target=3D"_=
blank">+86-25-52877345</a><br>
 Email: <a href=3D"mailto:zhai.hongjun@zte.com.cn" target=3D"_blank">zhai.h=
ongjun@zte.com.cn</a><br>
&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&qu=
ot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;=
&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&qu=
ot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;=
&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&qu=
ot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;<br>


<br>
</font>
<br>
<br>
<br>
<p></p><table width=3D"100%">
<tbody><tr valign=3D"top">
<td width=3D"36%"><font size=3D"1" face=3D"sans-serif"><b>Donald Eastlake &=
lt;<a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gmail.com</=
a>&gt;</b>
</font>
<br><font size=3D"1" face=3D"sans-serif">=B7=A2=BC=FE=C8=CB: &nbsp;<a href=
=3D"mailto:trill-bounces@ietf.org" target=3D"_blank">trill-bounces@ietf.org=
</a></font>
<p><font size=3D"1" face=3D"sans-serif">2013-01-11 04:15</font>
</p></td><td width=3D"63%">
<table width=3D"100%">
<tbody><tr valign=3D"top">
<td>
<div align=3D"right"><font size=3D"1" face=3D"sans-serif">=CA=D5=BC=FE=C8=
=CB</font></div>
</td><td><font size=3D"1" face=3D"sans-serif">Linda Dunbar &lt;<a href=3D"m=
ailto:linda.dunbar@huawei.com" target=3D"_blank">linda.dunbar@huawei.com</a=
>&gt;</font>
</td></tr><tr valign=3D"top">
<td>
<div align=3D"right"><font size=3D"1" face=3D"sans-serif">=B3=AD=CB=CD</fon=
t></div>
</td><td><font size=3D"1" face=3D"sans-serif">Radia Perlman &lt;<a href=3D"=
mailto:radiaperlman@gmail.com" target=3D"_blank">radiaperlman@gmail.com</a>=
&gt;,
&quot;Kesava_Vijaya_Krupak@Dell.com&quot; &lt;<a href=3D"mailto:Kesava_Vija=
ya_Krupak@dell.com" target=3D"_blank">Kesava_Vijaya_Krupak@dell.com</a>&gt;=
,
&quot;<a href=3D"mailto:hu.fangwei@zte.com.cn" target=3D"_blank">hu.fangwei=
@zte.com.cn</a>&quot; &lt;<a href=3D"mailto:hu.fangwei@zte.com.cn" target=
=3D"_blank">hu.fangwei@zte.com.cn</a>&gt;, &quot;<a href=3D"mailto:trill@ie=
tf.org" target=3D"_blank">trill@ietf.org</a>&quot;
&lt;<a href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ietf.org</a>&=
gt;</font>
</td></tr><tr valign=3D"top">
<td>
<div align=3D"right"><font size=3D"1" face=3D"sans-serif">=D6=F7=CC=E2</fon=
t></div>
</td><td><font size=3D"1" face=3D"sans-serif">Re: [trill] New version : TRI=
LL Smart
Endnodes draft</font></td></tr></tbody></table>
<br>
<table>
<tbody><tr valign=3D"top">
<td>
</td><td></td></tr></tbody></table>
<br></td></tr></tbody></table><div class=3D"HOEnZb"><div class=3D"h5">
<br>
<br>
<br><tt><font>There may also be reasons to communicate other TRILL
Data packet<br>
information with the smart end node, besides the ingress nickname<br>
needed for learning. For example, a smart end node might want to be<br>
able to send and receive TRILL Data packets with fine grained labels.<br>
<br>
Thanks,<br>
Donald<br>
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D<br>
 Donald E. Eastlake 3rd &nbsp; <a href=3D"tel:%2B1-508-333-2270" value=3D"+=
15083332270" target=3D"_blank">+1-508-333-2270</a> (cell)<br>
 155 Beaver Street, Milford, MA 01757 USA<br>
 <a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gmail.com</a>=
<br>
<br>
<br>
On Thu, Jan 10, 2013 at 1:24 PM, Radia Perlman &lt;<a href=3D"mailto:radiap=
erlman@gmail.com" target=3D"_blank">radiaperlman@gmail.com</a>&gt;
wrote:<br>
&gt; Yes, it&#39;s simpler, but it doesn&#39;t work, unless we want to say =
that
smart<br>
&gt; endnodes can never learn from data traffic...they can only work if
they are<br>
&gt; configured with (MAC, egress nickname) of all destinations they might
want<br>
&gt; to talk to, or must have a directory and request from the directory.<b=
r>
&gt;<br>
&gt; I think we need to allow smart endnodes to learn from data traffic,
in<br>
&gt; which, unfortunately (because as you point out, it&#39;s more complica=
ted)
the<br>
&gt; RB has to consciously know whether it&#39;s forwarding to a smart endn=
ode
or<br>
&gt; not, and then leave it encapsulated in that case.<br>
&gt;<br>
&gt; Radia<br>
&gt;<br>
&gt; On Thu, Jan 10, 2013 at 10:03 AM, Linda Dunbar &lt;<a href=3D"mailto:l=
inda.dunbar@huawei.com" target=3D"_blank">linda.dunbar@huawei.com</a>&gt;<b=
r>
&gt; wrote:<br>
&gt;&gt;<br>
&gt;&gt; It is definitely simpler for Egress RBridge to decapsulate all
received<br>
&gt;&gt; data frames:<br>
&gt;&gt;<br>
&gt;&gt; =A1=A4 &nbsp; &nbsp; &nbsp; &nbsp;So it doesn&rsquo;t need determi=
ne if
the target end node is smart &nbsp;or<br>
&gt;&gt; not<br>
&gt;&gt;<br>
&gt;&gt; =A1=A4 &nbsp; &nbsp; &nbsp; &nbsp;It doesn&rsquo;t have to attache=
d different
Ethernet Destination based<br>
&gt;&gt; on the inner header<br>
&gt;&gt;<br>
&gt;&gt; =A1=A4 &nbsp; &nbsp; &nbsp; &nbsp;It is consistent for all data fr=
ames
and all end nodes.<br>
&gt;&gt;<br>
&gt;&gt; =A1=A4 &nbsp; &nbsp; &nbsp; &nbsp;Smart End node can receive nativ=
e
Ethernet frames anyway. It<br>
&gt;&gt; doesn&rsquo;t cause any extra processing on the Smart End node eit=
her.<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; Linda<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; From: Linda Dunbar<br>
&gt;&gt; Sent: Thursday, January 10, 2013 11:43 AM<br>
&gt;&gt; To: &#39;Radia Perlman&#39;; Phanidhar Koganti<br>
&gt;&gt; Cc: <a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@g=
mail.com</a>; Kesava_Vijaya_Krupak@Dell.com;<br>
&gt;&gt; <a href=3D"mailto:hu.fangwei@zte.com.cn" target=3D"_blank">hu.fang=
wei@zte.com.cn</a>; <a href=3D"mailto:trill@ietf.org" target=3D"_blank">tri=
ll@ietf.org</a><br>
&gt;&gt; Subject: RE: [trill] New version : TRILL Smart Endnodes draft<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; Radia,<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; Questions inserted below:<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; a) When the packet reaches the egress RB I am presuming there
will be a<br>
&gt;&gt; inner-MAC lookup to deliver the packet to the egress end-node.<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; correct...when egress RB R has to look up the inner MAC address
E, which<br>
&gt;&gt; it has to do anyway without smart endnodes, in order to choose
which port to<br>
&gt;&gt; forward it on. &nbsp;With smart endnodes, when R looks up MAC
address E, it not<br>
&gt;&gt; only discovers which port to send it on, but that E is &quot;smart=
&quot;,
in which<br>
&gt;&gt; case R must leave the TRILL encapsulation<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; [Linda] Why can&rsquo;t the egress RB R simply decapsulates the TR=
ILL
header and<br>
&gt;&gt; forward the data frame towards the attached end nodes? Regardless
if the DA<br>
&gt;&gt; is smart node or not?<br>
&gt;&gt;<br>
&gt;&gt; So RB processing is much simpler. It doesn&rsquo;t have to attach
different<br>
&gt;&gt; Ethernet header based on the inner DA. The smart node can receive
Ethernet<br>
&gt;&gt; frames anyway (for communication within the directly attached
LAN).<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; Thanks, Linda<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; _______________________________________________<br>
&gt;&gt; trill mailing list<br>
&gt;&gt; <a href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ietf.org=
</a><br>
&gt;&gt; <a href=3D"https://www.ietf.org/mailman/listinfo/trill" target=3D"=
_blank">https://www.ietf.org/mailman/listinfo/trill</a><br>
&gt;&gt;<br>
&gt;<br>
_______________________________________________<br>
trill mailing list<br>
<a href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/trill" target=3D"_blank">h=
ttps://www.ietf.org/mailman/listinfo/trill</a><br>
</font></tt>
<br>
</div></div></blockquote></div><br></div>

--f46d044469ad85c94c04d2f925c8--

From radiaperlman@gmail.com  Thu Jan 10 17:34:47 2013
Return-Path: <radiaperlman@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B132F21F88E4 for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 17:34:47 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.455
X-Spam-Level: 
X-Spam-Status: No, score=-1.455 tagged_above=-999 required=5 tests=[AWL=-1.703, BAYES_00=-2.599, HTML_MESSAGE=0.001, MIME_CHARSET_FARAWAY=2.45, RCVD_IN_DNSWL_LOW=-1, WEIRD_QUOTING=1.396]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 4uEF4w6XiR6I for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 17:34:44 -0800 (PST)
Received: from mail-la0-f43.google.com (mail-la0-f43.google.com [209.85.215.43]) by ietfa.amsl.com (Postfix) with ESMTP id 6D6A921F88D1 for <trill@ietf.org>; Thu, 10 Jan 2013 17:34:43 -0800 (PST)
Received: by mail-la0-f43.google.com with SMTP id eg20so1293053lab.30 for <trill@ietf.org>; Thu, 10 Jan 2013 17:34:42 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=aeoE/iV5O/vsXJpWr/6rOYsxQ7iIV9zGNrY2DujCn2U=; b=02/2BJ4IDslD5l4bql5UoJNU6tc6qMjKeFtQ6Ap/n+AXrqggUfpMo2xoOJno8ECYg/ BvXyTcFYCBm/KIZnHPix+6hQpRbcu5kU6dLw+U5jc47/IPRNh1EP132ljMA9KJGQ9Sbw 73rXRjm8uxqIS2uPeh6GP8EDkydiDhTgKiSw1DFQm8N9WcY6nDSKBgguSwI0VCXd6bU8 MspPkTRBDR/PDv75wVJluR0NRm1iYt19K/0ccRNbSexpgjjWJjVFWx7WTVtR1jfOpWPy TLRW2MTxL3GkzbjFzAvMtcINMA3l7JzqQD+U6qCZFaUQzIct12MJKwkfdi4bl+ckDL6u l6TQ==
MIME-Version: 1.0
Received: by 10.152.125.7 with SMTP id mm7mr71786409lab.2.1357868081958; Thu, 10 Jan 2013 17:34:41 -0800 (PST)
Received: by 10.112.64.17 with HTTP; Thu, 10 Jan 2013 17:34:41 -0800 (PST)
In-Reply-To: <CAF4+nEF890TCnGiXV34eHrNE6kano76QcrR4Q8JJ5X5xxwz+xw@mail.gmail.com>
References: <CAF4+nEHdTX+g-A3kMykr+eaDP3KXf+zwh5fGvQOX7w=4xz31kw@mail.gmail.com> <OF537A9CFF.886515DC-ON48257AF0.00028D26-48257AF0.0004650E@zte.com.cn> <CAF4+nEF890TCnGiXV34eHrNE6kano76QcrR4Q8JJ5X5xxwz+xw@mail.gmail.com>
Date: Thu, 10 Jan 2013 17:34:41 -0800
Message-ID: <CAFOuuo5uQHQjAKb4rss0qr8Az4GBcWJfAhLvo_Yk4fmFxVkcNA@mail.gmail.com>
From: Radia Perlman <radiaperlman@gmail.com>
To: Donald Eastlake <d3e3e3@gmail.com>
Content-Type: multipart/alternative; boundary=f46d043743efad4d5a04d2f94c9c
Cc: zhai.hongjun@zte.com.cn, "trill@ietf.org" <trill@ietf.org>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, Linda Dunbar <linda.dunbar@huawei.com>
Subject: Re: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 11 Jan 2013 01:34:47 -0000

--f46d043743efad4d5a04d2f94c9c
Content-Type: text/plain; charset=GB2312
Content-Transfer-Encoding: quoted-printable

OK.  I've gotten confused here.  If a packet needs an FGL, and it is being
encapsulated by a smart endnode, the smart endnode needs to put on the FGL,
just like it would need to put on a VLAN.

So I'm not sure what the last 4 or 5 messages in the thread are about.

Radia



On Thu, Jan 10, 2013 at 5:23 PM, Donald Eastlake <d3e3e3@gmail.com> wrote:

> I believe FGL support should be announced in Hellos.
>
> I think the main reason for an end station to want to receive/send FGL
> TRILL Data packets is because it wants to handle data packets with a very
> large number of different labels. This strikes me as a bit unusual,
> probably a large gateway or server, so the network would normally be
> configured so that the TRILL switches it is connected with support FGL.
>
> Thanks,
> Donald
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D
>  Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
>  155 Beaver Street, Milford, MA 01757 USA
>  d3e3e3@gmail.com
>
>
> On Thu, Jan 10, 2013 at 7:44 PM, <zhai.hongjun@zte.com.cn> wrote:
>
>>
>> Hi Donald,
>>
>> > There may also be reasons to communicate other TRILL Data packet
>> > information with the smart end node, besides the ingress nickname
>> > needed for learning. For example, a smart end node might want to be
>> > able to send and receive TRILL Data packets with fine grained labels.
>>
>> As smartnodes have no Link State Database information for TRILL campus,
>> they do not know whether all the RBridges in TRILL campus support FGL
>> or not, or which RBridges support FGL and which does not support FGL.
>> Therefore, if a smartnode sends TRILL multi-destination data packets
>> with FGL, the packets might be discarded by some RBridgs that do not
>> support FGL. In this case, how such a sartnode makes sure those FGL
>> packets are able to arrive at their destination(s)?
>>
>>
>>
>> Best Regards,
>> Zhai Hongjun
>> """"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
>> Protocol Development Dept.VI, Central R&D Institute, ZTE Corporation
>> No. 68, Zijinghua Road, Yuhuatai District, Nanjing, P.R.China, 210012
>>
>> Zhai Hongjun
>>
>> Tel: +86-25-52877345
>> Email: zhai.hongjun@zte.com.cn
>> """""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
>>
>>
>>
>>
>>  *Donald Eastlake <d3e3e3@gmail.com>*
>> =B7=A2=BC=FE=C8=CB:  trill-bounces@ietf.org
>>
>> 2013-01-11 04:15
>>   =CA=D5=BC=FE=C8=CB
>> Linda Dunbar <linda.dunbar@huawei.com>
>> =B3=AD=CB=CD
>> Radia Perlman <radiaperlman@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com" =
<
>> Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <
>> hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
>> =D6=F7=CC=E2
>> Re: [trill] New version : TRILL Smart Endnodes draft
>>
>>
>>
>>
>> There may also be reasons to communicate other TRILL Data packet
>> information with the smart end node, besides the ingress nickname
>> needed for learning. For example, a smart end node might want to be
>> able to send and receive TRILL Data packets with fine grained labels.
>>
>> Thanks,
>> Donald
>> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D
>> Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
>> 155 Beaver Street, Milford, MA 01757 USA
>> d3e3e3@gmail.com
>>
>>
>> On Thu, Jan 10, 2013 at 1:24 PM, Radia Perlman <radiaperlman@gmail.com>
>> wrote:
>> > Yes, it's simpler, but it doesn't work, unless we want to say that sma=
rt
>> > endnodes can never learn from data traffic...they can only work if the=
y
>> are
>> > configured with (MAC, egress nickname) of all destinations they might
>> want
>> > to talk to, or must have a directory and request from the directory.
>> >
>> > I think we need to allow smart endnodes to learn from data traffic, in
>> > which, unfortunately (because as you point out, it's more complicated)
>> the
>> > RB has to consciously know whether it's forwarding to a smart endnode =
or
>> > not, and then leave it encapsulated in that case.
>> >
>> > Radia
>> >
>> > On Thu, Jan 10, 2013 at 10:03 AM, Linda Dunbar <linda.dunbar@huawei.co=
m
>> >
>> > wrote:
>> >>
>> >> It is definitely simpler for Egress RBridge to decapsulate all receiv=
ed
>> >> data frames:
>> >>
>> >> =A1=A4        So it doesn=A1=AFt need determine if the target end nod=
e is smart
>>  or
>> >> not
>> >>
>> >> =A1=A4        It doesn=A1=AFt have to attached different Ethernet Des=
tination
>> based
>> >> on the inner header
>> >>
>> >> =A1=A4        It is consistent for all data frames and all end nodes.
>> >>
>> >> =A1=A4        Smart End node can receive native Ethernet frames anywa=
y. It
>> >> doesn=A1=AFt cause any extra processing on the Smart End node either.
>> >>
>> >>
>> >>
>> >> Linda
>> >>
>> >>
>> >>
>> >> From: Linda Dunbar
>> >> Sent: Thursday, January 10, 2013 11:43 AM
>> >> To: 'Radia Perlman'; Phanidhar Koganti
>> >> Cc: d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com;
>> >> hu.fangwei@zte.com.cn; trill@ietf.org
>> >> Subject: RE: [trill] New version : TRILL Smart Endnodes draft
>> >>
>> >>
>> >>
>> >> Radia,
>> >>
>> >>
>> >>
>> >> Questions inserted below:
>> >>
>> >>
>> >>
>> >>
>> >>
>> >> a) When the packet reaches the egress RB I am presuming there will be=
 a
>> >> inner-MAC lookup to deliver the packet to the egress end-node.
>> >>
>> >>
>> >>
>> >> correct...when egress RB R has to look up the inner MAC address E,
>> which
>> >> it has to do anyway without smart endnodes, in order to choose which
>> port to
>> >> forward it on.  With smart endnodes, when R looks up MAC address E, i=
t
>> not
>> >> only discovers which port to send it on, but that E is "smart", in
>> which
>> >> case R must leave the TRILL encapsulation
>> >>
>> >>
>> >>
>> >> [Linda] Why can=A1=AFt the egress RB R simply decapsulates the TRILL =
header
>> and
>> >> forward the data frame towards the attached end nodes? Regardless if
>> the DA
>> >> is smart node or not?
>> >>
>> >> So RB processing is much simpler. It doesn=A1=AFt have to attach diff=
erent
>> >> Ethernet header based on the inner DA. The smart node can receive
>> Ethernet
>> >> frames anyway (for communication within the directly attached LAN).
>> >>
>> >>
>> >>
>> >> Thanks, Linda
>> >>
>> >>
>> >>
>> >>
>> >> _______________________________________________
>> >> trill mailing list
>> >> trill@ietf.org
>> >> https://www.ietf.org/mailman/listinfo/trill
>> >>
>> >
>> _______________________________________________
>> trill mailing list
>> trill@ietf.org
>> https://www.ietf.org/mailman/listinfo/trill
>>
>>
>

--f46d043743efad4d5a04d2f94c9c
Content-Type: text/html; charset=GB2312
Content-Transfer-Encoding: quoted-printable

OK. &nbsp;I&#39;ve gotten confused here. &nbsp;If a packet needs an FGL, an=
d it is being encapsulated by a smart endnode, the smart endnode needs to p=
ut on the FGL, just like it would need to put on a VLAN.<div><br></div><div=
>So I&#39;m not sure what the last 4 or 5 messages in the thread are about.=
</div>
<div><br></div><div>Radia</div><div><br></div><div><br><br><div class=3D"gm=
ail_quote">On Thu, Jan 10, 2013 at 5:23 PM, Donald Eastlake <span dir=3D"lt=
r">&lt;<a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gmail.c=
om</a>&gt;</span> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">I believe FGL support should be announced in=
 Hellos.<div><br></div><div>I think the main reason for an end station to w=
ant to receive/send FGL TRILL Data packets is because it wants to handle da=
ta packets with a very large number of different labels. This strikes me as=
 a bit unusual, probably a large gateway or server, so the network would no=
rmally be configured so that the TRILL switches it is connected with suppor=
t FGL.</div>


<div><div class=3D"im"><br clear=3D"all"><div>Thanks,<br>Donald<br>=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D<br>&nbsp;Donald E. Eastlake 3rd&nbsp;&nbsp; <a href=3D"tel:%2B1-508-=
333-2270" value=3D"+15083332270" target=3D"_blank">+1-508-333-2270</a> (cel=
l)<br>
&nbsp;155 Beaver Street,&nbsp;Milford, MA 01757 USA<br>&nbsp;<a href=3D"mai=
lto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gmail.com</a></div>


<br><br></div><div><div class=3D"h5"><div class=3D"gmail_quote">On Thu, Jan=
 10, 2013 at 7:44 PM,  <span dir=3D"ltr">&lt;<a href=3D"mailto:zhai.hongjun=
@zte.com.cn" target=3D"_blank">zhai.hongjun@zte.com.cn</a>&gt;</span> wrote=
:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">


<br><font face=3D"sans-serif">Hi Donald,</font>
<br><div>
<br><tt><font>&gt; There may also be reasons to communicate other
TRILL Data packet<br>
&gt; information with the smart end node, besides the ingress nickname<br>
&gt; needed for learning. For example, a smart end node might want to be<br=
>
&gt; able to send and receive TRILL Data packets with fine grained labels.<=
/font></tt>
<br>
<br></div><tt><font>As smartnodes have no Link State Database information
for TRILL campus, </font></tt>
<br><tt><font>they do not know whether all the RBridges in TRILL
campus support FGL </font></tt>
<br><tt><font>or not, or which RBridges support FGL and which does
not support FGL.</font></tt>
<br><tt><font>Therefore, if a smartnode sends TRILL multi-destination
data packets </font></tt>
<br><tt><font>with FGL, the packets might be discarded by some RBridgs
that do not </font></tt>
<br><tt><font>support FGL. In this case, how such a sartnode makes
sure those FGL </font></tt>
<br><tt><font>packets are able to arrive at their destination(s)?</font></t=
t>
<br><tt><font>&nbsp;<br>
</font></tt><font face=3D"sans-serif"><br>
<br>
Best Regards,<br>
Zhai Hongjun<br>
&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&qu=
ot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;=
&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&qu=
ot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;=
&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&qu=
ot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;<br>



 Protocol Development Dept.VI, Central R&amp;D Institute, ZTE Corporation<b=
r>
 No. 68, Zijinghua Road, Yuhuatai District, Nanjing, P.R.China, 210012<br>
 <br>
 Zhai Hongjun<br>
 <br>
 Tel: <a href=3D"tel:%2B86-25-52877345" value=3D"+862552877345" target=3D"_=
blank">+86-25-52877345</a><br>
 Email: <a href=3D"mailto:zhai.hongjun@zte.com.cn" target=3D"_blank">zhai.h=
ongjun@zte.com.cn</a><br>
&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&qu=
ot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;=
&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&qu=
ot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;=
&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&qu=
ot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;<br>



<br>
</font>
<br>
<br>
<br>
<p></p><table width=3D"100%">
<tbody><tr valign=3D"top">
<td width=3D"36%"><font size=3D"1" face=3D"sans-serif"><b>Donald Eastlake &=
lt;<a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gmail.com</=
a>&gt;</b>
</font>
<br><font size=3D"1" face=3D"sans-serif">=B7=A2=BC=FE=C8=CB: &nbsp;<a href=
=3D"mailto:trill-bounces@ietf.org" target=3D"_blank">trill-bounces@ietf.org=
</a></font>
<p><font size=3D"1" face=3D"sans-serif">2013-01-11 04:15</font>
</p></td><td width=3D"63%">
<table width=3D"100%">
<tbody><tr valign=3D"top">
<td>
<div align=3D"right"><font size=3D"1" face=3D"sans-serif">=CA=D5=BC=FE=C8=
=CB</font></div>
</td><td><font size=3D"1" face=3D"sans-serif">Linda Dunbar &lt;<a href=3D"m=
ailto:linda.dunbar@huawei.com" target=3D"_blank">linda.dunbar@huawei.com</a=
>&gt;</font>
</td></tr><tr valign=3D"top">
<td>
<div align=3D"right"><font size=3D"1" face=3D"sans-serif">=B3=AD=CB=CD</fon=
t></div>
</td><td><font size=3D"1" face=3D"sans-serif">Radia Perlman &lt;<a href=3D"=
mailto:radiaperlman@gmail.com" target=3D"_blank">radiaperlman@gmail.com</a>=
&gt;,
&quot;Kesava_Vijaya_Krupak@Dell.com&quot; &lt;<a href=3D"mailto:Kesava_Vija=
ya_Krupak@dell.com" target=3D"_blank">Kesava_Vijaya_Krupak@dell.com</a>&gt;=
,
&quot;<a href=3D"mailto:hu.fangwei@zte.com.cn" target=3D"_blank">hu.fangwei=
@zte.com.cn</a>&quot; &lt;<a href=3D"mailto:hu.fangwei@zte.com.cn" target=
=3D"_blank">hu.fangwei@zte.com.cn</a>&gt;, &quot;<a href=3D"mailto:trill@ie=
tf.org" target=3D"_blank">trill@ietf.org</a>&quot;
&lt;<a href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ietf.org</a>&=
gt;</font>
</td></tr><tr valign=3D"top">
<td>
<div align=3D"right"><font size=3D"1" face=3D"sans-serif">=D6=F7=CC=E2</fon=
t></div>
</td><td><font size=3D"1" face=3D"sans-serif">Re: [trill] New version : TRI=
LL Smart
Endnodes draft</font></td></tr></tbody></table>
<br>
<table>
<tbody><tr valign=3D"top">
<td>
</td><td></td></tr></tbody></table>
<br></td></tr></tbody></table><div><div>
<br>
<br>
<br><tt><font>There may also be reasons to communicate other TRILL
Data packet<br>
information with the smart end node, besides the ingress nickname<br>
needed for learning. For example, a smart end node might want to be<br>
able to send and receive TRILL Data packets with fine grained labels.<br>
<br>
Thanks,<br>
Donald<br>
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D<br>
 Donald E. Eastlake 3rd &nbsp; <a href=3D"tel:%2B1-508-333-2270" value=3D"+=
15083332270" target=3D"_blank">+1-508-333-2270</a> (cell)<br>
 155 Beaver Street, Milford, MA 01757 USA<br>
 <a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gmail.com</a>=
<br>
<br>
<br>
On Thu, Jan 10, 2013 at 1:24 PM, Radia Perlman &lt;<a href=3D"mailto:radiap=
erlman@gmail.com" target=3D"_blank">radiaperlman@gmail.com</a>&gt;
wrote:<br>
&gt; Yes, it&#39;s simpler, but it doesn&#39;t work, unless we want to say =
that
smart<br>
&gt; endnodes can never learn from data traffic...they can only work if
they are<br>
&gt; configured with (MAC, egress nickname) of all destinations they might
want<br>
&gt; to talk to, or must have a directory and request from the directory.<b=
r>
&gt;<br>
&gt; I think we need to allow smart endnodes to learn from data traffic,
in<br>
&gt; which, unfortunately (because as you point out, it&#39;s more complica=
ted)
the<br>
&gt; RB has to consciously know whether it&#39;s forwarding to a smart endn=
ode
or<br>
&gt; not, and then leave it encapsulated in that case.<br>
&gt;<br>
&gt; Radia<br>
&gt;<br>
&gt; On Thu, Jan 10, 2013 at 10:03 AM, Linda Dunbar &lt;<a href=3D"mailto:l=
inda.dunbar@huawei.com" target=3D"_blank">linda.dunbar@huawei.com</a>&gt;<b=
r>
&gt; wrote:<br>
&gt;&gt;<br>
&gt;&gt; It is definitely simpler for Egress RBridge to decapsulate all
received<br>
&gt;&gt; data frames:<br>
&gt;&gt;<br>
&gt;&gt; =A1=A4 &nbsp; &nbsp; &nbsp; &nbsp;So it doesn&rsquo;t need determi=
ne if
the target end node is smart &nbsp;or<br>
&gt;&gt; not<br>
&gt;&gt;<br>
&gt;&gt; =A1=A4 &nbsp; &nbsp; &nbsp; &nbsp;It doesn&rsquo;t have to attache=
d different
Ethernet Destination based<br>
&gt;&gt; on the inner header<br>
&gt;&gt;<br>
&gt;&gt; =A1=A4 &nbsp; &nbsp; &nbsp; &nbsp;It is consistent for all data fr=
ames
and all end nodes.<br>
&gt;&gt;<br>
&gt;&gt; =A1=A4 &nbsp; &nbsp; &nbsp; &nbsp;Smart End node can receive nativ=
e
Ethernet frames anyway. It<br>
&gt;&gt; doesn&rsquo;t cause any extra processing on the Smart End node eit=
her.<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; Linda<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; From: Linda Dunbar<br>
&gt;&gt; Sent: Thursday, January 10, 2013 11:43 AM<br>
&gt;&gt; To: &#39;Radia Perlman&#39;; Phanidhar Koganti<br>
&gt;&gt; Cc: <a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@g=
mail.com</a>; Kesava_Vijaya_Krupak@Dell.com;<br>
&gt;&gt; <a href=3D"mailto:hu.fangwei@zte.com.cn" target=3D"_blank">hu.fang=
wei@zte.com.cn</a>; <a href=3D"mailto:trill@ietf.org" target=3D"_blank">tri=
ll@ietf.org</a><br>
&gt;&gt; Subject: RE: [trill] New version : TRILL Smart Endnodes draft<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; Radia,<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; Questions inserted below:<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; a) When the packet reaches the egress RB I am presuming there
will be a<br>
&gt;&gt; inner-MAC lookup to deliver the packet to the egress end-node.<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; correct...when egress RB R has to look up the inner MAC address
E, which<br>
&gt;&gt; it has to do anyway without smart endnodes, in order to choose
which port to<br>
&gt;&gt; forward it on. &nbsp;With smart endnodes, when R looks up MAC
address E, it not<br>
&gt;&gt; only discovers which port to send it on, but that E is &quot;smart=
&quot;,
in which<br>
&gt;&gt; case R must leave the TRILL encapsulation<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; [Linda] Why can&rsquo;t the egress RB R simply decapsulates the TR=
ILL
header and<br>
&gt;&gt; forward the data frame towards the attached end nodes? Regardless
if the DA<br>
&gt;&gt; is smart node or not?<br>
&gt;&gt;<br>
&gt;&gt; So RB processing is much simpler. It doesn&rsquo;t have to attach
different<br>
&gt;&gt; Ethernet header based on the inner DA. The smart node can receive
Ethernet<br>
&gt;&gt; frames anyway (for communication within the directly attached
LAN).<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; Thanks, Linda<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; _______________________________________________<br>
&gt;&gt; trill mailing list<br>
&gt;&gt; <a href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ietf.org=
</a><br>
&gt;&gt; <a href=3D"https://www.ietf.org/mailman/listinfo/trill" target=3D"=
_blank">https://www.ietf.org/mailman/listinfo/trill</a><br>
&gt;&gt;<br>
&gt;<br>
_______________________________________________<br>
trill mailing list<br>
<a href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/trill" target=3D"_blank">h=
ttps://www.ietf.org/mailman/listinfo/trill</a><br>
</font></tt>
<br>
</div></div></blockquote></div><br></div></div></div>
</blockquote></div><br></div>

--f46d043743efad4d5a04d2f94c9c--

From d3e3e3@gmail.com  Thu Jan 10 17:48:35 2013
Return-Path: <d3e3e3@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 23D9721F88EE for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 17:48:35 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -101.423
X-Spam-Level: 
X-Spam-Status: No, score=-101.423 tagged_above=-999 required=5 tests=[AWL=-1.671, BAYES_00=-2.599, HTML_MESSAGE=0.001, MIME_CHARSET_FARAWAY=2.45, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100, WEIRD_QUOTING=1.396]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id bRYbAvLphWMG for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 17:48:31 -0800 (PST)
Received: from mail-oa0-f47.google.com (mail-oa0-f47.google.com [209.85.219.47]) by ietfa.amsl.com (Postfix) with ESMTP id 9B25F21F88E1 for <trill@ietf.org>; Thu, 10 Jan 2013 17:48:30 -0800 (PST)
Received: by mail-oa0-f47.google.com with SMTP id h1so1301431oag.6 for <trill@ietf.org>; Thu, 10 Jan 2013 17:48:30 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=di/U4PFWViuD1i1SN8OxBcJrRUcaWx1o26qByOfLqRk=; b=huqDx68me0ZSKA8hr/Oc18g+fIvwF36oS4iFU+fVoEM8mDsmPglyxAhPdXTyoeFCw5 y5wO2WxWZJSgRUpAEZwgsnIou0hN9OmvmbVVMFy/tPxOrOkj2xxq7FjXFCJegJN5vdhs 2ldYz40bpUJ7//7wU0fakYi4E7ie0tm5KR9DDQwkuYVAbSA9FWj6k6PnHtNfeIpv4/vc t3WFxLhrbtwqdVKZf5L4nx6RXsQ0ZhQrLqm0mCTbBXF6jA0N4mo/14li2WKpBoUQoxbW xjOLSbWXJm3YHMtFaz44eAQQkyRX4kt+P2KMRvYAf3TA9ywHQhQoQO/x4cF1FJYcM1Rg oYIA==
Received: by 10.60.169.207 with SMTP id ag15mr42689148oec.120.1357868909957; Thu, 10 Jan 2013 17:48:29 -0800 (PST)
MIME-Version: 1.0
Received: by 10.76.110.239 with HTTP; Thu, 10 Jan 2013 17:48:09 -0800 (PST)
In-Reply-To: <CAFOuuo5uQHQjAKb4rss0qr8Az4GBcWJfAhLvo_Yk4fmFxVkcNA@mail.gmail.com>
References: <CAF4+nEHdTX+g-A3kMykr+eaDP3KXf+zwh5fGvQOX7w=4xz31kw@mail.gmail.com> <OF537A9CFF.886515DC-ON48257AF0.00028D26-48257AF0.0004650E@zte.com.cn> <CAF4+nEF890TCnGiXV34eHrNE6kano76QcrR4Q8JJ5X5xxwz+xw@mail.gmail.com> <CAFOuuo5uQHQjAKb4rss0qr8Az4GBcWJfAhLvo_Yk4fmFxVkcNA@mail.gmail.com>
From: Donald Eastlake <d3e3e3@gmail.com>
Date: Thu, 10 Jan 2013 20:48:09 -0500
Message-ID: <CAF4+nEF7UGuK+tdSXBpV6S_ovUo1_JUYwfSWmDddtTzys+EcaA@mail.gmail.com>
To: Radia Perlman <radiaperlman@gmail.com>
Content-Type: multipart/alternative; boundary=bcaec550afdc07912404d2f97e18
Cc: zhai.hongjun@zte.com.cn, "trill@ietf.org" <trill@ietf.org>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, Linda Dunbar <linda.dunbar@huawei.com>
Subject: Re: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 11 Jan 2013 01:48:35 -0000

--bcaec550afdc07912404d2f97e18
Content-Type: text/plain; charset=GB2312
Content-Transfer-Encoding: quoted-printable

Maybe I should have used a TRILL Header extension as an example of
additional information that can be included by sending the TRILL Data
packet rather than egressing.

Some the latest messages in this thread are because all TRILL switches are
required to support VLANs but not necessarily fine grained labels. So if a
smart endnote wanted to use FGL TRILL Data frames and was connected to both
VL and FGL TRILL switches, it would want to know which was which.

Thanks,
Donald
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D
 Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
 155 Beaver Street, Milford, MA 01757 USA
 d3e3e3@gmail.com


On Thu, Jan 10, 2013 at 8:34 PM, Radia Perlman <radiaperlman@gmail.com>wrot=
e:

> OK.  I've gotten confused here.  If a packet needs an FGL, and it is bein=
g
> encapsulated by a smart endnode, the smart endnode needs to put on the FG=
L,
> just like it would need to put on a VLAN.
>
> So I'm not sure what the last 4 or 5 messages in the thread are about.
>
> Radia
>
>
>
> On Thu, Jan 10, 2013 at 5:23 PM, Donald Eastlake <d3e3e3@gmail.com> wrote=
:
>
>> I believe FGL support should be announced in Hellos.
>>
>> I think the main reason for an end station to want to receive/send FGL
>> TRILL Data packets is because it wants to handle data packets with a ver=
y
>> large number of different labels. This strikes me as a bit unusual,
>> probably a large gateway or server, so the network would normally be
>> configured so that the TRILL switches it is connected with support FGL.
>>
>> Thanks,
>> Donald
>> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D
>>  Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
>>  155 Beaver Street, Milford, MA 01757 USA
>>  d3e3e3@gmail.com
>>
>>
>> On Thu, Jan 10, 2013 at 7:44 PM, <zhai.hongjun@zte.com.cn> wrote:
>>
>>>
>>> Hi Donald,
>>>
>>> > There may also be reasons to communicate other TRILL Data packet
>>> > information with the smart end node, besides the ingress nickname
>>> > needed for learning. For example, a smart end node might want to be
>>> > able to send and receive TRILL Data packets with fine grained labels.
>>>
>>> As smartnodes have no Link State Database information for TRILL campus,
>>> they do not know whether all the RBridges in TRILL campus support FGL
>>> or not, or which RBridges support FGL and which does not support FGL.
>>> Therefore, if a smartnode sends TRILL multi-destination data packets
>>> with FGL, the packets might be discarded by some RBridgs that do not
>>> support FGL. In this case, how such a sartnode makes sure those FGL
>>> packets are able to arrive at their destination(s)?
>>>
>>>
>>>
>>> Best Regards,
>>> Zhai Hongjun
>>> """"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
>>> Protocol Development Dept.VI, Central R&D Institute, ZTE Corporation
>>> No. 68, Zijinghua Road, Yuhuatai District, Nanjing, P.R.China, 210012
>>>
>>> Zhai Hongjun
>>>
>>> Tel: +86-25-52877345
>>> Email: zhai.hongjun@zte.com.cn
>>> """""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
>>>
>>>
>>>
>>>
>>>  *Donald Eastlake <d3e3e3@gmail.com>*
>>> =B7=A2=BC=FE=C8=CB:  trill-bounces@ietf.org
>>>
>>> 2013-01-11 04:15
>>>   =CA=D5=BC=FE=C8=CB
>>> Linda Dunbar <linda.dunbar@huawei.com>
>>> =B3=AD=CB=CD
>>> Radia Perlman <radiaperlman@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com"
>>> <Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <
>>> hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
>>> =D6=F7=CC=E2
>>> Re: [trill] New version : TRILL Smart Endnodes draft
>>>
>>>
>>>
>>>
>>> There may also be reasons to communicate other TRILL Data packet
>>> information with the smart end node, besides the ingress nickname
>>> needed for learning. For example, a smart end node might want to be
>>> able to send and receive TRILL Data packets with fine grained labels.
>>>
>>> Thanks,
>>> Donald
>>> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D
>>> Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
>>> 155 Beaver Street, Milford, MA 01757 USA
>>> d3e3e3@gmail.com
>>>
>>>
>>> On Thu, Jan 10, 2013 at 1:24 PM, Radia Perlman <radiaperlman@gmail.com>
>>> wrote:
>>> > Yes, it's simpler, but it doesn't work, unless we want to say that
>>> smart
>>> > endnodes can never learn from data traffic...they can only work if
>>> they are
>>> > configured with (MAC, egress nickname) of all destinations they might
>>> want
>>> > to talk to, or must have a directory and request from the directory.
>>> >
>>> > I think we need to allow smart endnodes to learn from data traffic, i=
n
>>> > which, unfortunately (because as you point out, it's more complicated=
)
>>> the
>>> > RB has to consciously know whether it's forwarding to a smart endnode
>>> or
>>> > not, and then leave it encapsulated in that case.
>>> >
>>> > Radia
>>> >
>>> > On Thu, Jan 10, 2013 at 10:03 AM, Linda Dunbar <
>>> linda.dunbar@huawei.com>
>>> > wrote:
>>> >>
>>> >> It is definitely simpler for Egress RBridge to decapsulate all
>>> received
>>> >> data frames:
>>> >>
>>> >> =A1=A4        So it doesn=A1=AFt need determine if the target end no=
de is smart
>>>  or
>>> >> not
>>> >>
>>> >> =A1=A4        It doesn=A1=AFt have to attached different Ethernet De=
stination
>>> based
>>> >> on the inner header
>>> >>
>>> >> =A1=A4        It is consistent for all data frames and all end nodes=
.
>>> >>
>>> >> =A1=A4        Smart End node can receive native Ethernet frames anyw=
ay. It
>>> >> doesn=A1=AFt cause any extra processing on the Smart End node either=
.
>>> >>
>>> >>
>>> >>
>>> >> Linda
>>> >>
>>> >>
>>> >>
>>> >> From: Linda Dunbar
>>> >> Sent: Thursday, January 10, 2013 11:43 AM
>>> >> To: 'Radia Perlman'; Phanidhar Koganti
>>> >> Cc: d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com;
>>> >> hu.fangwei@zte.com.cn; trill@ietf.org
>>> >> Subject: RE: [trill] New version : TRILL Smart Endnodes draft
>>> >>
>>> >>
>>> >>
>>> >> Radia,
>>> >>
>>> >>
>>> >>
>>> >> Questions inserted below:
>>> >>
>>> >>
>>> >>
>>> >>
>>> >>
>>> >> a) When the packet reaches the egress RB I am presuming there will b=
e
>>> a
>>> >> inner-MAC lookup to deliver the packet to the egress end-node.
>>> >>
>>> >>
>>> >>
>>> >> correct...when egress RB R has to look up the inner MAC address E,
>>> which
>>> >> it has to do anyway without smart endnodes, in order to choose which
>>> port to
>>> >> forward it on.  With smart endnodes, when R looks up MAC address E,
>>> it not
>>> >> only discovers which port to send it on, but that E is "smart", in
>>> which
>>> >> case R must leave the TRILL encapsulation
>>> >>
>>> >>
>>> >>
>>> >> [Linda] Why can=A1=AFt the egress RB R simply decapsulates the TRILL
>>> header and
>>> >> forward the data frame towards the attached end nodes? Regardless if
>>> the DA
>>> >> is smart node or not?
>>> >>
>>> >> So RB processing is much simpler. It doesn=A1=AFt have to attach dif=
ferent
>>> >> Ethernet header based on the inner DA. The smart node can receive
>>> Ethernet
>>> >> frames anyway (for communication within the directly attached LAN).
>>> >>
>>> >>
>>> >>
>>> >> Thanks, Linda
>>> >>
>>> >>
>>> >>
>>> >>
>>> >> _______________________________________________
>>> >> trill mailing list
>>> >> trill@ietf.org
>>> >> https://www.ietf.org/mailman/listinfo/trill
>>> >>
>>> >
>>> _______________________________________________
>>> trill mailing list
>>> trill@ietf.org
>>> https://www.ietf.org/mailman/listinfo/trill
>>>
>>>
>>
>

--bcaec550afdc07912404d2f97e18
Content-Type: text/html; charset=GB2312
Content-Transfer-Encoding: quoted-printable

Maybe I should have used a TRILL Header extension as an example of addition=
al information that can be included by sending the TRILL Data packet rather=
 than egressing.<div><br></div><div>Some the latest messages in this thread=
 are because all TRILL switches are required to support VLANs but not neces=
sarily fine grained labels. So if a smart endnote wanted to use FGL TRILL D=
ata frames and was connected to both VL and FGL TRILL switches, it would wa=
nt to know which was which.</div>

<div><br clear=3D"all"><div>Thanks,<br>Donald<br>=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D<br>&nbsp;Do=
nald E. Eastlake 3rd&nbsp;&nbsp; +1-508-333-2270 (cell)<br>&nbsp;155 Beaver=
 Street,&nbsp;Milford, MA 01757 USA<br>&nbsp;<a href=3D"mailto:d3e3e3@gmail=
.com" target=3D"_blank">d3e3e3@gmail.com</a></div>


<br><br><div class=3D"gmail_quote">On Thu, Jan 10, 2013 at 8:34 PM, Radia P=
erlman <span dir=3D"ltr">&lt;<a href=3D"mailto:radiaperlman@gmail.com" targ=
et=3D"_blank">radiaperlman@gmail.com</a>&gt;</span> wrote:<br><blockquote c=
lass=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;=
padding-left:1ex">

OK. &nbsp;I&#39;ve gotten confused here. &nbsp;If a packet needs an FGL, an=
d it is being encapsulated by a smart endnode, the smart endnode needs to p=
ut on the FGL, just like it would need to put on a VLAN.<div><br></div><div=
>So I&#39;m not sure what the last 4 or 5 messages in the thread are about.=
</div>

<span class=3D"HOEnZb"><font color=3D"#888888">
<div><br></div><div>Radia</div></font></span><div class=3D"HOEnZb"><div cla=
ss=3D"h5"><div><br></div><div><br><br><div class=3D"gmail_quote">On Thu, Ja=
n 10, 2013 at 5:23 PM, Donald Eastlake <span dir=3D"ltr">&lt;<a href=3D"mai=
lto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gmail.com</a>&gt;</span> wro=
te:<br>


<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">I believe FGL support should be announced in=
 Hellos.<div><br></div><div>I think the main reason for an end station to w=
ant to receive/send FGL TRILL Data packets is because it wants to handle da=
ta packets with a very large number of different labels. This strikes me as=
 a bit unusual, probably a large gateway or server, so the network would no=
rmally be configured so that the TRILL switches it is connected with suppor=
t FGL.</div>




<div><div><br clear=3D"all"><div>Thanks,<br>Donald<br>=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D<br>&nbsp=
;Donald E. Eastlake 3rd&nbsp;&nbsp; <a href=3D"tel:%2B1-508-333-2270" value=
=3D"+15083332270" target=3D"_blank">+1-508-333-2270</a> (cell)<br>
&nbsp;155 Beaver Street,&nbsp;Milford, MA 01757 USA<br>&nbsp;<a href=3D"mai=
lto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gmail.com</a></div>


<br><br></div><div><div><div class=3D"gmail_quote">On Thu, Jan 10, 2013 at =
7:44 PM,  <span dir=3D"ltr">&lt;<a href=3D"mailto:zhai.hongjun@zte.com.cn" =
target=3D"_blank">zhai.hongjun@zte.com.cn</a>&gt;</span> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">


<br><font face=3D"sans-serif">Hi Donald,</font>
<br><div>
<br><tt><font>&gt; There may also be reasons to communicate other
TRILL Data packet<br>
&gt; information with the smart end node, besides the ingress nickname<br>
&gt; needed for learning. For example, a smart end node might want to be<br=
>
&gt; able to send and receive TRILL Data packets with fine grained labels.<=
/font></tt>
<br>
<br></div><tt><font>As smartnodes have no Link State Database information
for TRILL campus, </font></tt>
<br><tt><font>they do not know whether all the RBridges in TRILL
campus support FGL </font></tt>
<br><tt><font>or not, or which RBridges support FGL and which does
not support FGL.</font></tt>
<br><tt><font>Therefore, if a smartnode sends TRILL multi-destination
data packets </font></tt>
<br><tt><font>with FGL, the packets might be discarded by some RBridgs
that do not </font></tt>
<br><tt><font>support FGL. In this case, how such a sartnode makes
sure those FGL </font></tt>
<br><tt><font>packets are able to arrive at their destination(s)?</font></t=
t>
<br><tt><font>&nbsp;<br>
</font></tt><font face=3D"sans-serif"><br>
<br>
Best Regards,<br>
Zhai Hongjun<br>
&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&qu=
ot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;=
&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&qu=
ot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;=
&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&qu=
ot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;<br>





 Protocol Development Dept.VI, Central R&amp;D Institute, ZTE Corporation<b=
r>
 No. 68, Zijinghua Road, Yuhuatai District, Nanjing, P.R.China, 210012<br>
 <br>
 Zhai Hongjun<br>
 <br>
 Tel: <a href=3D"tel:%2B86-25-52877345" value=3D"+862552877345" target=3D"_=
blank">+86-25-52877345</a><br>
 Email: <a href=3D"mailto:zhai.hongjun@zte.com.cn" target=3D"_blank">zhai.h=
ongjun@zte.com.cn</a><br>
&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&qu=
ot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;=
&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&qu=
ot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;=
&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&qu=
ot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;&quot;<br>





<br>
</font>
<br>
<br>
<br>
<p></p><table width=3D"100%">
<tbody><tr valign=3D"top">
<td width=3D"36%"><font size=3D"1" face=3D"sans-serif"><b>Donald Eastlake &=
lt;<a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gmail.com</=
a>&gt;</b>
</font>
<br><font size=3D"1" face=3D"sans-serif">=B7=A2=BC=FE=C8=CB: &nbsp;<a href=
=3D"mailto:trill-bounces@ietf.org" target=3D"_blank">trill-bounces@ietf.org=
</a></font>
<p><font size=3D"1" face=3D"sans-serif">2013-01-11 04:15</font>
</p></td><td width=3D"63%">
<table width=3D"100%">
<tbody><tr valign=3D"top">
<td>
<div align=3D"right"><font size=3D"1" face=3D"sans-serif">=CA=D5=BC=FE=C8=
=CB</font></div>
</td><td><font size=3D"1" face=3D"sans-serif">Linda Dunbar &lt;<a href=3D"m=
ailto:linda.dunbar@huawei.com" target=3D"_blank">linda.dunbar@huawei.com</a=
>&gt;</font>
</td></tr><tr valign=3D"top">
<td>
<div align=3D"right"><font size=3D"1" face=3D"sans-serif">=B3=AD=CB=CD</fon=
t></div>
</td><td><font size=3D"1" face=3D"sans-serif">Radia Perlman &lt;<a href=3D"=
mailto:radiaperlman@gmail.com" target=3D"_blank">radiaperlman@gmail.com</a>=
&gt;,
&quot;Kesava_Vijaya_Krupak@Dell.com&quot; &lt;<a href=3D"mailto:Kesava_Vija=
ya_Krupak@dell.com" target=3D"_blank">Kesava_Vijaya_Krupak@dell.com</a>&gt;=
,
&quot;<a href=3D"mailto:hu.fangwei@zte.com.cn" target=3D"_blank">hu.fangwei=
@zte.com.cn</a>&quot; &lt;<a href=3D"mailto:hu.fangwei@zte.com.cn" target=
=3D"_blank">hu.fangwei@zte.com.cn</a>&gt;, &quot;<a href=3D"mailto:trill@ie=
tf.org" target=3D"_blank">trill@ietf.org</a>&quot;
&lt;<a href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ietf.org</a>&=
gt;</font>
</td></tr><tr valign=3D"top">
<td>
<div align=3D"right"><font size=3D"1" face=3D"sans-serif">=D6=F7=CC=E2</fon=
t></div>
</td><td><font size=3D"1" face=3D"sans-serif">Re: [trill] New version : TRI=
LL Smart
Endnodes draft</font></td></tr></tbody></table>
<br>
<table>
<tbody><tr valign=3D"top">
<td>
</td><td></td></tr></tbody></table>
<br></td></tr></tbody></table><div><div>
<br>
<br>
<br><tt><font>There may also be reasons to communicate other TRILL
Data packet<br>
information with the smart end node, besides the ingress nickname<br>
needed for learning. For example, a smart end node might want to be<br>
able to send and receive TRILL Data packets with fine grained labels.<br>
<br>
Thanks,<br>
Donald<br>
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D<br>
 Donald E. Eastlake 3rd &nbsp; <a href=3D"tel:%2B1-508-333-2270" value=3D"+=
15083332270" target=3D"_blank">+1-508-333-2270</a> (cell)<br>
 155 Beaver Street, Milford, MA 01757 USA<br>
 <a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gmail.com</a>=
<br>
<br>
<br>
On Thu, Jan 10, 2013 at 1:24 PM, Radia Perlman &lt;<a href=3D"mailto:radiap=
erlman@gmail.com" target=3D"_blank">radiaperlman@gmail.com</a>&gt;
wrote:<br>
&gt; Yes, it&#39;s simpler, but it doesn&#39;t work, unless we want to say =
that
smart<br>
&gt; endnodes can never learn from data traffic...they can only work if
they are<br>
&gt; configured with (MAC, egress nickname) of all destinations they might
want<br>
&gt; to talk to, or must have a directory and request from the directory.<b=
r>
&gt;<br>
&gt; I think we need to allow smart endnodes to learn from data traffic,
in<br>
&gt; which, unfortunately (because as you point out, it&#39;s more complica=
ted)
the<br>
&gt; RB has to consciously know whether it&#39;s forwarding to a smart endn=
ode
or<br>
&gt; not, and then leave it encapsulated in that case.<br>
&gt;<br>
&gt; Radia<br>
&gt;<br>
&gt; On Thu, Jan 10, 2013 at 10:03 AM, Linda Dunbar &lt;<a href=3D"mailto:l=
inda.dunbar@huawei.com" target=3D"_blank">linda.dunbar@huawei.com</a>&gt;<b=
r>
&gt; wrote:<br>
&gt;&gt;<br>
&gt;&gt; It is definitely simpler for Egress RBridge to decapsulate all
received<br>
&gt;&gt; data frames:<br>
&gt;&gt;<br>
&gt;&gt; =A1=A4 &nbsp; &nbsp; &nbsp; &nbsp;So it doesn&rsquo;t need determi=
ne if
the target end node is smart &nbsp;or<br>
&gt;&gt; not<br>
&gt;&gt;<br>
&gt;&gt; =A1=A4 &nbsp; &nbsp; &nbsp; &nbsp;It doesn&rsquo;t have to attache=
d different
Ethernet Destination based<br>
&gt;&gt; on the inner header<br>
&gt;&gt;<br>
&gt;&gt; =A1=A4 &nbsp; &nbsp; &nbsp; &nbsp;It is consistent for all data fr=
ames
and all end nodes.<br>
&gt;&gt;<br>
&gt;&gt; =A1=A4 &nbsp; &nbsp; &nbsp; &nbsp;Smart End node can receive nativ=
e
Ethernet frames anyway. It<br>
&gt;&gt; doesn&rsquo;t cause any extra processing on the Smart End node eit=
her.<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; Linda<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; From: Linda Dunbar<br>
&gt;&gt; Sent: Thursday, January 10, 2013 11:43 AM<br>
&gt;&gt; To: &#39;Radia Perlman&#39;; Phanidhar Koganti<br>
&gt;&gt; Cc: <a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@g=
mail.com</a>; Kesava_Vijaya_Krupak@Dell.com;<br>
&gt;&gt; <a href=3D"mailto:hu.fangwei@zte.com.cn" target=3D"_blank">hu.fang=
wei@zte.com.cn</a>; <a href=3D"mailto:trill@ietf.org" target=3D"_blank">tri=
ll@ietf.org</a><br>
&gt;&gt; Subject: RE: [trill] New version : TRILL Smart Endnodes draft<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; Radia,<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; Questions inserted below:<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; a) When the packet reaches the egress RB I am presuming there
will be a<br>
&gt;&gt; inner-MAC lookup to deliver the packet to the egress end-node.<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; correct...when egress RB R has to look up the inner MAC address
E, which<br>
&gt;&gt; it has to do anyway without smart endnodes, in order to choose
which port to<br>
&gt;&gt; forward it on. &nbsp;With smart endnodes, when R looks up MAC
address E, it not<br>
&gt;&gt; only discovers which port to send it on, but that E is &quot;smart=
&quot;,
in which<br>
&gt;&gt; case R must leave the TRILL encapsulation<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; [Linda] Why can&rsquo;t the egress RB R simply decapsulates the TR=
ILL
header and<br>
&gt;&gt; forward the data frame towards the attached end nodes? Regardless
if the DA<br>
&gt;&gt; is smart node or not?<br>
&gt;&gt;<br>
&gt;&gt; So RB processing is much simpler. It doesn&rsquo;t have to attach
different<br>
&gt;&gt; Ethernet header based on the inner DA. The smart node can receive
Ethernet<br>
&gt;&gt; frames anyway (for communication within the directly attached
LAN).<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; Thanks, Linda<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; _______________________________________________<br>
&gt;&gt; trill mailing list<br>
&gt;&gt; <a href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ietf.org=
</a><br>
&gt;&gt; <a href=3D"https://www.ietf.org/mailman/listinfo/trill" target=3D"=
_blank">https://www.ietf.org/mailman/listinfo/trill</a><br>
&gt;&gt;<br>
&gt;<br>
_______________________________________________<br>
trill mailing list<br>
<a href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/trill" target=3D"_blank">h=
ttps://www.ietf.org/mailman/listinfo/trill</a><br>
</font></tt>
<br>
</div></div></blockquote></div><br></div></div></div>
</blockquote></div><br></div>
</div></div></blockquote></div><br></div>

--bcaec550afdc07912404d2f97e18--

From zhai.hongjun@zte.com.cn  Thu Jan 10 18:32:49 2013
Return-Path: <zhai.hongjun@zte.com.cn>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 83C5621F8941 for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 18:32:49 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -98.056
X-Spam-Level: 
X-Spam-Status: No, score=-98.056 tagged_above=-999 required=5 tests=[AWL=2.546, BAYES_00=-2.599, HTML_MESSAGE=0.001, J_CHICKENPOX_92=0.6, USER_IN_WHITELIST=-100, WEIRD_QUOTING=1.396]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id cTiHMA80b-UE for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 18:32:48 -0800 (PST)
Received: from zte.com.cn (mx6.zte.com.cn [95.130.199.165]) by ietfa.amsl.com (Postfix) with ESMTP id 5646D21F8939 for <trill@ietf.org>; Thu, 10 Jan 2013 18:32:47 -0800 (PST)
Received: from zte.com.cn (unknown [192.168.168.119]) by Websense Email Security Gateway with ESMTP id 4DB292E3A for <trill@ietf.org>; Fri, 11 Jan 2013 10:32:34 +0800 (CST)
Received: from mse02.zte.com.cn (unknown [10.30.3.21]) by Websense Email Security Gateway with ESMTPS id A9602DF4939; Fri, 11 Jan 2013 10:22:25 +0800 (CST)
Received: from notes_smtp.zte.com.cn ([10.30.1.239]) by mse02.zte.com.cn with ESMTP id r0B2WSie059259; Fri, 11 Jan 2013 10:32:28 +0800 (GMT-8) (envelope-from zhai.hongjun@zte.com.cn)
In-Reply-To: <CAF4+nEF7UGuK+tdSXBpV6S_ovUo1_JUYwfSWmDddtTzys+EcaA@mail.gmail.com>
To: Donald Eastlake <d3e3e3@gmail.com>
MIME-Version: 1.0
X-KeepSent: 32C21609:022566FE-48257AF0:000A5535; type=4; name=$KeepSent
X-Mailer: Lotus Notes Release 6.5.6 March 06, 2007
Message-ID: <OF32C21609.022566FE-ON48257AF0.000A5535-48257AF0.000E42B1@zte.com.cn>
From: zhai.hongjun@zte.com.cn
Date: Fri, 11 Jan 2013 10:32:32 +0800
X-MIMETrack: Serialize by Router on notes_smtp/zte_ltd(Release 8.5.3FP1 HF212|May 23, 2012) at 2013-01-11 10:32:22, Serialize complete at 2013-01-11 10:32:22
Content-Type: multipart/alternative; boundary="=_alternative 000E42A648257AF0_="
X-MAIL: mse02.zte.com.cn r0B2WSie059259
Cc: "trill@ietf.org" <trill@ietf.org>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, Radia Perlman <radiaperlman@gmail.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, Linda Dunbar <linda.dunbar@huawei.com>
Subject: Re: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 11 Jan 2013 02:32:49 -0000

This is a multipart message in MIME format.
--=_alternative 000E42A648257AF0_=
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: base64
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--=_alternative 000E42A648257AF0_=
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: base64
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--=_alternative 000E42A648257AF0_=--

From d3e3e3@gmail.com  Thu Jan 10 18:55:15 2013
Return-Path: <d3e3e3@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 39FDB21F87F3 for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 18:55:15 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -100.972
X-Spam-Level: 
X-Spam-Status: No, score=-100.972 tagged_above=-999 required=5 tests=[AWL=-1.819, BAYES_00=-2.599, J_CHICKENPOX_92=0.6, MIME_CHARSET_FARAWAY=2.45, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100, WEIRD_QUOTING=1.396]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id yMl5i-8MwAqR for <trill@ietfa.amsl.com>; Thu, 10 Jan 2013 18:55:14 -0800 (PST)
Received: from mail-oa0-f48.google.com (mail-oa0-f48.google.com [209.85.219.48]) by ietfa.amsl.com (Postfix) with ESMTP id 2040F21F87ED for <trill@ietf.org>; Thu, 10 Jan 2013 18:55:14 -0800 (PST)
Received: by mail-oa0-f48.google.com with SMTP id h2so1329753oag.7 for <trill@ietf.org>; Thu, 10 Jan 2013 18:55:13 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type:content-transfer-encoding; bh=em/xKbfGKTiQl+bBkXQRrMAE5qN+0G39JxGxKaLhLkQ=; b=Nvu1YaxPIbXc9Hp5Q3q5Qh9n+c4qnN1IFPQ6as7q5ERx8u2LAX3wi6fmcK3X0z6+Z3 DlZqPa+7MGMnlnFKxN1+9++HyX8SHlJ1FC8ajkPV31TYTT6RNxv69sMN+5mMS6LlvNXm WFIMNQrb+TQaFUaM7DrQpudPECPAS+h7If5qeuIIibArgMELKfLdAA2AyNNwtsABvh68 f6prB0qPrka+XWHPUnez3paGy9tAZAyZZc+ikpvWazJ5Jj33tdyWOCpnvRB/rVRVkHyc 2wEyrAiHpB+rVD4tWeh5a3sKh9fwPT9cMfM8TRyfFh8w8dt72TPbJTSgAPqYnSX1kIV2 Jg2w==
Received: by 10.60.169.240 with SMTP id ah16mr44742144oec.9.1357872913520; Thu, 10 Jan 2013 18:55:13 -0800 (PST)
MIME-Version: 1.0
Received: by 10.76.110.239 with HTTP; Thu, 10 Jan 2013 18:54:53 -0800 (PST)
In-Reply-To: <OF32C21609.022566FE-ON48257AF0.000A5535-48257AF0.000E42B1@zte.com.cn>
References: <CAF4+nEF7UGuK+tdSXBpV6S_ovUo1_JUYwfSWmDddtTzys+EcaA@mail.gmail.com> <OF32C21609.022566FE-ON48257AF0.000A5535-48257AF0.000E42B1@zte.com.cn>
From: Donald Eastlake <d3e3e3@gmail.com>
Date: Thu, 10 Jan 2013 21:54:53 -0500
Message-ID: <CAF4+nEEjLXvJRVDV6N5hgo6EaFFRBN66ufFJqW7mnhdZvA31HQ@mail.gmail.com>
To: zhai.hongjun@zte.com.cn
Content-Type: text/plain; charset=GB2312
Content-Transfer-Encoding: quoted-printable
Cc: "trill@ietf.org" <trill@ietf.org>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, Radia Perlman <radiaperlman@gmail.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, Linda Dunbar <linda.dunbar@huawei.com>
Subject: Re: [trill] New version : TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 11 Jan 2013 02:55:15 -0000

Hi,

On Thu, Jan 10, 2013 at 9:32 PM, <zhai.hongjun@zte.com.cn> wrote:
>
> > Some the latest messages in this thread are because all TRILL switches =
are required to
> > support VLANs but not necessarily fine grained labels. So if a smart en=
dnote wanted to
> > use FGL TRILL Data frames and was connected to both VL and FGL TRILL sw=
itches, it would
> > want to know which was which.
>
> Maybe I have not made my meaning clear, so I try to make it clearer using=
 an example.
>
> Assume E is an smartnode connected to TRILL campus through RB1. There are=
 2,000
> RBridges in TRILL campus where 1,000 of them support FGL, and the remaind=
er 1,000
> Rbridges do not support FGL. In this scenario, when E encapsulates a non-=
unicast
> frame into TRILL data frames with FGL and transmit in TRILL, it MUST know=
 exactly
> which RBridges support FGL, then it has to TRILL unicast such a frame to =
all
> relevant egress RBridges (of the remainder 1,000 RBridges). Otherwise,if =
it does not
> know the relevant egress RBridges, it has to employ a tree to multicast t=
he frame.
> And that frame will be discarded by non-FGL RBridges and not able to reac=
h the
> interested end nodes

Under the current FGL draft, there is no data topology adjacency
between FGL and VL TRILL switches. So the distribution tree or trees
built by FGL TRILL switches will be bounded by the contiguous area of
FGL switches. Sending a multi-destination FGL TRILL Data frame on such
a tree will get to every FGL TRILL switch in that area without passing
through any VL TRILL switches. So flooding unknown unicast or using a
distribution tree for multicast should work fine.

Thanks,
Donald
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D
 Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
 155 Beaver Street, Milford, MA 01757 USA
 d3e3e3@gmail.com

> So if E wants to send TRILL Data packets(including multi-destination pack=
ets) with fine
> grained labels, it must learn which RBridges support FGL, which do not su=
pport FGL.
> Maybe RB1 can tell E the 1,000 FGL-RBridges by its hellos to E1 if Hello =
packet can
> accommodate so many RBridges nicknames.
>
> Best Regards,
> Zhai Hongjun
> """"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
> Protocol Development Dept.VI, Central R&D Institute, ZTE Corporation
> No. 68, Zijinghua Road, Yuhuatai District, Nanjing, P.R.China, 210012
>
> Zhai Hongjun
>
> Tel: +86-25-52877345
> Email: zhai.hongjun@zte.com.cn
> """""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
>
>
>
>
> Donald Eastlake <d3e3e3@gmail.com>
>
> 2013-01-11 09:48
>
> =CA=D5=BC=FE=C8=CB
> Radia Perlman <radiaperlman@gmail.com>
> =B3=AD=CB=CD
> zhai.hongjun@zte.com.cn, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>,=
 "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, Linda Dun=
bar <linda.dunbar@huawei.com>, "trill@ietf.org" <trill@ietf.org>
> =D6=F7=CC=E2
> Re: [trill] New version : TRILL Smart Endnodes draft
>
>
>
>
>
> Maybe I should have used a TRILL Header extension as an example of additi=
onal information that can be included by sending the TRILL Data packet rath=
er than egressing.
>
> Some the latest messages in this thread are because all TRILL switches ar=
e required to support VLANs but not necessarily fine grained labels. So if =
a smart endnote wanted to use FGL TRILL Data frames and was connected to bo=
th VL and FGL TRILL switches, it would want to know which was which.
>
> Thanks,
> Donald
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D
> Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
> 155 Beaver Street, Milford, MA 01757 USA
> d3e3e3@gmail.com
>
>
> On Thu, Jan 10, 2013 at 8:34 PM, Radia Perlman <radiaperlman@gmail.com> w=
rote:
> OK.  I've gotten confused here.  If a packet needs an FGL, and it is bein=
g encapsulated by a smart endnode, the smart endnode needs to put on the FG=
L, just like it would need to put on a VLAN.
>
> So I'm not sure what the last 4 or 5 messages in the thread are about.
>
> Radia
>
>
>
> On Thu, Jan 10, 2013 at 5:23 PM, Donald Eastlake <d3e3e3@gmail.com> wrote=
:
> I believe FGL support should be announced in Hellos.
>
> I think the main reason for an end station to want to receive/send FGL TR=
ILL Data packets is because it wants to handle data packets with a very lar=
ge number of different labels. This strikes me as a bit unusual, probably a=
 large gateway or server, so the network would normally be configured so th=
at the TRILL switches it is connected with support FGL.
>
> Thanks,
> Donald
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D
> Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
> 155 Beaver Street, Milford, MA 01757 USA
> d3e3e3@gmail.com
>
>
> On Thu, Jan 10, 2013 at 7:44 PM, <zhai.hongjun@zte.com.cn> wrote:
>
> Hi Donald,
>
> > There may also be reasons to communicate other TRILL Data packet
> > information with the smart end node, besides the ingress nickname
> > needed for learning. For example, a smart end node might want to be
> > able to send and receive TRILL Data packets with fine grained labels.
>
> As smartnodes have no Link State Database information for TRILL campus,
> they do not know whether all the RBridges in TRILL campus support FGL
> or not, or which RBridges support FGL and which does not support FGL.
> Therefore, if a smartnode sends TRILL multi-destination data packets
> with FGL, the packets might be discarded by some RBridgs that do not
> support FGL. In this case, how such a sartnode makes sure those FGL
> packets are able to arrive at their destination(s)?
>
>
>
> Best Regards,
> Zhai Hongjun
> """"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
> Protocol Development Dept.VI, Central R&D Institute, ZTE Corporation
> No. 68, Zijinghua Road, Yuhuatai District, Nanjing, P.R.China, 210012
>
> Zhai Hongjun
>
> Tel: +86-25-52877345
> Email: zhai.hongjun@zte.com.cn
> """""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
>
>
>
> Donald Eastlake <d3e3e3@gmail.com>
> =B7=A2=BC=FE=C8=CB:  trill-bounces@ietf.org
>
> 2013-01-11 04:15
>
>
> =CA=D5=BC=FE=C8=CB
> Linda Dunbar <linda.dunbar@huawei.com>
> =B3=AD=CB=CD
> Radia Perlman <radiaperlman@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com" <=
Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com=
.cn>, "trill@ietf.org" <trill@ietf.org>
> =D6=F7=CC=E2
> Re: [trill] New version : TRILL Smart Endnodes draft
>
>
>
>
>
>
>
> There may also be reasons to communicate other TRILL Data packet
> information with the smart end node, besides the ingress nickname
> needed for learning. For example, a smart end node might want to be
> able to send and receive TRILL Data packets with fine grained labels.
>
> Thanks,
> Donald
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D
> Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
> 155 Beaver Street, Milford, MA 01757 USA
> d3e3e3@gmail.com
>
>
> On Thu, Jan 10, 2013 at 1:24 PM, Radia Perlman <radiaperlman@gmail.com> w=
rote:
> > Yes, it's simpler, but it doesn't work, unless we want to say that smar=
t
> > endnodes can never learn from data traffic...they can only work if they=
 are
> > configured with (MAC, egress nickname) of all destinations they might w=
ant
> > to talk to, or must have a directory and request from the directory.
> >
> > I think we need to allow smart endnodes to learn from data traffic, in
> > which, unfortunately (because as you point out, it's more complicated) =
the
> > RB has to consciously know whether it's forwarding to a smart endnode o=
r
> > not, and then leave it encapsulated in that case.
> >
> > Radia
> >
> > On Thu, Jan 10, 2013 at 10:03 AM, Linda Dunbar <linda.dunbar@huawei.com=
>
> > wrote:
> >>
> >> It is definitely simpler for Egress RBridge to decapsulate all receive=
d
> >> data frames:
> >>
> >> =A1=A4        So it doesn=A1=AFt need determine if the target end node=
 is smart  or
> >> not
> >>
> >> =A1=A4        It doesn=A1=AFt have to attached different Ethernet Dest=
ination based
> >> on the inner header
> >>
> >> =A1=A4        It is consistent for all data frames and all end nodes.
> >>
> >> =A1=A4        Smart End node can receive native Ethernet frames anyway=
. It
> >> doesn=A1=AFt cause any extra processing on the Smart End node either.
> >>
> >>
> >>
> >> Linda
> >>
> >>
> >>
> >> From: Linda Dunbar
> >> Sent: Thursday, January 10, 2013 11:43 AM
> >> To: 'Radia Perlman'; Phanidhar Koganti
> >> Cc: d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com;
> >> hu.fangwei@zte.com.cn; trill@ietf.org
> >> Subject: RE: [trill] New version : TRILL Smart Endnodes draft
> >>
> >>
> >>
> >> Radia,
> >>
> >>
> >>
> >> Questions inserted below:
> >>
> >>
> >>
> >>
> >>
> >> a) When the packet reaches the egress RB I am presuming there will be =
a
> >> inner-MAC lookup to deliver the packet to the egress end-node.
> >>
> >>
> >>
> >> correct...when egress RB R has to look up the inner MAC address E, whi=
ch
> >> it has to do anyway without smart endnodes, in order to choose which p=
ort to
> >> forward it on.  With smart endnodes, when R looks up MAC address E, it=
 not
> >> only discovers which port to send it on, but that E is "smart", in whi=
ch
> >> case R must leave the TRILL encapsulation
> >>
> >>
> >>
> >> [Linda] Why can=A1=AFt the egress RB R simply decapsulates the TRILL h=
eader and
> >> forward the data frame towards the attached end nodes? Regardless if t=
he DA
> >> is smart node or not?
> >>
> >> So RB processing is much simpler. It doesn=A1=AFt have to attach diffe=
rent
> >> Ethernet header based on the inner DA. The smart node can receive Ethe=
rnet
> >> frames anyway (for communication within the directly attached LAN).
> >>
> >>
> >>
> >> Thanks, Linda
> >>
> >>
> >>
> >>
> >> _______________________________________________
> >> trill mailing list
> >> trill@ietf.org
> >> https://www.ietf.org/mailman/listinfo/trill
> >>
> >
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill
>
>
>
>

From linda.dunbar@huawei.com  Fri Jan 11 12:21:36 2013
Return-Path: <linda.dunbar@huawei.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 55BD721F8A93 for <trill@ietfa.amsl.com>; Fri, 11 Jan 2013 12:21:35 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.498
X-Spam-Level: 
X-Spam-Status: No, score=-6.498 tagged_above=-999 required=5 tests=[AWL=0.100,  BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id kXhKHzA5O9on for <trill@ietfa.amsl.com>; Fri, 11 Jan 2013 12:21:33 -0800 (PST)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) by ietfa.amsl.com (Postfix) with ESMTP id 225A021F89FD for <trill@ietf.org>; Fri, 11 Jan 2013 12:21:27 -0800 (PST)
Received: from 172.18.7.190 (EHLO lhreml204-edg.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.5-GA FastPath queued) with ESMTP id AOQ87584; Fri, 11 Jan 2013 20:21:26 +0000 (GMT)
Received: from LHREML406-HUB.china.huawei.com (10.201.5.243) by lhreml204-edg.china.huawei.com (172.18.7.223) with Microsoft SMTP Server (TLS) id 14.1.323.3; Fri, 11 Jan 2013 20:20:19 +0000
Received: from DFWEML408-HUB.china.huawei.com (10.193.5.134) by lhreml406-hub.china.huawei.com (10.201.5.243) with Microsoft SMTP Server (TLS) id 14.1.323.3; Fri, 11 Jan 2013 20:21:23 +0000
Received: from DFWEML505-MBX.china.huawei.com ([10.124.31.100]) by dfweml408-hub.china.huawei.com ([10.193.5.134]) with mapi id 14.01.0323.003; Fri, 11 Jan 2013 12:21:19 -0800
From: Linda Dunbar <linda.dunbar@huawei.com>
To: Radia Perlman <radiaperlman@gmail.com>
Thread-Topic: [trill] Questions to TRILL Smart Endnodes draft
Thread-Index: AQHN7rkrS7AADs7TvEa0mW1AYaXweZhEk6ZQ
Date: Fri, 11 Jan 2013 20:21:19 +0000
Message-ID: <4A95BA014132FF49AE685FAB4B9F17F6450E0B9A@dfweml505-mbx>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com> <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com> <6895EAE0CA8DEE40B92D7CA88BB521F32414007BC1@HQ1-EXCH02.corp.brocade.com> <CAFOuuo626hw-c0kM4-1Fp-w2gByPenRnkFP3xNT+cTB4YcJ2kg@mail.gmail.com> <4A95BA014132FF49AE685FAB4B9F17F6450E01E4@dfweml505-mbx> <CAFOuuo6=wDeKiG-zjSnkcO3B2iTbazpsGHNYJLSfOxYZsh9kFA@mail.gmail.com>
In-Reply-To: <CAFOuuo6=wDeKiG-zjSnkcO3B2iTbazpsGHNYJLSfOxYZsh9kFA@mail.gmail.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.192.11.236]
Content-Type: multipart/alternative; boundary="_000_4A95BA014132FF49AE685FAB4B9F17F6450E0B9Adfweml505mbx_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Cc: "d3e3e3@gmail.com" <d3e3e3@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Questions to TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 11 Jan 2013 20:21:36 -0000

--_000_4A95BA014132FF49AE685FAB4B9F17F6450E0B9Adfweml505mbx_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Radia,

Questions inserted below:

From: Radia Perlman [mailto:radiaperlman@gmail.com]
Sent: Wednesday, January 09, 2013 4:32 PM
To: Linda Dunbar
Cc: Phanidhar Koganti; d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com; hu.=
fangwei@zte.com.cn; trill@ietf.org
Subject: Re: [trill] Questions to TRILL Smart Endnodes draft


If there is a smart endnode on a shared link with other endnodes, it learns=
 about its endnode neighbors, and sends directly to them (without encapsula=
tion).  If smart E wants to talk to unknown destination D, E has to send tw=
o copies; one (without encapsulation), multicast on the link, and the other=
 (with encapsulation), forwarded to the attached RBridge R.
[Linda] Agree

R has been warned by E of all the MAC addresses E owns.  R must drop all pa=
ckets it receives on the link if either the source or destination MAC is on=
e of the ones owned by a smart endnode on the link.

[Linda] What if multiple "smart endnode" sharing one link? For example, the=
re could be a bridge aggregating traffic from multiple end nodes (smart and=
 regular), or the multiple smart nodes as VMs sharing a physical server?

Linda


--_000_4A95BA014132FF49AE685FAB4B9F17F6450E0B9Adfweml505mbx_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micr=
osoft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:x=3D"urn:schemas-microsoft-com:office:excel" xmlns:p=3D"urn:schemas-m=
icrosoft-com:office:powerpoint" xmlns:a=3D"urn:schemas-microsoft-com:office=
:access" xmlns:dt=3D"uuid:C2F41010-65B3-11d1-A29F-00AA00C14882" xmlns:s=3D"=
uuid:BDC6E3F0-6DA3-11d1-A2A3-00AA00C14882" xmlns:rs=3D"urn:schemas-microsof=
t-com:rowset" xmlns:z=3D"#RowsetSchema" xmlns:b=3D"urn:schemas-microsoft-co=
m:office:publisher" xmlns:ss=3D"urn:schemas-microsoft-com:office:spreadshee=
t" xmlns:c=3D"urn:schemas-microsoft-com:office:component:spreadsheet" xmlns=
:odc=3D"urn:schemas-microsoft-com:office:odc" xmlns:oa=3D"urn:schemas-micro=
soft-com:office:activation" xmlns:html=3D"http://www.w3.org/TR/REC-html40" =
xmlns:q=3D"http://schemas.xmlsoap.org/soap/envelope/" xmlns:rtc=3D"http://m=
icrosoft.com/officenet/conferencing" xmlns:D=3D"DAV:" xmlns:Repl=3D"http://=
schemas.microsoft.com/repl/" xmlns:mt=3D"http://schemas.microsoft.com/share=
point/soap/meetings/" xmlns:x2=3D"http://schemas.microsoft.com/office/excel=
/2003/xml" xmlns:ppda=3D"http://www.passport.com/NameSpace.xsd" xmlns:ois=
=3D"http://schemas.microsoft.com/sharepoint/soap/ois/" xmlns:dir=3D"http://=
schemas.microsoft.com/sharepoint/soap/directory/" xmlns:ds=3D"http://www.w3=
.org/2000/09/xmldsig#" xmlns:dsp=3D"http://schemas.microsoft.com/sharepoint=
/dsp" xmlns:udc=3D"http://schemas.microsoft.com/data/udc" xmlns:xsd=3D"http=
://www.w3.org/2001/XMLSchema" xmlns:sub=3D"http://schemas.microsoft.com/sha=
repoint/soap/2002/1/alerts/" xmlns:ec=3D"http://www.w3.org/2001/04/xmlenc#"=
 xmlns:sp=3D"http://schemas.microsoft.com/sharepoint/" xmlns:sps=3D"http://=
schemas.microsoft.com/sharepoint/soap/" xmlns:xsi=3D"http://www.w3.org/2001=
/XMLSchema-instance" xmlns:udcs=3D"http://schemas.microsoft.com/data/udc/so=
ap" xmlns:udcxf=3D"http://schemas.microsoft.com/data/udc/xmlfile" xmlns:udc=
p2p=3D"http://schemas.microsoft.com/data/udc/parttopart" xmlns:wf=3D"http:/=
/schemas.microsoft.com/sharepoint/soap/workflow/" xmlns:dsss=3D"http://sche=
mas.microsoft.com/office/2006/digsig-setup" xmlns:dssi=3D"http://schemas.mi=
crosoft.com/office/2006/digsig" xmlns:mdssi=3D"http://schemas.openxmlformat=
s.org/package/2006/digital-signature" xmlns:mver=3D"http://schemas.openxmlf=
ormats.org/markup-compatibility/2006" xmlns:m=3D"http://schemas.microsoft.c=
om/office/2004/12/omml" xmlns:mrels=3D"http://schemas.openxmlformats.org/pa=
ckage/2006/relationships" xmlns:spwp=3D"http://microsoft.com/sharepoint/web=
partpages" xmlns:ex12t=3D"http://schemas.microsoft.com/exchange/services/20=
06/types" xmlns:ex12m=3D"http://schemas.microsoft.com/exchange/services/200=
6/messages" xmlns:pptsl=3D"http://schemas.microsoft.com/sharepoint/soap/Sli=
deLibrary/" xmlns:spsl=3D"http://microsoft.com/webservices/SharePointPortal=
Server/PublishedLinksService" xmlns:Z=3D"urn:schemas-microsoft-com:" xmlns:=
st=3D"&#1;" xmlns=3D"http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Dus-ascii"=
>
<meta name=3D"Generator" content=3D"Microsoft Word 12 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
	{font-family:SimSun;
	panose-1:2 1 6 0 3 1 1 1 1 1;}
@font-face
	{font-family:"Cambria Math";
	panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:Tahoma;
	panose-1:2 11 6 4 3 5 4 4 2 4;}
@font-face
	{font-family:SimSun;
	panose-1:2 1 6 0 3 1 1 1 1 1;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman","serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
span.EmailStyle17
	{mso-style-type:personal-reply;
	font-family:"Calibri","sans-serif";
	color:#1F497D;}
.MsoChpDefault
	{mso-style-type:export-only;}
@page WordSection1
	{size:8.5in 11.0in;
	margin:1.0in 1.25in 1.0in 1.25in;}
div.WordSection1
	{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang=3D"EN-US" link=3D"blue" vlink=3D"purple">
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">Radia,
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">Questions inserted below:=
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<div style=3D"border:none;border-left:solid blue 1.5pt;padding:0in 0in 0in =
4.0pt">
<div>
<div style=3D"border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:10.0pt;font-family:&quot=
;Tahoma&quot;,&quot;sans-serif&quot;">From:</span></b><span style=3D"font-s=
ize:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;"> Radia Pe=
rlman [mailto:radiaperlman@gmail.com]
<br>
<b>Sent:</b> Wednesday, January 09, 2013 4:32 PM<br>
<b>To:</b> Linda Dunbar<br>
<b>Cc:</b> Phanidhar Koganti; d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.c=
om; hu.fangwei@zte.com.cn; trill@ietf.org<br>
<b>Subject:</b> Re: [trill] Questions to TRILL Smart Endnodes draft<o:p></o=
:p></span></p>
</div>
</div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal">If there is a smart endnode on a shared link with ot=
her endnodes, it learns about its endnode neighbors, and sends directly to =
them (without encapsulation). &nbsp;If smart E wants to talk to unknown des=
tination D, E has to send two copies; one
 (without encapsulation), multicast on the link, and the other (with encaps=
ulation), forwarded to the attached RBridge R.<o:p></o:p></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">[Linda] Agree<o:p></o:p><=
/span></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal">R has been warned by E of all the MAC addresses E ow=
ns. &nbsp;R must drop all packets it receives on the link if either the sou=
rce or destination MAC is one of the ones owned by a smart endnode on the l=
ink.<o:p></o:p></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">[Linda] What if multiple =
&#8220;smart endnode&#8221; sharing one link? For example, there could be a=
 bridge aggregating traffic from multiple end nodes (smart and regular),
 or the multiple smart nodes as VMs sharing a physical server? <o:p></o:p><=
/span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">Linda
<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
</div>
</div>
</body>
</html>

--_000_4A95BA014132FF49AE685FAB4B9F17F6450E0B9Adfweml505mbx_--

From radiaperlman@gmail.com  Fri Jan 11 12:37:51 2013
Return-Path: <radiaperlman@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5C17E21F871C for <trill@ietfa.amsl.com>; Fri, 11 Jan 2013 12:37:51 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.264
X-Spam-Level: 
X-Spam-Status: No, score=-3.264 tagged_above=-999 required=5 tests=[AWL=0.334,  BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id xbl7rPsj4SCu for <trill@ietfa.amsl.com>; Fri, 11 Jan 2013 12:37:50 -0800 (PST)
Received: from mail-la0-f46.google.com (mail-la0-f46.google.com [209.85.215.46]) by ietfa.amsl.com (Postfix) with ESMTP id E40CA21F872C for <trill@ietf.org>; Fri, 11 Jan 2013 12:37:49 -0800 (PST)
Received: by mail-la0-f46.google.com with SMTP id fq13so2141286lab.5 for <trill@ietf.org>; Fri, 11 Jan 2013 12:37:48 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=b7fd+aH5d9e8SnYVzHOeoUIgoOFvG408aBWaqsFaU/M=; b=jSDTUQ0+l2w6h1ckU6qWdBxTMiUuJ3zWn4ortH3mdaZkHyYVWL1+alH3MR75H/hu3l bRlYEBFH04AiGqQJUaYAS8Fi895Kerxfzq438GR3IEh15HPFu+mmuMQVsbtvvGThXloI uWm5RlA4mw/jBC2dSJC3hy+Ydmwd5NYvC7I8rUIW9MnxRemTrLYk2H7qgAiuhIeJr4Su 2n8dYvhyjugLdEfKkBcAHmAY/YB0OXkUjHZJrrcEeW1BqEWwmtWpIS1qObkLSjC5ckxm cYpFLypmfh6j2058MMduyq8KlLpCjMRz5LY6PHPiytd6sQqpZYOWym7DLU5KenCvWJgz 9g8w==
MIME-Version: 1.0
Received: by 10.152.123.49 with SMTP id lx17mr22056261lab.52.1357936668662; Fri, 11 Jan 2013 12:37:48 -0800 (PST)
Received: by 10.112.64.17 with HTTP; Fri, 11 Jan 2013 12:37:48 -0800 (PST)
In-Reply-To: <4A95BA014132FF49AE685FAB4B9F17F6450E0B9A@dfweml505-mbx>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com> <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com> <6895EAE0CA8DEE40B92D7CA88BB521F32414007BC1@HQ1-EXCH02.corp.brocade.com> <CAFOuuo626hw-c0kM4-1Fp-w2gByPenRnkFP3xNT+cTB4YcJ2kg@mail.gmail.com> <4A95BA014132FF49AE685FAB4B9F17F6450E01E4@dfweml505-mbx> <CAFOuuo6=wDeKiG-zjSnkcO3B2iTbazpsGHNYJLSfOxYZsh9kFA@mail.gmail.com> <4A95BA014132FF49AE685FAB4B9F17F6450E0B9A@dfweml505-mbx>
Date: Fri, 11 Jan 2013 12:37:48 -0800
Message-ID: <CAFOuuo5kVnBw3ODP-GfpF9hhE=VK2rKFMaF_WHf82zWP2PR=Tg@mail.gmail.com>
From: Radia Perlman <radiaperlman@gmail.com>
To: Linda Dunbar <linda.dunbar@huawei.com>
Content-Type: multipart/alternative; boundary=f46d044270d0c35bc204d3094484
Cc: "d3e3e3@gmail.com" <d3e3e3@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Questions to TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 11 Jan 2013 20:37:51 -0000

--f46d044270d0c35bc204d3094484
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

I believe that the smart endnode draft already discusses that.  It
considers having a shared link that has any collection of smart and not
smart endnodes.

Endnodes on that link will talk to each other using non-encapsulated
frames, whether or not they are smart endnodes.

So for a few examples, let D1 and D2 be "not smart" endnodes, and S1 and S2
be "smart" endnodes.  And let R be the attached RBridge.

D1 and D2 always transmit and receive native (non-encapsulated).  If R does
not know where D2 is, then R might encapsulate and flood a packet D1 sends
to D2, but once R learns (from seeing traffic from D2 on the link) that D2
is on the link, R will drop native packets it sees on the link with
destination D2.

Now let S1 want to talk to D2.  (As stated in the draft), if S1 does not
know where D2 is, S1 must send two copies:
1) native on the link, in case D2 is there
2) encapsulated as a to-be-flooded packet to R, in case D2 is elsewhere on
the campus

If D2 wants to talk to S1, D2 just sends it native, and S1 receives it just
fine (S1 has to be able to receive native as well as encapsulated packets).

R drops the packet because S1 has informed R of all of the MAC addresses
that S1 owns.

Now let's say S1 wants to talk to S2.  If S1 has received traffic from S2,
S1 will know that S2 is on the link, and will only send a single copy to
S2; a native copy.

 If S1 does not know where S2 is, it sends two copies; one encapsulated,
and one native.  R has already learned S2's MAC addresses, so R knows it
should drop the packet.  S2, however, will receive two copies of the
packet; one native, and one encapsulated, and S2 will receive both of them.
 But S2 will learn that S1 is on the link, and when it sends something to
S1, S1 will learn that S2 is on the link.

So once in a blue moon, a smart endnode might receive two copies of a
packet, from another smart node on the link, until they learn about the
other one.  I suppose we could have the smart endnodes on the link listen
to the Hellos of other smart endnodes, in order to learn all the MAC
addresses on the link owned by smart endnodes, in order to avoid the
once-in-a-blue-moon receipt of a double packet, but I prefer not adding
that complexity.

Radia


On Fri, Jan 11, 2013 at 12:21 PM, Linda Dunbar <linda.dunbar@huawei.com>wro=
te:

>  Radia, ****
>
> ** **
>
> Questions inserted below:****
>
> ** **
>
> *From:* Radia Perlman [mailto:radiaperlman@gmail.com]
> *Sent:* Wednesday, January 09, 2013 4:32 PM
> *To:* Linda Dunbar
> *Cc:* Phanidhar Koganti; d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com;
> hu.fangwei@zte.com.cn; trill@ietf.org
> *Subject:* Re: [trill] Questions to TRILL Smart Endnodes draft****
>
> ** **
>
> ** **
>
> If there is a smart endnode on a shared link with other endnodes, it
> learns about its endnode neighbors, and sends directly to them (without
> encapsulation).  If smart E wants to talk to unknown destination D, E has
> to send two copies; one (without encapsulation), multicast on the link, a=
nd
> the other (with encapsulation), forwarded to the attached RBridge R.****
>
> [Linda] Agree****
>
> ** **
>
> R has been warned by E of all the MAC addresses E owns.  R must drop all
> packets it receives on the link if either the source or destination MAC i=
s
> one of the ones owned by a smart endnode on the link.****
>
> ** **
>
> [Linda] What if multiple =93smart endnode=94 sharing one link? For exampl=
e,
> there could be a bridge aggregating traffic from multiple end nodes (smar=
t
> and regular), or the multiple smart nodes as VMs sharing a physical serve=
r?
> ****
>
> ** **
>
> Linda ****
>
> ** **
>
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill
>
>

--f46d044270d0c35bc204d3094484
Content-Type: text/html; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

I believe that the smart endnode draft already discusses that. =A0It consid=
ers having a shared link that has any collection of smart and not smart end=
nodes.=A0<div><br></div><div>Endnodes on that link will talk to each other =
using non-encapsulated frames, whether or not they are smart endnodes.</div=
>
<div><br></div><div><div>So for a few examples, let D1 and D2 be &quot;not =
smart&quot; endnodes, and S1 and S2 be &quot;smart&quot; endnodes. =A0And l=
et R be the attached RBridge.</div><div><br></div><div>D1 and D2 always tra=
nsmit and receive native (non-encapsulated). =A0If R does not know where D2=
 is, then R might encapsulate and flood a packet D1 sends to D2, but once R=
 learns (from seeing traffic from D2 on the link) that D2 is on the link, R=
 will drop native packets it sees on the link with destination D2.</div>
<div><br></div><div>Now let S1 want to talk to D2. =A0(As stated in the dra=
ft), if S1 does not know where D2 is, S1 must send two copies:</div><div>1)=
 native on the link, in case D2 is there</div><div>2) encapsulated as a to-=
be-flooded packet to R, in case D2 is elsewhere on the campus</div>
<div><br></div><div>If D2 wants to talk to S1, D2 just sends it native, and=
 S1 receives it just fine (S1 has to be able to receive native as well as e=
ncapsulated packets).</div><div><br></div><div>R drops the packet because S=
1 has informed R of all of the MAC addresses that S1 owns.</div>
<div><br></div><div>Now let&#39;s say S1 wants to talk to S2. =A0If S1 has =
received traffic from S2, S1 will know that S2 is on the link, and will onl=
y send a single copy to S2; a native copy.</div><div><br></div><div>=A0If S=
1 does not know where S2 is, it sends two copies; one encapsulated, and one=
 native. =A0R has already learned S2&#39;s MAC addresses, so R knows it sho=
uld drop the packet. =A0S2, however, will receive two copies of the packet;=
 one native, and one encapsulated, and S2 will receive both of them. =A0But=
 S2 will learn that S1 is on the link, and when it sends something to S1, S=
1 will learn that S2 is on the link.</div>
<div><br></div><div>So once in a blue moon, a smart endnode might receive t=
wo copies of a packet, from another smart node on the link, until they lear=
n about the other one. =A0I suppose we could have the smart endnodes on the=
 link listen to the Hellos of other smart endnodes, in order to learn all t=
he MAC addresses on the link owned by smart endnodes, in order to avoid the=
 once-in-a-blue-moon receipt of a double packet, but I prefer not adding th=
at complexity.</div>
<div><br></div><div>Radia</div><div><br></div><div><br><div class=3D"gmail_=
quote">On Fri, Jan 11, 2013 at 12:21 PM, Linda Dunbar <span dir=3D"ltr">&lt=
;<a href=3D"mailto:linda.dunbar@huawei.com" target=3D"_blank">linda.dunbar@=
huawei.com</a>&gt;</span> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">





<div lang=3D"EN-US" link=3D"blue" vlink=3D"purple">
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">Radia,
<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">Questions inserted below:=
<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
<div style=3D"border:none;border-left:solid blue 1.5pt;padding:0in 0in 0in =
4.0pt">
<div>
<div style=3D"border:none;border-top:solid #b5c4df 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:10.0pt;font-family:&quot=
;Tahoma&quot;,&quot;sans-serif&quot;">From:</span></b><span style=3D"font-s=
ize:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;"> Radia Pe=
rlman [mailto:<a href=3D"mailto:radiaperlman@gmail.com" target=3D"_blank">r=
adiaperlman@gmail.com</a>]
<br>
<b>Sent:</b> Wednesday, January 09, 2013 4:32 PM<br>
<b>To:</b> Linda Dunbar<br>
<b>Cc:</b> Phanidhar Koganti; <a href=3D"mailto:d3e3e3@gmail.com" target=3D=
"_blank">d3e3e3@gmail.com</a>; Kesava_Vijaya_Krupak@Dell.com; <a href=3D"ma=
ilto:hu.fangwei@zte.com.cn" target=3D"_blank">hu.fangwei@zte.com.cn</a>; <a=
 href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ietf.org</a><br>

<b>Subject:</b> Re: [trill] Questions to TRILL Smart Endnodes draft<u></u><=
u></u></span></p>
</div>
</div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
<div><div class=3D"im">
<p class=3D"MsoNormal">If there is a smart endnode on a shared link with ot=
her endnodes, it learns about its endnode neighbors, and sends directly to =
them (without encapsulation). =A0If smart E wants to talk to unknown destin=
ation D, E has to send two copies; one
 (without encapsulation), multicast on the link, and the other (with encaps=
ulation), forwarded to the attached RBridge R.<u></u><u></u></p>
</div><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&q=
uot;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">[Linda] Agree<u></u=
><u></u></span></p>
</div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
<div><div class=3D"im">
<p class=3D"MsoNormal">R has been warned by E of all the MAC addresses E ow=
ns. =A0R must drop all packets it receives on the link if either the source=
 or destination MAC is one of the ones owned by a smart endnode on the link=
.<u></u><u></u></p>

<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
</div><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&q=
uot;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">[Linda] What if mul=
tiple =93smart endnode=94 sharing one link? For example, there could be a b=
ridge aggregating traffic from multiple end nodes (smart and regular),
 or the multiple smart nodes as VMs sharing a physical server? <span class=
=3D"HOEnZb"><font color=3D"#888888"><u></u><u></u></font></span></span></p>=
<span class=3D"HOEnZb"><font color=3D"#888888">
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">Linda
<u></u><u></u></span></p>
</font></span></div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
</div>
</div>
</div>

<br>_______________________________________________<br>
trill mailing list<br>
<a href=3D"mailto:trill@ietf.org">trill@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/trill" target=3D"_blank">h=
ttps://www.ietf.org/mailman/listinfo/trill</a><br>
<br></blockquote></div><br></div></div>

--f46d044270d0c35bc204d3094484--

From linda.dunbar@huawei.com  Fri Jan 11 13:00:18 2013
Return-Path: <linda.dunbar@huawei.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id DA09F21F8A0C for <trill@ietfa.amsl.com>; Fri, 11 Jan 2013 13:00:17 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.512
X-Spam-Level: 
X-Spam-Status: No, score=-6.512 tagged_above=-999 required=5 tests=[AWL=0.086,  BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 9sqgWRw6e8kq for <trill@ietfa.amsl.com>; Fri, 11 Jan 2013 13:00:14 -0800 (PST)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) by ietfa.amsl.com (Postfix) with ESMTP id 4C11921F87D6 for <trill@ietf.org>; Fri, 11 Jan 2013 13:00:13 -0800 (PST)
Received: from 172.18.7.190 (EHLO lhreml203-edg.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.5-GA FastPath queued) with ESMTP id ANK00893; Fri, 11 Jan 2013 21:00:11 +0000 (GMT)
Received: from LHREML405-HUB.china.huawei.com (10.201.5.242) by lhreml203-edg.huawei.com (172.18.7.221) with Microsoft SMTP Server (TLS) id 14.1.323.3; Fri, 11 Jan 2013 20:58:53 +0000
Received: from DFWEML406-HUB.china.huawei.com (10.193.5.131) by lhreml405-hub.china.huawei.com (10.201.5.242) with Microsoft SMTP Server (TLS) id 14.1.323.3; Fri, 11 Jan 2013 21:00:10 +0000
Received: from DFWEML505-MBX.china.huawei.com ([10.124.31.100]) by dfweml406-hub.china.huawei.com ([10.193.5.131]) with mapi id 14.01.0323.003; Fri, 11 Jan 2013 13:00:07 -0800
From: Linda Dunbar <linda.dunbar@huawei.com>
To: Radia Perlman <radiaperlman@gmail.com>
Thread-Topic: [trill] Questions to TRILL Smart Endnodes draft
Thread-Index: AQHN7rkrS7AADs7TvEa0mW1AYaXweZhEk6ZQgACMDAD//34Q8A==
Date: Fri, 11 Jan 2013 21:00:06 +0000
Message-ID: <4A95BA014132FF49AE685FAB4B9F17F6450E0BED@dfweml505-mbx>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com> <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com> <6895EAE0CA8DEE40B92D7CA88BB521F32414007BC1@HQ1-EXCH02.corp.brocade.com> <CAFOuuo626hw-c0kM4-1Fp-w2gByPenRnkFP3xNT+cTB4YcJ2kg@mail.gmail.com> <4A95BA014132FF49AE685FAB4B9F17F6450E01E4@dfweml505-mbx> <CAFOuuo6=wDeKiG-zjSnkcO3B2iTbazpsGHNYJLSfOxYZsh9kFA@mail.gmail.com> <4A95BA014132FF49AE685FAB4B9F17F6450E0B9A@dfweml505-mbx> <CAFOuuo5kVnBw3ODP-GfpF9hhE=VK2rKFMaF_WHf82zWP2PR=Tg@mail.gmail.com>
In-Reply-To: <CAFOuuo5kVnBw3ODP-GfpF9hhE=VK2rKFMaF_WHf82zWP2PR=Tg@mail.gmail.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.192.11.236]
Content-Type: multipart/alternative; boundary="_000_4A95BA014132FF49AE685FAB4B9F17F6450E0BEDdfweml505mbx_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Cc: "d3e3e3@gmail.com" <d3e3e3@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Questions to TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 11 Jan 2013 21:00:18 -0000

--_000_4A95BA014132FF49AE685FAB4B9F17F6450E0BEDdfweml505mbx_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Radia,

Thank you very much for the detailed description. Yes most of your descript=
ion are in the draft.

I am sorry for not stating my question clearly. What I want to ask is the p=
rocessing of Egress R receiving TRILL frames from the TRILL domain and ther=
e are multiple End Nodes attached via the same physical port.  I don't thin=
k it is described clearly in the draft.

Since multiple Smart Nodes on the LAN sharing the same Nickname, Egress nod=
e has to add another layer Ethernet header before sending the data to the c=
orrect Smart End node. Basically the inner Ethernet header has to be repeat=
ed on the outer header (with all the header information, e.g. VLAN, etc) in=
 order for the frame to reach the correct Smart Node.

Linda



From: Radia Perlman [mailto:radiaperlman@gmail.com]
Sent: Friday, January 11, 2013 2:38 PM
To: Linda Dunbar
Cc: d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com; hu.fangwei@zte.com.cn;=
 trill@ietf.org
Subject: Re: [trill] Questions to TRILL Smart Endnodes draft

I believe that the smart endnode draft already discusses that.  It consider=
s having a shared link that has any collection of smart and not smart endno=
des.

Endnodes on that link will talk to each other using non-encapsulated frames=
, whether or not they are smart endnodes.

So for a few examples, let D1 and D2 be "not smart" endnodes, and S1 and S2=
 be "smart" endnodes.  And let R be the attached RBridge.

D1 and D2 always transmit and receive native (non-encapsulated).  If R does=
 not know where D2 is, then R might encapsulate and flood a packet D1 sends=
 to D2, but once R learns (from seeing traffic from D2 on the link) that D2=
 is on the link, R will drop native packets it sees on the link with destin=
ation D2.

Now let S1 want to talk to D2.  (As stated in the draft), if S1 does not kn=
ow where D2 is, S1 must send two copies:
1) native on the link, in case D2 is there
2) encapsulated as a to-be-flooded packet to R, in case D2 is elsewhere on =
the campus

If D2 wants to talk to S1, D2 just sends it native, and S1 receives it just=
 fine (S1 has to be able to receive native as well as encapsulated packets)=
.

R drops the packet because S1 has informed R of all of the MAC addresses th=
at S1 owns.

Now let's say S1 wants to talk to S2.  If S1 has received traffic from S2, =
S1 will know that S2 is on the link, and will only send a single copy to S2=
; a native copy.

 If S1 does not know where S2 is, it sends two copies; one encapsulated, an=
d one native.  R has already learned S2's MAC addresses, so R knows it shou=
ld drop the packet.  S2, however, will receive two copies of the packet; on=
e native, and one encapsulated, and S2 will receive both of them.  But S2 w=
ill learn that S1 is on the link, and when it sends something to S1, S1 wil=
l learn that S2 is on the link.

So once in a blue moon, a smart endnode might receive two copies of a packe=
t, from another smart node on the link, until they learn about the other on=
e.  I suppose we could have the smart endnodes on the link listen to the He=
llos of other smart endnodes, in order to learn all the MAC addresses on th=
e link owned by smart endnodes, in order to avoid the once-in-a-blue-moon r=
eceipt of a double packet, but I prefer not adding that complexity.

Radia


On Fri, Jan 11, 2013 at 12:21 PM, Linda Dunbar <linda.dunbar@huawei.com<mai=
lto:linda.dunbar@huawei.com>> wrote:
Radia,

Questions inserted below:

From: Radia Perlman [mailto:radiaperlman@gmail.com<mailto:radiaperlman@gmai=
l.com>]
Sent: Wednesday, January 09, 2013 4:32 PM
To: Linda Dunbar
Cc: Phanidhar Koganti; d3e3e3@gmail.com<mailto:d3e3e3@gmail.com>; Kesava_Vi=
jaya_Krupak@Dell.com; hu.fangwei@zte.com.cn<mailto:hu.fangwei@zte.com.cn>; =
trill@ietf.org<mailto:trill@ietf.org>
Subject: Re: [trill] Questions to TRILL Smart Endnodes draft


If there is a smart endnode on a shared link with other endnodes, it learns=
 about its endnode neighbors, and sends directly to them (without encapsula=
tion).  If smart E wants to talk to unknown destination D, E has to send tw=
o copies; one (without encapsulation), multicast on the link, and the other=
 (with encapsulation), forwarded to the attached RBridge R.
[Linda] Agree

R has been warned by E of all the MAC addresses E owns.  R must drop all pa=
ckets it receives on the link if either the source or destination MAC is on=
e of the ones owned by a smart endnode on the link.

[Linda] What if multiple "smart endnode" sharing one link? For example, the=
re could be a bridge aggregating traffic from multiple end nodes (smart and=
 regular), or the multiple smart nodes as VMs sharing a physical server?

Linda


_______________________________________________
trill mailing list
trill@ietf.org<mailto:trill@ietf.org>
https://www.ietf.org/mailman/listinfo/trill


--_000_4A95BA014132FF49AE685FAB4B9F17F6450E0BEDdfweml505mbx_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micr=
osoft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:x=3D"urn:schemas-microsoft-com:office:excel" xmlns:p=3D"urn:schemas-m=
icrosoft-com:office:powerpoint" xmlns:a=3D"urn:schemas-microsoft-com:office=
:access" xmlns:dt=3D"uuid:C2F41010-65B3-11d1-A29F-00AA00C14882" xmlns:s=3D"=
uuid:BDC6E3F0-6DA3-11d1-A2A3-00AA00C14882" xmlns:rs=3D"urn:schemas-microsof=
t-com:rowset" xmlns:z=3D"#RowsetSchema" xmlns:b=3D"urn:schemas-microsoft-co=
m:office:publisher" xmlns:ss=3D"urn:schemas-microsoft-com:office:spreadshee=
t" xmlns:c=3D"urn:schemas-microsoft-com:office:component:spreadsheet" xmlns=
:odc=3D"urn:schemas-microsoft-com:office:odc" xmlns:oa=3D"urn:schemas-micro=
soft-com:office:activation" xmlns:html=3D"http://www.w3.org/TR/REC-html40" =
xmlns:q=3D"http://schemas.xmlsoap.org/soap/envelope/" xmlns:rtc=3D"http://m=
icrosoft.com/officenet/conferencing" xmlns:D=3D"DAV:" xmlns:Repl=3D"http://=
schemas.microsoft.com/repl/" xmlns:mt=3D"http://schemas.microsoft.com/share=
point/soap/meetings/" xmlns:x2=3D"http://schemas.microsoft.com/office/excel=
/2003/xml" xmlns:ppda=3D"http://www.passport.com/NameSpace.xsd" xmlns:ois=
=3D"http://schemas.microsoft.com/sharepoint/soap/ois/" xmlns:dir=3D"http://=
schemas.microsoft.com/sharepoint/soap/directory/" xmlns:ds=3D"http://www.w3=
.org/2000/09/xmldsig#" xmlns:dsp=3D"http://schemas.microsoft.com/sharepoint=
/dsp" xmlns:udc=3D"http://schemas.microsoft.com/data/udc" xmlns:xsd=3D"http=
://www.w3.org/2001/XMLSchema" xmlns:sub=3D"http://schemas.microsoft.com/sha=
repoint/soap/2002/1/alerts/" xmlns:ec=3D"http://www.w3.org/2001/04/xmlenc#"=
 xmlns:sp=3D"http://schemas.microsoft.com/sharepoint/" xmlns:sps=3D"http://=
schemas.microsoft.com/sharepoint/soap/" xmlns:xsi=3D"http://www.w3.org/2001=
/XMLSchema-instance" xmlns:udcs=3D"http://schemas.microsoft.com/data/udc/so=
ap" xmlns:udcxf=3D"http://schemas.microsoft.com/data/udc/xmlfile" xmlns:udc=
p2p=3D"http://schemas.microsoft.com/data/udc/parttopart" xmlns:wf=3D"http:/=
/schemas.microsoft.com/sharepoint/soap/workflow/" xmlns:dsss=3D"http://sche=
mas.microsoft.com/office/2006/digsig-setup" xmlns:dssi=3D"http://schemas.mi=
crosoft.com/office/2006/digsig" xmlns:mdssi=3D"http://schemas.openxmlformat=
s.org/package/2006/digital-signature" xmlns:mver=3D"http://schemas.openxmlf=
ormats.org/markup-compatibility/2006" xmlns:m=3D"http://schemas.microsoft.c=
om/office/2004/12/omml" xmlns:mrels=3D"http://schemas.openxmlformats.org/pa=
ckage/2006/relationships" xmlns:spwp=3D"http://microsoft.com/sharepoint/web=
partpages" xmlns:ex12t=3D"http://schemas.microsoft.com/exchange/services/20=
06/types" xmlns:ex12m=3D"http://schemas.microsoft.com/exchange/services/200=
6/messages" xmlns:pptsl=3D"http://schemas.microsoft.com/sharepoint/soap/Sli=
deLibrary/" xmlns:spsl=3D"http://microsoft.com/webservices/SharePointPortal=
Server/PublishedLinksService" xmlns:Z=3D"urn:schemas-microsoft-com:" xmlns:=
st=3D"&#1;" xmlns=3D"http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Dus-ascii"=
>
<meta name=3D"Generator" content=3D"Microsoft Word 12 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
	{font-family:SimSun;
	panose-1:2 1 6 0 3 1 1 1 1 1;}
@font-face
	{font-family:"Cambria Math";
	panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:Tahoma;
	panose-1:2 11 6 4 3 5 4 4 2 4;}
@font-face
	{font-family:SimSun;
	panose-1:2 1 6 0 3 1 1 1 1 1;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman","serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
span.hoenzb
	{mso-style-name:hoenzb;}
span.EmailStyle18
	{mso-style-type:personal-reply;
	font-family:"Calibri","sans-serif";
	color:#1F497D;}
.MsoChpDefault
	{mso-style-type:export-only;}
@page WordSection1
	{size:8.5in 11.0in;
	margin:1.0in 1.25in 1.0in 1.25in;}
div.WordSection1
	{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang=3D"EN-US" link=3D"blue" vlink=3D"purple">
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">Radia,
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">Thank you very much for t=
he detailed description. Yes most of your description are in the draft.
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">I am sorry for not statin=
g my question clearly. What I want to ask is the processing of Egress R rec=
eiving TRILL frames from the TRILL domain and there are
 multiple End Nodes attached via the same physical port. &nbsp;I don&#8217;=
t think it is described clearly in the draft.
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">Since multiple Smart Node=
s on the LAN sharing the same Nickname, Egress node has to add another laye=
r Ethernet header before sending the data to the correct
 Smart End node. Basically the inner Ethernet header has to be repeated on =
the outer header (with all the header information, e.g. VLAN, etc) in order=
 for the frame to reach the correct Smart Node.
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">Linda<o:p></o:p></span></=
p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<div style=3D"border:none;border-left:solid blue 1.5pt;padding:0in 0in 0in =
4.0pt">
<div>
<div style=3D"border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:10.0pt;font-family:&quot=
;Tahoma&quot;,&quot;sans-serif&quot;">From:</span></b><span style=3D"font-s=
ize:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;"> Radia Pe=
rlman [mailto:radiaperlman@gmail.com]
<br>
<b>Sent:</b> Friday, January 11, 2013 2:38 PM<br>
<b>To:</b> Linda Dunbar<br>
<b>Cc:</b> d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com; hu.fangwei@zte.=
com.cn; trill@ietf.org<br>
<b>Subject:</b> Re: [trill] Questions to TRILL Smart Endnodes draft<o:p></o=
:p></span></p>
</div>
</div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">I believe that the smart endnode draft already discu=
sses that. &nbsp;It considers having a shared link that has any collection =
of smart and not smart endnodes.&nbsp;<o:p></o:p></p>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal">Endnodes on that link will talk to each other using =
non-encapsulated frames, whether or not they are smart endnodes.<o:p></o:p>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<div>
<p class=3D"MsoNormal">So for a few examples, let D1 and D2 be &quot;not sm=
art&quot; endnodes, and S1 and S2 be &quot;smart&quot; endnodes. &nbsp;And =
let R be the attached RBridge.<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal">D1 and D2 always transmit and receive native (non-en=
capsulated). &nbsp;If R does not know where D2 is, then R might encapsulate=
 and flood a packet D1 sends to D2, but once R learns (from seeing traffic =
from D2 on the link) that D2 is on the
 link, R will drop native packets it sees on the link with destination D2.<=
o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal">Now let S1 want to talk to D2. &nbsp;(As stated in t=
he draft), if S1 does not know where D2 is, S1 must send two copies:<o:p></=
o:p></p>
</div>
<div>
<p class=3D"MsoNormal">1) native on the link, in case D2 is there<o:p></o:p=
></p>
</div>
<div>
<p class=3D"MsoNormal">2) encapsulated as a to-be-flooded packet to R, in c=
ase D2 is elsewhere on the campus<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal">If D2 wants to talk to S1, D2 just sends it native, =
and S1 receives it just fine (S1 has to be able to receive native as well a=
s encapsulated packets).<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal">R drops the packet because S1 has informed R of all =
of the MAC addresses that S1 owns.<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal">Now let's say S1 wants to talk to S2. &nbsp;If S1 ha=
s received traffic from S2, S1 will know that S2 is on the link, and will o=
nly send a single copy to S2; a native copy.<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal">&nbsp;If S1 does not know where S2 is, it sends two =
copies; one encapsulated, and one native. &nbsp;R has already learned S2's =
MAC addresses, so R knows it should drop the packet. &nbsp;S2, however, wil=
l receive two copies of the packet; one native, and
 one encapsulated, and S2 will receive both of them. &nbsp;But S2 will lear=
n that S1 is on the link, and when it sends something to S1, S1 will learn =
that S2 is on the link.<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal">So once in a blue moon, a smart endnode might receiv=
e two copies of a packet, from another smart node on the link, until they l=
earn about the other one. &nbsp;I suppose we could have the smart endnodes =
on the link listen to the Hellos of other
 smart endnodes, in order to learn all the MAC addresses on the link owned =
by smart endnodes, in order to avoid the once-in-a-blue-moon receipt of a d=
ouble packet, but I prefer not adding that complexity.<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal">Radia<o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<div>
<p class=3D"MsoNormal">On Fri, Jan 11, 2013 at 12:21 PM, Linda Dunbar &lt;<=
a href=3D"mailto:linda.dunbar@huawei.com" target=3D"_blank">linda.dunbar@hu=
awei.com</a>&gt; wrote:<o:p></o:p></p>
<div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">Radia,
</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">&nbsp;</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">Questions inserted below:</span><o:p></=
o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">&nbsp;</span><o:p></o:p></p>
<div style=3D"border:none;border-left:solid blue 1.5pt;padding:0in 0in 0in =
4.0pt">
<div>
<div style=3D"border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><b><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,=
&quot;sans-serif&quot;">From:</span></b><span style=3D"font-size:10.0pt;fon=
t-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;"> Radia Perlman [mailto:=
<a href=3D"mailto:radiaperlman@gmail.com" target=3D"_blank">radiaperlman@gm=
ail.com</a>]
<br>
<b>Sent:</b> Wednesday, January 09, 2013 4:32 PM<br>
<b>To:</b> Linda Dunbar<br>
<b>Cc:</b> Phanidhar Koganti; <a href=3D"mailto:d3e3e3@gmail.com" target=3D=
"_blank">d3e3e3@gmail.com</a>; Kesava_Vijaya_Krupak@Dell.com;
<a href=3D"mailto:hu.fangwei@zte.com.cn" target=3D"_blank">hu.fangwei@zte.c=
om.cn</a>;
<a href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ietf.org</a><br>
<b>Subject:</b> Re: [trill] Questions to TRILL Smart Endnodes draft</span><=
o:p></o:p></p>
</div>
</div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
</div>
<div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">If there is a smart endnode on a shared link with other endnodes, =
it learns about its endnode neighbors, and sends directly to them (without =
encapsulation). &nbsp;If smart E wants to
 talk to unknown destination D, E has to send two copies; one (without enca=
psulation), multicast on the link, and the other (with encapsulation), forw=
arded to the attached RBridge R.<o:p></o:p></p>
</div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">[Linda] Agree</span><o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
</div>
<div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">R has been warned by E of all the MAC addresses E owns. &nbsp;R mu=
st drop all packets it receives on the link if either the source or destina=
tion MAC is one of the ones owned by a smart
 endnode on the link.<o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">&nbsp;</span><o:p></o:p></p>
</div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">[Linda] What if multiple &#8220;smart e=
ndnode&#8221; sharing one link? For example, there could be a bridge aggreg=
ating
 traffic from multiple end nodes (smart and regular), or the multiple smart=
 nodes as VMs sharing a physical server?
</span><o:p></o:p></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">&nbsp;</span><span style=3D"color:#8888=
88"><o:p></o:p></span></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&q=
uot;sans-serif&quot;;color:#1F497D">Linda
</span><span style=3D"color:#888888"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:auto;mso-margin-bottom-a=
lt:auto">&nbsp;<o:p></o:p></p>
</div>
</div>
</div>
</div>
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><br>
_______________________________________________<br>
trill mailing list<br>
<a href=3D"mailto:trill@ietf.org">trill@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/trill" target=3D"_blank">h=
ttps://www.ietf.org/mailman/listinfo/trill</a><o:p></o:p></p>
</div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
</div>
</div>
</div>
</body>
</html>

--_000_4A95BA014132FF49AE685FAB4B9F17F6450E0BEDdfweml505mbx_--

From d3e3e3@gmail.com  Fri Jan 11 14:46:55 2013
Return-Path: <d3e3e3@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2A0D521F892C for <trill@ietfa.amsl.com>; Fri, 11 Jan 2013 14:46:55 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.743
X-Spam-Level: 
X-Spam-Status: No, score=-102.743 tagged_above=-999 required=5 tests=[AWL=0.256, BAYES_00=-2.599, J_CHICKENPOX_55=0.6, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id mTQVfNCN8V65 for <trill@ietfa.amsl.com>; Fri, 11 Jan 2013 14:46:54 -0800 (PST)
Received: from mail-ob0-f176.google.com (mail-ob0-f176.google.com [209.85.214.176]) by ietfa.amsl.com (Postfix) with ESMTP id F2F2621F8460 for <trill@ietf.org>; Fri, 11 Jan 2013 14:46:53 -0800 (PST)
Received: by mail-ob0-f176.google.com with SMTP id un3so2249415obb.35 for <trill@ietf.org>; Fri, 11 Jan 2013 14:46:53 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type:content-transfer-encoding; bh=UYG9QXQvs7vzkG9+RHB7na+eQnCXDzzg4k1+CT9RWYo=; b=wBP6mshOC3b0FwkhJydX6EvAmHA/3mrthfjHZ0MGtNuP6WtXdrqmpTmwS2drh/V/8J 27Yiei2rpslrAX3uyEkFzuvFCyEEYM0RCYvg+eHi1TM30vdbJMJ4Tvq0T0XnT80n66YY RMw6oWGCzbcuww9pwgEg+WyRYhsZ62rk0ctFhj1rDsTqkk17ln5ZvEJS/45K91VxPVGZ kA7CHiNDo0belH0ny82eVNqvtemeaAe7/7lcO/z2n0PNhugoNdeS4iOQIOymn4eSbacZ U7thFMzCVoLSUFSXHlefwXU5Hj1EJ7ojz13DrQ5B2UDhkfDaVP3C7rbiN1SWdwJS1V/7 Zn4Q==
Received: by 10.182.212.2 with SMTP id ng2mr54832632obc.81.1357944413402; Fri, 11 Jan 2013 14:46:53 -0800 (PST)
MIME-Version: 1.0
Received: by 10.76.110.239 with HTTP; Fri, 11 Jan 2013 14:46:33 -0800 (PST)
In-Reply-To: <4A95BA014132FF49AE685FAB4B9F17F6450E0BED@dfweml505-mbx>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com> <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com> <6895EAE0CA8DEE40B92D7CA88BB521F32414007BC1@HQ1-EXCH02.corp.brocade.com> <CAFOuuo626hw-c0kM4-1Fp-w2gByPenRnkFP3xNT+cTB4YcJ2kg@mail.gmail.com> <4A95BA014132FF49AE685FAB4B9F17F6450E01E4@dfweml505-mbx> <CAFOuuo6=wDeKiG-zjSnkcO3B2iTbazpsGHNYJLSfOxYZsh9kFA@mail.gmail.com> <4A95BA014132FF49AE685FAB4B9F17F6450E0B9A@dfweml505-mbx> <CAFOuuo5kVnBw3ODP-GfpF9hhE=VK2rKFMaF_WHf82zWP2PR=Tg@mail.gmail.com> <4A95BA014132FF49AE685FAB4B9F17F6450E0BED@dfweml505-mbx>
From: Donald Eastlake <d3e3e3@gmail.com>
Date: Fri, 11 Jan 2013 17:46:33 -0500
Message-ID: <CAF4+nEG5X_OHw1oZZ_DhkdgVj-N6u1HQdOk1gtvgPbzb1muy4A@mail.gmail.com>
To: Linda Dunbar <linda.dunbar@huawei.com>
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable
Cc: "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, Radia Perlman <radiaperlman@gmail.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Questions to TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 11 Jan 2013 22:46:55 -0000

Hi Linda,

On Fri, Jan 11, 2013 at 4:00 PM, Linda Dunbar <linda.dunbar@huawei.com> wro=
te:
> Radia,
>
>
> Thank you very much for the detailed description. Yes most of your
> description are in the draft.
>
>
> I am sorry for not stating my question clearly. What I want to ask is the
> processing of Egress R receiving TRILL frames from the TRILL domain and
> there are multiple End Nodes attached via the same physical port.  I don=
=92t
> think it is described clearly in the draft.
>
>
> Since multiple Smart Nodes on the LAN sharing the same Nickname, Egress n=
ode
> has to add another layer Ethernet header before sending the data to the
> correct Smart End node. Basically the inner Ethernet header has to be
> repeated on the outer header (with all the header information, e.g. VLAN,
> etc) in order for the frame to reach the correct Smart Node.

Assuming we are talking just about Ethernet links, TRILL switches
normally change the outer MAC source and destination addresses and, if
necessary, the outer VLAN label, on every hop. In this case, instead
of setting the Outer.MacDA to the address of the next hop TRILL switch
it copies the Inner.MacDA to the Outer.MacDA. I think it needs to to
do this for unicast packets regardless of whether there is one Smart
Endnode out a port or lots of them. I don't see how this is adding
another layer of Ethernet header.

Thanks,
Donald
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D
 Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
 155 Beaver Street, Milford, MA 01757 USA
 d3e3e3@gmail.com

> Linda
>
>
>
>
>
>
>
> From: Radia Perlman [mailto:radiaperlman@gmail.com]
> Sent: Friday, January 11, 2013 2:38 PM
> To: Linda Dunbar
> Cc: d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com; hu.fangwei@zte.com.c=
n;
> trill@ietf.org
>
>
> Subject: Re: [trill] Questions to TRILL Smart Endnodes draft
>
>
>
> I believe that the smart endnode draft already discusses that.  It consid=
ers
> having a shared link that has any collection of smart and not smart
> endnodes.
>
>
>
> Endnodes on that link will talk to each other using non-encapsulated fram=
es,
> whether or not they are smart endnodes.
>
>
>
> So for a few examples, let D1 and D2 be "not smart" endnodes, and S1 and =
S2
> be "smart" endnodes.  And let R be the attached RBridge.
>
>
>
> D1 and D2 always transmit and receive native (non-encapsulated).  If R do=
es
> not know where D2 is, then R might encapsulate and flood a packet D1 send=
s
> to D2, but once R learns (from seeing traffic from D2 on the link) that D=
2
> is on the link, R will drop native packets it sees on the link with
> destination D2.
>
>
>
> Now let S1 want to talk to D2.  (As stated in the draft), if S1 does not
> know where D2 is, S1 must send two copies:
>
> 1) native on the link, in case D2 is there
>
> 2) encapsulated as a to-be-flooded packet to R, in case D2 is elsewhere o=
n
> the campus
>
>
>
> If D2 wants to talk to S1, D2 just sends it native, and S1 receives it ju=
st
> fine (S1 has to be able to receive native as well as encapsulated packets=
).
>
>
>
> R drops the packet because S1 has informed R of all of the MAC addresses
> that S1 owns.
>
>
>
> Now let's say S1 wants to talk to S2.  If S1 has received traffic from S2=
,
> S1 will know that S2 is on the link, and will only send a single copy to =
S2;
> a native copy.
>
>
>
>  If S1 does not know where S2 is, it sends two copies; one encapsulated, =
and
> one native.  R has already learned S2's MAC addresses, so R knows it shou=
ld
> drop the packet.  S2, however, will receive two copies of the packet; one
> native, and one encapsulated, and S2 will receive both of them.  But S2 w=
ill
> learn that S1 is on the link, and when it sends something to S1, S1 will
> learn that S2 is on the link.
>
>
>
> So once in a blue moon, a smart endnode might receive two copies of a
> packet, from another smart node on the link, until they learn about the
> other one.  I suppose we could have the smart endnodes on the link listen=
 to
> the Hellos of other smart endnodes, in order to learn all the MAC address=
es
> on the link owned by smart endnodes, in order to avoid the
> once-in-a-blue-moon receipt of a double packet, but I prefer not adding t=
hat
> complexity.
>
>
>
> Radia
>
>
>
>
>
> On Fri, Jan 11, 2013 at 12:21 PM, Linda Dunbar <linda.dunbar@huawei.com>
> wrote:
>
> Radia,
>
>
>
> Questions inserted below:
>
>
>
> From: Radia Perlman [mailto:radiaperlman@gmail.com]
> Sent: Wednesday, January 09, 2013 4:32 PM
> To: Linda Dunbar
> Cc: Phanidhar Koganti; d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com;
> hu.fangwei@zte.com.cn; trill@ietf.org
> Subject: Re: [trill] Questions to TRILL Smart Endnodes draft
>
>
>
>
>
> If there is a smart endnode on a shared link with other endnodes, it lear=
ns
> about its endnode neighbors, and sends directly to them (without
> encapsulation).  If smart E wants to talk to unknown destination D, E has=
 to
> send two copies; one (without encapsulation), multicast on the link, and =
the
> other (with encapsulation), forwarded to the attached RBridge R.
>
> [Linda] Agree
>
>
>
> R has been warned by E of all the MAC addresses E owns.  R must drop all
> packets it receives on the link if either the source or destination MAC i=
s
> one of the ones owned by a smart endnode on the link.
>
>
>
> [Linda] What if multiple =93smart endnode=94 sharing one link? For exampl=
e,
> there could be a bridge aggregating traffic from multiple end nodes (smar=
t
> and regular), or the multiple smart nodes as VMs sharing a physical serve=
r?
>
>
>
> Linda
>
>
>
>
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill
>
>
>
>
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill
>

From linda.dunbar@huawei.com  Fri Jan 11 16:01:01 2013
Return-Path: <linda.dunbar@huawei.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B284121F88FB for <trill@ietfa.amsl.com>; Fri, 11 Jan 2013 16:01:00 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.224
X-Spam-Level: 
X-Spam-Status: No, score=-6.224 tagged_above=-999 required=5 tests=[AWL=-0.225, BAYES_00=-2.599, J_CHICKENPOX_55=0.6, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 1IftdFfQOUlH for <trill@ietfa.amsl.com>; Fri, 11 Jan 2013 16:00:59 -0800 (PST)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) by ietfa.amsl.com (Postfix) with ESMTP id 837D921F8881 for <trill@ietf.org>; Fri, 11 Jan 2013 16:00:58 -0800 (PST)
Received: from 172.18.7.190 (EHLO lhreml203-edg.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.5-GA FastPath queued) with ESMTP id AOQ94602; Sat, 12 Jan 2013 00:00:54 +0000 (GMT)
Received: from LHREML401-HUB.china.huawei.com (10.201.5.240) by lhreml203-edg.huawei.com (172.18.7.221) with Microsoft SMTP Server (TLS) id 14.1.323.3; Fri, 11 Jan 2013 23:59:35 +0000
Received: from DFWEML408-HUB.china.huawei.com (10.193.5.134) by lhreml401-hub.china.huawei.com (10.201.5.240) with Microsoft SMTP Server (TLS) id 14.1.323.3; Sat, 12 Jan 2013 00:00:52 +0000
Received: from DFWEML505-MBX.china.huawei.com ([10.124.31.100]) by dfweml408-hub.china.huawei.com ([10.193.5.134]) with mapi id 14.01.0323.003; Fri, 11 Jan 2013 16:00:46 -0800
From: Linda Dunbar <linda.dunbar@huawei.com>
To: Donald Eastlake <d3e3e3@gmail.com>
Thread-Topic: [trill] Questions to TRILL Smart Endnodes draft
Thread-Index: AQHN7rkrS7AADs7TvEa0mW1AYaXweZhEk6ZQgACMDAD//34Q8IAApemA//+M7FA=
Date: Sat, 12 Jan 2013 00:00:45 +0000
Message-ID: <4A95BA014132FF49AE685FAB4B9F17F6450E0D29@dfweml505-mbx>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com> <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com> <6895EAE0CA8DEE40B92D7CA88BB521F32414007BC1@HQ1-EXCH02.corp.brocade.com> <CAFOuuo626hw-c0kM4-1Fp-w2gByPenRnkFP3xNT+cTB4YcJ2kg@mail.gmail.com> <4A95BA014132FF49AE685FAB4B9F17F6450E01E4@dfweml505-mbx> <CAFOuuo6=wDeKiG-zjSnkcO3B2iTbazpsGHNYJLSfOxYZsh9kFA@mail.gmail.com> <4A95BA014132FF49AE685FAB4B9F17F6450E0B9A@dfweml505-mbx> <CAFOuuo5kVnBw3ODP-GfpF9hhE=VK2rKFMaF_WHf82zWP2PR=Tg@mail.gmail.com> <4A95BA014132FF49AE685FAB4B9F17F6450E0BED@dfweml505-mbx> <CAF4+nEG5X_OHw1oZZ_DhkdgVj-N6u1HQdOk1gtvgPbzb1muy4A@mail.gmail.com>
In-Reply-To: <CAF4+nEG5X_OHw1oZZ_DhkdgVj-N6u1HQdOk1gtvgPbzb1muy4A@mail.gmail.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.192.11.236]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Cc: "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, Radia Perlman <radiaperlman@gmail.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Questions to TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 12 Jan 2013 00:01:01 -0000

Donald,=20

Thanks for the explanation.=20
Normally the Egress RBridge simply decapsulates the TRILL header and forwar=
d the "inner Ethernet frames" to its attached LAN.=20

When there are  Smart endnodes attached to an RBridge, the behavior on RBri=
dge is more like intermediate RBridge, less like Egress RBridge.=20

Therefore, it may be more accurate to use the term "Slave RBridge", (i.e. a=
 group of RBridges sharing their master's nickname) instead of smart "endno=
de".=20

Linda

> -----Original Message-----
> From: Donald Eastlake [mailto:d3e3e3@gmail.com]
> Sent: Friday, January 11, 2013 4:47 PM
> To: Linda Dunbar
> Cc: Radia Perlman; Kesava_Vijaya_Krupak@Dell.com; hu.fangwei@zte.com.cn;
> trill@ietf.org
> Subject: Re: [trill] Questions to TRILL Smart Endnodes draft
>=20
> Hi Linda,
>=20
> On Fri, Jan 11, 2013 at 4:00 PM, Linda Dunbar <linda.dunbar@huawei.com>
> wrote:
> > Radia,
> >
> >
> > Thank you very much for the detailed description. Yes most of your
> > description are in the draft.
> >
> >
> > I am sorry for not stating my question clearly. What I want to ask is
> the
> > processing of Egress R receiving TRILL frames from the TRILL domain
> and
> > there are multiple End Nodes attached via the same physical port.  I
> don't
> > think it is described clearly in the draft.
> >
> >
> > Since multiple Smart Nodes on the LAN sharing the same Nickname,
> Egress node
> > has to add another layer Ethernet header before sending the data to
> the
> > correct Smart End node. Basically the inner Ethernet header has to be
> > repeated on the outer header (with all the header information, e.g.
> VLAN,
> > etc) in order for the frame to reach the correct Smart Node.
>=20
> Assuming we are talking just about Ethernet links, TRILL switches
> normally change the outer MAC source and destination addresses and, if
> necessary, the outer VLAN label, on every hop. In this case, instead
> of setting the Outer.MacDA to the address of the next hop TRILL switch
> it copies the Inner.MacDA to the Outer.MacDA. I think it needs to to
> do this for unicast packets regardless of whether there is one Smart
> Endnode out a port or lots of them. I don't see how this is adding
> another layer of Ethernet header.
>=20
> Thanks,
> Donald
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D
>  Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
>  155 Beaver Street, Milford, MA 01757 USA
>  d3e3e3@gmail.com
>=20
> > Linda
> >
> >
> >
> >
> >
> >
> >
> > From: Radia Perlman [mailto:radiaperlman@gmail.com]
> > Sent: Friday, January 11, 2013 2:38 PM
> > To: Linda Dunbar
> > Cc: d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com;
> hu.fangwei@zte.com.cn;
> > trill@ietf.org
> >
> >
> > Subject: Re: [trill] Questions to TRILL Smart Endnodes draft
> >
> >
> >
> > I believe that the smart endnode draft already discusses that.  It
> considers
> > having a shared link that has any collection of smart and not smart
> > endnodes.
> >
> >
> >
> > Endnodes on that link will talk to each other using non-encapsulated
> frames,
> > whether or not they are smart endnodes.
> >
> >
> >
> > So for a few examples, let D1 and D2 be "not smart" endnodes, and S1
> and S2
> > be "smart" endnodes.  And let R be the attached RBridge.
> >
> >
> >
> > D1 and D2 always transmit and receive native (non-encapsulated).  If
> R does
> > not know where D2 is, then R might encapsulate and flood a packet D1
> sends
> > to D2, but once R learns (from seeing traffic from D2 on the link)
> that D2
> > is on the link, R will drop native packets it sees on the link with
> > destination D2.
> >
> >
> >
> > Now let S1 want to talk to D2.  (As stated in the draft), if S1 does
> not
> > know where D2 is, S1 must send two copies:
> >
> > 1) native on the link, in case D2 is there
> >
> > 2) encapsulated as a to-be-flooded packet to R, in case D2 is
> elsewhere on
> > the campus
> >
> >
> >
> > If D2 wants to talk to S1, D2 just sends it native, and S1 receives
> it just
> > fine (S1 has to be able to receive native as well as encapsulated
> packets).
> >
> >
> >
> > R drops the packet because S1 has informed R of all of the MAC
> addresses
> > that S1 owns.
> >
> >
> >
> > Now let's say S1 wants to talk to S2.  If S1 has received traffic
> from S2,
> > S1 will know that S2 is on the link, and will only send a single copy
> to S2;
> > a native copy.
> >
> >
> >
> >  If S1 does not know where S2 is, it sends two copies; one
> encapsulated, and
> > one native.  R has already learned S2's MAC addresses, so R knows it
> should
> > drop the packet.  S2, however, will receive two copies of the packet;
> one
> > native, and one encapsulated, and S2 will receive both of them.  But
> S2 will
> > learn that S1 is on the link, and when it sends something to S1, S1
> will
> > learn that S2 is on the link.
> >
> >
> >
> > So once in a blue moon, a smart endnode might receive two copies of a
> > packet, from another smart node on the link, until they learn about
> the
> > other one.  I suppose we could have the smart endnodes on the link
> listen to
> > the Hellos of other smart endnodes, in order to learn all the MAC
> addresses
> > on the link owned by smart endnodes, in order to avoid the
> > once-in-a-blue-moon receipt of a double packet, but I prefer not
> adding that
> > complexity.
> >
> >
> >
> > Radia
> >
> >
> >
> >
> >
> > On Fri, Jan 11, 2013 at 12:21 PM, Linda Dunbar
> <linda.dunbar@huawei.com>
> > wrote:
> >
> > Radia,
> >
> >
> >
> > Questions inserted below:
> >
> >
> >
> > From: Radia Perlman [mailto:radiaperlman@gmail.com]
> > Sent: Wednesday, January 09, 2013 4:32 PM
> > To: Linda Dunbar
> > Cc: Phanidhar Koganti; d3e3e3@gmail.com;
> Kesava_Vijaya_Krupak@Dell.com;
> > hu.fangwei@zte.com.cn; trill@ietf.org
> > Subject: Re: [trill] Questions to TRILL Smart Endnodes draft
> >
> >
> >
> >
> >
> > If there is a smart endnode on a shared link with other endnodes, it
> learns
> > about its endnode neighbors, and sends directly to them (without
> > encapsulation).  If smart E wants to talk to unknown destination D, E
> has to
> > send two copies; one (without encapsulation), multicast on the link,
> and the
> > other (with encapsulation), forwarded to the attached RBridge R.
> >
> > [Linda] Agree
> >
> >
> >
> > R has been warned by E of all the MAC addresses E owns.  R must drop
> all
> > packets it receives on the link if either the source or destination
> MAC is
> > one of the ones owned by a smart endnode on the link.
> >
> >
> >
> > [Linda] What if multiple "smart endnode" sharing one link? For
> example,
> > there could be a bridge aggregating traffic from multiple end nodes
> (smart
> > and regular), or the multiple smart nodes as VMs sharing a physical
> server?
> >
> >
> >
> > Linda
> >
> >
> >
> >
> > _______________________________________________
> > trill mailing list
> > trill@ietf.org
> > https://www.ietf.org/mailman/listinfo/trill
> >
> >
> >
> >
> > _______________________________________________
> > trill mailing list
> > trill@ietf.org
> > https://www.ietf.org/mailman/listinfo/trill
> >

From radiaperlman@gmail.com  Fri Jan 11 21:36:51 2013
Return-Path: <radiaperlman@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id DCB6321F8596 for <trill@ietfa.amsl.com>; Fri, 11 Jan 2013 21:36:50 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.335
X-Spam-Level: 
X-Spam-Status: No, score=-3.335 tagged_above=-999 required=5 tests=[AWL=0.263,  BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id zt4DH8-J12zz for <trill@ietfa.amsl.com>; Fri, 11 Jan 2013 21:36:50 -0800 (PST)
Received: from mail-la0-f53.google.com (mail-la0-f53.google.com [209.85.215.53]) by ietfa.amsl.com (Postfix) with ESMTP id A561521F858E for <trill@ietf.org>; Fri, 11 Jan 2013 21:36:49 -0800 (PST)
Received: by mail-la0-f53.google.com with SMTP id fn20so2372830lab.26 for <trill@ietf.org>; Fri, 11 Jan 2013 21:36:48 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :content-type; bh=O/tTiwPh/hZJCyeh3jFQX3tiqi4ht7uvFuRB7qwc6zQ=; b=awoPN4vGEpTIkOFb48wybxjP9NL5dfE618FbkE4Nx28MRdYp3jKRt7PwuNtvVWKn6U JDSf6yy5YixluPDJtAC64H5OvbVg/j4G2duXX5+1t8AELJxPcKxM4UWQUHm6aP74rY2C TczZCEyYcfYtsYv7+BNVG5sojgGwmkTxryj5pk8LsQKWEmzHOWBrCys8mJEy7r6z9v8b Mv1ira46hLN/X2GAZh8CFgJFDog5tAX7S+SuPWtHM1R73wP7lYDW+/JNVd+kP7ru9dnC 3qZ/e/Q7AeMQtxrnI87UDR042/IISs36y/m429/GT0D3fz8BC+goiGahpkne8ElKfK5a agPg==
MIME-Version: 1.0
Received: by 10.112.83.100 with SMTP id p4mr32058525lby.96.1357969008508; Fri, 11 Jan 2013 21:36:48 -0800 (PST)
Received: by 10.112.64.17 with HTTP; Fri, 11 Jan 2013 21:36:48 -0800 (PST)
In-Reply-To: <CAFOuuo6DNJ16-fwExwoAe2gAHnDN7yXQp5TVPQCZCbEZRQ0Smw@mail.gmail.com>
References: <CAFOuuo6DNJ16-fwExwoAe2gAHnDN7yXQp5TVPQCZCbEZRQ0Smw@mail.gmail.com>
Date: Fri, 11 Jan 2013 21:36:48 -0800
Message-ID: <CAFOuuo7o-eVM9y++8+eWcfETzZUmQMSub4h2vRPRpXHuEFAWvA@mail.gmail.com>
From: Radia Perlman <radiaperlman@gmail.com>
To: trill@ietf.org
Content-Type: multipart/alternative; boundary=f46d040167f15e402304d310ccf0
Subject: [trill] Smart Endnodes and FGL Interaction
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 12 Jan 2013 05:36:51 -0000

--f46d040167f15e402304d310ccf0
Content-Type: text/plain; charset=ISO-8859-1

I got an off-line question that might be worth explicitly answering in the
draft, next time we do an update of the smart endnode draft.

The question came about because of someone wondering howthe  smart endnodes
design interacst with the FGL design, and what relationship the outer VLAN
is to the FGL.

I believe that a TRILL encapsulated packet to/from a smart endnode should
look exactly the same as if it were being forwarded to an RBridge, still
encapsulated.

So the answer is that the outer VLAN is the Designated VLAN for the link.
 (as has been the case for TRILL for years, each link is assigned a
Designated VLAN by the DRB, and all TRILL-encapsulated packets are
encapsulated with that VL as the outer VLAN).  The Designated VLAN will
always be a VL, not an FGL.

The inner VLAN will either be VL or FGL, but it will be filled in by the
smart endnode (and delivered to the smart endnode) exactly as the first RB
would have encapsulated it.  If it's supposed to be a 24-bit VLAN, it will
be an FGL.  If it's supposed to be a 12-bit VLAN, it will be a VL.

Hope that adequately answers the question that nobody on the mailing list
asked.  :-)

Radia

--f46d040167f15e402304d310ccf0
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

I got an off-line question that might be worth explicitly answering in the =
draft, next time we do an update of the smart endnode draft.<br><div class=
=3D"gmail_quote"><div><br></div><div>The question came about because of som=
eone wondering howthe =A0smart endnodes design interacst with the FGL desig=
n, and what relationship the outer VLAN is to the FGL.</div>

<div><br></div><div>I believe that a TRILL encapsulated packet to/from a sm=
art endnode should look exactly the same as if it were being forwarded to a=
n RBridge, still encapsulated.</div><div><br></div><div>So the answer is th=
at the outer VLAN is the Designated VLAN for the link. =A0(as has been the =
case for TRILL for years, each link is assigned a Designated VLAN by the DR=
B, and all TRILL-encapsulated packets are encapsulated with that VL as the =
outer VLAN). =A0The Designated VLAN will always be a VL, not an FGL.</div>

<div><br></div><div>The inner VLAN will either be VL or FGL, but it will be=
 filled in by the smart endnode (and delivered to the smart endnode) exactl=
y as the first RB would have encapsulated it. =A0If it&#39;s supposed to be=
 a 24-bit VLAN, it will be an FGL. =A0If it&#39;s supposed to be a 12-bit V=
LAN, it will be a VL.</div>

<div><br></div><div>Hope that adequately answers the question that nobody o=
n the mailing list asked. =A0:-)</div><span class=3D"HOEnZb"><font color=3D=
"#888888"><div><br></div><div>Radia</div><div><br></div><div><br></div>
</font></span></div><br>

--f46d040167f15e402304d310ccf0--

From radiaperlman@gmail.com  Sun Jan 13 11:37:19 2013
Return-Path: <radiaperlman@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0A8F321F87F3 for <trill@ietfa.amsl.com>; Sun, 13 Jan 2013 11:37:19 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.348
X-Spam-Level: 
X-Spam-Status: No, score=-3.348 tagged_above=-999 required=5 tests=[AWL=0.250,  BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id RLGy0ev9PWhW for <trill@ietfa.amsl.com>; Sun, 13 Jan 2013 11:37:14 -0800 (PST)
Received: from mail-la0-f48.google.com (mail-la0-f48.google.com [209.85.215.48]) by ietfa.amsl.com (Postfix) with ESMTP id 538B321F8692 for <trill@ietf.org>; Sun, 13 Jan 2013 11:37:14 -0800 (PST)
Received: by mail-la0-f48.google.com with SMTP id ej20so3223684lab.35 for <trill@ietf.org>; Sun, 13 Jan 2013 11:37:13 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=WYFk+BZFRYP8NEYbBj3jfRbNrx5Du8lb17J+oQKWVhw=; b=F8/c/17eq9LTK3D4NE5cHd7CXZ4ktnGFVJKWSv7PNXMRgc0xx3bQYdjSPdx0/oYEK6 RE5Ri9BI+khpc/muzvxLOgZjeurz87YZPwCx3VXl1UuP5fsbGIr4pDvQovnTKiUM+/6H 1Y6CaMXlD2wilVFcRNMaBPULDZWORejowvYopWy4geMPfuDr7uVGbHJBWED3IDu98i/U 8MwRRPlbM+dCGWhUliZJm4B9vDTcPD3ntcQN6HUQS5nJh+D5h1v9x2hjj8VHecypJBn9 zPQPf/wNk3evCkeyGctRRQZIemY1P3qXlLExjYJxd9cM21sVDlpUJyQtgLtkEOwVPHzD ybmQ==
MIME-Version: 1.0
Received: by 10.152.108.12 with SMTP id hg12mr19618940lab.43.1358105832949; Sun, 13 Jan 2013 11:37:12 -0800 (PST)
Received: by 10.112.64.17 with HTTP; Sun, 13 Jan 2013 11:37:12 -0800 (PST)
In-Reply-To: <CAFOuuo5kVnBw3ODP-GfpF9hhE=VK2rKFMaF_WHf82zWP2PR=Tg@mail.gmail.com>
References: <D5A6F3355F664C40AFB65BB1277D8D450186C7F7AD@MAAX7MCDC101.APAC.DELL.COM> <6895EAE0CA8DEE40B92D7CA88BB521F32414007827@HQ1-EXCH02.corp.brocade.com> <CAFOuuo7t8xt8Z1mx78G2SsE8tkmAMNs44xd11Mc43S1_WfH6RA@mail.gmail.com> <6895EAE0CA8DEE40B92D7CA88BB521F32414007BC1@HQ1-EXCH02.corp.brocade.com> <CAFOuuo626hw-c0kM4-1Fp-w2gByPenRnkFP3xNT+cTB4YcJ2kg@mail.gmail.com> <4A95BA014132FF49AE685FAB4B9F17F6450E01E4@dfweml505-mbx> <CAFOuuo6=wDeKiG-zjSnkcO3B2iTbazpsGHNYJLSfOxYZsh9kFA@mail.gmail.com> <4A95BA014132FF49AE685FAB4B9F17F6450E0B9A@dfweml505-mbx> <CAFOuuo5kVnBw3ODP-GfpF9hhE=VK2rKFMaF_WHf82zWP2PR=Tg@mail.gmail.com>
Date: Sun, 13 Jan 2013 11:37:12 -0800
Message-ID: <CAFOuuo676+mRNQZJC0i7jLF6z+J478yMxb7oGswgHzTVGw7KAw@mail.gmail.com>
From: Radia Perlman <radiaperlman@gmail.com>
To: Linda Dunbar <linda.dunbar@huawei.com>
Content-Type: multipart/alternative; boundary=bcaec54eef0abd898c04d330a7ea
Cc: "d3e3e3@gmail.com" <d3e3e3@gmail.com>, "Kesava_Vijaya_Krupak@Dell.com" <Kesava_Vijaya_Krupak@dell.com>, "hu.fangwei@zte.com.cn" <hu.fangwei@zte.com.cn>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Questions to TRILL Smart Endnodes draft
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 13 Jan 2013 19:37:19 -0000

--bcaec54eef0abd898c04d330a7ea
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

Actually, I wrote this too quickly (I really haven't had time to answer
carefully...still don't...but I want to correct one thing.  A smart endnode
would not receive two copies of the packet in the case I said.  All the
cases should work just fine.  When smart S1 is transmitting to S2, and S1
and S2 are on the same link, S2 will only receive the native copy from S1.
 The encapsulated one will only go to RB.

Radia

On Fri, Jan 11, 2013 at 12:37 PM, Radia Perlman <radiaperlman@gmail.com>wro=
te:

> I believe that the smart endnode draft already discusses that.  It
> considers having a shared link that has any collection of smart and not
> smart endnodes.
>
> Endnodes on that link will talk to each other using non-encapsulated
> frames, whether or not they are smart endnodes.
>
> So for a few examples, let D1 and D2 be "not smart" endnodes, and S1 and
> S2 be "smart" endnodes.  And let R be the attached RBridge.
>
> D1 and D2 always transmit and receive native (non-encapsulated).  If R
> does not know where D2 is, then R might encapsulate and flood a packet D1
> sends to D2, but once R learns (from seeing traffic from D2 on the link)
> that D2 is on the link, R will drop native packets it sees on the link wi=
th
> destination D2.
>
> Now let S1 want to talk to D2.  (As stated in the draft), if S1 does not
> know where D2 is, S1 must send two copies:
> 1) native on the link, in case D2 is there
> 2) encapsulated as a to-be-flooded packet to R, in case D2 is elsewhere o=
n
> the campus
>
> If D2 wants to talk to S1, D2 just sends it native, and S1 receives it
> just fine (S1 has to be able to receive native as well as encapsulated
> packets).
>
> R drops the packet because S1 has informed R of all of the MAC addresses
> that S1 owns.
>
> Now let's say S1 wants to talk to S2.  If S1 has received traffic from S2=
,
> S1 will know that S2 is on the link, and will only send a single copy to
> S2; a native copy.
>
>  If S1 does not know where S2 is, it sends two copies; one encapsulated,
> and one native.  R has already learned S2's MAC addresses, so R knows it
> should drop the packet.  S2, however, will receive two copies of the
> packet; one native, and one encapsulated, and S2 will receive both of the=
m.
>  But S2 will learn that S1 is on the link, and when it sends something to
> S1, S1 will learn that S2 is on the link.
>
> So once in a blue moon, a smart endnode might receive two copies of a
> packet, from another smart node on the link, until they learn about the
> other one.  I suppose we could have the smart endnodes on the link listen
> to the Hellos of other smart endnodes, in order to learn all the MAC
> addresses on the link owned by smart endnodes, in order to avoid the
> once-in-a-blue-moon receipt of a double packet, but I prefer not adding
> that complexity.
>
> Radia
>
>
> On Fri, Jan 11, 2013 at 12:21 PM, Linda Dunbar <linda.dunbar@huawei.com>w=
rote:
>
>>  Radia, ****
>>
>> ** **
>>
>> Questions inserted below:****
>>
>> ** **
>>
>> *From:* Radia Perlman [mailto:radiaperlman@gmail.com]
>> *Sent:* Wednesday, January 09, 2013 4:32 PM
>> *To:* Linda Dunbar
>> *Cc:* Phanidhar Koganti; d3e3e3@gmail.com; Kesava_Vijaya_Krupak@Dell.com=
;
>> hu.fangwei@zte.com.cn; trill@ietf.org
>> *Subject:* Re: [trill] Questions to TRILL Smart Endnodes draft****
>>
>> ** **
>>
>> ** **
>>
>> If there is a smart endnode on a shared link with other endnodes, it
>> learns about its endnode neighbors, and sends directly to them (without
>> encapsulation).  If smart E wants to talk to unknown destination D, E ha=
s
>> to send two copies; one (without encapsulation), multicast on the link, =
and
>> the other (with encapsulation), forwarded to the attached RBridge R.****
>>
>> [Linda] Agree****
>>
>> ** **
>>
>> R has been warned by E of all the MAC addresses E owns.  R must drop all
>> packets it receives on the link if either the source or destination MAC =
is
>> one of the ones owned by a smart endnode on the link.****
>>
>> ** **
>>
>> [Linda] What if multiple =93smart endnode=94 sharing one link? For examp=
le,
>> there could be a bridge aggregating traffic from multiple end nodes (sma=
rt
>> and regular), or the multiple smart nodes as VMs sharing a physical serv=
er?
>> ****
>>
>> ** **
>>
>> Linda ****
>>
>> ** **
>>
>> _______________________________________________
>> trill mailing list
>> trill@ietf.org
>> https://www.ietf.org/mailman/listinfo/trill
>>
>>
>

--bcaec54eef0abd898c04d330a7ea
Content-Type: text/html; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

Actually, I wrote this too quickly (I really haven&#39;t had time to answer=
 carefully...still don&#39;t...but I want to correct one thing. =A0A smart =
endnode would not receive two copies of the packet in the case I said. =A0A=
ll the cases should work just fine. =A0When smart S1 is transmitting to S2,=
 and S1 and S2 are on the same link, S2 will only receive the native copy f=
rom S1. =A0The encapsulated one will only go to RB.<div>
<br></div><div>Radia<br><br><div class=3D"gmail_quote">On Fri, Jan 11, 2013=
 at 12:37 PM, Radia Perlman <span dir=3D"ltr">&lt;<a href=3D"mailto:radiape=
rlman@gmail.com" target=3D"_blank">radiaperlman@gmail.com</a>&gt;</span> wr=
ote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">I believe that the smart endnode draft alrea=
dy discusses that. =A0It considers having a shared link that has any collec=
tion of smart and not smart endnodes.=A0<div>
<br></div><div>Endnodes on that link will talk to each other using non-enca=
psulated frames, whether or not they are smart endnodes.</div>
<div><br></div><div><div>So for a few examples, let D1 and D2 be &quot;not =
smart&quot; endnodes, and S1 and S2 be &quot;smart&quot; endnodes. =A0And l=
et R be the attached RBridge.</div><div><br></div><div>D1 and D2 always tra=
nsmit and receive native (non-encapsulated). =A0If R does not know where D2=
 is, then R might encapsulate and flood a packet D1 sends to D2, but once R=
 learns (from seeing traffic from D2 on the link) that D2 is on the link, R=
 will drop native packets it sees on the link with destination D2.</div>

<div><br></div><div>Now let S1 want to talk to D2. =A0(As stated in the dra=
ft), if S1 does not know where D2 is, S1 must send two copies:</div><div>1)=
 native on the link, in case D2 is there</div><div>2) encapsulated as a to-=
be-flooded packet to R, in case D2 is elsewhere on the campus</div>

<div><br></div><div>If D2 wants to talk to S1, D2 just sends it native, and=
 S1 receives it just fine (S1 has to be able to receive native as well as e=
ncapsulated packets).</div><div><br></div><div>R drops the packet because S=
1 has informed R of all of the MAC addresses that S1 owns.</div>

<div><br></div><div>Now let&#39;s say S1 wants to talk to S2. =A0If S1 has =
received traffic from S2, S1 will know that S2 is on the link, and will onl=
y send a single copy to S2; a native copy.</div><div><br></div><div>=A0If S=
1 does not know where S2 is, it sends two copies; one encapsulated, and one=
 native. =A0R has already learned S2&#39;s MAC addresses, so R knows it sho=
uld drop the packet. =A0S2, however, will receive two copies of the packet;=
 one native, and one encapsulated, and S2 will receive both of them. =A0But=
 S2 will learn that S1 is on the link, and when it sends something to S1, S=
1 will learn that S2 is on the link.</div>

<div><br></div><div>So once in a blue moon, a smart endnode might receive t=
wo copies of a packet, from another smart node on the link, until they lear=
n about the other one. =A0I suppose we could have the smart endnodes on the=
 link listen to the Hellos of other smart endnodes, in order to learn all t=
he MAC addresses on the link owned by smart endnodes, in order to avoid the=
 once-in-a-blue-moon receipt of a double packet, but I prefer not adding th=
at complexity.</div>
<span class=3D"HOEnZb"><font color=3D"#888888">
<div><br></div><div>Radia</div><div><br></div></font></span><div><br><div c=
lass=3D"gmail_quote"><div><div class=3D"h5">On Fri, Jan 11, 2013 at 12:21 P=
M, Linda Dunbar <span dir=3D"ltr">&lt;<a href=3D"mailto:linda.dunbar@huawei=
.com" target=3D"_blank">linda.dunbar@huawei.com</a>&gt;</span> wrote:<br>

</div></div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;bo=
rder-left:1px #ccc solid;padding-left:1ex"><div><div class=3D"h5">





<div lang=3D"EN-US" link=3D"blue" vlink=3D"purple">
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">Radia,
<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">Questions inserted below:=
<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
<div style=3D"border:none;border-left:solid blue 1.5pt;padding:0in 0in 0in =
4.0pt">
<div>
<div style=3D"border:none;border-top:solid #b5c4df 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:10.0pt;font-family:&quot=
;Tahoma&quot;,&quot;sans-serif&quot;">From:</span></b><span style=3D"font-s=
ize:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;"> Radia Pe=
rlman [mailto:<a href=3D"mailto:radiaperlman@gmail.com" target=3D"_blank">r=
adiaperlman@gmail.com</a>]
<br>
<b>Sent:</b> Wednesday, January 09, 2013 4:32 PM<br>
<b>To:</b> Linda Dunbar<br>
<b>Cc:</b> Phanidhar Koganti; <a href=3D"mailto:d3e3e3@gmail.com" target=3D=
"_blank">d3e3e3@gmail.com</a>; Kesava_Vijaya_Krupak@Dell.com; <a href=3D"ma=
ilto:hu.fangwei@zte.com.cn" target=3D"_blank">hu.fangwei@zte.com.cn</a>; <a=
 href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ietf.org</a><br>


<b>Subject:</b> Re: [trill] Questions to TRILL Smart Endnodes draft<u></u><=
u></u></span></p>
</div>
</div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
<div><div>
<p class=3D"MsoNormal">If there is a smart endnode on a shared link with ot=
her endnodes, it learns about its endnode neighbors, and sends directly to =
them (without encapsulation). =A0If smart E wants to talk to unknown destin=
ation D, E has to send two copies; one
 (without encapsulation), multicast on the link, and the other (with encaps=
ulation), forwarded to the attached RBridge R.<u></u><u></u></p>
</div><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&q=
uot;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">[Linda] Agree<u></u=
><u></u></span></p>
</div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
<div><div>
<p class=3D"MsoNormal">R has been warned by E of all the MAC addresses E ow=
ns. =A0R must drop all packets it receives on the link if either the source=
 or destination MAC is one of the ones owned by a smart endnode on the link=
.<u></u><u></u></p>


<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
</div><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&q=
uot;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">[Linda] What if mul=
tiple =93smart endnode=94 sharing one link? For example, there could be a b=
ridge aggregating traffic from multiple end nodes (smart and regular),
 or the multiple smart nodes as VMs sharing a physical server? <span><font =
color=3D"#888888"><u></u><u></u></font></span></span></p><span><font color=
=3D"#888888">
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">Linda
<u></u><u></u></span></p>
</font></span></div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
</div>
</div>
</div>
</div>

<br></div></div><div class=3D"im">_________________________________________=
______<br>
trill mailing list<br>
<a href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/trill" target=3D"_blank">h=
ttps://www.ietf.org/mailman/listinfo/trill</a><br>
<br></div></blockquote></div><br></div></div>
</blockquote></div><br></div>

--bcaec54eef0abd898c04d330a7ea--

From internet-drafts@ietf.org  Mon Jan 14 07:51:12 2013
Return-Path: <internet-drafts@ietf.org>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E95C621F8925; Mon, 14 Jan 2013 07:51:11 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.525
X-Spam-Level: 
X-Spam-Status: No, score=-102.525 tagged_above=-999 required=5 tests=[AWL=0.074, BAYES_00=-2.599, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id TApPjX4pfePm; Mon, 14 Jan 2013 07:51:11 -0800 (PST)
Received: from ietfa.amsl.com (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 76E2D21F8919; Mon, 14 Jan 2013 07:51:11 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
From: internet-drafts@ietf.org
To: i-d-announce@ietf.org
X-Test-IDTracker: no
X-IETF-IDTracker: 4.37
Message-ID: <20130114155111.1010.16060.idtracker@ietfa.amsl.com>
Date: Mon, 14 Jan 2013 07:51:11 -0800
Cc: trill@ietf.org
Subject: [trill] I-D Action: draft-ietf-trill-directory-framework-02.txt
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 14 Jan 2013 15:51:17 -0000

A New Internet-Draft is available from the on-line Internet-Drafts director=
ies.
 This draft is a work item of the Transparent Interconnection of Lots of Li=
nks Working Group of the IETF.

	Title           : TRILL (Transparent Interconnection of Lots of Links) Edg=
e Directory Assistance Framework
	Author(s)       : Linda Dunbar
                          Donald Eastlake
                          Intel Labs
                          Igor Gashinsky
	Filename        : draft-ietf-trill-directory-framework-02.txt
	Pages           : 14
	Date            : 2013-01-14

Abstract:
   Edge RBridges currently learn the mapping between MAC addresses and
   their egress RBridges by observing the data packets ingressed or
   egressed and by the TRILL ESADI protocol. When ingress RBridge
   receives a data frame whose destination address (MAC&VLAN) that
   RBridge does not know, the data frame is flooded within the VLAN
   across the TRILL campus.

   This draft describes the framework for using directory services to
   assist edge RBridges by reducing multi-destination frames,
   particularly unknown unicast frames flooding, and ARP/ND, improving
   TRILL (Transparent Interconnection of Lots of Links) network
   scalability in environments such as data centers.


The IETF datatracker status page for this draft is:
https://datatracker.ietf.org/doc/draft-ietf-trill-directory-framework

There's also a htmlized version available at:
http://tools.ietf.org/html/draft-ietf-trill-directory-framework-02

A diff from the previous version is available at:
http://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-trill-directory-framework-02


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


From linda.dunbar@huawei.com  Mon Jan 14 08:10:42 2013
Return-Path: <linda.dunbar@huawei.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2D19721F8925; Mon, 14 Jan 2013 08:10:42 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.499
X-Spam-Level: 
X-Spam-Status: No, score=-6.499 tagged_above=-999 required=5 tests=[AWL=0.100,  BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id YlD3zFmjdc6w; Mon, 14 Jan 2013 08:10:40 -0800 (PST)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) by ietfa.amsl.com (Postfix) with ESMTP id 5214A21F88E4; Mon, 14 Jan 2013 08:10:39 -0800 (PST)
Received: from 172.18.7.190 (EHLO lhreml203-edg.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.5-GA FastPath queued) with ESMTP id AOT54149; Mon, 14 Jan 2013 16:10:37 +0000 (GMT)
Received: from LHREML405-HUB.china.huawei.com (10.201.5.242) by lhreml203-edg.huawei.com (172.18.7.221) with Microsoft SMTP Server (TLS) id 14.1.323.3; Mon, 14 Jan 2013 16:10:28 +0000
Received: from DFWEML405-HUB.china.huawei.com (10.193.5.102) by lhreml405-hub.china.huawei.com (10.201.5.242) with Microsoft SMTP Server (TLS) id 14.1.323.3; Mon, 14 Jan 2013 16:10:22 +0000
Received: from DFWEML505-MBB.china.huawei.com ([169.254.1.58]) by dfweml405-hub.china.huawei.com ([10.193.5.102]) with mapi id 14.01.0323.003; Mon, 14 Jan 2013 08:10:20 -0800
From: Linda Dunbar <linda.dunbar@huawei.com>
To: "internet-drafts@ietf.org" <internet-drafts@ietf.org>, "i-d-announce@ietf.org" <i-d-announce@ietf.org>
Thread-Topic: [trill] I-D Action: draft-ietf-trill-directory-framework-02.txt
Thread-Index: AQHN8m8ckhVnPAyL30CBHCQ1kNw3i5hI/WyQ
Date: Mon, 14 Jan 2013 16:10:20 +0000
Message-ID: <4A95BA014132FF49AE685FAB4B9F17F6450EB7A7@dfweml505-mbb.china.huawei.com>
References: <20130114155111.1010.16060.idtracker@ietfa.amsl.com>
In-Reply-To: <20130114155111.1010.16060.idtracker@ietfa.amsl.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.136.33]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Cc: "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] I-D Action: draft-ietf-trill-directory-framework-02.txt
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 14 Jan 2013 16:10:42 -0000

Thanks to everyone who provided the valuable comments to "draft-ietf-trill-=
directory-framework" during the last WGLC period. This updated draft has in=
corporated all the comments received. Please review again and let us know i=
f you have more issues or comments.=20

Thanks, Linda

> -----Original Message-----
> From: trill-bounces@ietf.org [mailto:trill-bounces@ietf.org] On Behalf
> Of internet-drafts@ietf.org
> Sent: Monday, January 14, 2013 9:51 AM
> To: i-d-announce@ietf.org
> Cc: trill@ietf.org
> Subject: [trill] I-D Action: draft-ietf-trill-directory-framework-
> 02.txt
>=20
>=20
> A New Internet-Draft is available from the on-line Internet-Drafts
> directories.
>  This draft is a work item of the Transparent Interconnection of Lots
> of Links Working Group of the IETF.
>=20
> 	Title           : TRILL (Transparent Interconnection of Lots of
> Links) Edge Directory Assistance Framework
> 	Author(s)       : Linda Dunbar
>                           Donald Eastlake
>                           Intel Labs
>                           Igor Gashinsky
> 	Filename        : draft-ietf-trill-directory-framework-02.txt
> 	Pages           : 14
> 	Date            : 2013-01-14
>=20
> Abstract:
>    Edge RBridges currently learn the mapping between MAC addresses and
>    their egress RBridges by observing the data packets ingressed or
>    egressed and by the TRILL ESADI protocol. When ingress RBridge
>    receives a data frame whose destination address (MAC&VLAN) that
>    RBridge does not know, the data frame is flooded within the VLAN
>    across the TRILL campus.
>=20
>    This draft describes the framework for using directory services to
>    assist edge RBridges by reducing multi-destination frames,
>    particularly unknown unicast frames flooding, and ARP/ND, improving
>    TRILL (Transparent Interconnection of Lots of Links) network
>    scalability in environments such as data centers.
>=20
>=20
> The IETF datatracker status page for this draft is:
> https://datatracker.ietf.org/doc/draft-ietf-trill-directory-framework
>=20
> There's also a htmlized version available at:
> http://tools.ietf.org/html/draft-ietf-trill-directory-framework-02
>=20
> A diff from the previous version is available at:
> http://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-trill-directory-framework-
> 02
>=20
>=20
> Internet-Drafts are also available by anonymous FTP at:
> ftp://ftp.ietf.org/internet-drafts/
>=20
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill

From liyizhou@huawei.com  Mon Jan 14 19:01:48 2013
Return-Path: <liyizhou@huawei.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2E17B11E80A5 for <trill@ietfa.amsl.com>; Mon, 14 Jan 2013 19:01:48 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.599
X-Spam-Level: 
X-Spam-Status: No, score=-6.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0JtElGAhieEx for <trill@ietfa.amsl.com>; Mon, 14 Jan 2013 19:01:47 -0800 (PST)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) by ietfa.amsl.com (Postfix) with ESMTP id 9BC6C11E80A3 for <trill@ietf.org>; Mon, 14 Jan 2013 19:01:46 -0800 (PST)
Received: from 172.18.7.190 (EHLO lhreml203-edg.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.5-GA FastPath queued) with ESMTP id ANN81275; Tue, 15 Jan 2013 03:01:45 +0000 (GMT)
Received: from LHREML405-HUB.china.huawei.com (10.201.5.242) by lhreml203-edg.huawei.com (172.18.7.221) with Microsoft SMTP Server (TLS) id 14.1.323.3; Tue, 15 Jan 2013 03:01:35 +0000
Received: from SZXEML417-HUB.china.huawei.com (10.82.67.156) by lhreml405-hub.china.huawei.com (10.201.5.242) with Microsoft SMTP Server (TLS) id 14.1.323.3; Tue, 15 Jan 2013 03:01:42 +0000
Received: from l63746 (10.138.122.43) by smtpscn.huawei.com (10.82.67.156) with Microsoft SMTP Server (TLS) id 14.1.323.3; Tue, 15 Jan 2013 11:01:38 +0800
From: Yizhou Li <liyizhou@huawei.com>
To: <trill@ietf.org>
Date: Tue, 15 Jan 2013 11:01:38 +0800
Message-ID: <048b01cdf2cc$a3476df0$e9d649d0$@com>
MIME-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
X-Mailer: Microsoft Office Outlook 12.0
Thread-Index: Ac3yy49L7yb8hTXkS3SH6tOLj4p7kgAAIAZA
Content-Language: zh-cn
X-Originating-IP: [10.138.122.43]
X-CFilter-Loop: Reflected
Subject: [trill] FW: New Version Notification for draft-yizhou-trill-tree-selection-02.txt
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 15 Jan 2013 03:01:48 -0000

Dear folks,

We have updated the draft. A new section (section 5) was added to talk =
about another data plane approach, aggregated tree nickname, based on =
comments received.=20
The co-authors would like to ask for more comments on both approaches =
for the multicast forwarding table size reduction. Thank you.

Yizhou

-----Original Message-----
From: internet-drafts@ietf.org [mailto:internet-drafts@ietf.org]=20
Sent: Tuesday, January 15, 2013 10:54 AM
To: liyizhou@huawei.com
Cc: sunny.rajagopalan@us.ibm.com; haoweiguo@huawei.com; =
somnath.chatterjee01@gmail.com; radia@alum.mit.edu; naveen@broadcom.com
Subject: New Version Notification for =
draft-yizhou-trill-tree-selection-02.txt


A new version of I-D, draft-yizhou-trill-tree-selection-02.txt
has been successfully submitted by Yizhou Li and posted to the
IETF repository.

Filename:	 draft-yizhou-trill-tree-selection
Revision:	 02
Title:		 VLAN based Tree Selection for Multi-destination Frames
Creation date:	 2013-01-15
WG ID:		 Individual Submission
Number of pages: 17
URL:             =
http://www.ietf.org/internet-drafts/draft-yizhou-trill-tree-selection-02.=
txt
Status:          =
http://datatracker.ietf.org/doc/draft-yizhou-trill-tree-selection
Htmlized:        =
http://tools.ietf.org/html/draft-yizhou-trill-tree-selection-02
Diff:            =
http://www.ietf.org/rfcdiff?url2=3Ddraft-yizhou-trill-tree-selection-02

Abstract:
   TRILL uses the distribution trees to deliver multi-destination =
frames.
   Multiple trees can be used by an ingress RBridge for different flow
   based on VLAN and/or multicast group. Different ingress RBridges may
   choose different distribution trees for the same VLAN and/or
   multicast group traffic. Distribution trees are normally pruned based
   on VLAN.

   For any RBridge RBn, if RBn has downstream receivers of VLAN x in a
   distribution tree t, there will be an entry of (t, x, port list) in
   the multicast forwarding table on RBn. If there are n trees and m
   VLANs, the multicast forwarding table size on RBn is typically n*m
   entries. The value of m is up to 4096 and n is the total number of
   distribution trees in the campus. If fine grained labeling is
   implemented or finer granularity filtering such as VLAN plus L2/L3
   multicast address is used for pruning, the multicast forwarding table
   size further increases dramatically. TRILL multicast forwarding table
   size is limited by hardware and L3 multicasting may share the same
   table with it in hardware implementations. Therefore multicast table
   entry is a precious resource. This document specifies a VLAN based
   tree selection mechanism to reduce the TRILL multicast forwarding
   table size on RBridge.

                                                                         =
        =20


The IETF Secretariat


From ostokes@extremenetworks.com  Wed Jan 23 08:25:16 2013
Return-Path: <ostokes@extremenetworks.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4B23621F8753 for <trill@ietfa.amsl.com>; Wed, 23 Jan 2013 08:25:16 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.598
X-Spam-Level: 
X-Spam-Status: No, score=-2.598 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id YwfhiLeTPGzb for <trill@ietfa.amsl.com>; Wed, 23 Jan 2013 08:25:15 -0800 (PST)
Received: from ussc-casht-p1.extremenetworks.com (ussc-casht-p1.extremenetworks.com [207.179.9.62]) by ietfa.amsl.com (Postfix) with ESMTP id 0C5E921F85EE for <trill@ietf.org>; Wed, 23 Jan 2013 08:25:15 -0800 (PST)
Received: from USEXCHANGE.corp.extremenetworks.com ([10.0.4.74]) by ussc-casht-p1.corp.extremenetworks.com ([10.0.4.73]) with mapi; Wed, 23 Jan 2013 08:25:14 -0800
From: Olen Stokes <ostokes@extremenetworks.com>
To: "trill@ietf.org" <trill@ietf.org>
Date: Wed, 23 Jan 2013 08:25:13 -0800
Thread-Topic: Network connectivity and AF status
Thread-Index: Ac35ciNMsWIDM9nvTbazyk2L7v4dlQ==
Message-ID: <A3C4E51A53661B4EBEE7C5F5E6FCDEB50275CEB1556D@USEXCHANGE.corp.extremenetworks.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US
Content-Type: multipart/alternative; boundary="_000_A3C4E51A53661B4EBEE7C5F5E6FCDEB50275CEB1556DUSEXCHANGEc_"
MIME-Version: 1.0
Subject: [trill] Network connectivity and AF status
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 23 Jan 2013 16:25:16 -0000

--_000_A3C4E51A53661B4EBEE7C5F5E6FCDEB50275CEB1556DUSEXCHANGEc_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Take a simple network where RB1 and RB2 are connected to a TRILL network an=
d also connected together over an access link (bridged network).  Assume th=
at RB1 is DRB on the access link.  Assume that RB1 is the AF for half of th=
e access VLANs and that RB2 is the AF for the other half of the access VLAN=
s.

One scenario of interest is to assume that RB1 loses all of its connectivit=
y to the TRILL network (all of its trunk ports go down).  RB1 can simply gi=
ve RB2 AF status for all of the VLANs and access traffic can all go through=
 RB2 to use the TRILL network.

A second scenario of interest is to assume that RB2 loses all of its connec=
tivity to the TRILL network.  In this case, RB2 would like for RB1 to becom=
e AF for all of the VLANs.  The question becomes, how does RB2 signal to DR=
B RB1 to revoke AF from RB2 for all of its AF VLANs?   The following method=
s come to mind:


1)      Set the "disable port" bit in RB2's Hellos to RB1

2)      Set the "end-station service disable (trunk port)" bit in RB2

3)      Set the "P2P" bit in RB2

4)      Send a null set of Announcing VLANs

Setting the "disable port" bit seems to be the most appropriate.  However, =
I cannot find where this is discussed in RFC 6325 or RFC 6439.  Have I miss=
ed it someplace?  Is there another more appropriate method to accomplish th=
is?

Cheers,
Olen

--_000_A3C4E51A53661B4EBEE7C5F5E6FCDEB50275CEB1556DUSEXCHANGEc_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micr=
osoft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" xmlns=3D"http:=
//www.w3.org/TR/REC-html40"><head><meta http-equiv=3DContent-Type content=
=3D"text/html; charset=3Dus-ascii"><meta name=3DGenerator content=3D"Micros=
oft Word 14 (filtered medium)"><style><!--
/* Font Definitions */
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:11.0pt;
	font-family:"Calibri","sans-serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
p.MsoListParagraph, li.MsoListParagraph, div.MsoListParagraph
	{mso-style-priority:34;
	margin-top:0in;
	margin-right:0in;
	margin-bottom:0in;
	margin-left:.5in;
	margin-bottom:.0001pt;
	font-size:11.0pt;
	font-family:"Calibri","sans-serif";}
span.EmailStyle17
	{mso-style-type:personal-compose;
	font-family:"Calibri","sans-serif";
	color:windowtext;}
.MsoChpDefault
	{mso-style-type:export-only;
	font-family:"Calibri","sans-serif";}
@page WordSection1
	{size:8.5in 11.0in;
	margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
	{page:WordSection1;}
/* List Definitions */
@list l0
	{mso-list-id:855731808;
	mso-list-type:hybrid;
	mso-list-template-ids:-682189932 67698705 67698713 67698715 67698703 67698=
713 67698715 67698703 67698713 67698715;}
@list l0:level1
	{mso-level-text:"%1\)";
	mso-level-tab-stop:none;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level2
	{mso-level-number-format:alpha-lower;
	mso-level-tab-stop:none;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level3
	{mso-level-number-format:roman-lower;
	mso-level-tab-stop:none;
	mso-level-number-position:right;
	text-indent:-9.0pt;}
@list l0:level4
	{mso-level-tab-stop:none;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level5
	{mso-level-number-format:alpha-lower;
	mso-level-tab-stop:none;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level6
	{mso-level-number-format:roman-lower;
	mso-level-tab-stop:none;
	mso-level-number-position:right;
	text-indent:-9.0pt;}
@list l0:level7
	{mso-level-tab-stop:none;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level8
	{mso-level-number-format:alpha-lower;
	mso-level-tab-stop:none;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level9
	{mso-level-number-format:roman-lower;
	mso-level-tab-stop:none;
	mso-level-number-position:right;
	text-indent:-9.0pt;}
ol
	{margin-bottom:0in;}
ul
	{margin-bottom:0in;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]--></head><body lang=3DEN-US link=3Dblue vli=
nk=3Dpurple><div class=3DWordSection1><p class=3DMsoNormal>Take a simple ne=
twork where RB1 and RB2 are connected to a TRILL network and also connected=
 together over an access link (bridged network).&nbsp; Assume that RB1 is D=
RB on the access link. &nbsp;Assume that RB1 is the AF for half of the acce=
ss VLANs and that RB2 is the AF for the other half of the access VLANs.<o:p=
></o:p></p><p class=3DMsoNormal><o:p>&nbsp;</o:p></p><p class=3DMsoNormal>O=
ne scenario of interest is to assume that RB1 loses all of its connectivity=
 to the TRILL network (all of its trunk ports go down).&nbsp; RB1 can simpl=
y give RB2 AF status for all of the VLANs and access traffic can all go thr=
ough RB2 to use the TRILL network.<o:p></o:p></p><p class=3DMsoNormal><o:p>=
&nbsp;</o:p></p><p class=3DMsoNormal>A second scenario of interest is to as=
sume that RB2 loses all of its connectivity to the TRILL network.&nbsp; In =
this case, RB2 would like for RB1 to become AF for all of the VLANs.&nbsp; =
The question becomes, how does RB2 signal to DRB RB1 to revoke AF from RB2 =
for all of its AF VLANs?&nbsp; &nbsp;The following methods come to mind:<o:=
p></o:p></p><p class=3DMsoNormal><o:p>&nbsp;</o:p></p><p class=3DMsoListPar=
agraph style=3D'text-indent:-.25in;mso-list:l0 level1 lfo1'><![if !supportL=
ists]><span style=3D'mso-list:Ignore'>1)<span style=3D'font:7.0pt "Times Ne=
w Roman"'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; </span></span><![endif]>Set the &#=
8220;disable port&#8221; bit in RB2&#8217;s Hellos to RB1<o:p></o:p></p><p =
class=3DMsoListParagraph style=3D'text-indent:-.25in;mso-list:l0 level1 lfo=
1'><![if !supportLists]><span style=3D'mso-list:Ignore'>2)<span style=3D'fo=
nt:7.0pt "Times New Roman"'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; </span></span><!=
[endif]>Set the &#8220;end-station service disable (trunk port)&#8221; bit =
in RB2<o:p></o:p></p><p class=3DMsoListParagraph style=3D'text-indent:-.25i=
n;mso-list:l0 level1 lfo1'><![if !supportLists]><span style=3D'mso-list:Ign=
ore'>3)<span style=3D'font:7.0pt "Times New Roman"'>&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp; </span></span><![endif]>Set the &#8220;P2P&#8221; bit in RB2<o:p></=
o:p></p><p class=3DMsoListParagraph style=3D'text-indent:-.25in;mso-list:l0=
 level1 lfo1'><![if !supportLists]><span style=3D'mso-list:Ignore'>4)<span =
style=3D'font:7.0pt "Times New Roman"'>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; </spa=
n></span><![endif]>Send a null set of Announcing VLANs<o:p></o:p></p><p cla=
ss=3DMsoNormal><o:p>&nbsp;</o:p></p><p class=3DMsoNormal>Setting the &#8220=
;disable port&#8221; bit seems to be the most appropriate.&nbsp; However, I=
 cannot find where this is discussed in RFC 6325 or RFC 6439.&nbsp; Have I =
missed it someplace?&nbsp; Is there another more appropriate method to acco=
mplish this?<o:p></o:p></p><p class=3DMsoNormal><o:p>&nbsp;</o:p></p><p cla=
ss=3DMsoNormal>Cheers,<o:p></o:p></p><p class=3DMsoNormal>Olen<o:p></o:p></=
p></div></body></html>=

--_000_A3C4E51A53661B4EBEE7C5F5E6FCDEB50275CEB1556DUSEXCHANGEc_--

From d3e3e3@gmail.com  Wed Jan 23 10:38:37 2013
Return-Path: <d3e3e3@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8356721F85D4 for <trill@ietfa.amsl.com>; Wed, 23 Jan 2013 10:38:37 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.599
X-Spam-Level: 
X-Spam-Status: No, score=-103.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 5t8I8CVP8pmN for <trill@ietfa.amsl.com>; Wed, 23 Jan 2013 10:38:36 -0800 (PST)
Received: from mail-qa0-f48.google.com (mail-qa0-f48.google.com [209.85.216.48]) by ietfa.amsl.com (Postfix) with ESMTP id C215C21F85B2 for <trill@ietf.org>; Wed, 23 Jan 2013 10:38:36 -0800 (PST)
Received: by mail-qa0-f48.google.com with SMTP id l8so766708qaq.0 for <trill@ietf.org>; Wed, 23 Jan 2013 10:38:36 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=x-received:mime-version:in-reply-to:references:from:date:message-id :subject:to:cc:content-type:content-transfer-encoding; bh=8PgahCXqqlIqEFIKzpA5mw8M4lr9ZrqINtI7SxvUzOk=; b=O1FrTfbi7cF/5+V7IE3PJW/drDE6FC+BnF6d7NVt/Iz2DzdSQF3tVQwi9mjc6RrGeJ IClgOKKgvxEG5J26otgwd3kBlpUAOv81c//fbc1fSEVu6Ty7nOMShXUJdpvVSmx/4MCn b6w1Rl/Kas7YNKg2YmXpxUJaosa2e+LDHLgt8lcHDEeGi2vMEiWrxi7iXZEj15V3hbb8 PoAebozJ9PeIaRx/i1t4qp0ygxQsTwgS3xTNJXSRca+L+uuxFY70r5NmuWf5xrvMmoQB iy7hUMEBMYtUkwXiXOvA4KsDOPki/YhsK2i9waP2r8EiduQAise3sWUYaES6C6rbAHWR yR3A==
X-Received: by 10.49.118.162 with SMTP id kn2mr2750729qeb.65.1358966316252; Wed, 23 Jan 2013 10:38:36 -0800 (PST)
MIME-Version: 1.0
Received: by 10.49.116.140 with HTTP; Wed, 23 Jan 2013 10:38:16 -0800 (PST)
In-Reply-To: <A3C4E51A53661B4EBEE7C5F5E6FCDEB50275CEB1556D@USEXCHANGE.corp.extremenetworks.com>
References: <A3C4E51A53661B4EBEE7C5F5E6FCDEB50275CEB1556D@USEXCHANGE.corp.extremenetworks.com>
From: Donald Eastlake <d3e3e3@gmail.com>
Date: Wed, 23 Jan 2013 13:38:16 -0500
Message-ID: <CAF4+nEF978eQviPE7kTEbFj8_Tnwvb6R7oPubJ-5njqOsQO45A@mail.gmail.com>
To: Olen Stokes <ostokes@extremenetworks.com>
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable
Cc: "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Network connectivity and AF status
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 23 Jan 2013 18:38:37 -0000

Hi Olen,

On Wed, Jan 23, 2013 at 11:25 AM, Olen Stokes
<ostokes@extremenetworks.com> wrote:
> Take a simple network where RB1 and RB2 are connected to a TRILL network =
and
> also connected together over an access link (bridged network).  Assume th=
at
> RB1 is DRB on the access link.  Assume that RB1 is the AF for half of the
> access VLANs and that RB2 is the AF for the other half of the access VLAN=
s.
>
> One scenario of interest is to assume that RB1 loses all of its connectiv=
ity
> to the TRILL network (all of its trunk ports go down).  RB1 can simply gi=
ve
> RB2 AF status for all of the VLANs and access traffic can all go through =
RB2
> to use the TRILL network.
>
> A second scenario of interest is to assume that RB2 loses all of its
> connectivity to the TRILL network.  In this case, RB2 would like for RB1 =
to
> become AF for all of the VLANs.  The question becomes, how does RB2 signa=
l
> to DRB RB1 to revoke AF from RB2 for all of its AF VLANs?   The following
> methods come to mind:
>
> 1)      Set the =93disable port=94 bit in RB2=92s Hellos to RB1

I believe there actually isn't such a bit in Hellos. (It would be in
the Special VLANs and Flags sub-TLV.) The disable bit is a
configuration bit associated with a TRILL switch port. As shown in
Section 6 of draft-ietf-trill-clear-correct-06.txt, turning on the
disable bit will stop any Hellos from being sent. RB1 will conclude
that RB2 is down and presumably start handling all native traffic
itself. Lack of Hellos is a bit slow, but if RB1 wants to find out
more quickly when RB2 goes down, if should be using BFD
(draft-ietf-trill-rbridge-bfd-07.txt) or the like.

> 2)      Set the =93end-station service disable (trunk port)=94 bit in RB2

That seem like a good way to me. It will notify RB1 more rapidly. It
takes, typically, three Hellos times to decide the other end has died
while the "end-station service disable" bit could be noticed in the
very next Hello. RB1 would be specifically prohibited from appointing
RB2 as an Appointed Forwarder by the text about this configuration bit
on page 72 of RFC 6325,

> 3)      Set the =93P2P=94 bit in RB2

That is also a bit not in the Hello. It would cause RB2 to start
sending IS-IS P2P Hellos. RB1 should ignore these Hellos (unless it is
also configured for P2P, which is inconsistent with the situation as
described) as provided in Section 4.2.5 of RFC 6325. So, logically,
this is the same as 1. RB1 stops seeing any Hellos that it can accept
from RB2 and the adjacency falls away.

> 4)      Send a null set of Announcing VLANs

Well, you could use the Enabled VLANs sub-TLV to announce a null set
of VLANs enabled for end-station service.

The Announcing VLANs set is an internal configuration at a TRILL
switch port. It is used to optionally reduce the number of Hellos that
TRILL sends on a port.

> Setting the =93disable port=94 bit seems to be the most appropriate.  How=
ever, I
> cannot find where this is discussed in RFC 6325 or RFC 6439.  Have I miss=
ed
> it someplace?  Is there another more appropriate method to accomplish thi=
s?

See discussion above.

By the way, the draft-ietf-trill-clear-correct-06.txt,
draft-ietf-trill-rbridge-bfd-07.txt, and
draft-ietf-trill-rbridge-channel-08.txt drafts are all fully approved
and in the RFC Editor's queue. They are just as binding as published
RFCs.

Thanks,
Donald
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D
 Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
 155 Beaver Street, Milford, MA 01757 USA
 d3e3e3@gmail.com

> Cheers,
>
> Olen
>
>
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill
>

From d3e3e3@gmail.com  Wed Jan 23 14:41:19 2013
Return-Path: <d3e3e3@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A8B4521F85AC for <trill@ietfa.amsl.com>; Wed, 23 Jan 2013 14:41:19 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.599
X-Spam-Level: 
X-Spam-Status: No, score=-103.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id fI6RghOWvxuf for <trill@ietfa.amsl.com>; Wed, 23 Jan 2013 14:41:19 -0800 (PST)
Received: from mail-oa0-f52.google.com (mail-oa0-f52.google.com [209.85.219.52]) by ietfa.amsl.com (Postfix) with ESMTP id E3E7221F8550 for <trill@ietf.org>; Wed, 23 Jan 2013 14:41:18 -0800 (PST)
Received: by mail-oa0-f52.google.com with SMTP id o6so9220322oag.11 for <trill@ietf.org>; Wed, 23 Jan 2013 14:41:18 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=x-received:mime-version:in-reply-to:references:from:date:message-id :subject:to:cc:content-type:content-transfer-encoding; bh=EnG3ZFAthI/5h12yuyAZgMr2cDtU9ZDwUv9VVrOnFZA=; b=VGLlsobWDELXOaOQE99RxwxX+Jby5EViESpnrb8NyJmcFk19VXXiJDPIRQwqqtgWiS G5boxhWdSWgUx7M8t7Gg7qcn5yW/sbME2ja0aAru/AGyAzLVzxIotaFCr0yD3kX66Gk3 UCC7EP23FXH7I22CoUcH8DQlKdCQp0lBApb62Zi77JAP1Ys/I0X0nQt3hzPz8eD5IGrv CsT2WsTXJzsnrSijsMt2m/g22KCflKl0w/MXLquYQnQlNl46eOmD1fHf1bjD2K0gCz/G oAK7mEWx0j9lqihU69QhMQY+pxDTcwWfk/f8ixMs3NnwqevG4uBeoKf0h8M26rU9O2yJ 5VWA==
X-Received: by 10.60.31.206 with SMTP id c14mr2410311oei.88.1358980878452; Wed, 23 Jan 2013 14:41:18 -0800 (PST)
MIME-Version: 1.0
Received: by 10.76.110.239 with HTTP; Wed, 23 Jan 2013 14:40:58 -0800 (PST)
In-Reply-To: <CAF4+nEF978eQviPE7kTEbFj8_Tnwvb6R7oPubJ-5njqOsQO45A@mail.gmail.com>
References: <A3C4E51A53661B4EBEE7C5F5E6FCDEB50275CEB1556D@USEXCHANGE.corp.extremenetworks.com> <CAF4+nEF978eQviPE7kTEbFj8_Tnwvb6R7oPubJ-5njqOsQO45A@mail.gmail.com>
From: Donald Eastlake <d3e3e3@gmail.com>
Date: Wed, 23 Jan 2013 17:40:58 -0500
Message-ID: <CAF4+nEHwFUPf_bMUj7pfGMeN7v-7fjB8Mz79faVBURn2jSaG2w@mail.gmail.com>
To: Olen Stokes <ostokes@extremenetworks.com>
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable
Cc: "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Network connectivity and AF status
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 23 Jan 2013 22:41:19 -0000

An additional thought and a correction:

The correction is that reference below to section 4.2.5 of RFC 6325
should be to Section 4.2.4.

The thought is that RFC 6325 deliberately puts the criterion the DRB
uses to decide who to appoint a forwarder as out of scope to give
different implementers freedom to come up with different strategies.
But one reasonable factor to use is such decisions is the connectivity
to the campus of the TRILL switch being considered for appointment.
RB1 will learn from former neighbors of RB2 that they are no longer
connected to RB2 and it would be reasonable for RB1 to use that as the
basis for revoking any forwarder appointments of RB2.

Thanks,
Donald
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D
 Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
 155 Beaver Street, Milford, MA 01757 USA
 d3e3e3@gmail.com


On Wed, Jan 23, 2013 at 1:38 PM, Donald Eastlake <d3e3e3@gmail.com> wrote:
> Hi Olen,
>
> On Wed, Jan 23, 2013 at 11:25 AM, Olen Stokes
> <ostokes@extremenetworks.com> wrote:
>> Take a simple network where RB1 and RB2 are connected to a TRILL network=
 and
>> also connected together over an access link (bridged network).  Assume t=
hat
>> RB1 is DRB on the access link.  Assume that RB1 is the AF for half of th=
e
>> access VLANs and that RB2 is the AF for the other half of the access VLA=
Ns.
>>
>> One scenario of interest is to assume that RB1 loses all of its connecti=
vity
>> to the TRILL network (all of its trunk ports go down).  RB1 can simply g=
ive
>> RB2 AF status for all of the VLANs and access traffic can all go through=
 RB2
>> to use the TRILL network.
>>
>> A second scenario of interest is to assume that RB2 loses all of its
>> connectivity to the TRILL network.  In this case, RB2 would like for RB1=
 to
>> become AF for all of the VLANs.  The question becomes, how does RB2 sign=
al
>> to DRB RB1 to revoke AF from RB2 for all of its AF VLANs?   The followin=
g
>> methods come to mind:
>>
>> 1)      Set the =93disable port=94 bit in RB2=92s Hellos to RB1
>
> I believe there actually isn't such a bit in Hellos. (It would be in
> the Special VLANs and Flags sub-TLV.) The disable bit is a
> configuration bit associated with a TRILL switch port. As shown in
> Section 6 of draft-ietf-trill-clear-correct-06.txt, turning on the
> disable bit will stop any Hellos from being sent. RB1 will conclude
> that RB2 is down and presumably start handling all native traffic
> itself. Lack of Hellos is a bit slow, but if RB1 wants to find out
> more quickly when RB2 goes down, if should be using BFD
> (draft-ietf-trill-rbridge-bfd-07.txt) or the like.
>
>> 2)      Set the =93end-station service disable (trunk port)=94 bit in RB=
2
>
> That seem like a good way to me. It will notify RB1 more rapidly. It
> takes, typically, three Hellos times to decide the other end has died
> while the "end-station service disable" bit could be noticed in the
> very next Hello. RB1 would be specifically prohibited from appointing
> RB2 as an Appointed Forwarder by the text about this configuration bit
> on page 72 of RFC 6325,
>
>> 3)      Set the =93P2P=94 bit in RB2
>
> That is also a bit not in the Hello. It would cause RB2 to start
> sending IS-IS P2P Hellos. RB1 should ignore these Hellos (unless it is
> also configured for P2P, which is inconsistent with the situation as
> described) as provided in Section 4.2.5 of RFC 6325. So, logically,
> this is the same as 1. RB1 stops seeing any Hellos that it can accept
> from RB2 and the adjacency falls away.
>
>> 4)      Send a null set of Announcing VLANs
>
> Well, you could use the Enabled VLANs sub-TLV to announce a null set
> of VLANs enabled for end-station service.
>
> The Announcing VLANs set is an internal configuration at a TRILL
> switch port. It is used to optionally reduce the number of Hellos that
> TRILL sends on a port.
>
>> Setting the =93disable port=94 bit seems to be the most appropriate.  Ho=
wever, I
>> cannot find where this is discussed in RFC 6325 or RFC 6439.  Have I mis=
sed
>> it someplace?  Is there another more appropriate method to accomplish th=
is?
>
> See discussion above.
>
> By the way, the draft-ietf-trill-clear-correct-06.txt,
> draft-ietf-trill-rbridge-bfd-07.txt, and
> draft-ietf-trill-rbridge-channel-08.txt drafts are all fully approved
> and in the RFC Editor's queue. They are just as binding as published
> RFCs.
>
> Thanks,
> Donald
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D
>  Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
>  155 Beaver Street, Milford, MA 01757 USA
>  d3e3e3@gmail.com
>
>> Cheers,
>>
>> Olen
>>
>>
>> _______________________________________________
>> trill mailing list
>> trill@ietf.org
>> https://www.ietf.org/mailman/listinfo/trill
>>

From ostokes@extremenetworks.com  Wed Jan 23 14:50:01 2013
Return-Path: <ostokes@extremenetworks.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2E6D621F8786 for <trill@ietfa.amsl.com>; Wed, 23 Jan 2013 14:50:01 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.599
X-Spam-Level: 
X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[AWL=0.001,  BAYES_00=-2.599]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 3za7GNVfJZNs for <trill@ietfa.amsl.com>; Wed, 23 Jan 2013 14:50:00 -0800 (PST)
Received: from ussc-casht-p1.extremenetworks.com (ussc-casht-p1.extremenetworks.com [207.179.9.62]) by ietfa.amsl.com (Postfix) with ESMTP id 71FFD21F874F for <trill@ietf.org>; Wed, 23 Jan 2013 14:50:00 -0800 (PST)
Received: from USEXCHANGE.corp.extremenetworks.com ([10.0.4.74]) by ussc-casht-p1.corp.extremenetworks.com ([10.0.4.73]) with mapi; Wed, 23 Jan 2013 14:50:00 -0800
From: Olen Stokes <ostokes@extremenetworks.com>
To: Donald Eastlake <d3e3e3@gmail.com>
Date: Wed, 23 Jan 2013 14:49:59 -0800
Thread-Topic: [trill] Network connectivity and AF status
Thread-Index: Ac35usNdaUisa5apRWyCyqphCf4v4AAAN5eg
Message-ID: <A3C4E51A53661B4EBEE7C5F5E6FCDEB50275CEB1580A@USEXCHANGE.corp.extremenetworks.com>
References: <A3C4E51A53661B4EBEE7C5F5E6FCDEB50275CEB1556D@USEXCHANGE.corp.extremenetworks.com> <CAF4+nEF978eQviPE7kTEbFj8_Tnwvb6R7oPubJ-5njqOsQO45A@mail.gmail.com> <CAF4+nEHwFUPf_bMUj7pfGMeN7v-7fjB8Mz79faVBURn2jSaG2w@mail.gmail.com>
In-Reply-To: <CAF4+nEHwFUPf_bMUj7pfGMeN7v-7fjB8Mz79faVBURn2jSaG2w@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Cc: "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Network connectivity and AF status
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 23 Jan 2013 22:50:01 -0000

Agreed.  I was looking for something that could be signaled proactively by =
the RBridge that detects the loss of connectivity.

Thanks for the both the initial reply and this follow-up message.

Cheers,
Olen

-----Original Message-----
From: Donald Eastlake [mailto:d3e3e3@gmail.com]=20
Sent: Wednesday, January 23, 2013 5:41 PM
To: Olen Stokes
Cc: trill@ietf.org
Subject: Re: [trill] Network connectivity and AF status

An additional thought and a correction:

The correction is that reference below to section 4.2.5 of RFC 6325 should =
be to Section 4.2.4.

The thought is that RFC 6325 deliberately puts the criterion the DRB uses t=
o decide who to appoint a forwarder as out of scope to give different imple=
menters freedom to come up with different strategies.
But one reasonable factor to use is such decisions is the connectivity to t=
he campus of the TRILL switch being considered for appointment.
RB1 will learn from former neighbors of RB2 that they are no longer connect=
ed to RB2 and it would be reasonable for RB1 to use that as the basis for r=
evoking any forwarder appointments of RB2.

Thanks,
Donald
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D
 Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
 155 Beaver Street, Milford, MA 01757 USA  d3e3e3@gmail.com


On Wed, Jan 23, 2013 at 1:38 PM, Donald Eastlake <d3e3e3@gmail.com> wrote:
> Hi Olen,
>
> On Wed, Jan 23, 2013 at 11:25 AM, Olen Stokes=20
> <ostokes@extremenetworks.com> wrote:
>> Take a simple network where RB1 and RB2 are connected to a TRILL=20
>> network and also connected together over an access link (bridged=20
>> network).  Assume that
>> RB1 is DRB on the access link.  Assume that RB1 is the AF for half of=20
>> the access VLANs and that RB2 is the AF for the other half of the access=
 VLANs.
>>
>> One scenario of interest is to assume that RB1 loses all of its=20
>> connectivity to the TRILL network (all of its trunk ports go down). =20
>> RB1 can simply give
>> RB2 AF status for all of the VLANs and access traffic can all go=20
>> through RB2 to use the TRILL network.
>>
>> A second scenario of interest is to assume that RB2 loses all of its=20
>> connectivity to the TRILL network.  In this case, RB2 would like for=20
>> RB1 to become AF for all of the VLANs.  The question becomes, how does R=
B2 signal
>> to DRB RB1 to revoke AF from RB2 for all of its AF VLANs?   The followin=
g
>> methods come to mind:
>>
>> 1)      Set the "disable port" bit in RB2's Hellos to RB1
>
> I believe there actually isn't such a bit in Hellos. (It would be in=20
> the Special VLANs and Flags sub-TLV.) The disable bit is a=20
> configuration bit associated with a TRILL switch port. As shown in=20
> Section 6 of draft-ietf-trill-clear-correct-06.txt, turning on the=20
> disable bit will stop any Hellos from being sent. RB1 will conclude=20
> that RB2 is down and presumably start handling all native traffic=20
> itself. Lack of Hellos is a bit slow, but if RB1 wants to find out=20
> more quickly when RB2 goes down, if should be using BFD
> (draft-ietf-trill-rbridge-bfd-07.txt) or the like.
>
>> 2)      Set the "end-station service disable (trunk port)" bit in RB2
>
> That seem like a good way to me. It will notify RB1 more rapidly. It=20
> takes, typically, three Hellos times to decide the other end has died=20
> while the "end-station service disable" bit could be noticed in the=20
> very next Hello. RB1 would be specifically prohibited from appointing
> RB2 as an Appointed Forwarder by the text about this configuration bit=20
> on page 72 of RFC 6325,
>
>> 3)      Set the "P2P" bit in RB2
>
> That is also a bit not in the Hello. It would cause RB2 to start=20
> sending IS-IS P2P Hellos. RB1 should ignore these Hellos (unless it is=20
> also configured for P2P, which is inconsistent with the situation as
> described) as provided in Section 4.2.5 of RFC 6325. So, logically,=20
> this is the same as 1. RB1 stops seeing any Hellos that it can accept=20
> from RB2 and the adjacency falls away.
>
>> 4)      Send a null set of Announcing VLANs
>
> Well, you could use the Enabled VLANs sub-TLV to announce a null set=20
> of VLANs enabled for end-station service.
>
> The Announcing VLANs set is an internal configuration at a TRILL=20
> switch port. It is used to optionally reduce the number of Hellos that=20
> TRILL sends on a port.
>
>> Setting the "disable port" bit seems to be the most appropriate. =20
>> However, I cannot find where this is discussed in RFC 6325 or RFC=20
>> 6439.  Have I missed it someplace?  Is there another more appropriate me=
thod to accomplish this?
>
> See discussion above.
>
> By the way, the draft-ietf-trill-clear-correct-06.txt,
> draft-ietf-trill-rbridge-bfd-07.txt, and=20
> draft-ietf-trill-rbridge-channel-08.txt drafts are all fully approved=20
> and in the RFC Editor's queue. They are just as binding as published=20
> RFCs.
>
> Thanks,
> Donald
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D
>  Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
>  155 Beaver Street, Milford, MA 01757 USA  d3e3e3@gmail.com
>
>> Cheers,
>>
>> Olen
>>
>>
>> _______________________________________________
>> trill mailing list
>> trill@ietf.org
>> https://www.ietf.org/mailman/listinfo/trill
>>

From d3e3e3@gmail.com  Wed Jan 23 20:40:19 2013
Return-Path: <d3e3e3@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id AEA2A21F8462 for <trill@ietfa.amsl.com>; Wed, 23 Jan 2013 20:40:19 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.599
X-Spam-Level: 
X-Spam-Status: No, score=-103.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 96kRwmPHUTaf for <trill@ietfa.amsl.com>; Wed, 23 Jan 2013 20:40:19 -0800 (PST)
Received: from mail-oa0-f50.google.com (mail-oa0-f50.google.com [209.85.219.50]) by ietfa.amsl.com (Postfix) with ESMTP id 0886621F8716 for <trill@ietf.org>; Wed, 23 Jan 2013 20:40:18 -0800 (PST)
Received: by mail-oa0-f50.google.com with SMTP id n16so9433950oag.9 for <trill@ietf.org>; Wed, 23 Jan 2013 20:40:18 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=x-received:mime-version:from:date:message-id:subject:to :content-type; bh=xZ75+thQ3SDVt7cqdi28A7VfZxXYqmBTchMBOXKC9yg=; b=YGHmCzvOCdOxUF/CKpJuTJ9BmcUTwOMpWxcg1BHuCC7lgUWmYyhjJc2a9KDrahMb0l /k7Jb2wrN0S0mqdddPo3vtY2NeSBAzdwxKE0CG/GYWRS6DdaCNC4nVBfXWFLkvbkaIFm 26WPw2RXoz/6/4/EJbLZ7lMT8BMNSwDr76NLhP+HiFkYP8gUX7rll9JVQ3R17uSjvUme i+XPZAOsuc47og6g9ioIisYVnunDPvvUJgOChJr/e66i2Oehf3wlyn6V6ZkcRSbRMAMM /nA9XtGuv+Ad5LQyKev8gHfCEK0ubMJA6+JKgv/Z7En2r98OSmNha0119DmHiRU1bA8S eV9w==
X-Received: by 10.60.169.240 with SMTP id ah16mr522792oec.9.1359002418584; Wed, 23 Jan 2013 20:40:18 -0800 (PST)
MIME-Version: 1.0
Received: by 10.76.110.239 with HTTP; Wed, 23 Jan 2013 20:39:58 -0800 (PST)
From: Donald Eastlake <d3e3e3@gmail.com>
Date: Wed, 23 Jan 2013 23:39:58 -0500
Message-ID: <CAF4+nEGQfV5zGLx_FtYN+tfpqmufrtTxYVO43kXjJsyHuv207Q@mail.gmail.com>
To: trill@ietf.org
Content-Type: text/plain; charset=ISO-8859-1
Subject: [trill] FGL tweak for improved migration
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 24 Jan 2013 04:40:19 -0000

Hi,

As described in draft-ietf-trill-fine-labeling-04, a VL TRILL switch
cannot safely handle an FGL TRILL Data packet: it might egress such a
frame out an incorrect port in the wrong VLAN or re-order frames due
to incorrect ECMP flow classification or pre-maturely prune
distribution if the frame is multi-destination. The draft solves this
by blocking data communication between VL and FGL TRILL switches.
However, this makes migration from VL to FGL more difficult.

Since I believe that all, or almost all, existing VL RBridges can be
upgraded to Limited FGL by a software change, a small relaxation of
this blocking makes it possible to migrate more smoothly. The data
adjacency blocking can be relaxed so that VL and Limited FGL TRILL
switches are adjacent as long as there are no Full FGL TRILL switches
in the topology. This permits a smooth migration from VL to Limited
FGL by upgrading VL TRILL switches and adding Limited FGL TRILL
switches. After which Full FGL TRILL switches could be added, or
Limited FGL upgraded to Full. This safely maintains general data
connectivity.

Thanks,
Donald
=============================
 Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
 155 Beaver Street, Milford, MA 01757 USA
 d3e3e3@gmail.com

From d3e3e3@gmail.com  Thu Jan 24 20:40:19 2013
Return-Path: <d3e3e3@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id DB3E611E80DE for <trill@ietfa.amsl.com>; Thu, 24 Jan 2013 20:40:19 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.599
X-Spam-Level: 
X-Spam-Status: No, score=-103.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 2uqtPQTvWD6B for <trill@ietfa.amsl.com>; Thu, 24 Jan 2013 20:40:19 -0800 (PST)
Received: from mail-oa0-f46.google.com (mail-oa0-f46.google.com [209.85.219.46]) by ietfa.amsl.com (Postfix) with ESMTP id 5CD6911E80BF for <trill@ietf.org>; Thu, 24 Jan 2013 20:40:16 -0800 (PST)
Received: by mail-oa0-f46.google.com with SMTP id h16so11065094oag.33 for <trill@ietf.org>; Thu, 24 Jan 2013 20:40:16 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=x-received:mime-version:from:date:message-id:subject:to :content-type; bh=sLORa0wx5kbB1Ha/1cbmtYYp2WTo//HCEPPyJP3Kqt8=; b=YzviFSV6SShpz8atW2Wy+wBVIit8fPcwBV6jaOc5wcIbH56K/3DI2zdVwyzuRyhCVv XcZqdN7AmK/wVSCDvea7iG/sBD7h+4GTsEwzHQRomFAyElw8H2ueEOFdKfgPsv2Yko+M CmOqj9W/33lM2IL3Fh94IqCODMIIMix882vDz3ZbYSZYpjfWhlLbM2HAaE+BcVWOpa2W UPMwYuwl/LbToYOR9EStcOs9gikGA8pVSfaMyBNYqQ0PpuGq4AN56jXTxfbi4IevHFFA qsqDiuZkP8/0geJtmqnHZicISmv8XO/q1pk1ViooTYrUl8GFLoCgOrRHXZk5sIXx+0I9 CUgw==
X-Received: by 10.182.136.68 with SMTP id py4mr3505582obb.38.1359088815915; Thu, 24 Jan 2013 20:40:15 -0800 (PST)
MIME-Version: 1.0
Received: by 10.76.110.239 with HTTP; Thu, 24 Jan 2013 20:39:55 -0800 (PST)
From: Donald Eastlake <d3e3e3@gmail.com>
Date: Thu, 24 Jan 2013 23:39:55 -0500
Message-ID: <CAF4+nEHuZUYAoZQgrvsEQgXNht4GyKikdaAKLaOHyuzDOOnosA@mail.gmail.com>
To: trill@ietf.org
Content-Type: text/plain; charset=ISO-8859-1
Subject: [trill] RBridge Channel Ethertype allocated
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 25 Jan 2013 04:40:20 -0000

Ethertype 0x8946 has been allocated for RBridge Channel.
See draft-ietf-trill-rbridge-channel-08.txt.

Thanks,
Donald
=============================
 Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
 155 Beaver Street, Milford, MA 01757 USA
 d3e3e3@gmail.com

From radiaperlman@gmail.com  Fri Jan 25 11:00:33 2013
Return-Path: <radiaperlman@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7460221F87C3 for <trill@ietfa.amsl.com>; Fri, 25 Jan 2013 11:00:33 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.598
X-Spam-Level: 
X-Spam-Status: No, score=-3.598 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id rT0TqM7h5GGx for <trill@ietfa.amsl.com>; Fri, 25 Jan 2013 11:00:32 -0800 (PST)
Received: from mail-lb0-f180.google.com (mail-lb0-f180.google.com [209.85.217.180]) by ietfa.amsl.com (Postfix) with ESMTP id 32E9B21F88E2 for <trill@ietf.org>; Fri, 25 Jan 2013 11:00:26 -0800 (PST)
Received: by mail-lb0-f180.google.com with SMTP id q12so1255986lbc.39 for <trill@ietf.org>; Fri, 25 Jan 2013 11:00:25 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:x-received:date:message-id:subject:from:to :content-type; bh=MDGHsPzLwUS+HHLrkItRa+uqzrdT4WJBk5Je/1SsfQo=; b=MVZgTWzDXg4zU9YO6MnaI89c3yoDb6n1+yYmzSdUwRbCdQiAyyl4fOQwfwyzHdU3T3 AJYwJ4wt7HZCIOk1C/7XmFhNrqR5m5DgCvH53lVNvviDlAHEW9WJ35B0eLLLmx89CUW0 UN3xXkHwYxmWLZQelFs3c26NvDAhBwbf/SP2kF77ZqDuY6jjeV/4bXVxfLi6xWsrQAKt l45a/HUfFBwX42pmUFho2NoLN5nN0ze1ljX+i1AMnShEIzn0gnjf9TpAH+n78yvUPuU0 EwfC4OSnhmu8ieBQWU9FNkMtGiAU6cheeXjFP/vQlnErj1zkmhFiDyslxELyiFj86fL5 HYVw==
MIME-Version: 1.0
X-Received: by 10.152.121.212 with SMTP id lm20mr5979269lab.42.1359140425085;  Fri, 25 Jan 2013 11:00:25 -0800 (PST)
Received: by 10.112.64.17 with HTTP; Fri, 25 Jan 2013 11:00:24 -0800 (PST)
Date: Fri, 25 Jan 2013 11:00:24 -0800
Message-ID: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com>
From: Radia Perlman <radiaperlman@gmail.com>
To: trill@ietf.org
Content-Type: multipart/alternative; boundary=f46d04374ee13cb32e04d4218ad5
Subject: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 25 Jan 2013 19:00:33 -0000

--f46d04374ee13cb32e04d4218ad5
Content-Type: text/plain; charset=ISO-8859-1

I'm going to summarize three options for phasing-in FGL, and explain the
tradeoffs, to help people have informed opinions about the implications of
the three approaches.  There are infinite variations, of course, but as I
said, I'll describe three main ones.

Option 1:  Original draft, no changes.  There are two types of RBridges.
The first type is "VL" which only understands VLAN tags, and have unknown
and possibly dangerous behavior when they receive an FGL-labeled packet.
(e.g., decapsulate a packet onto a link where it is not allowed, or
mistakenly prune a multicast so it does not reach everywhere it should).
The second type is "FGL".  FGL guys need to do two things: 1) understand
FGL tags and do the right thing with them, and 2) ostrasize VL
guys...meaning that an FGL guy refuses to form an adjacency with a VL guy.
 This option is very simple for implementers of FGL, but the implication is
that you have to upgrade your entire campus to FGL at once.  There is no
coexistence.

Option 2: Two types of RBridges, but the 2nd type is different from option
1.  VL guys of course are the same as in option 1...they cannot be trusted
with FGL frames. The second type I will call "FGL-safe". An FGL-safe
RBridge must advertise in its LSP that it is FGL-safe, and it must not "do
anything bad" with FGL frames, meaning that it is allowed to ignore pruning
of FGL (or even VL frames) entirely...it just can't falsely drop FGL
frames.  And it must not decapsulate an FGL frame onto a link for which
that FGL frame doesn't belong. In this option, all RBridges must be
upgraded to FGL-safe, but it need not happen instantaneously...it's fine to
mix FGL-safe RBridges with VL RBridges...it's just not safe to inject FGL
frames yet.  Once all the RBridges have been upgraded to be FGL-safe, then
edge RBridges can start announcing they are connected to an FGL link, and
can start injecting FGL packets.  It is considered a misconfiguration if
you start injecting FGL frames before all the RBridges are upgraded to
FGL-safe, so an additional chore for an FGL-safe RBridge R1 is to examine
LSPs, and if any RBridge claims to be attached to an FGL link, then R1 must
ostrasize any VL neighbors.  (don't start ostrasizing VL guys until it is
necessary because of actually starting to use FGL frames, in other words).
This option is more work for the upgraded RBridges than option 1, and there
still isn't good coexistence with VL guys long-term (as option 3 will), but
it does allow upgrading RBridges one by one in a working campus without
causing disruption.

Option 3: Two types of RBridges.  VL, of course, is the same as in options
1 and 2.  This option makes FGL guys do more work, but allows maximal long
term coexistence of VL and FGL guys. In this option, FGL guys calculate
unicast paths to FGL edge guys that avoid any VL guys, and calculate at
least one FGL-friendly multicast tree that also avoids any VL guys. So
let's say R1 is an FGL guy.  R1 discards all LSPs from VL guys (ones that
don't advertise FGL capability in their LSP), when calculating paths to
other FGL RBridges. Then R1 calculates paths to the VL guys using all LSPs.
 Likewise, when calculating an FGL-friendly tree, R1 calculates a tree
through only FGL guys.   This option is more work for the upgraded RBridges
(because they have to calculate Dijkstra in two different ways, one for
reaching VL guys, and one for reaching FGL guys).  However, it does allow
having long-term coexistence with VL guys.  For instance, you could forever
keep some VL edge RBridges that communicate just fine through an FGL core.
 They can stay there forever, and still be able to communicate through the
core to all the links attaching to their VLAN.

--f46d04374ee13cb32e04d4218ad5
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div style=3D"color:rgb(34,34,34);font-family:arial,sans-serif;font-size:13=
px;background-color:rgb(255,255,255)">I&#39;m going to summarize three opti=
ons for phasing-in FGL, and explain the tradeoffs, to help people have info=
rmed opinions about the implications of the three approaches. =A0There are =
infinite variations, of course, but as I said, I&#39;ll describe three main=
 ones.</div>
<div style=3D"color:rgb(34,34,34);font-family:arial,sans-serif;font-size:13=
px;background-color:rgb(255,255,255)"><br></div><div style=3D"color:rgb(34,=
34,34);font-family:arial,sans-serif;font-size:13px;background-color:rgb(255=
,255,255)">
Option 1: =A0Original draft, no changes. =A0There are two types of RBridges=
. The first type is &quot;VL&quot; which only understands VLAN tags, and ha=
ve unknown and possibly dangerous behavior when they receive an FGL-labeled=
 packet. (e.g., decapsulate a packet onto a link where it is not allowed, o=
r mistakenly prune a multicast so it does not reach everywhere it should). =
The second type is &quot;FGL&quot;. =A0FGL guys need to do two things: 1) u=
nderstand FGL tags and do the right thing with them, and 2) ostrasize VL gu=
ys...meaning that an FGL guy refuses to form an adjacency with a VL guy. =
=A0This option is very simple for implementers of FGL, but the implication =
is that you have to upgrade your entire campus to FGL at once. =A0There is =
no coexistence.</div>
<div style=3D"color:rgb(34,34,34);font-family:arial,sans-serif;font-size:13=
px;background-color:rgb(255,255,255)"><br></div><div style=3D"color:rgb(34,=
34,34);font-family:arial,sans-serif;font-size:13px;background-color:rgb(255=
,255,255)">
Option 2: Two types of RBridges, but the 2nd type is different from option =
1. =A0VL guys of course are the same as in option 1...they cannot be truste=
d with FGL frames. The second type I will call &quot;FGL-safe&quot;. An FGL=
-safe RBridge must advertise in its LSP that it is FGL-safe, and it must no=
t &quot;do anything bad&quot; with FGL frames, meaning that it is allowed t=
o ignore pruning of FGL (or even VL frames) entirely...it just can&#39;t fa=
lsely drop FGL frames. =A0And it must not decapsulate an FGL frame onto a l=
ink for which that FGL frame doesn&#39;t belong. In this option, all RBridg=
es must be upgraded to FGL-safe, but it need not happen instantaneously...i=
t&#39;s fine to mix FGL-safe RBridges with VL RBridges...it&#39;s just not =
safe to inject FGL frames yet. =A0Once all the RBridges have been upgraded =
to be FGL-safe, then edge RBridges can start announcing they are connected =
to an FGL link, and can start injecting FGL packets. =A0It is considered a =
misconfiguration if you start injecting FGL frames before all the RBridges =
are upgraded to FGL-safe, so an additional chore for an FGL-safe RBridge R1=
 is to examine LSPs, and if any RBridge claims to be attached to an FGL lin=
k, then R1 must ostrasize any VL neighbors. =A0(don&#39;t start ostrasizing=
 VL guys until it is necessary because of actually starting to use FGL fram=
es, in other words). This option is more work for the upgraded RBridges tha=
n option 1, and there still isn&#39;t good coexistence with VL guys long-te=
rm (as option 3 will), but it does allow upgrading RBridges one by one in a=
 working campus without causing disruption.</div>
<div style=3D"color:rgb(34,34,34);font-family:arial,sans-serif;font-size:13=
px;background-color:rgb(255,255,255)"><br></div><div style=3D"color:rgb(34,=
34,34);font-family:arial,sans-serif;font-size:13px;background-color:rgb(255=
,255,255)">
Option 3: Two types of RBridges. =A0VL, of course, is the same as in option=
s 1 and 2. =A0This option makes FGL guys do more work, but allows maximal l=
ong term coexistence of VL and FGL guys. In this option, FGL guys calculate=
 unicast paths to FGL edge guys that avoid any VL guys, and calculate at le=
ast one FGL-friendly multicast tree that also avoids any VL guys. So let&#3=
9;s say R1 is an FGL guy. =A0R1 discards all LSPs from VL guys (ones that d=
on&#39;t advertise FGL capability in their LSP), when calculating paths to =
other FGL RBridges. Then R1 calculates paths to the VL guys using all LSPs.=
 =A0Likewise, when calculating an FGL-friendly tree, R1 calculates a tree t=
hrough only FGL guys. =A0 This option is more work for the upgraded RBridge=
s (because they have to calculate Dijkstra in two different ways, one for r=
eaching VL guys, and one for reaching FGL guys). =A0However, it does allow =
having long-term coexistence with VL guys. =A0For instance, you could forev=
er keep some VL edge RBridges that communicate just fine through an FGL cor=
e. =A0They can stay there forever, and still be able to communicate through=
 the core to all the links attaching to their VLAN.</div>

--f46d04374ee13cb32e04d4218ad5--

From internet-drafts@ietf.org  Sat Jan 26 07:47:13 2013
Return-Path: <internet-drafts@ietf.org>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8BF7521F88BD; Sat, 26 Jan 2013 07:47:13 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.512
X-Spam-Level: 
X-Spam-Status: No, score=-102.512 tagged_above=-999 required=5 tests=[AWL=0.087, BAYES_00=-2.599, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id o8IE7P64AnQN; Sat, 26 Jan 2013 07:47:12 -0800 (PST)
Received: from ietfa.amsl.com (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CA30321F8782; Sat, 26 Jan 2013 07:47:12 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
From: internet-drafts@ietf.org
To: i-d-announce@ietf.org
X-Test-IDTracker: no
X-IETF-IDTracker: 4.37
Message-ID: <20130126154712.20770.73781.idtracker@ietfa.amsl.com>
Date: Sat, 26 Jan 2013 07:47:12 -0800
Cc: trill@ietf.org
Subject: [trill] I-D Action: draft-ietf-trill-oam-req-05.txt
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 26 Jan 2013 15:47:13 -0000

A New Internet-Draft is available from the on-line Internet-Drafts director=
ies.
 This draft is a work item of the Transparent Interconnection of Lots of Li=
nks Working Group of the IETF.

	Title           : Requirements for Operations, Administration and Maintena=
nce (OAM) in TRILL (Transparent Interconnection of Lots of Links)
	Author(s)       : Tissa Senevirathne
                          David Bond
                          Sam Aldrin
                          Yizhou Li
                          Rohit Watve
	Filename        : draft-ietf-trill-oam-req-05.txt
	Pages           : 14
	Date            : 2013-01-26

Abstract:
   OAM (Operations, Administration and Maintenance) is a general term
   used to identify functions and toolsets to troubleshoot and monitor
   networks. This document presents OAM Requirements applicable to
   TRILL.


The IETF datatracker status page for this draft is:
https://datatracker.ietf.org/doc/draft-ietf-trill-oam-req

There's also a htmlized version available at:
http://tools.ietf.org/html/draft-ietf-trill-oam-req-05

A diff from the previous version is available at:
http://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-trill-oam-req-05


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


From wwwrun@rfc-editor.org  Mon Jan 28 17:34:31 2013
Return-Path: <wwwrun@rfc-editor.org>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C8E1421E805F; Mon, 28 Jan 2013 17:34:31 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.154
X-Spam-Level: 
X-Spam-Status: No, score=-102.154 tagged_above=-999 required=5 tests=[AWL=-0.154, BAYES_00=-2.599, J_CHICKENPOX_93=0.6, NO_RELAYS=-0.001, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 4Cmw3GY1sspn; Mon, 28 Jan 2013 17:34:31 -0800 (PST)
Received: from rfc-editor.org (rfc-editor.org [IPv6:2001:1890:123a::1:2f]) by ietfa.amsl.com (Postfix) with ESMTP id 4804F21E805E; Mon, 28 Jan 2013 17:34:31 -0800 (PST)
Received: by rfc-editor.org (Postfix, from userid 30) id C9D96B1E003; Mon, 28 Jan 2013 17:22:57 -0800 (PST)
To: ietf-announce@ietf.org, rfc-dist@rfc-editor.org
From: rfc-editor@rfc-editor.org
Message-Id: <20130129012257.C9D96B1E003@rfc-editor.org>
Date: Mon, 28 Jan 2013 17:22:57 -0800 (PST)
Cc: trill@ietf.org, rfc-editor@rfc-editor.org
Subject: [trill] RFC 6850 on Definitions of Managed Objects for Routing Bridges (RBridges)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 29 Jan 2013 01:34:31 -0000

A new Request for Comments is now available in online RFC libraries.

        
        RFC 6850

        Title:      Definitions of Managed Objects for 
                    Routing Bridges (RBridges) 
        Author:     A. Rijhsinghani, K. Zebrose
        Status:     Standards Track
        Stream:     IETF
        Date:       January 2013
        Mailbox:    anil@charter.net, 
                    zebrose@alum.mit.edu
        Pages:      59
        Characters: 104287
        Updates/Obsoletes/SeeAlso:   None

        I-D Tag:    draft-ietf-trill-rbridge-mib-10.txt

        URL:        http://www.rfc-editor.org/rfc/rfc6850.txt

This memo defines a portion of the Management Information Base (MIB)
for use with network management protocols.  In particular, it defines
objects for managing a Routing Bridge (RBridge), also known as a
TRILL Switch, based on the IETF TRILL (Transparent Interconnection of
Lots of Links) protocol.  [STANDARDS-TRACK]

This document is a product of the Transparent Interconnection of Lots of Links Working Group of the IETF.

This is now a Proposed Standard.

STANDARDS TRACK: This document specifies an Internet standards track
protocol for the Internet community,and requests discussion and suggestions
for improvements.  Please refer to the current edition of the Internet
Official Protocol Standards (STD 1) for the standardization state and
status of this protocol.  Distribution of this memo is unlimited.

This announcement is sent to the IETF-Announce and rfc-dist lists.
To subscribe or unsubscribe, see
  http://www.ietf.org/mailman/listinfo/ietf-announce
  http://mailman.rfc-editor.org/mailman/listinfo/rfc-dist

For searching the RFC series, see http://www.rfc-editor.org/rfcsearch.html.
For downloading RFCs, see http://www.rfc-editor.org/rfc.html.

Requests for special distribution should be addressed to either the
author of the RFC in question, or to rfc-editor@rfc-editor.org.  Unless
specifically noted otherwise on the RFC itself, all RFCs are for
unlimited distribution.


The RFC Editor Team
Association Management Solutions, LLC



From haoweiguo@huawei.com  Mon Jan 28 22:59:30 2013
Return-Path: <haoweiguo@huawei.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0D8C821F8915 for <trill@ietfa.amsl.com>; Mon, 28 Jan 2013 22:59:30 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 2.45
X-Spam-Level: **
X-Spam-Status: No, score=2.45 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, CHARSET_FARAWAY_HEADER=3.2, HTML_MESSAGE=0.001, MIME_8BIT_HEADER=0.3, MIME_BASE64_TEXT=1.753, MIME_CHARSET_FARAWAY=2.45, RCVD_IN_DNSWL_MED=-4, SARE_SUB_ENC_GB2312=1.345]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id PkOAIsYNMDa4 for <trill@ietfa.amsl.com>; Mon, 28 Jan 2013 22:59:29 -0800 (PST)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) by ietfa.amsl.com (Postfix) with ESMTP id 6F0CF21F8904 for <trill@ietf.org>; Mon, 28 Jan 2013 22:59:28 -0800 (PST)
Received: from 172.18.7.190 (EHLO lhreml204-edg.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.5-GA FastPath queued) with ESMTP id APE96868; Tue, 29 Jan 2013 06:59:25 +0000 (GMT)
Received: from LHREML404-HUB.china.huawei.com (10.201.5.218) by lhreml204-edg.china.huawei.com (172.18.7.223) with Microsoft SMTP Server (TLS) id 14.1.323.7; Tue, 29 Jan 2013 06:58:53 +0000
Received: from nkgeml405-hub.china.huawei.com (10.98.56.36) by lhreml404-hub.china.huawei.com (10.201.5.218) with Microsoft SMTP Server (TLS) id 14.1.323.7; Tue, 29 Jan 2013 14:59:22 +0800
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.101]) by nkgeml405-hub.china.huawei.com ([10.98.56.36]) with mapi id 14.01.0323.007; Tue, 29 Jan 2013 14:59:17 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: Radia Perlman <radiaperlman@gmail.com>, "trill@ietf.org" <trill@ietf.org>
Thread-Topic: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
Thread-Index: AQHN+zaweNKBMJg2i02ybQUizdQo4Zhf5BsZ
Date: Tue, 29 Jan 2013 06:59:17 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F651415@nkgeml501-mbs.china.huawei.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com>
In-Reply-To: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com>
Accept-Language: zh-CN, en-US
Content-Language: zh-CN
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.195.137.52]
Content-Type: multipart/alternative; boundary="_000_DD5FC8DE455C3348B94340C0AB5517334F651415nkgeml501mbschi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Subject: [trill] =?gb2312?b?tPC4tDogIEV4cGxhaW5pbmcgdGhyZWUgb3B0aW9ucyBm?= =?gb2312?b?b3IgdXBncmFkaW5nIHRvIEZHTAkoZmluZS1ncmFpbmVkLWxhYmVsaW5nKQ==?=
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 29 Jan 2013 06:59:30 -0000

--_000_DD5FC8DE455C3348B94340C0AB5517334F651415nkgeml501mbschi_
Content-Type: text/plain; charset="gb2312"
Content-Transfer-Encoding: base64

SGkgUmFkaWEsDQoNCkluIG9wdGlvbjM6DQoNCndoeSBGR0wgc3dpdGNoZXMgY2FsY3VsYXRlIHVu
aWNhc3QgcGF0aHMgdG8gRkdMIGVkZ2UgZ3V5cyB0aGF0IGF2b2lkIGFueSBWTCBndXlzLCBhbmQg
Y2FsY3VsYXRlIGF0IGxlYXN0IG9uZSBGR0wtZnJpZW5kbHkgbXVsdGljYXN0IHRyZWUgdGhhdCBh
bHNvIGF2b2lkcyBhbnkgVkwgZ3V5cz8gSWYgdHdvIEZHTCBzd2l0Y2hlcyBjYW4gb25seSB0cmFu
c21pdCBkYXRhIHRocm91Z2ggVkwgc3dpdGNoLCB0aGUgbWV0aG9kIHdpbGwgZGlzY29ubmVjdCBk
YXRhIGNoYW5uZWwgYmV0d2VlbiBGR0wgc3dpdGNoZXMuIEkgdGhpbmsgVkwgc3dpdGNoIGlzIHNh
ZmUgYXMgdHJhbnNpdCBub2RlIHRvIHRyYW5taXQgdW5pY2FzdCBGR0wgZGF0YSBwYWNrZXQuDQoN
CnRoYW5rcw0KDQp3ZWlndW8NCg0KX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18NCg0K
t6K8/sjLOiBSYWRpYSBQZXJsbWFuIFtyYWRpYXBlcmxtYW5AZ21haWwuY29tXQ0Kt6LLzcqxvOQ6
IDIwMTPE6jHUwjI2yNUgMzowMA0Ktb06IHRyaWxsQGlldGYub3JnDQrW98ziOiBbdHJpbGxdIEV4
cGxhaW5pbmcgdGhyZWUgb3B0aW9ucyBmb3IgdXBncmFkaW5nIHRvIEZHTCAoZmluZS1ncmFpbmVk
LWxhYmVsaW5nKQ0KDQpJJ20gZ29pbmcgdG8gc3VtbWFyaXplIHRocmVlIG9wdGlvbnMgZm9yIHBo
YXNpbmctaW4gRkdMLCBhbmQgZXhwbGFpbiB0aGUgdHJhZGVvZmZzLCB0byBoZWxwIHBlb3BsZSBo
YXZlIGluZm9ybWVkIG9waW5pb25zIGFib3V0IHRoZSBpbXBsaWNhdGlvbnMgb2YgdGhlIHRocmVl
IGFwcHJvYWNoZXMuICBUaGVyZSBhcmUgaW5maW5pdGUgdmFyaWF0aW9ucywgb2YgY291cnNlLCBi
dXQgYXMgSSBzYWlkLCBJJ2xsIGRlc2NyaWJlIHRocmVlIG1haW4gb25lcy4NCg0KT3B0aW9uIDE6
ICBPcmlnaW5hbCBkcmFmdCwgbm8gY2hhbmdlcy4gIFRoZXJlIGFyZSB0d28gdHlwZXMgb2YgUkJy
aWRnZXMuIFRoZSBmaXJzdCB0eXBlIGlzICJWTCIgd2hpY2ggb25seSB1bmRlcnN0YW5kcyBWTEFO
IHRhZ3MsIGFuZCBoYXZlIHVua25vd24gYW5kIHBvc3NpYmx5IGRhbmdlcm91cyBiZWhhdmlvciB3
aGVuIHRoZXkgcmVjZWl2ZSBhbiBGR0wtbGFiZWxlZCBwYWNrZXQuIChlLmcuLCBkZWNhcHN1bGF0
ZSBhIHBhY2tldCBvbnRvIGEgbGluayB3aGVyZSBpdCBpcyBub3QgYWxsb3dlZCwgb3IgbWlzdGFr
ZW5seSBwcnVuZSBhIG11bHRpY2FzdCBzbyBpdCBkb2VzIG5vdCByZWFjaCBldmVyeXdoZXJlIGl0
IHNob3VsZCkuIFRoZSBzZWNvbmQgdHlwZSBpcyAiRkdMIi4gIEZHTCBndXlzIG5lZWQgdG8gZG8g
dHdvIHRoaW5nczogMSkgdW5kZXJzdGFuZCBGR0wgdGFncyBhbmQgZG8gdGhlIHJpZ2h0IHRoaW5n
IHdpdGggdGhlbSwgYW5kIDIpIG9zdHJhc2l6ZSBWTCBndXlzLi4ubWVhbmluZyB0aGF0IGFuIEZH
TCBndXkgcmVmdXNlcyB0byBmb3JtIGFuIGFkamFjZW5jeSB3aXRoIGEgVkwgZ3V5LiAgVGhpcyBv
cHRpb24gaXMgdmVyeSBzaW1wbGUgZm9yIGltcGxlbWVudGVycyBvZiBGR0wsIGJ1dCB0aGUgaW1w
bGljYXRpb24gaXMgdGhhdCB5b3UgaGF2ZSB0byB1cGdyYWRlIHlvdXIgZW50aXJlIGNhbXB1cyB0
byBGR0wgYXQgb25jZS4gIFRoZXJlIGlzIG5vIGNvZXhpc3RlbmNlLg0KDQpPcHRpb24gMjogVHdv
IHR5cGVzIG9mIFJCcmlkZ2VzLCBidXQgdGhlIDJuZCB0eXBlIGlzIGRpZmZlcmVudCBmcm9tIG9w
dGlvbiAxLiAgVkwgZ3V5cyBvZiBjb3Vyc2UgYXJlIHRoZSBzYW1lIGFzIGluIG9wdGlvbiAxLi4u
dGhleSBjYW5ub3QgYmUgdHJ1c3RlZCB3aXRoIEZHTCBmcmFtZXMuIFRoZSBzZWNvbmQgdHlwZSBJ
IHdpbGwgY2FsbCAiRkdMLXNhZmUiLiBBbiBGR0wtc2FmZSBSQnJpZGdlIG11c3QgYWR2ZXJ0aXNl
IGluIGl0cyBMU1AgdGhhdCBpdCBpcyBGR0wtc2FmZSwgYW5kIGl0IG11c3Qgbm90ICJkbyBhbnl0
aGluZyBiYWQiIHdpdGggRkdMIGZyYW1lcywgbWVhbmluZyB0aGF0IGl0IGlzIGFsbG93ZWQgdG8g
aWdub3JlIHBydW5pbmcgb2YgRkdMIChvciBldmVuIFZMIGZyYW1lcykgZW50aXJlbHkuLi5pdCBq
dXN0IGNhbid0IGZhbHNlbHkgZHJvcCBGR0wgZnJhbWVzLiAgQW5kIGl0IG11c3Qgbm90IGRlY2Fw
c3VsYXRlIGFuIEZHTCBmcmFtZSBvbnRvIGEgbGluayBmb3Igd2hpY2ggdGhhdCBGR0wgZnJhbWUg
ZG9lc24ndCBiZWxvbmcuIEluIHRoaXMgb3B0aW9uLCBhbGwgUkJyaWRnZXMgbXVzdCBiZSB1cGdy
YWRlZCB0byBGR0wtc2FmZSwgYnV0IGl0IG5lZWQgbm90IGhhcHBlbiBpbnN0YW50YW5lb3VzbHku
Li5pdCdzIGZpbmUgdG8gbWl4IEZHTC1zYWZlIFJCcmlkZ2VzIHdpdGggVkwgUkJyaWRnZXMuLi5p
dCdzIGp1c3Qgbm90IHNhZmUgdG8gaW5qZWN0IEZHTCBmcmFtZXMgeWV0LiAgT25jZSBhbGwgdGhl
IFJCcmlkZ2VzIGhhdmUgYmVlbiB1cGdyYWRlZCB0byBiZSBGR0wtc2FmZSwgdGhlbiBlZGdlIFJC
cmlkZ2VzIGNhbiBzdGFydCBhbm5vdW5jaW5nIHRoZXkgYXJlIGNvbm5lY3RlZCB0byBhbiBGR0wg
bGluaywgYW5kIGNhbiBzdGFydCBpbmplY3RpbmcgRkdMIHBhY2tldHMuICBJdCBpcyBjb25zaWRl
cmVkIGEgbWlzY29uZmlndXJhdGlvbiBpZiB5b3Ugc3RhcnQgaW5qZWN0aW5nIEZHTCBmcmFtZXMg
YmVmb3JlIGFsbCB0aGUgUkJyaWRnZXMgYXJlIHVwZ3JhZGVkIHRvIEZHTC1zYWZlLCBzbyBhbiBh
ZGRpdGlvbmFsIGNob3JlIGZvciBhbiBGR0wtc2FmZSBSQnJpZGdlIFIxIGlzIHRvIGV4YW1pbmUg
TFNQcywgYW5kIGlmIGFueSBSQnJpZGdlIGNsYWltcyB0byBiZSBhdHRhY2hlZCB0byBhbiBGR0wg
bGluaywgdGhlbiBSMSBtdXN0IG9zdHJhc2l6ZSBhbnkgVkwgbmVpZ2hib3JzLiAgKGRvbid0IHN0
YXJ0IG9zdHJhc2l6aW5nIFZMIGd1eXMgdW50aWwgaXQgaXMgbmVjZXNzYXJ5IGJlY2F1c2Ugb2Yg
YWN0dWFsbHkgc3RhcnRpbmcgdG8gdXNlIEZHTCBmcmFtZXMsIGluIG90aGVyIHdvcmRzKS4gVGhp
cyBvcHRpb24gaXMgbW9yZSB3b3JrIGZvciB0aGUgdXBncmFkZWQgUkJyaWRnZXMgdGhhbiBvcHRp
b24gMSwgYW5kIHRoZXJlIHN0aWxsIGlzbid0IGdvb2QgY29leGlzdGVuY2Ugd2l0aCBWTCBndXlz
IGxvbmctdGVybSAoYXMgb3B0aW9uIDMgd2lsbCksIGJ1dCBpdCBkb2VzIGFsbG93IHVwZ3JhZGlu
ZyBSQnJpZGdlcyBvbmUgYnkgb25lIGluIGEgd29ya2luZyBjYW1wdXMgd2l0aG91dCBjYXVzaW5n
IGRpc3J1cHRpb24uDQoNCk9wdGlvbiAzOiBUd28gdHlwZXMgb2YgUkJyaWRnZXMuICBWTCwgb2Yg
Y291cnNlLCBpcyB0aGUgc2FtZSBhcyBpbiBvcHRpb25zIDEgYW5kIDIuICBUaGlzIG9wdGlvbiBt
YWtlcyBGR0wgZ3V5cyBkbyBtb3JlIHdvcmssIGJ1dCBhbGxvd3MgbWF4aW1hbCBsb25nIHRlcm0g
Y29leGlzdGVuY2Ugb2YgVkwgYW5kIEZHTCBndXlzLiBJbiB0aGlzIG9wdGlvbiwgRkdMIGd1eXMg
Y2FsY3VsYXRlIHVuaWNhc3QgcGF0aHMgdG8gRkdMIGVkZ2UgZ3V5cyB0aGF0IGF2b2lkIGFueSBW
TCBndXlzLCBhbmQgY2FsY3VsYXRlIGF0IGxlYXN0IG9uZSBGR0wtZnJpZW5kbHkgbXVsdGljYXN0
IHRyZWUgdGhhdCBhbHNvIGF2b2lkcyBhbnkgVkwgZ3V5cy4gU28gbGV0J3Mgc2F5IFIxIGlzIGFu
IEZHTCBndXkuICBSMSBkaXNjYXJkcyBhbGwgTFNQcyBmcm9tIFZMIGd1eXMgKG9uZXMgdGhhdCBk
b24ndCBhZHZlcnRpc2UgRkdMIGNhcGFiaWxpdHkgaW4gdGhlaXIgTFNQKSwgd2hlbiBjYWxjdWxh
dGluZyBwYXRocyB0byBvdGhlciBGR0wgUkJyaWRnZXMuIFRoZW4gUjEgY2FsY3VsYXRlcyBwYXRo
cyB0byB0aGUgVkwgZ3V5cyB1c2luZyBhbGwgTFNQcy4gIExpa2V3aXNlLCB3aGVuIGNhbGN1bGF0
aW5nIGFuIEZHTC1mcmllbmRseSB0cmVlLCBSMSBjYWxjdWxhdGVzIGEgdHJlZSB0aHJvdWdoIG9u
bHkgRkdMIGd1eXMuICAgVGhpcyBvcHRpb24gaXMgbW9yZSB3b3JrIGZvciB0aGUgdXBncmFkZWQg
UkJyaWRnZXMgKGJlY2F1c2UgdGhleSBoYXZlIHRvIGNhbGN1bGF0ZSBEaWprc3RyYSBpbiB0d28g
ZGlmZmVyZW50IHdheXMsIG9uZSBmb3IgcmVhY2hpbmcgVkwgZ3V5cywgYW5kIG9uZSBmb3IgcmVh
Y2hpbmcgRkdMIGd1eXMpLiAgSG93ZXZlciwgaXQgZG9lcyBhbGxvdyBoYXZpbmcgbG9uZy10ZXJt
IGNvZXhpc3RlbmNlIHdpdGggVkwgZ3V5cy4gIEZvciBpbnN0YW5jZSwgeW91IGNvdWxkIGZvcmV2
ZXIga2VlcCBzb21lIFZMIGVkZ2UgUkJyaWRnZXMgdGhhdCBjb21tdW5pY2F0ZSBqdXN0IGZpbmUg
dGhyb3VnaCBhbiBGR0wgY29yZS4gIFRoZXkgY2FuIHN0YXkgdGhlcmUgZm9yZXZlciwgYW5kIHN0
aWxsIGJlIGFibGUgdG8gY29tbXVuaWNhdGUgdGhyb3VnaCB0aGUgY29yZSB0byBhbGwgdGhlIGxp
bmtzIGF0dGFjaGluZyB0byB0aGVpciBWTEFOLg0K

--_000_DD5FC8DE455C3348B94340C0AB5517334F651415nkgeml501mbschi_
Content-Type: text/html; charset="gb2312"
Content-Transfer-Encoding: quoted-printable

<html dir=3D"ltr">
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Dgb2312">
<style id=3D"owaParaStyle">P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
</style>
</head>
<body fPStyle=3D"1" ocsi=3D"0">
<div style=3D"direction: ltr;font-family: Tahoma;color: #000000;font-size: =
10pt;">
<p>Hi Radia,</p>
<p>In option3:</p>
<p>why FGL switches calculate unicast paths to FGL edge guys that avoid any=
 VL guys, and calculate at least one FGL-friendly multicast tree that also =
avoids any VL guys? If two FGL switches can only transmit data through VL s=
witch, the method will disconnect
 data channel between FGL switches. I think VL switch is safe as transit no=
de to tranmit unicast FGL data packet.</p>
<p>thanks</p>
<p>weiguo</p>
<p></p>
<hr tabindex=3D"-1">
<p></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<div style=3D"DIRECTION: ltr" id=3D"divRpF81253"><font color=3D"#000000" si=
ze=3D"2" face=3D"Tahoma"><b>=B7=A2=BC=FE=C8=CB:</b> Radia Perlman [radiaper=
lman@gmail.com]<br>
<b>=B7=A2=CB=CD=CA=B1=BC=E4:</b> 2013=C4=EA1=D4=C226=C8=D5 3:00<br>
<b>=B5=BD:</b> trill@ietf.org<br>
<b>=D6=F7=CC=E2:</b> [trill] Explaining three options for upgrading to FGL =
(fine-grained-labeling)<br>
</font><br>
</div>
<div></div>
<div>
<div style=3D"BACKGROUND-COLOR: rgb(255,255,255); FONT-FAMILY: arial,sans-s=
erif; COLOR: rgb(34,34,34); FONT-SIZE: 13px">
I'm going to summarize three options for phasing-in FGL, and explain the tr=
adeoffs, to help people have informed opinions about the implications of th=
e three approaches. &nbsp;There are infinite variations, of course, but as =
I said, I'll describe three main ones.</div>
<div style=3D"BACKGROUND-COLOR: rgb(255,255,255); FONT-FAMILY: arial,sans-s=
erif; COLOR: rgb(34,34,34); FONT-SIZE: 13px">
<br>
</div>
<div style=3D"BACKGROUND-COLOR: rgb(255,255,255); FONT-FAMILY: arial,sans-s=
erif; COLOR: rgb(34,34,34); FONT-SIZE: 13px">
Option 1: &nbsp;Original draft, no changes. &nbsp;There are two types of RB=
ridges. The first type is &quot;VL&quot; which only understands VLAN tags, =
and have unknown and possibly dangerous behavior when they receive an FGL-l=
abeled packet. (e.g., decapsulate a packet onto a link
 where it is not allowed, or mistakenly prune a multicast so it does not re=
ach everywhere it should). The second type is &quot;FGL&quot;. &nbsp;FGL gu=
ys need to do two things: 1) understand FGL tags and do the right thing wit=
h them, and 2) ostrasize VL guys...meaning that
 an FGL guy refuses to form an adjacency with a VL guy. &nbsp;This option i=
s very simple for implementers of FGL, but the implication is that you have=
 to upgrade your entire campus to FGL at once. &nbsp;There is no coexistenc=
e.</div>
<div style=3D"BACKGROUND-COLOR: rgb(255,255,255); FONT-FAMILY: arial,sans-s=
erif; COLOR: rgb(34,34,34); FONT-SIZE: 13px">
<br>
</div>
<div style=3D"BACKGROUND-COLOR: rgb(255,255,255); FONT-FAMILY: arial,sans-s=
erif; COLOR: rgb(34,34,34); FONT-SIZE: 13px">
Option 2: Two types of RBridges, but the 2nd type is different from option =
1. &nbsp;VL guys of course are the same as in option 1...they cannot be tru=
sted with FGL frames. The second type I will call &quot;FGL-safe&quot;. An =
FGL-safe RBridge must advertise in its LSP that
 it is FGL-safe, and it must not &quot;do anything bad&quot; with FGL frame=
s, meaning that it is allowed to ignore pruning of FGL (or even VL frames) =
entirely...it just can't falsely drop FGL frames. &nbsp;And it must not dec=
apsulate an FGL frame onto a link for which that
 FGL frame doesn't belong. In this option, all RBridges must be upgraded to=
 FGL-safe, but it need not happen instantaneously...it's fine to mix FGL-sa=
fe RBridges with VL RBridges...it's just not safe to inject FGL frames yet.=
 &nbsp;Once all the RBridges have been
 upgraded to be FGL-safe, then edge RBridges can start announcing they are =
connected to an FGL link, and can start injecting FGL packets. &nbsp;It is =
considered a misconfiguration if you start injecting FGL frames before all =
the RBridges are upgraded to FGL-safe,
 so an additional chore for an FGL-safe RBridge R1 is to examine LSPs, and =
if any RBridge claims to be attached to an FGL link, then R1 must ostrasize=
 any VL neighbors. &nbsp;(don't start ostrasizing VL guys until it is neces=
sary because of actually starting to
 use FGL frames, in other words). This option is more work for the upgraded=
 RBridges than option 1, and there still isn't good coexistence with VL guy=
s long-term (as option 3 will), but it does allow upgrading RBridges one by=
 one in a working campus without
 causing disruption.</div>
<div style=3D"BACKGROUND-COLOR: rgb(255,255,255); FONT-FAMILY: arial,sans-s=
erif; COLOR: rgb(34,34,34); FONT-SIZE: 13px">
<br>
</div>
<div style=3D"BACKGROUND-COLOR: rgb(255,255,255); FONT-FAMILY: arial,sans-s=
erif; COLOR: rgb(34,34,34); FONT-SIZE: 13px">
Option 3: Two types of RBridges. &nbsp;VL, of course, is the same as in opt=
ions 1 and 2. &nbsp;This option makes FGL guys do more work, but allows max=
imal long term coexistence of VL and FGL guys. In this option, FGL guys cal=
culate unicast paths to FGL edge guys that
 avoid any VL guys, and calculate at least one FGL-friendly multicast tree =
that also avoids any VL guys. So let's say R1 is an FGL guy. &nbsp;R1 disca=
rds all LSPs from VL guys (ones that don't advertise FGL capability in thei=
r LSP), when calculating paths to other
 FGL RBridges. Then R1 calculates paths to the VL guys using all LSPs. &nbs=
p;Likewise, when calculating an FGL-friendly tree, R1 calculates a tree thr=
ough only FGL guys. &nbsp; This option is more work for the upgraded RBridg=
es (because they have to calculate Dijkstra
 in two different ways, one for reaching VL guys, and one for reaching FGL =
guys). &nbsp;However, it does allow having long-term coexistence with VL gu=
ys. &nbsp;For instance, you could forever keep some VL edge RBridges that c=
ommunicate just fine through an FGL core.
 &nbsp;They can stay there forever, and still be able to communicate throug=
h the core to all the links attaching to their VLAN.</div>
</div>
</div>
</div>
</body>
</html>

--_000_DD5FC8DE455C3348B94340C0AB5517334F651415nkgeml501mbschi_--

From radiaperlman@gmail.com  Tue Jan 29 02:59:12 2013
Return-Path: <radiaperlman@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D998021F86FD for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 02:59:12 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 0.05
X-Spam-Level: 
X-Spam-Status: No, score=0.05 tagged_above=-999 required=5 tests=[AWL=-3.647,  BAYES_00=-2.599, CHARSET_FARAWAY_HEADER=3.2, HTML_MESSAGE=0.001, MIME_8BIT_HEADER=0.3, MIME_CHARSET_FARAWAY=2.45, RCVD_IN_DNSWL_LOW=-1,  SARE_SUB_ENC_GB2312=1.345]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id d+gUGXNUBKLh for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 02:59:12 -0800 (PST)
Received: from mail-lb0-f178.google.com (mail-lb0-f178.google.com [209.85.217.178]) by ietfa.amsl.com (Postfix) with ESMTP id 7B08621F86F8 for <trill@ietf.org>; Tue, 29 Jan 2013 02:59:11 -0800 (PST)
Received: by mail-lb0-f178.google.com with SMTP id n1so533051lba.23 for <trill@ietf.org>; Tue, 29 Jan 2013 02:59:10 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:x-received:in-reply-to:references:date:message-id :subject:from:to:cc:content-type; bh=TS9eR/N6o43wrpkSg9DPJvhR34dH4uUhMRnIrqHUFpE=; b=Rjq/ygUftv+0m/aLwkRe80vd288YmhHKRFFb0+ws7n1q6PvJrsmFjmdEKhK+cjaJP7 rrAm0bmuhtU6UtN4Du5BlB9B17NLA3dXPSEuquRTTgnWWLkkY4xOxF+bbHl8yYs83j61 HkO0An5RozcpZEBAsxFBo7pPM9AcjnxWvNgUTG/A/knglSznwKaYkzkxIjOJlMwZ+8mh O9fa86q18gpXYGAm0XayPNWBM7U+Lh1H79ML2eVCSWGPEuIfnzw+Tfsaig2eatrKSdbu nUwAUK84rih2IbrH9ojISC/E6+E7qXov0ftPA/WpFpVihjw7TmeseV8BO5jE0nrZvfG1 m4Mw==
MIME-Version: 1.0
X-Received: by 10.112.50.109 with SMTP id b13mr322683lbo.8.1359457150285; Tue, 29 Jan 2013 02:59:10 -0800 (PST)
Received: by 10.112.64.17 with HTTP; Tue, 29 Jan 2013 02:59:10 -0800 (PST)
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F651415@nkgeml501-mbs.china.huawei.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <DD5FC8DE455C3348B94340C0AB5517334F651415@nkgeml501-mbs.china.huawei.com>
Date: Tue, 29 Jan 2013 02:59:10 -0800
Message-ID: <CAFOuuo597npOuDLT1di-0V+Q1Fh7nYvKD3jHG3bbZPB81hc=Tw@mail.gmail.com>
From: Radia Perlman <radiaperlman@gmail.com>
To: Haoweiguo <haoweiguo@huawei.com>
Content-Type: multipart/alternative; boundary=f46d0401fbc987b7aa04d46b4826
Cc: "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] =?gb2312?b?tPC4tDogIEV4cGxhaW5pbmcgdGhyZWUgb3B0aW9ucyBm?= =?gb2312?b?b3IgdXBncmFkaW5nIHRvIEZHTCAoZmluZS1ncmFpbmVkLWxhYmVs?= =?gb2312?b?aW5nKQ==?=
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 29 Jan 2013 10:59:13 -0000

--f46d0401fbc987b7aa04d46b4826
Content-Type: text/plain; charset=GB2312
Content-Transfer-Encoding: quoted-printable

Hi Weiguo,

 If we knew that it was already safe to transit FGL frames through VL
nodes, that would be great.  We could basically just start sending FGL
frames, not worry about ostrasizing VL RBridges, not worry about
calculating two forwarding databases.

 But,  I was assuming that there were implementations of VL that might do
bad things if they saw an FGL packet.  For instance, for unicast, they
might drop the packet, or they might decapsulate it onto a link that should
not receive packets for that FGL.  For multicast, they might falsely filter
the packet.  If that assumption is true, then the safe thing is to avoid VL
RBridges for FGL packets, which means calculating paths that do not include
VL RBridges.
.
Thanks,

Radia



On Mon, Jan 28, 2013 at 10:59 PM, Haoweiguo <haoweiguo@huawei.com> wrote:

>  Hi Radia,
>
> In option3:
>
> why FGL switches calculate unicast paths to FGL edge guys that avoid any
> VL guys, and calculate at least one FGL-friendly multicast tree that also
> avoids any VL guys? If two FGL switches can only transmit data through VL
> switch, the method will disconnect data channel between FGL switches. I
> think VL switch is safe as transit node to tranmit unicast FGL data packe=
t.
>
> thanks
>
> weiguo
>
> ------------------------------
>
>  *=B7=A2=BC=FE=C8=CB:* Radia Perlman [radiaperlman@gmail.com]
> *=B7=A2=CB=CD=CA=B1=BC=E4:* 2013=C4=EA1=D4=C226=C8=D5 3:00
> *=B5=BD:* trill@ietf.org
> *=D6=F7=CC=E2:* [trill] Explaining three options for upgrading to FGL
> (fine-grained-labeling)
>
>   I'm going to summarize three options for phasing-in FGL, and explain
> the tradeoffs, to help people have informed opinions about the implicatio=
ns
> of the three approaches.  There are infinite variations, of course, but a=
s
> I said, I'll describe three main ones.
>
>  Option 1:  Original draft, no changes.  There are two types of RBridges.
> The first type is "VL" which only understands VLAN tags, and have unknown
> and possibly dangerous behavior when they receive an FGL-labeled packet.
> (e.g., decapsulate a packet onto a link where it is not allowed, or
> mistakenly prune a multicast so it does not reach everywhere it should).
> The second type is "FGL".  FGL guys need to do two things: 1) understand
> FGL tags and do the right thing with them, and 2) ostrasize VL
> guys...meaning that an FGL guy refuses to form an adjacency with a VL guy=
.
>  This option is very simple for implementers of FGL, but the implication =
is
> that you have to upgrade your entire campus to FGL at once.  There is no
> coexistence.
>
>  Option 2: Two types of RBridges, but the 2nd type is different from
> option 1.  VL guys of course are the same as in option 1...they cannot be
> trusted with FGL frames. The second type I will call "FGL-safe". An
> FGL-safe RBridge must advertise in its LSP that it is FGL-safe, and it mu=
st
> not "do anything bad" with FGL frames, meaning that it is allowed to igno=
re
> pruning of FGL (or even VL frames) entirely...it just can't falsely drop
> FGL frames.  And it must not decapsulate an FGL frame onto a link for whi=
ch
> that FGL frame doesn't belong. In this option, all RBridges must be
> upgraded to FGL-safe, but it need not happen instantaneously...it's fine =
to
> mix FGL-safe RBridges with VL RBridges...it's just not safe to inject FGL
> frames yet.  Once all the RBridges have been upgraded to be FGL-safe, the=
n
> edge RBridges can start announcing they are connected to an FGL link, and
> can start injecting FGL packets.  It is considered a misconfiguration if
> you start injecting FGL frames before all the RBridges are upgraded to
> FGL-safe, so an additional chore for an FGL-safe RBridge R1 is to examine
> LSPs, and if any RBridge claims to be attached to an FGL link, then R1 mu=
st
> ostrasize any VL neighbors.  (don't start ostrasizing VL guys until it is
> necessary because of actually starting to use FGL frames, in other words)=
.
> This option is more work for the upgraded RBridges than option 1, and the=
re
> still isn't good coexistence with VL guys long-term (as option 3 will), b=
ut
> it does allow upgrading RBridges one by one in a working campus without
> causing disruption.
>
>  Option 3: Two types of RBridges.  VL, of course, is the same as in
> options 1 and 2.  This option makes FGL guys do more work, but allows
> maximal long term coexistence of VL and FGL guys. In this option, FGL guy=
s
> calculate unicast paths to FGL edge guys that avoid any VL guys, and
> calculate at least one FGL-friendly multicast tree that also avoids any V=
L
> guys. So let's say R1 is an FGL guy.  R1 discards all LSPs from VL guys
> (ones that don't advertise FGL capability in their LSP), when calculating
> paths to other FGL RBridges. Then R1 calculates paths to the VL guys usin=
g
> all LSPs.  Likewise, when calculating an FGL-friendly tree, R1 calculates=
 a
> tree through only FGL guys.   This option is more work for the upgraded
> RBridges (because they have to calculate Dijkstra in two different ways,
> one for reaching VL guys, and one for reaching FGL guys).  However, it do=
es
> allow having long-term coexistence with VL guys.  For instance, you could
> forever keep some VL edge RBridges that communicate just fine through an
> FGL core.  They can stay there forever, and still be able to communicate
> through the core to all the links attaching to their VLAN.
>

--f46d0401fbc987b7aa04d46b4826
Content-Type: text/html; charset=GB2312
Content-Transfer-Encoding: quoted-printable

Hi W<span style=3D"font-family:Tahoma;font-size:13px">eiguo,</span><div><br=
></div><div>&nbsp;If we knew that it was already safe to transit FGL frames=
 through VL nodes, that would be great. &nbsp;We could basically just start=
 sending FGL frames, not worry about ostrasizing VL RBridges, not worry abo=
ut calculating two forwarding databases.</div>
<div><br></div><div>&nbsp;But, &nbsp;I was assuming that there were impleme=
ntations of VL that might do bad things if they saw an FGL packet. &nbsp;Fo=
r instance, for unicast, they might drop the packet, or they might decapsul=
ate it onto a link that should not receive packets for that FGL. &nbsp;For =
multicast, they might falsely filter the packet. &nbsp;If that assumption i=
s true, then the safe thing is to avoid VL RBridges for FGL packets, which =
means calculating paths that do not include VL RBridges.<div>
.</div><div>Thanks,</div><div><br></div><div>Radia</div><div><br></div><div=
><br><br><div class=3D"gmail_quote">On Mon, Jan 28, 2013 at 10:59 PM, Haowe=
iguo <span dir=3D"ltr">&lt;<a href=3D"mailto:haoweiguo@huawei.com" target=
=3D"_blank">haoweiguo@huawei.com</a>&gt;</span> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">




<div>
<div style=3D"direction:ltr;font-size:10pt;font-family:Tahoma">
<p>Hi Radia,</p>
<p>In option3:</p>
<p>why FGL switches calculate unicast paths to FGL edge guys that avoid any=
 VL guys, and calculate at least one FGL-friendly multicast tree that also =
avoids any VL guys? If two FGL switches can only transmit data through VL s=
witch, the method will disconnect
 data channel between FGL switches. I think VL switch is safe as transit no=
de to tranmit unicast FGL data packet.</p>
<p>thanks</p>
<p>weiguo</p>
<p></p>
<hr>
<p></p>
<div style=3D"font-size:16px;font-family:Times New Roman">
<div style=3D"DIRECTION:ltr"><font color=3D"#000000" face=3D"Tahoma"><b>=B7=
=A2=BC=FE=C8=CB:</b> Radia Perlman [<a href=3D"mailto:radiaperlman@gmail.co=
m" target=3D"_blank">radiaperlman@gmail.com</a>]<br>
<b>=B7=A2=CB=CD=CA=B1=BC=E4:</b> 2013=C4=EA1=D4=C226=C8=D5 3:00<br>
<b>=B5=BD:</b> <a href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ie=
tf.org</a><br>
<b>=D6=F7=CC=E2:</b> [trill] Explaining three options for upgrading to FGL =
(fine-grained-labeling)<br>
</font><br>
</div><div><div class=3D"h5">
<div></div>
<div>
<div style=3D"color:rgb(34,34,34);font-size:13px;font-family:arial,sans-ser=
if">
I&#39;m going to summarize three options for phasing-in FGL, and explain th=
e tradeoffs, to help people have informed opinions about the implications o=
f the three approaches. &nbsp;There are infinite variations, of course, but=
 as I said, I&#39;ll describe three main ones.</div>

<div style=3D"color:rgb(34,34,34);font-size:13px;font-family:arial,sans-ser=
if">
<br>
</div>
<div style=3D"color:rgb(34,34,34);font-size:13px;font-family:arial,sans-ser=
if">
Option 1: &nbsp;Original draft, no changes. &nbsp;There are two types of RB=
ridges. The first type is &quot;VL&quot; which only understands VLAN tags, =
and have unknown and possibly dangerous behavior when they receive an FGL-l=
abeled packet. (e.g., decapsulate a packet onto a link
 where it is not allowed, or mistakenly prune a multicast so it does not re=
ach everywhere it should). The second type is &quot;FGL&quot;. &nbsp;FGL gu=
ys need to do two things: 1) understand FGL tags and do the right thing wit=
h them, and 2) ostrasize VL guys...meaning that
 an FGL guy refuses to form an adjacency with a VL guy. &nbsp;This option i=
s very simple for implementers of FGL, but the implication is that you have=
 to upgrade your entire campus to FGL at once. &nbsp;There is no coexistenc=
e.</div>

<div style=3D"color:rgb(34,34,34);font-size:13px;font-family:arial,sans-ser=
if">
<br>
</div>
<div style=3D"color:rgb(34,34,34);font-size:13px;font-family:arial,sans-ser=
if">
Option 2: Two types of RBridges, but the 2nd type is different from option =
1. &nbsp;VL guys of course are the same as in option 1...they cannot be tru=
sted with FGL frames. The second type I will call &quot;FGL-safe&quot;. An =
FGL-safe RBridge must advertise in its LSP that
 it is FGL-safe, and it must not &quot;do anything bad&quot; with FGL frame=
s, meaning that it is allowed to ignore pruning of FGL (or even VL frames) =
entirely...it just can&#39;t falsely drop FGL frames. &nbsp;And it must not=
 decapsulate an FGL frame onto a link for which that
 FGL frame doesn&#39;t belong. In this option, all RBridges must be upgrade=
d to FGL-safe, but it need not happen instantaneously...it&#39;s fine to mi=
x FGL-safe RBridges with VL RBridges...it&#39;s just not safe to inject FGL=
 frames yet. &nbsp;Once all the RBridges have been
 upgraded to be FGL-safe, then edge RBridges can start announcing they are =
connected to an FGL link, and can start injecting FGL packets. &nbsp;It is =
considered a misconfiguration if you start injecting FGL frames before all =
the RBridges are upgraded to FGL-safe,
 so an additional chore for an FGL-safe RBridge R1 is to examine LSPs, and =
if any RBridge claims to be attached to an FGL link, then R1 must ostrasize=
 any VL neighbors. &nbsp;(don&#39;t start ostrasizing VL guys until it is n=
ecessary because of actually starting to
 use FGL frames, in other words). This option is more work for the upgraded=
 RBridges than option 1, and there still isn&#39;t good coexistence with VL=
 guys long-term (as option 3 will), but it does allow upgrading RBridges on=
e by one in a working campus without
 causing disruption.</div>
<div style=3D"color:rgb(34,34,34);font-size:13px;font-family:arial,sans-ser=
if">
<br>
</div>
<div style=3D"color:rgb(34,34,34);font-size:13px;font-family:arial,sans-ser=
if">
Option 3: Two types of RBridges. &nbsp;VL, of course, is the same as in opt=
ions 1 and 2. &nbsp;This option makes FGL guys do more work, but allows max=
imal long term coexistence of VL and FGL guys. In this option, FGL guys cal=
culate unicast paths to FGL edge guys that
 avoid any VL guys, and calculate at least one FGL-friendly multicast tree =
that also avoids any VL guys. So let&#39;s say R1 is an FGL guy. &nbsp;R1 d=
iscards all LSPs from VL guys (ones that don&#39;t advertise FGL capability=
 in their LSP), when calculating paths to other
 FGL RBridges. Then R1 calculates paths to the VL guys using all LSPs. &nbs=
p;Likewise, when calculating an FGL-friendly tree, R1 calculates a tree thr=
ough only FGL guys. &nbsp; This option is more work for the upgraded RBridg=
es (because they have to calculate Dijkstra
 in two different ways, one for reaching VL guys, and one for reaching FGL =
guys). &nbsp;However, it does allow having long-term coexistence with VL gu=
ys. &nbsp;For instance, you could forever keep some VL edge RBridges that c=
ommunicate just fine through an FGL core.
 &nbsp;They can stay there forever, and still be able to communicate throug=
h the core to all the links attaching to their VLAN.</div>
</div>
</div></div></div>
</div>
</div>

</blockquote></div><br></div></div>

--f46d0401fbc987b7aa04d46b4826--

From haoweiguo@huawei.com  Tue Jan 29 05:23:22 2013
Return-Path: <haoweiguo@huawei.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D428B21F8891 for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 05:23:22 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 2.75
X-Spam-Level: **
X-Spam-Status: No, score=2.75 tagged_above=-999 required=5 tests=[AWL=-0.300,  BAYES_00=-2.599, CHARSET_FARAWAY_HEADER=3.2, HTML_MESSAGE=0.001, J_CHICKENPOX_74=0.6, MIME_8BIT_HEADER=0.3, MIME_BASE64_TEXT=1.753, MIME_CHARSET_FARAWAY=2.45, RCVD_IN_DNSWL_MED=-4, SARE_SUB_ENC_GB2312=1.345]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id WPVVnIoOBgVB for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 05:23:21 -0800 (PST)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) by ietfa.amsl.com (Postfix) with ESMTP id 074B821F8415 for <trill@ietf.org>; Tue, 29 Jan 2013 05:23:20 -0800 (PST)
Received: from 172.18.7.190 (EHLO lhreml203-edg.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.5-GA FastPath queued) with ESMTP id ANY97794; Tue, 29 Jan 2013 13:23:20 +0000 (GMT)
Received: from LHREML403-HUB.china.huawei.com (10.201.5.217) by lhreml203-edg.huawei.com (172.18.7.221) with Microsoft SMTP Server (TLS) id 14.1.323.7; Tue, 29 Jan 2013 13:22:45 +0000
Received: from NKGEML401-HUB.china.huawei.com (10.98.56.32) by lhreml403-hub.china.huawei.com (10.201.5.217) with Microsoft SMTP Server (TLS) id 14.1.323.7; Tue, 29 Jan 2013 13:23:18 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.101]) by nkgeml401-hub.china.huawei.com ([10.98.56.32]) with mapi id 14.01.0323.007; Tue, 29 Jan 2013 21:23:13 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: Radia Perlman <radiaperlman@gmail.com>
Thread-Topic: =?gb2312?B?tPC4tDogW3RyaWxsXSBFeHBsYWluaW5nIHRocmVlIG9wdGlvbnMgZm9yIHVw?= =?gb2312?Q?grading_to_FGL_(fine-grained-labeling)?=
Thread-Index: AQHN/g+vw5yqGjG9+U2vlPyKn8v+s5hgROR6
Date: Tue, 29 Jan 2013 13:23:13 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F6514AD@nkgeml501-mbs.china.huawei.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <DD5FC8DE455C3348B94340C0AB5517334F651415@nkgeml501-mbs.china.huawei.com>, <CAFOuuo597npOuDLT1di-0V+Q1Fh7nYvKD3jHG3bbZPB81hc=Tw@mail.gmail.com>
In-Reply-To: <CAFOuuo597npOuDLT1di-0V+Q1Fh7nYvKD3jHG3bbZPB81hc=Tw@mail.gmail.com>
Accept-Language: zh-CN, en-US
Content-Language: zh-CN
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.108.133]
Content-Type: multipart/alternative; boundary="_000_DD5FC8DE455C3348B94340C0AB5517334F6514ADnkgeml501mbschi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Cc: "trill@ietf.org" <trill@ietf.org>
Subject: [trill] =?gb2312?b?tPC4tDogtPC4tDogIEV4cGxhaW5pbmcgdGhyZWUgb3B0?= =?gb2312?b?aW9ucyBmb3IgdXBncmFkaW5nIHRvIEZHTCAoZmluZS1ncmFpbmVkLWxhYmVs?= =?gb2312?b?aW5nKQ==?=
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 29 Jan 2013 13:23:23 -0000

--_000_DD5FC8DE455C3348B94340C0AB5517334F6514ADnkgeml501mbschi_
Content-Type: text/plain; charset="gb2312"
Content-Transfer-Encoding: base64
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--_000_DD5FC8DE455C3348B94340C0AB5517334F6514ADnkgeml501mbschi_
Content-Type: text/html; charset="gb2312"
Content-Transfer-Encoding: quoted-printable

<html dir=3D"ltr">
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Dgb2312">
<style id=3D"owaParaStyle">P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
</style>
</head>
<body fPStyle=3D"1" ocsi=3D"0">
<div style=3D"direction: ltr;font-family: Tahoma;color: #000000;font-size: =
10pt;">
<p>Hi Radia,</p>
<p>For unicast, I think TRILL switch don't care about inner VLAN or FGL and=
 they forward the unicast data frame just by egress nickname. So I think VL=
 switch can support unicast FGL data frame forwarding and have no problem.<=
/p>
<p>For multicast, VL switch can't perform distribution tree pruning by FGL =
and VL switch has following possible behaviors:</p>
<p>1,VL switch drops FGL multicast data frame because VL switch cares about=
 inner ethtype.The VL switch only support inner ethtype of 0x8100.</p>
<p>2,VL switch don't implement VLAN pruning and forward multicast data fram=
e by broadcast distribution tree.The forwarding key of the broadcast distri=
bution tree is nickname of DST root.The VL switch don't care about inner VL=
AN or FGL also.Multicast FGL data
 frame can be forwarded by this kind of VL switch.</p>
<p>3,VL switch implements VLAN pruning and forward multicast data frame by =
VLAN pruning tree. The forwarding key of the pruning tree is root nickname&=
#43;inner VLAN.&nbsp; If the switch cares about accurate inner ethtype(0x81=
00),FGL data frame will be dropped. If the
 switch allows random or multiple ethtype,then FGL data frame can be forwar=
ded higher 12 bit of FGL.</p>
<p>thanks</p>
<p>weiguo</p>
<p></p>
<hr tabindex=3D"-1">
<p></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<div style=3D"DIRECTION: ltr" id=3D"divRpF13387"><font color=3D"#000000" si=
ze=3D"2" face=3D"Tahoma"><b>=B7=A2=BC=FE=C8=CB:</b> Radia Perlman [radiaper=
lman@gmail.com]<br>
<b>=B7=A2=CB=CD=CA=B1=BC=E4:</b> 2013=C4=EA1=D4=C229=C8=D5 18:59<br>
<b>=B5=BD:</b> Haoweiguo<br>
<b>Cc:</b> trill@ietf.org<br>
<b>=D6=F7=CC=E2:</b> Re: =B4=F0=B8=B4: [trill] Explaining three options for=
 upgrading to FGL (fine-grained-labeling)<br>
</font><br>
</div>
<div></div>
<div>Hi W<span style=3D"FONT-FAMILY: Tahoma; FONT-SIZE: 13px">eiguo,</span>
<div><br>
</div>
<div>&nbsp;If we knew that it was already safe to transit FGL frames throug=
h VL nodes, that would be great. &nbsp;We could basically just start sendin=
g FGL frames, not worry about ostrasizing VL RBridges, not worry about calc=
ulating two forwarding databases.</div>
<div><br>
</div>
<div>&nbsp;But, &nbsp;I was assuming that there were implementations of VL =
that might do bad things if they saw an FGL packet. &nbsp;For instance, for=
 unicast, they might drop the packet, or they might decapsulate it onto a l=
ink that should not receive packets for that FGL.
 &nbsp;For multicast, they might falsely filter the packet. &nbsp;If that a=
ssumption is true, then the safe thing is to avoid VL RBridges for FGL pack=
ets, which means calculating paths that do not include VL RBridges.
<div>.</div>
<div>Thanks,</div>
<div><br>
</div>
<div>Radia</div>
<div><br>
</div>
<div><br>
<br>
<div class=3D"gmail_quote">On Mon, Jan 28, 2013 at 10:59 PM, Haoweiguo <spa=
n dir=3D"ltr">
&lt;<a href=3D"mailto:haoweiguo@huawei.com" target=3D"_blank">haoweiguo@hua=
wei.com</a>&gt;</span> wrote:<br>
<blockquote style=3D"BORDER-LEFT: #ccc 1px solid; MARGIN: 0px 0px 0px 0.8ex=
; PADDING-LEFT: 1ex" class=3D"gmail_quote">
<div>
<div style=3D"FONT-FAMILY: Tahoma; DIRECTION: ltr; FONT-SIZE: 10pt">
<p>Hi Radia,</p>
<p>In option3:</p>
<p>why FGL switches calculate unicast paths to FGL edge guys that avoid any=
 VL guys, and calculate at least one FGL-friendly multicast tree that also =
avoids any VL guys? If two FGL switches can only transmit data through VL s=
witch, the method will disconnect
 data channel between FGL switches. I think VL switch is safe as transit no=
de to tranmit unicast FGL data packet.</p>
<p>thanks</p>
<p>weiguo</p>
<p></p>
<hr>
<p></p>
<div style=3D"FONT-FAMILY: Times New Roman; FONT-SIZE: 16px">
<div style=3D"DIRECTION: ltr"><font color=3D"#000000" face=3D"Tahoma"><b>=
=B7=A2=BC=FE=C8=CB:</b> Radia Perlman [<a href=3D"mailto:radiaperlman@gmail=
.com" target=3D"_blank">radiaperlman@gmail.com</a>]<br>
<b>=B7=A2=CB=CD=CA=B1=BC=E4:</b> 2013=C4=EA1=D4=C226=C8=D5 3:00<br>
<b>=B5=BD:</b> <a href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ie=
tf.org</a><br>
<b>=D6=F7=CC=E2:</b> [trill] Explaining three options for upgrading to FGL =
(fine-grained-labeling)<br>
</font><br>
</div>
<div>
<div class=3D"h5">
<div></div>
<div>
<div style=3D"FONT-FAMILY: arial,sans-serif; COLOR: rgb(34,34,34); FONT-SIZ=
E: 13px">
I'm going to summarize three options for phasing-in FGL, and explain the tr=
adeoffs, to help people have informed opinions about the implications of th=
e three approaches. &nbsp;There are infinite variations, of course, but as =
I said, I'll describe three main ones.</div>
<div style=3D"FONT-FAMILY: arial,sans-serif; COLOR: rgb(34,34,34); FONT-SIZ=
E: 13px">
<br>
</div>
<div style=3D"FONT-FAMILY: arial,sans-serif; COLOR: rgb(34,34,34); FONT-SIZ=
E: 13px">
Option 1: &nbsp;Original draft, no changes. &nbsp;There are two types of RB=
ridges. The first type is &quot;VL&quot; which only understands VLAN tags, =
and have unknown and possibly dangerous behavior when they receive an FGL-l=
abeled packet. (e.g., decapsulate a packet onto a link
 where it is not allowed, or mistakenly prune a multicast so it does not re=
ach everywhere it should). The second type is &quot;FGL&quot;. &nbsp;FGL gu=
ys need to do two things: 1) understand FGL tags and do the right thing wit=
h them, and 2) ostrasize VL guys...meaning that
 an FGL guy refuses to form an adjacency with a VL guy. &nbsp;This option i=
s very simple for implementers of FGL, but the implication is that you have=
 to upgrade your entire campus to FGL at once. &nbsp;There is no coexistenc=
e.</div>
<div style=3D"FONT-FAMILY: arial,sans-serif; COLOR: rgb(34,34,34); FONT-SIZ=
E: 13px">
<br>
</div>
<div style=3D"FONT-FAMILY: arial,sans-serif; COLOR: rgb(34,34,34); FONT-SIZ=
E: 13px">
Option 2: Two types of RBridges, but the 2nd type is different from option =
1. &nbsp;VL guys of course are the same as in option 1...they cannot be tru=
sted with FGL frames. The second type I will call &quot;FGL-safe&quot;. An =
FGL-safe RBridge must advertise in its LSP that
 it is FGL-safe, and it must not &quot;do anything bad&quot; with FGL frame=
s, meaning that it is allowed to ignore pruning of FGL (or even VL frames) =
entirely...it just can't falsely drop FGL frames. &nbsp;And it must not dec=
apsulate an FGL frame onto a link for which that
 FGL frame doesn't belong. In this option, all RBridges must be upgraded to=
 FGL-safe, but it need not happen instantaneously...it's fine to mix FGL-sa=
fe RBridges with VL RBridges...it's just not safe to inject FGL frames yet.=
 &nbsp;Once all the RBridges have been
 upgraded to be FGL-safe, then edge RBridges can start announcing they are =
connected to an FGL link, and can start injecting FGL packets. &nbsp;It is =
considered a misconfiguration if you start injecting FGL frames before all =
the RBridges are upgraded to FGL-safe,
 so an additional chore for an FGL-safe RBridge R1 is to examine LSPs, and =
if any RBridge claims to be attached to an FGL link, then R1 must ostrasize=
 any VL neighbors. &nbsp;(don't start ostrasizing VL guys until it is neces=
sary because of actually starting to
 use FGL frames, in other words). This option is more work for the upgraded=
 RBridges than option 1, and there still isn't good coexistence with VL guy=
s long-term (as option 3 will), but it does allow upgrading RBridges one by=
 one in a working campus without
 causing disruption.</div>
<div style=3D"FONT-FAMILY: arial,sans-serif; COLOR: rgb(34,34,34); FONT-SIZ=
E: 13px">
<br>
</div>
<div style=3D"FONT-FAMILY: arial,sans-serif; COLOR: rgb(34,34,34); FONT-SIZ=
E: 13px">
Option 3: Two types of RBridges. &nbsp;VL, of course, is the same as in opt=
ions 1 and 2. &nbsp;This option makes FGL guys do more work, but allows max=
imal long term coexistence of VL and FGL guys. In this option, FGL guys cal=
culate unicast paths to FGL edge guys that
 avoid any VL guys, and calculate at least one FGL-friendly multicast tree =
that also avoids any VL guys. So let's say R1 is an FGL guy. &nbsp;R1 disca=
rds all LSPs from VL guys (ones that don't advertise FGL capability in thei=
r LSP), when calculating paths to other
 FGL RBridges. Then R1 calculates paths to the VL guys using all LSPs. &nbs=
p;Likewise, when calculating an FGL-friendly tree, R1 calculates a tree thr=
ough only FGL guys. &nbsp; This option is more work for the upgraded RBridg=
es (because they have to calculate Dijkstra
 in two different ways, one for reaching VL guys, and one for reaching FGL =
guys). &nbsp;However, it does allow having long-term coexistence with VL gu=
ys. &nbsp;For instance, you could forever keep some VL edge RBridges that c=
ommunicate just fine through an FGL core.
 &nbsp;They can stay there forever, and still be able to communicate throug=
h the core to all the links attaching to their VLAN.</div>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</div>
<br>
</div>
</div>
</div>
</div>
</div>
</body>
</html>

--_000_DD5FC8DE455C3348B94340C0AB5517334F6514ADnkgeml501mbschi_--

From ghanwani@gmail.com  Tue Jan 29 06:42:46 2013
Return-Path: <ghanwani@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B07DE21F86BC for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 06:42:46 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.978
X-Spam-Level: 
X-Spam-Status: No, score=-1.978 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, NO_RELAYS=-0.001]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id eclsaE+rps19 for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 06:42:46 -0800 (PST)
Received: from mail-ia0-x229.google.com (mail-ia0-x229.google.com [IPv6:2607:f8b0:4001:c02::229]) by ietfa.amsl.com (Postfix) with ESMTP id 0C21721F866F for <trill@ietf.org>; Tue, 29 Jan 2013 06:42:45 -0800 (PST)
Received: by mail-ia0-f169.google.com with SMTP id j5so661937iaf.28 for <trill@ietf.org>; Tue, 29 Jan 2013 06:42:45 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:x-received:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type; bh=702hqBayd/Ys2WK/HzeRqmGGfMGv7zaQ9sget92IBwU=; b=YOWxFZ8wxUzbXigzkVSxopEc8Gqk44eJRAF9EoCXq+CojQBn0c3aISBfVJan1rMefm 3iPu32av47nY5zaFN1X/vrto+DUjP7Wo5t+XHEpj0cfECOw7kHbGm9XmgEun8rzL5beP GLQ/f8UkK5hyOc7X3fM7quYYOskHpTAb688mAPsxTJHhBcQAFJWUgBZPmbtejTAOCaTv Q1/X/1l3fmO/IESp9Q5kWJH7refx+Yvey5DKRCQaAZ10YVylyAC6tmFWN/ag60/HjwHl REg0h/8VRQb4Y6gumjCuGk0XHZe8065imoFlMFGujReDX3KXO8FLLluJeHVOnAUIvMXE +zOA==
MIME-Version: 1.0
X-Received: by 10.50.193.234 with SMTP id hr10mr896505igc.21.1359470564407; Tue, 29 Jan 2013 06:42:44 -0800 (PST)
Sender: ghanwani@gmail.com
Received: by 10.50.182.201 with HTTP; Tue, 29 Jan 2013 06:42:44 -0800 (PST)
In-Reply-To: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com>
Date: Tue, 29 Jan 2013 06:42:44 -0800
X-Google-Sender-Auth: jEFRTrJ5cdud41XfphAP73OUdcE
Message-ID: <CA+-tSzx0E3J4i5Dvfbj4egruCeGi7z1se1Jn5RXtx3RedQLPdg@mail.gmail.com>
From: Anoop Ghanwani <anoop@alumni.duke.edu>
To: Radia Perlman <radiaperlman@gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Cc: trill@ietf.org
Subject: Re: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 29 Jan 2013 14:42:46 -0000

At noted, option 1 looks dangerous so that should be ruled out.

Option 2 seems reasonable.  It might be useful to clarify a couple of things:
- Can the FGL-safe campus have both VLAN & FGL frames going around?
- Do all RBridges need to agree on when the campus is FGL-safe?  I would
  hope not.  But the problem I'm thinking about is where RB1 thinks
the campus is
  FGL-safe and starts announcing attachment to FGL links, but RB2 hadn't yet
  concluded that process and it see RB3 which happens to be VL.

Option 3 seems to introduce more complexity to solve this problem than
is necessary.

Anoop

On Fri, Jan 25, 2013 at 11:00 AM, Radia Perlman <radiaperlman@gmail.com> wrote:
> I'm going to summarize three options for phasing-in FGL, and explain the
> tradeoffs, to help people have informed opinions about the implications of
> the three approaches.  There are infinite variations, of course, but as I
> said, I'll describe three main ones.
>
> Option 1:  Original draft, no changes.  There are two types of RBridges. The
> first type is "VL" which only understands VLAN tags, and have unknown and
> possibly dangerous behavior when they receive an FGL-labeled packet. (e.g.,
> decapsulate a packet onto a link where it is not allowed, or mistakenly
> prune a multicast so it does not reach everywhere it should). The second
> type is "FGL".  FGL guys need to do two things: 1) understand FGL tags and
> do the right thing with them, and 2) ostrasize VL guys...meaning that an FGL
> guy refuses to form an adjacency with a VL guy.  This option is very simple
> for implementers of FGL, but the implication is that you have to upgrade
> your entire campus to FGL at once.  There is no coexistence.
>
> Option 2: Two types of RBridges, but the 2nd type is different from option
> 1.  VL guys of course are the same as in option 1...they cannot be trusted
> with FGL frames. The second type I will call "FGL-safe". An FGL-safe RBridge
> must advertise in its LSP that it is FGL-safe, and it must not "do anything
> bad" with FGL frames, meaning that it is allowed to ignore pruning of FGL
> (or even VL frames) entirely...it just can't falsely drop FGL frames.  And
> it must not decapsulate an FGL frame onto a link for which that FGL frame
> doesn't belong. In this option, all RBridges must be upgraded to FGL-safe,
> but it need not happen instantaneously...it's fine to mix FGL-safe RBridges
> with VL RBridges...it's just not safe to inject FGL frames yet.  Once all
> the RBridges have been upgraded to be FGL-safe, then edge RBridges can start
> announcing they are connected to an FGL link, and can start injecting FGL
> packets.  It is considered a misconfiguration if you start injecting FGL
> frames before all the RBridges are upgraded to FGL-safe, so an additional
> chore for an FGL-safe RBridge R1 is to examine LSPs, and if any RBridge
> claims to be attached to an FGL link, then R1 must ostrasize any VL
> neighbors.  (don't start ostrasizing VL guys until it is necessary because
> of actually starting to use FGL frames, in other words). This option is more
> work for the upgraded RBridges than option 1, and there still isn't good
> coexistence with VL guys long-term (as option 3 will), but it does allow
> upgrading RBridges one by one in a working campus without causing
> disruption.
>
> Option 3: Two types of RBridges.  VL, of course, is the same as in options 1
> and 2.  This option makes FGL guys do more work, but allows maximal long
> term coexistence of VL and FGL guys. In this option, FGL guys calculate
> unicast paths to FGL edge guys that avoid any VL guys, and calculate at
> least one FGL-friendly multicast tree that also avoids any VL guys. So let's
> say R1 is an FGL guy.  R1 discards all LSPs from VL guys (ones that don't
> advertise FGL capability in their LSP), when calculating paths to other FGL
> RBridges. Then R1 calculates paths to the VL guys using all LSPs.  Likewise,
> when calculating an FGL-friendly tree, R1 calculates a tree through only FGL
> guys.   This option is more work for the upgraded RBridges (because they
> have to calculate Dijkstra in two different ways, one for reaching VL guys,
> and one for reaching FGL guys).  However, it does allow having long-term
> coexistence with VL guys.  For instance, you could forever keep some VL edge
> RBridges that communicate just fine through an FGL core.  They can stay
> there forever, and still be able to communicate through the core to all the
> links attaching to their VLAN.
>
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill
>

From d3e3e3@gmail.com  Tue Jan 29 07:09:23 2013
Return-Path: <d3e3e3@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CDA9E21F8840 for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 07:09:23 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.599
X-Spam-Level: 
X-Spam-Status: No, score=-103.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id k7dRzhsRxxNu for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 07:09:22 -0800 (PST)
Received: from mail-ob0-f169.google.com (mail-ob0-f169.google.com [209.85.214.169]) by ietfa.amsl.com (Postfix) with ESMTP id C101F21F84D1 for <trill@ietf.org>; Tue, 29 Jan 2013 07:09:22 -0800 (PST)
Received: by mail-ob0-f169.google.com with SMTP id ta14so504094obb.28 for <trill@ietf.org>; Tue, 29 Jan 2013 07:09:22 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=x-received:mime-version:in-reply-to:references:from:date:message-id :subject:to:cc:content-type; bh=a5dBmccTr7Mi7pXcL/z3D1SQdIXTAzE7By9XFDXYxuk=; b=uJtjPIO22DFpsfciQDm3znnTgv5ULJ6xpngBRYCiZM+hnwUA+t4ztzEimwiy0VVZ53 aqx6OgM6wRB0XN15t1YQCTeJxc6mkkh63rPHTP+3S3LmRTwVC4ubevasK2dM8hfsDVie SfToOIrrYUFslux4glRF26ZFnUrRFlUZne61Ri7GhvlcdASqF527FqqyGtj9+aqHlmRZ 9QHGSxSFIMwr+uPHo5AkIxPe24U4T3LGc3DoOJdheZM0VZT64nIwZvjuH/myqxPAcQ5Y Kbe49kxd4SsKZe1YIZsh6qpKvxy3Z90LAyA1pKHawm8EZDObmIKNXIpoagxAk4F+y3yW zHXg==
X-Received: by 10.182.78.233 with SMTP id e9mr37701obx.68.1359472162320; Tue, 29 Jan 2013 07:09:22 -0800 (PST)
MIME-Version: 1.0
Received: by 10.76.98.168 with HTTP; Tue, 29 Jan 2013 07:09:02 -0800 (PST)
In-Reply-To: <CA+-tSzx0E3J4i5Dvfbj4egruCeGi7z1se1Jn5RXtx3RedQLPdg@mail.gmail.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <CA+-tSzx0E3J4i5Dvfbj4egruCeGi7z1se1Jn5RXtx3RedQLPdg@mail.gmail.com>
From: Donald Eastlake <d3e3e3@gmail.com>
Date: Tue, 29 Jan 2013 10:09:02 -0500
Message-ID: <CAF4+nEGO0BDHOAb9Byb_-5anFNwxJ5wLtscyqb=ewV=qA630Yw@mail.gmail.com>
To: Anoop Ghanwani <anoop@alumni.duke.edu>
Content-Type: text/plain; charset=ISO-8859-1
Cc: Radia Perlman <radiaperlman@gmail.com>, trill@ietf.org
Subject: Re: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 29 Jan 2013 15:09:23 -0000

Hi Anoop,

On Tue, Jan 29, 2013 at 9:42 AM, Anoop Ghanwani <anoop@alumni.duke.edu> wrote:
> At noted, option 1 looks dangerous so that should be ruled out.
>
> Option 2 seems reasonable.  It might be useful to clarify a couple of things:
> - Can the FGL-safe campus have both VLAN & FGL frames going around?

Yes.

That is part of the migration idea. There may be FGL-safe TRILL
switches (probably former VL switches with a software upgrade) that
cannot ingress or egress FGL. For an end station attached to such an
FGL-safe TRILL switch to communicate with an end station attached to
some new Full FGL switch, it needs to use VL frames. Or you could have
two end stations attached to RB1 and RB99 that were communicating fine
in a VL campus but you are now adding new FGL switches and upgrading
the core to FGL. Until you actually introduce FGL traffic by
configuring FGL ports, these end stations can continue to communicate
using VL frames through their VL TRILL switches and then through this
new FGL core. As soon as you want to actually introduce FGL traffic
(including multi-destination which can get distributed to every switch
with data connectivity to the campus) you would have to upgrade RB1
and RB99 to be FGL-safe, but the end stations can continue to talk to
each other with VL frames.

> - Do all RBridges need to agree on when the campus is FGL-safe?  I would
>   hope not.  But the problem I'm thinking about is where RB1 thinks
> the campus is
>   FGL-safe and starts announcing attachment to FGL links, but RB2 hadn't yet
>   concluded that process and it see RB3 which happens to be VL.

Well, if you are migrating to FGL, you should upgrade all your VL
TRILL switches to be at least FGL-safe, possibly add new full FGL
switches, and then introduce FGL traffic. It obviously takes a little
time for information to propagate as each VL TRILL switch is upgrade
to be FGL-safe and later that there is a source of true FGL frames at
a TRILL switch in the campus. But all TRILL switches should agree
pretty quickly converge to the same link state database.

> Option 3 seems to introduce more complexity to solve this problem than
> is necessary.

I agree.

Thanks,
Donald
=============================
 Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
 155 Beaver Street, Milford, MA 01757 USA
 d3e3e3@gmail.com

> Anoop
>
> On Fri, Jan 25, 2013 at 11:00 AM, Radia Perlman <radiaperlman@gmail.com> wrote:
>> I'm going to summarize three options for phasing-in FGL, and explain the
>> tradeoffs, to help people have informed opinions about the implications of
>> the three approaches.  There are infinite variations, of course, but as I
>> said, I'll describe three main ones.
>>
>> Option 1:  Original draft, no changes.  There are two types of RBridges. The
>> first type is "VL" which only understands VLAN tags, and have unknown and
>> possibly dangerous behavior when they receive an FGL-labeled packet. (e.g.,
>> decapsulate a packet onto a link where it is not allowed, or mistakenly
>> prune a multicast so it does not reach everywhere it should). The second
>> type is "FGL".  FGL guys need to do two things: 1) understand FGL tags and
>> do the right thing with them, and 2) ostrasize VL guys...meaning that an FGL
>> guy refuses to form an adjacency with a VL guy.  This option is very simple
>> for implementers of FGL, but the implication is that you have to upgrade
>> your entire campus to FGL at once.  There is no coexistence.
>>
>> Option 2: Two types of RBridges, but the 2nd type is different from option
>> 1.  VL guys of course are the same as in option 1...they cannot be trusted
>> with FGL frames. The second type I will call "FGL-safe". An FGL-safe RBridge
>> must advertise in its LSP that it is FGL-safe, and it must not "do anything
>> bad" with FGL frames, meaning that it is allowed to ignore pruning of FGL
>> (or even VL frames) entirely...it just can't falsely drop FGL frames.  And
>> it must not decapsulate an FGL frame onto a link for which that FGL frame
>> doesn't belong. In this option, all RBridges must be upgraded to FGL-safe,
>> but it need not happen instantaneously...it's fine to mix FGL-safe RBridges
>> with VL RBridges...it's just not safe to inject FGL frames yet.  Once all
>> the RBridges have been upgraded to be FGL-safe, then edge RBridges can start
>> announcing they are connected to an FGL link, and can start injecting FGL
>> packets.  It is considered a misconfiguration if you start injecting FGL
>> frames before all the RBridges are upgraded to FGL-safe, so an additional
>> chore for an FGL-safe RBridge R1 is to examine LSPs, and if any RBridge
>> claims to be attached to an FGL link, then R1 must ostrasize any VL
>> neighbors.  (don't start ostrasizing VL guys until it is necessary because
>> of actually starting to use FGL frames, in other words). This option is more
>> work for the upgraded RBridges than option 1, and there still isn't good
>> coexistence with VL guys long-term (as option 3 will), but it does allow
>> upgrading RBridges one by one in a working campus without causing
>> disruption.
>>
>> Option 3: Two types of RBridges.  VL, of course, is the same as in options 1
>> and 2.  This option makes FGL guys do more work, but allows maximal long
>> term coexistence of VL and FGL guys. In this option, FGL guys calculate
>> unicast paths to FGL edge guys that avoid any VL guys, and calculate at
>> least one FGL-friendly multicast tree that also avoids any VL guys. So let's
>> say R1 is an FGL guy.  R1 discards all LSPs from VL guys (ones that don't
>> advertise FGL capability in their LSP), when calculating paths to other FGL
>> RBridges. Then R1 calculates paths to the VL guys using all LSPs.  Likewise,
>> when calculating an FGL-friendly tree, R1 calculates a tree through only FGL
>> guys.   This option is more work for the upgraded RBridges (because they
>> have to calculate Dijkstra in two different ways, one for reaching VL guys,
>> and one for reaching FGL guys).  However, it does allow having long-term
>> coexistence with VL guys.  For instance, you could forever keep some VL edge
>> RBridges that communicate just fine through an FGL core.  They can stay
>> there forever, and still be able to communicate through the core to all the
>> links attaching to their VLAN.
>>
>> _______________________________________________
>> trill mailing list
>> trill@ietf.org
>> https://www.ietf.org/mailman/listinfo/trill
>>
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill

From d3e3e3@gmail.com  Tue Jan 29 08:38:39 2013
Return-Path: <d3e3e3@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B57A821F88A6 for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 08:38:39 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -99.951
X-Spam-Level: 
X-Spam-Status: No, score=-99.951 tagged_above=-999 required=5 tests=[AWL=-3.647, BAYES_00=-2.599, CHARSET_FARAWAY_HEADER=3.2, MIME_8BIT_HEADER=0.3, MIME_CHARSET_FARAWAY=2.45, RCVD_IN_DNSWL_LOW=-1,  SARE_SUB_ENC_GB2312=1.345, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id tmk3c9mHkA-3 for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 08:38:39 -0800 (PST)
Received: from mail-ob0-f178.google.com (mail-ob0-f178.google.com [209.85.214.178]) by ietfa.amsl.com (Postfix) with ESMTP id CC43421F88A9 for <trill@ietf.org>; Tue, 29 Jan 2013 08:38:38 -0800 (PST)
Received: by mail-ob0-f178.google.com with SMTP id wd20so609646obb.23 for <trill@ietf.org>; Tue, 29 Jan 2013 08:38:38 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=x-received:mime-version:in-reply-to:references:from:date:message-id :subject:to:cc:content-type:content-transfer-encoding; bh=nxVGEVx/FGQomFlCBsx9gS5TyXxPrKb176smSnVdOi0=; b=MsYeG9CHqggJwog7LlVEK2CGVudks8S5XKOVucIQ9en7PgJayR55Q+/6MYNtHcs9C9 ilSGdGrHVrhYteGEwDGKE4Tth5r09tDb6RZsn7ORwiMcjxXdRyC3a5HqYB4esaenJAcE VGH4lcNqczMM0BkLyFxpwFcxdSOOHR2/wDJhwlg7K54bP664u1/l5gHimCwfqzJ4xZ7c dtjXrIl7d+V3PIsr1bgP9HU1ATzf4J8MepUtlDxVOAdMKgw8LB3S9+HTYnfbXnRkJbO5 PQLXEbhmhGn+JsErDK6S7CGxwWjy7Cq3YKtgo0KXW8+NsUTnApIpBO4TAo6rytmG1WXT Aamw==
X-Received: by 10.182.40.3 with SMTP id t3mr1183032obk.56.1359477518350; Tue, 29 Jan 2013 08:38:38 -0800 (PST)
MIME-Version: 1.0
Received: by 10.76.98.168 with HTTP; Tue, 29 Jan 2013 08:38:17 -0800 (PST)
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F6514AD@nkgeml501-mbs.china.huawei.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <DD5FC8DE455C3348B94340C0AB5517334F651415@nkgeml501-mbs.china.huawei.com> <CAFOuuo597npOuDLT1di-0V+Q1Fh7nYvKD3jHG3bbZPB81hc=Tw@mail.gmail.com> <DD5FC8DE455C3348B94340C0AB5517334F6514AD@nkgeml501-mbs.china.huawei.com>
From: Donald Eastlake <d3e3e3@gmail.com>
Date: Tue, 29 Jan 2013 11:38:17 -0500
Message-ID: <CAF4+nEHR9SadJoAFXGqD-s_=K-js4Z3E7ONsLWxBSHjPMN+hzQ@mail.gmail.com>
To: Haoweiguo <haoweiguo@huawei.com>
Content-Type: text/plain; charset=GB2312
Content-Transfer-Encoding: quoted-printable
Cc: Radia Perlman <radiaperlman@gmail.com>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] =?gb2312?b?tPC4tDogtPC4tDogRXhwbGFpbmluZyB0aHJlZSBvcHRp?= =?gb2312?b?b25zIGZvciB1cGdyYWRpbmcgdG8gRkdMIChmaW5lLWdyYWluZWQt?= =?gb2312?b?bGFiZWxpbmcp?=
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 29 Jan 2013 16:38:39 -0000

Hi Weiguo,

On Tue, Jan 29, 2013 at 8:23 AM, Haoweiguo <haoweiguo@huawei.com>
wrote:
> Hi Radia,
>
> For unicast, I think TRILL switch don't care about inner VLAN or FGL
> and they forward the unicast data frame just by egress nickname. So
> I think VL switch can support unicast FGL data frame forwarding and
> have no problem.

It is true that the case for unicast is better than the case for
multi-destination. But a VL TRILL switch might (a) drop an FGL frame
as malformed because it does not have 0x8100 after the inner MAC
addresses or (b) do incorrect flow classification for ECMP, resulting
in persistent re-ordering, because it is trying to look deeper into
the frame past the data label and gets confused by an FGL. If it does
either (a) or (b) it is not FGL-safe (although (b) is not as bad as
(a)) and you have to route FGL frames around such VL switches.

> For multicast, VL switch can't perform distribution tree pruning by
> FGL and VL switch has following possible behaviors:
>
> 1,VL switch drops FGL multicast data frame because VL switch cares
> about inner ethtype. The VL switch only support inner ethtype of
> 0x8100.

If it does this, it is not FGL-safe and you have to route FGL traffic
around VL switches.

> 2,VL switch don't implement VLAN pruning and forward multicast data
> frame by broadcast distribution tree. The forwarding key of the
> broadcast distribution tree is nickname of DST root. The VL switch
> don't care about inner VLAN or FGL also Multicast FGL data frame can
> be forwarded by this kind of VL switch.

As far as I know, all TRILL products with silicon fast path support do
implement pruning.

> 3,VL switch implements VLAN pruning and forward multicast data frame
> by VLAN pruning tree. The forwarding key of the pruning tree is root
> nickname+inner VLAN.  If the switch cares about accurate inner
> ethtype(0x8100), FGL data frame will be dropped. If the switch
> allows random or multiple ethtype, then FGL data frame can be
> forwarded higher 12 bit of FGL.

If the VL switch drops frames without the 0x8100, they it is not
FGL-safe and you have to steer FGL traffic around VL switches. If the
VL switch pruning is based on believing that the top 12 bits of the
FGL are a VLAN ID, it will prune incorrect and will not be FGL-safe,
so you have to route FGL traffic around such a VL switch.

And then there is a fourth behavior:

4, VL switch ignores the Ethertype and uses the top 12 bits of the FGL
as a VLAN ID. As a result, it egresses multi-destination frames out
wrong ports in the wrong VLAN and violates security considerations.
This may be the worst of the possible behaviors. Since you cannot tell
if a VL TRILL switch might do this, you cannot trust VL switches to
properly handle FGL traffic.

However, as far as I know, existing silicon based VL TRILL switches
can be upgraded with software changes to be FGL-safe.

Thanks,
Donald
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D
 Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
 155 Beaver Street, Milford, MA 01757 USA
 d3e3e3@gmail.com

> thanks
>
> weiguo
>
> ________________________________
>
> =B7=A2=BC=FE=C8=CB: Radia Perlman [radiaperlman@gmail.com]
> =B7=A2=CB=CD=CA=B1=BC=E4: 2013=C4=EA1=D4=C229=C8=D5 18:59
> =B5=BD: Haoweiguo
> Cc: trill@ietf.org
> =D6=F7=CC=E2: Re: =B4=F0=B8=B4: [trill] Explaining three options for upgr=
ading to FGL
> (fine-grained-labeling)
>
> Hi Weiguo,
>
>  If we knew that it was already safe to transit FGL frames through VL nod=
es,
> that would be great.  We could basically just start sending FGL frames, n=
ot
> worry about ostrasizing VL RBridges, not worry about calculating two
> forwarding databases.
>
>  But,  I was assuming that there were implementations of VL that might do
> bad things if they saw an FGL packet.  For instance, for unicast, they mi=
ght
> drop the packet, or they might decapsulate it onto a link that should not
> receive packets for that FGL.  For multicast, they might falsely filter t=
he
> packet.  If that assumption is true, then the safe thing is to avoid VL
> RBridges for FGL packets, which means calculating paths that do not inclu=
de
> VL RBridges.
> .
> Thanks,
>
> Radia
>
>
>
> On Mon, Jan 28, 2013 at 10:59 PM, Haoweiguo <haoweiguo@huawei.com> wrote:
>>
>> Hi Radia,
>>
>> In option3:
>>
>> why FGL switches calculate unicast paths to FGL edge guys that avoid any
>> VL guys, and calculate at least one FGL-friendly multicast tree that als=
o
>> avoids any VL guys? If two FGL switches can only transmit data through V=
L
>> switch, the method will disconnect data channel between FGL switches. I
>> think VL switch is safe as transit node to tranmit unicast FGL data pack=
et.
>>
>> thanks
>>
>> weiguo
>>
>> ________________________________
>>
>> =B7=A2=BC=FE=C8=CB: Radia Perlman [radiaperlman@gmail.com]
>> =B7=A2=CB=CD=CA=B1=BC=E4: 2013=C4=EA1=D4=C226=C8=D5 3:00
>> =B5=BD: trill@ietf.org
>> =D6=F7=CC=E2: [trill] Explaining three options for upgrading to FGL
>> (fine-grained-labeling)
>>
>> I'm going to summarize three options for phasing-in FGL, and explain the
>> tradeoffs, to help people have informed opinions about the implications =
of
>> the three approaches.  There are infinite variations, of course, but as =
I
>> said, I'll describe three main ones.
>>
>> Option 1:  Original draft, no changes.  There are two types of RBridges.
>> The first type is "VL" which only understands VLAN tags, and have unknow=
n
>> and possibly dangerous behavior when they receive an FGL-labeled packet.
>> (e.g., decapsulate a packet onto a link where it is not allowed, or
>> mistakenly prune a multicast so it does not reach everywhere it should).=
 The
>> second type is "FGL".  FGL guys need to do two things: 1) understand FGL
>> tags and do the right thing with them, and 2) ostrasize VL guys...meanin=
g
>> that an FGL guy refuses to form an adjacency with a VL guy.  This option=
 is
>> very simple for implementers of FGL, but the implication is that you hav=
e to
>> upgrade your entire campus to FGL at once.  There is no coexistence.
>>
>> Option 2: Two types of RBridges, but the 2nd type is different from opti=
on
>> 1.  VL guys of course are the same as in option 1...they cannot be trust=
ed
>> with FGL frames. The second type I will call "FGL-safe". An FGL-safe RBr=
idge
>> must advertise in its LSP that it is FGL-safe, and it must not "do anyth=
ing
>> bad" with FGL frames, meaning that it is allowed to ignore pruning of FG=
L
>> (or even VL frames) entirely...it just can't falsely drop FGL frames.  A=
nd
>> it must not decapsulate an FGL frame onto a link for which that FGL fram=
e
>> doesn't belong. In this option, all RBridges must be upgraded to FGL-saf=
e,
>> but it need not happen instantaneously...it's fine to mix FGL-safe RBrid=
ges
>> with VL RBridges...it's just not safe to inject FGL frames yet.  Once al=
l
>> the RBridges have been upgraded to be FGL-safe, then edge RBridges can s=
tart
>> announcing they are connected to an FGL link, and can start injecting FG=
L
>> packets.  It is considered a misconfiguration if you start injecting FGL
>> frames before all the RBridges are upgraded to FGL-safe, so an additiona=
l
>> chore for an FGL-safe RBridge R1 is to examine LSPs, and if any RBridge
>> claims to be attached to an FGL link, then R1 must ostrasize any VL
>> neighbors.  (don't start ostrasizing VL guys until it is necessary becau=
se
>> of actually starting to use FGL frames, in other words). This option is =
more
>> work for the upgraded RBridges than option 1, and there still isn't good
>> coexistence with VL guys long-term (as option 3 will), but it does allow
>> upgrading RBridges one by one in a working campus without causing
>> disruption.
>>
>> Option 3: Two types of RBridges.  VL, of course, is the same as in optio=
ns
>> 1 and 2.  This option makes FGL guys do more work, but allows maximal lo=
ng
>> term coexistence of VL and FGL guys. In this option, FGL guys calculate
>> unicast paths to FGL edge guys that avoid any VL guys, and calculate at
>> least one FGL-friendly multicast tree that also avoids any VL guys. So l=
et's
>> say R1 is an FGL guy.  R1 discards all LSPs from VL guys (ones that don'=
t
>> advertise FGL capability in their LSP), when calculating paths to other =
FGL
>> RBridges. Then R1 calculates paths to the VL guys using all LSPs.  Likew=
ise,
>> when calculating an FGL-friendly tree, R1 calculates a tree through only=
 FGL
>> guys.   This option is more work for the upgraded RBridges (because they
>> have to calculate Dijkstra in two different ways, one for reaching VL gu=
ys,
>> and one for reaching FGL guys).  However, it does allow having long-term
>> coexistence with VL guys.  For instance, you could forever keep some VL =
edge
>> RBridges that communicate just fine through an FGL core.  They can stay
>> there forever, and still be able to communicate through the core to all =
the
>> links attaching to their VLAN.

From ayabaner@gmail.com  Tue Jan 29 11:45:04 2013
Return-Path: <ayabaner@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D9F7721F875C for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 11:45:04 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.598
X-Spam-Level: 
X-Spam-Status: No, score=-3.598 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id U3dPLvPcES8F for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 11:45:03 -0800 (PST)
Received: from mail-vb0-f52.google.com (mail-vb0-f52.google.com [209.85.212.52]) by ietfa.amsl.com (Postfix) with ESMTP id A494121F8722 for <trill@ietf.org>; Tue, 29 Jan 2013 11:45:03 -0800 (PST)
Received: by mail-vb0-f52.google.com with SMTP id fa15so513341vbb.25 for <trill@ietf.org>; Tue, 29 Jan 2013 11:45:03 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:x-received:in-reply-to:references:date:message-id :subject:from:to:cc:content-type; bh=QdQAE3gJPM+MPL4V55Pgnfz4hZ5N5i1+Tk0NPvtF5MA=; b=oRVX8pTO2AvmrPAgtf/0MIN1VHzwKsmGd0HS7QhoGJVDs2RvWDEcUI1hLbiIzHCp0r ahxFK5JdKDdaDPtFaug/kFeyFG/AdaMm71r04VTXCblzT0/y4p5WbCPIECbaWGINwy57 UW6jKWiEtJB4pOCJvsmnxhKswu93VjuqQi105KBIh71DyFvdF1Jb3Lnbwb9b06aopEHT poeJ5OhWIJlVYO8n6rhidBmJNhhWSOSuQHeBbFX1ytjj4jvmLuaP9ECKPqM26d8Jtze0 uDCcv83D6fOKDC2eyuW+rPC/8fyO+A7tM/R2yyRcTE8zGrsBigRH+ikr9nANEtVUG/d5 MTNQ==
MIME-Version: 1.0
X-Received: by 10.52.66.14 with SMTP id b14mr2170734vdt.0.1359488703070; Tue, 29 Jan 2013 11:45:03 -0800 (PST)
Received: by 10.220.144.78 with HTTP; Tue, 29 Jan 2013 11:45:02 -0800 (PST)
In-Reply-To: <CA+-tSzx0E3J4i5Dvfbj4egruCeGi7z1se1Jn5RXtx3RedQLPdg@mail.gmail.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <CA+-tSzx0E3J4i5Dvfbj4egruCeGi7z1se1Jn5RXtx3RedQLPdg@mail.gmail.com>
Date: Tue, 29 Jan 2013 11:45:02 -0800
Message-ID: <CAHD03N89RNX4=XxRZpD+Y2PCktYf-FtVjyYJ5oFS3dui3SsXaA@mail.gmail.com>
From: Ayan Banerjee <ayabaner@gmail.com>
To: Anoop Ghanwani <anoop@alumni.duke.edu>
Content-Type: multipart/alternative; boundary=20cf3071c8aa3903ab04d472a115
Cc: Radia Perlman <radiaperlman@gmail.com>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 29 Jan 2013 19:45:05 -0000

--20cf3071c8aa3903ab04d472a115
Content-Type: text/plain; charset=ISO-8859-1

I agree with Anoop here.

I think that we are describing a network that is FGL-safe and ready and
then switch to FGL-mode. However, we need some statements to account for a
VL-node coming-in/leaving-the network due to a flaky link.

Thanks,
Ayan



On Tue, Jan 29, 2013 at 6:42 AM, Anoop Ghanwani <anoop@alumni.duke.edu>wrote:

> At noted, option 1 looks dangerous so that should be ruled out.
>
> Option 2 seems reasonable.  It might be useful to clarify a couple of
> things:
> - Can the FGL-safe campus have both VLAN & FGL frames going around?
> - Do all RBridges need to agree on when the campus is FGL-safe?  I would
>   hope not.  But the problem I'm thinking about is where RB1 thinks
> the campus is
>   FGL-safe and starts announcing attachment to FGL links, but RB2 hadn't
> yet
>   concluded that process and it see RB3 which happens to be VL.
>
> Option 3 seems to introduce more complexity to solve this problem than
> is necessary.
>
> Anoop
>
> On Fri, Jan 25, 2013 at 11:00 AM, Radia Perlman <radiaperlman@gmail.com>
> wrote:
> > I'm going to summarize three options for phasing-in FGL, and explain the
> > tradeoffs, to help people have informed opinions about the implications
> of
> > the three approaches.  There are infinite variations, of course, but as I
> > said, I'll describe three main ones.
> >
> > Option 1:  Original draft, no changes.  There are two types of RBridges.
> The
> > first type is "VL" which only understands VLAN tags, and have unknown and
> > possibly dangerous behavior when they receive an FGL-labeled packet.
> (e.g.,
> > decapsulate a packet onto a link where it is not allowed, or mistakenly
> > prune a multicast so it does not reach everywhere it should). The second
> > type is "FGL".  FGL guys need to do two things: 1) understand FGL tags
> and
> > do the right thing with them, and 2) ostrasize VL guys...meaning that an
> FGL
> > guy refuses to form an adjacency with a VL guy.  This option is very
> simple
> > for implementers of FGL, but the implication is that you have to upgrade
> > your entire campus to FGL at once.  There is no coexistence.
> >
> > Option 2: Two types of RBridges, but the 2nd type is different from
> option
> > 1.  VL guys of course are the same as in option 1...they cannot be
> trusted
> > with FGL frames. The second type I will call "FGL-safe". An FGL-safe
> RBridge
> > must advertise in its LSP that it is FGL-safe, and it must not "do
> anything
> > bad" with FGL frames, meaning that it is allowed to ignore pruning of FGL
> > (or even VL frames) entirely...it just can't falsely drop FGL frames.
>  And
> > it must not decapsulate an FGL frame onto a link for which that FGL frame
> > doesn't belong. In this option, all RBridges must be upgraded to
> FGL-safe,
> > but it need not happen instantaneously...it's fine to mix FGL-safe
> RBridges
> > with VL RBridges...it's just not safe to inject FGL frames yet.  Once all
> > the RBridges have been upgraded to be FGL-safe, then edge RBridges can
> start
> > announcing they are connected to an FGL link, and can start injecting FGL
> > packets.  It is considered a misconfiguration if you start injecting FGL
> > frames before all the RBridges are upgraded to FGL-safe, so an additional
> > chore for an FGL-safe RBridge R1 is to examine LSPs, and if any RBridge
> > claims to be attached to an FGL link, then R1 must ostrasize any VL
> > neighbors.  (don't start ostrasizing VL guys until it is necessary
> because
> > of actually starting to use FGL frames, in other words). This option is
> more
> > work for the upgraded RBridges than option 1, and there still isn't good
> > coexistence with VL guys long-term (as option 3 will), but it does allow
> > upgrading RBridges one by one in a working campus without causing
> > disruption.
> >
> > Option 3: Two types of RBridges.  VL, of course, is the same as in
> options 1
> > and 2.  This option makes FGL guys do more work, but allows maximal long
> > term coexistence of VL and FGL guys. In this option, FGL guys calculate
> > unicast paths to FGL edge guys that avoid any VL guys, and calculate at
> > least one FGL-friendly multicast tree that also avoids any VL guys. So
> let's
> > say R1 is an FGL guy.  R1 discards all LSPs from VL guys (ones that don't
> > advertise FGL capability in their LSP), when calculating paths to other
> FGL
> > RBridges. Then R1 calculates paths to the VL guys using all LSPs.
>  Likewise,
> > when calculating an FGL-friendly tree, R1 calculates a tree through only
> FGL
> > guys.   This option is more work for the upgraded RBridges (because they
> > have to calculate Dijkstra in two different ways, one for reaching VL
> guys,
> > and one for reaching FGL guys).  However, it does allow having long-term
> > coexistence with VL guys.  For instance, you could forever keep some VL
> edge
> > RBridges that communicate just fine through an FGL core.  They can stay
> > there forever, and still be able to communicate through the core to all
> the
> > links attaching to their VLAN.
> >
> > _______________________________________________
> > trill mailing list
> > trill@ietf.org
> > https://www.ietf.org/mailman/listinfo/trill
> >
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill
>

--20cf3071c8aa3903ab04d472a115
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>I agree with Anoop here. <br><br></div>I think that w=
e are describing a network that is FGL-safe and ready and then switch to FG=
L-mode. However, we need some statements to account for a VL-node coming-in=
/leaving-the network due to a flaky link. <br>
<div><br>Thanks,<br>Ayan<br><br></div></div><div class=3D"gmail_extra"><br>=
<br><div class=3D"gmail_quote">On Tue, Jan 29, 2013 at 6:42 AM, Anoop Ghanw=
ani <span dir=3D"ltr">&lt;<a href=3D"mailto:anoop@alumni.duke.edu" target=
=3D"_blank">anoop@alumni.duke.edu</a>&gt;</span> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">At noted, option 1 looks dangerous so that s=
hould be ruled out.<br>
<br>
Option 2 seems reasonable. =A0It might be useful to clarify a couple of thi=
ngs:<br>
- Can the FGL-safe campus have both VLAN &amp; FGL frames going around?<br>
- Do all RBridges need to agree on when the campus is FGL-safe? =A0I would<=
br>
=A0 hope not. =A0But the problem I&#39;m thinking about is where RB1 thinks=
<br>
the campus is<br>
=A0 FGL-safe and starts announcing attachment to FGL links, but RB2 hadn&#3=
9;t yet<br>
=A0 concluded that process and it see RB3 which happens to be VL.<br>
<br>
Option 3 seems to introduce more complexity to solve this problem than<br>
is necessary.<br>
<br>
Anoop<br>
<div><div class=3D"h5"><br>
On Fri, Jan 25, 2013 at 11:00 AM, Radia Perlman &lt;<a href=3D"mailto:radia=
perlman@gmail.com">radiaperlman@gmail.com</a>&gt; wrote:<br>
&gt; I&#39;m going to summarize three options for phasing-in FGL, and expla=
in the<br>
&gt; tradeoffs, to help people have informed opinions about the implication=
s of<br>
&gt; the three approaches. =A0There are infinite variations, of course, but=
 as I<br>
&gt; said, I&#39;ll describe three main ones.<br>
&gt;<br>
&gt; Option 1: =A0Original draft, no changes. =A0There are two types of RBr=
idges. The<br>
&gt; first type is &quot;VL&quot; which only understands VLAN tags, and hav=
e unknown and<br>
&gt; possibly dangerous behavior when they receive an FGL-labeled packet. (=
e.g.,<br>
&gt; decapsulate a packet onto a link where it is not allowed, or mistakenl=
y<br>
&gt; prune a multicast so it does not reach everywhere it should). The seco=
nd<br>
&gt; type is &quot;FGL&quot;. =A0FGL guys need to do two things: 1) underst=
and FGL tags and<br>
&gt; do the right thing with them, and 2) ostrasize VL guys...meaning that =
an FGL<br>
&gt; guy refuses to form an adjacency with a VL guy. =A0This option is very=
 simple<br>
&gt; for implementers of FGL, but the implication is that you have to upgra=
de<br>
&gt; your entire campus to FGL at once. =A0There is no coexistence.<br>
&gt;<br>
&gt; Option 2: Two types of RBridges, but the 2nd type is different from op=
tion<br>
&gt; 1. =A0VL guys of course are the same as in option 1...they cannot be t=
rusted<br>
&gt; with FGL frames. The second type I will call &quot;FGL-safe&quot;. An =
FGL-safe RBridge<br>
&gt; must advertise in its LSP that it is FGL-safe, and it must not &quot;d=
o anything<br>
&gt; bad&quot; with FGL frames, meaning that it is allowed to ignore prunin=
g of FGL<br>
&gt; (or even VL frames) entirely...it just can&#39;t falsely drop FGL fram=
es. =A0And<br>
&gt; it must not decapsulate an FGL frame onto a link for which that FGL fr=
ame<br>
&gt; doesn&#39;t belong. In this option, all RBridges must be upgraded to F=
GL-safe,<br>
&gt; but it need not happen instantaneously...it&#39;s fine to mix FGL-safe=
 RBridges<br>
&gt; with VL RBridges...it&#39;s just not safe to inject FGL frames yet. =
=A0Once all<br>
&gt; the RBridges have been upgraded to be FGL-safe, then edge RBridges can=
 start<br>
&gt; announcing they are connected to an FGL link, and can start injecting =
FGL<br>
&gt; packets. =A0It is considered a misconfiguration if you start injecting=
 FGL<br>
&gt; frames before all the RBridges are upgraded to FGL-safe, so an additio=
nal<br>
&gt; chore for an FGL-safe RBridge R1 is to examine LSPs, and if any RBridg=
e<br>
&gt; claims to be attached to an FGL link, then R1 must ostrasize any VL<br=
>
&gt; neighbors. =A0(don&#39;t start ostrasizing VL guys until it is necessa=
ry because<br>
&gt; of actually starting to use FGL frames, in other words). This option i=
s more<br>
&gt; work for the upgraded RBridges than option 1, and there still isn&#39;=
t good<br>
&gt; coexistence with VL guys long-term (as option 3 will), but it does all=
ow<br>
&gt; upgrading RBridges one by one in a working campus without causing<br>
&gt; disruption.<br>
&gt;<br>
&gt; Option 3: Two types of RBridges. =A0VL, of course, is the same as in o=
ptions 1<br>
&gt; and 2. =A0This option makes FGL guys do more work, but allows maximal =
long<br>
&gt; term coexistence of VL and FGL guys. In this option, FGL guys calculat=
e<br>
&gt; unicast paths to FGL edge guys that avoid any VL guys, and calculate a=
t<br>
&gt; least one FGL-friendly multicast tree that also avoids any VL guys. So=
 let&#39;s<br>
&gt; say R1 is an FGL guy. =A0R1 discards all LSPs from VL guys (ones that =
don&#39;t<br>
&gt; advertise FGL capability in their LSP), when calculating paths to othe=
r FGL<br>
&gt; RBridges. Then R1 calculates paths to the VL guys using all LSPs. =A0L=
ikewise,<br>
&gt; when calculating an FGL-friendly tree, R1 calculates a tree through on=
ly FGL<br>
&gt; guys. =A0 This option is more work for the upgraded RBridges (because =
they<br>
&gt; have to calculate Dijkstra in two different ways, one for reaching VL =
guys,<br>
&gt; and one for reaching FGL guys). =A0However, it does allow having long-=
term<br>
&gt; coexistence with VL guys. =A0For instance, you could forever keep some=
 VL edge<br>
&gt; RBridges that communicate just fine through an FGL core. =A0They can s=
tay<br>
&gt; there forever, and still be able to communicate through the core to al=
l the<br>
&gt; links attaching to their VLAN.<br>
&gt;<br>
</div></div>&gt; _______________________________________________<br>
&gt; trill mailing list<br>
&gt; <a href=3D"mailto:trill@ietf.org">trill@ietf.org</a><br>
&gt; <a href=3D"https://www.ietf.org/mailman/listinfo/trill" target=3D"_bla=
nk">https://www.ietf.org/mailman/listinfo/trill</a><br>
&gt;<br>
_______________________________________________<br>
trill mailing list<br>
<a href=3D"mailto:trill@ietf.org">trill@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/trill" target=3D"_blank">h=
ttps://www.ietf.org/mailman/listinfo/trill</a><br>
</blockquote></div><br></div>

--20cf3071c8aa3903ab04d472a115--

From d3e3e3@gmail.com  Tue Jan 29 12:50:20 2013
Return-Path: <d3e3e3@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1DAE221F892F for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 12:50:20 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.991
X-Spam-Level: 
X-Spam-Status: No, score=-102.991 tagged_above=-999 required=5 tests=[AWL=0.608, BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id R69uIXVDljGG for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 12:50:19 -0800 (PST)
Received: from mail-ob0-f180.google.com (mail-ob0-f180.google.com [209.85.214.180]) by ietfa.amsl.com (Postfix) with ESMTP id 3922121F891D for <trill@ietf.org>; Tue, 29 Jan 2013 12:50:19 -0800 (PST)
Received: by mail-ob0-f180.google.com with SMTP id ef5so878319obb.25 for <trill@ietf.org>; Tue, 29 Jan 2013 12:50:18 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=x-received:mime-version:in-reply-to:references:from:date:message-id :subject:to:cc:content-type; bh=f/qtXwy7NOIw47caDtkR4l6fn8uR7tBspFs5cQXbQJ8=; b=LomFr/ZE8fEFN2pRgCe3jwA5WtcyNUS0Sd3pdrnlimbdRDyxpyIPKWWOcwF3GHOKU5 3lQINBkjFaB58lBSMP5Vg3SuLnAmxIjtO1+cVcYVmCYvHxHaROsqxVkS2BDYtSKjWfh7 YBw8thfyEBtSIts1pBDytJHrn3M2rXjssCDvjGHAC/bkrRhrsQ4ejUMmNCdiF/JT1rtD iX1wHZdwIETfjrVh8DNkhaMRS2Mapt1I+9FNisLpwmX/dIOe3bxNzxQNpFOC50PrWkjW EaPUsdqreBHyaNZx9PAm4He6WkvBhEbT6FljFiH7A0hZswWA2xTCOmX3EMdeaq6rkGJG R0kw==
X-Received: by 10.182.48.37 with SMTP id i5mr1887042obn.18.1359492618680; Tue, 29 Jan 2013 12:50:18 -0800 (PST)
MIME-Version: 1.0
Received: by 10.76.98.168 with HTTP; Tue, 29 Jan 2013 12:49:58 -0800 (PST)
In-Reply-To: <CAHD03N89RNX4=XxRZpD+Y2PCktYf-FtVjyYJ5oFS3dui3SsXaA@mail.gmail.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <CA+-tSzx0E3J4i5Dvfbj4egruCeGi7z1se1Jn5RXtx3RedQLPdg@mail.gmail.com> <CAHD03N89RNX4=XxRZpD+Y2PCktYf-FtVjyYJ5oFS3dui3SsXaA@mail.gmail.com>
From: Donald Eastlake <d3e3e3@gmail.com>
Date: Tue, 29 Jan 2013 15:49:58 -0500
Message-ID: <CAF4+nEG7YKGzGmObAMaOxackivjJsALymqHX=6PKBd5qU-Z+zQ@mail.gmail.com>
To: Ayan Banerjee <ayabaner@gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Cc: Anoop Ghanwani <anoop@alumni.duke.edu>, Radia Perlman <radiaperlman@gmail.com>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 29 Jan 2013 20:50:20 -0000

Hi Ayan,

On Tue, Jan 29, 2013 at 2:45 PM, Ayan Banerjee <ayabaner@gmail.com> wrote:
> I agree with Anoop here.
>
> I think that we are describing a network that is FGL-safe and ready and then
> switch to FGL-mode. However, we need some statements to account for a
> VL-node coming-in/leaving-the network due to a flaky link.

The general assumption is that you are migrating from VL to FGL. First
you migrate to FGL-safe. Then you unleash the FGL traffic, by
configuring FGL ports, which isolates any remaining (FGL-unsafe) VL
switches while still being able to handle VL traffic between VL ports
on FGL-safe switches.

After you have unleashed FGL traffic, it would certainly be possible
that a VL switch could be attached to the network or come and go due
to a flakey link, but this would have almost no effect. With FGL
traffic being handled, VL switches are isolated.

Thanks,
Donald
=============================
 Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
 155 Beaver Street, Milford, MA 01757 USA
 d3e3e3@gmail.com

> Thanks,
> Ayan
>
>
>
> On Tue, Jan 29, 2013 at 6:42 AM, Anoop Ghanwani <anoop@alumni.duke.edu>
> wrote:
>>
>> At noted, option 1 looks dangerous so that should be ruled out.
>>
>> Option 2 seems reasonable.  It might be useful to clarify a couple of
>> things:
>> - Can the FGL-safe campus have both VLAN & FGL frames going around?
>> - Do all RBridges need to agree on when the campus is FGL-safe?  I would
>>   hope not.  But the problem I'm thinking about is where RB1 thinks
>> the campus is
>>   FGL-safe and starts announcing attachment to FGL links, but RB2 hadn't
>> yet
>>   concluded that process and it see RB3 which happens to be VL.
>>
>> Option 3 seems to introduce more complexity to solve this problem than
>> is necessary.
>>
>> Anoop
>>
>> On Fri, Jan 25, 2013 at 11:00 AM, Radia Perlman <radiaperlman@gmail.com>
>> wrote:
>> > I'm going to summarize three options for phasing-in FGL, and explain the
>> > tradeoffs, to help people have informed opinions about the implications
>> > of
>> > the three approaches.  There are infinite variations, of course, but as
>> > I
>> > said, I'll describe three main ones.
>> >
>> > Option 1:  Original draft, no changes.  There are two types of RBridges.
>> > The
>> > first type is "VL" which only understands VLAN tags, and have unknown
>> > and
>> > possibly dangerous behavior when they receive an FGL-labeled packet.
>> > (e.g.,
>> > decapsulate a packet onto a link where it is not allowed, or mistakenly
>> > prune a multicast so it does not reach everywhere it should). The second
>> > type is "FGL".  FGL guys need to do two things: 1) understand FGL tags
>> > and
>> > do the right thing with them, and 2) ostrasize VL guys...meaning that an
>> > FGL
>> > guy refuses to form an adjacency with a VL guy.  This option is very
>> > simple
>> > for implementers of FGL, but the implication is that you have to upgrade
>> > your entire campus to FGL at once.  There is no coexistence.
>> >
>> > Option 2: Two types of RBridges, but the 2nd type is different from
>> > option
>> > 1.  VL guys of course are the same as in option 1...they cannot be
>> > trusted
>> > with FGL frames. The second type I will call "FGL-safe". An FGL-safe
>> > RBridge
>> > must advertise in its LSP that it is FGL-safe, and it must not "do
>> > anything
>> > bad" with FGL frames, meaning that it is allowed to ignore pruning of
>> > FGL
>> > (or even VL frames) entirely...it just can't falsely drop FGL frames.
>> > And
>> > it must not decapsulate an FGL frame onto a link for which that FGL
>> > frame
>> > doesn't belong. In this option, all RBridges must be upgraded to
>> > FGL-safe,
>> > but it need not happen instantaneously...it's fine to mix FGL-safe
>> > RBridges
>> > with VL RBridges...it's just not safe to inject FGL frames yet.  Once
>> > all
>> > the RBridges have been upgraded to be FGL-safe, then edge RBridges can
>> > start
>> > announcing they are connected to an FGL link, and can start injecting
>> > FGL
>> > packets.  It is considered a misconfiguration if you start injecting FGL
>> > frames before all the RBridges are upgraded to FGL-safe, so an
>> > additional
>> > chore for an FGL-safe RBridge R1 is to examine LSPs, and if any RBridge
>> > claims to be attached to an FGL link, then R1 must ostrasize any VL
>> > neighbors.  (don't start ostrasizing VL guys until it is necessary
>> > because
>> > of actually starting to use FGL frames, in other words). This option is
>> > more
>> > work for the upgraded RBridges than option 1, and there still isn't good
>> > coexistence with VL guys long-term (as option 3 will), but it does allow
>> > upgrading RBridges one by one in a working campus without causing
>> > disruption.
>> >
>> > Option 3: Two types of RBridges.  VL, of course, is the same as in
>> > options 1
>> > and 2.  This option makes FGL guys do more work, but allows maximal long
>> > term coexistence of VL and FGL guys. In this option, FGL guys calculate
>> > unicast paths to FGL edge guys that avoid any VL guys, and calculate at
>> > least one FGL-friendly multicast tree that also avoids any VL guys. So
>> > let's
>> > say R1 is an FGL guy.  R1 discards all LSPs from VL guys (ones that
>> > don't
>> > advertise FGL capability in their LSP), when calculating paths to other
>> > FGL
>> > RBridges. Then R1 calculates paths to the VL guys using all LSPs.
>> > Likewise,
>> > when calculating an FGL-friendly tree, R1 calculates a tree through only
>> > FGL
>> > guys.   This option is more work for the upgraded RBridges (because they
>> > have to calculate Dijkstra in two different ways, one for reaching VL
>> > guys,
>> > and one for reaching FGL guys).  However, it does allow having long-term
>> > coexistence with VL guys.  For instance, you could forever keep some VL
>> > edge
>> > RBridges that communicate just fine through an FGL core.  They can stay
>> > there forever, and still be able to communicate through the core to all
>> > the
>> > links attaching to their VLAN.
>> >
>> > _______________________________________________
>> > trill mailing list
>> > trill@ietf.org
>> > https://www.ietf.org/mailman/listinfo/trill
>> >
>> _______________________________________________
>> trill mailing list
>> trill@ietf.org
>> https://www.ietf.org/mailman/listinfo/trill
>
>
>
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill
>

From ayabaner@gmail.com  Tue Jan 29 13:15:33 2013
Return-Path: <ayabaner@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E605C21F888C for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 13:15:33 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.598
X-Spam-Level: 
X-Spam-Status: No, score=-3.598 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id NVVO6dmV3pAC for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 13:15:32 -0800 (PST)
Received: from mail-vb0-f54.google.com (mail-vb0-f54.google.com [209.85.212.54]) by ietfa.amsl.com (Postfix) with ESMTP id 7038F21F875A for <trill@ietf.org>; Tue, 29 Jan 2013 13:15:32 -0800 (PST)
Received: by mail-vb0-f54.google.com with SMTP id l1so578662vba.13 for <trill@ietf.org>; Tue, 29 Jan 2013 13:15:31 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:x-received:in-reply-to:references:date:message-id :subject:from:to:cc:content-type; bh=UUPdEgz/077f54qkOBSgaA4co8UYd8H3eiS2QOlyCXs=; b=gxWXusHNxsmjEqm6Nnz6jD5+zhryoxIPRuVls3c5o8uzHit46JgkzYZ5E2bIrxc1Rd 7N2gJvi9S7/bvdPkFYli4lOj2D8RBnCEX/X5ij7+YyEMS1N8zo8tIhHK9TY5vM83jmE+ FhH77Ia3NqELAZyoCysQJ1nPv4WtzXxOWMibS1DOC96wXAtM7SO9KALwO5DRumCfFcLt QGhiDFLo3Jgh+6oKYiNKm3vZmc3NNQgMQP97FOuktXsQDcZYSa36dXb8XJUzrqz0X2It IlbSFLq9w2nCkfiJdxuLrx9phPexH3EWdazH2R7xxBo5XwyPrsDmiqUX0McJch/wOwpI jUgQ==
MIME-Version: 1.0
X-Received: by 10.52.69.176 with SMTP id f16mr2310491vdu.97.1359494131522; Tue, 29 Jan 2013 13:15:31 -0800 (PST)
Received: by 10.220.144.78 with HTTP; Tue, 29 Jan 2013 13:15:31 -0800 (PST)
In-Reply-To: <CAF4+nEG7YKGzGmObAMaOxackivjJsALymqHX=6PKBd5qU-Z+zQ@mail.gmail.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <CA+-tSzx0E3J4i5Dvfbj4egruCeGi7z1se1Jn5RXtx3RedQLPdg@mail.gmail.com> <CAHD03N89RNX4=XxRZpD+Y2PCktYf-FtVjyYJ5oFS3dui3SsXaA@mail.gmail.com> <CAF4+nEG7YKGzGmObAMaOxackivjJsALymqHX=6PKBd5qU-Z+zQ@mail.gmail.com>
Date: Tue, 29 Jan 2013 13:15:31 -0800
Message-ID: <CAHD03N8HMnMY9p2dm60qCnTfNXL-rawf0hkY0a=cKF_B+jknXA@mail.gmail.com>
From: Ayan Banerjee <ayabaner@gmail.com>
To: Donald Eastlake <d3e3e3@gmail.com>
Content-Type: multipart/alternative; boundary=20cf307f3aa0c89db704d473e408
Cc: Anoop Ghanwani <anoop@alumni.duke.edu>, Radia Perlman <radiaperlman@gmail.com>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 29 Jan 2013 21:15:34 -0000

--20cf307f3aa0c89db704d473e408
Content-Type: text/plain; charset=ISO-8859-1

Donald,

Yes, realize that would be way to do it - not sure if these options are
being captured in a document, if so it would be great if this also added
there.

Thanks,
Ayan



On Tue, Jan 29, 2013 at 12:49 PM, Donald Eastlake <d3e3e3@gmail.com> wrote:

> Hi Ayan,
>
> On Tue, Jan 29, 2013 at 2:45 PM, Ayan Banerjee <ayabaner@gmail.com> wrote:
> > I agree with Anoop here.
> >
> > I think that we are describing a network that is FGL-safe and ready and
> then
> > switch to FGL-mode. However, we need some statements to account for a
> > VL-node coming-in/leaving-the network due to a flaky link.
>
> The general assumption is that you are migrating from VL to FGL. First
> you migrate to FGL-safe. Then you unleash the FGL traffic, by
> configuring FGL ports, which isolates any remaining (FGL-unsafe) VL
> switches while still being able to handle VL traffic between VL ports
> on FGL-safe switches.
>
> After you have unleashed FGL traffic, it would certainly be possible
> that a VL switch could be attached to the network or come and go due
> to a flakey link, but this would have almost no effect. With FGL
> traffic being handled, VL switches are isolated.
>
> Thanks,
> Donald
> =============================
>  Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
>  155 Beaver Street, Milford, MA 01757 USA
>  d3e3e3@gmail.com
>
> > Thanks,
> > Ayan
> >
> >
> >
> > On Tue, Jan 29, 2013 at 6:42 AM, Anoop Ghanwani <anoop@alumni.duke.edu>
> > wrote:
> >>
> >> At noted, option 1 looks dangerous so that should be ruled out.
> >>
> >> Option 2 seems reasonable.  It might be useful to clarify a couple of
> >> things:
> >> - Can the FGL-safe campus have both VLAN & FGL frames going around?
> >> - Do all RBridges need to agree on when the campus is FGL-safe?  I would
> >>   hope not.  But the problem I'm thinking about is where RB1 thinks
> >> the campus is
> >>   FGL-safe and starts announcing attachment to FGL links, but RB2 hadn't
> >> yet
> >>   concluded that process and it see RB3 which happens to be VL.
> >>
> >> Option 3 seems to introduce more complexity to solve this problem than
> >> is necessary.
> >>
> >> Anoop
> >>
> >> On Fri, Jan 25, 2013 at 11:00 AM, Radia Perlman <radiaperlman@gmail.com
> >
> >> wrote:
> >> > I'm going to summarize three options for phasing-in FGL, and explain
> the
> >> > tradeoffs, to help people have informed opinions about the
> implications
> >> > of
> >> > the three approaches.  There are infinite variations, of course, but
> as
> >> > I
> >> > said, I'll describe three main ones.
> >> >
> >> > Option 1:  Original draft, no changes.  There are two types of
> RBridges.
> >> > The
> >> > first type is "VL" which only understands VLAN tags, and have unknown
> >> > and
> >> > possibly dangerous behavior when they receive an FGL-labeled packet.
> >> > (e.g.,
> >> > decapsulate a packet onto a link where it is not allowed, or
> mistakenly
> >> > prune a multicast so it does not reach everywhere it should). The
> second
> >> > type is "FGL".  FGL guys need to do two things: 1) understand FGL tags
> >> > and
> >> > do the right thing with them, and 2) ostrasize VL guys...meaning that
> an
> >> > FGL
> >> > guy refuses to form an adjacency with a VL guy.  This option is very
> >> > simple
> >> > for implementers of FGL, but the implication is that you have to
> upgrade
> >> > your entire campus to FGL at once.  There is no coexistence.
> >> >
> >> > Option 2: Two types of RBridges, but the 2nd type is different from
> >> > option
> >> > 1.  VL guys of course are the same as in option 1...they cannot be
> >> > trusted
> >> > with FGL frames. The second type I will call "FGL-safe". An FGL-safe
> >> > RBridge
> >> > must advertise in its LSP that it is FGL-safe, and it must not "do
> >> > anything
> >> > bad" with FGL frames, meaning that it is allowed to ignore pruning of
> >> > FGL
> >> > (or even VL frames) entirely...it just can't falsely drop FGL frames.
> >> > And
> >> > it must not decapsulate an FGL frame onto a link for which that FGL
> >> > frame
> >> > doesn't belong. In this option, all RBridges must be upgraded to
> >> > FGL-safe,
> >> > but it need not happen instantaneously...it's fine to mix FGL-safe
> >> > RBridges
> >> > with VL RBridges...it's just not safe to inject FGL frames yet.  Once
> >> > all
> >> > the RBridges have been upgraded to be FGL-safe, then edge RBridges can
> >> > start
> >> > announcing they are connected to an FGL link, and can start injecting
> >> > FGL
> >> > packets.  It is considered a misconfiguration if you start injecting
> FGL
> >> > frames before all the RBridges are upgraded to FGL-safe, so an
> >> > additional
> >> > chore for an FGL-safe RBridge R1 is to examine LSPs, and if any
> RBridge
> >> > claims to be attached to an FGL link, then R1 must ostrasize any VL
> >> > neighbors.  (don't start ostrasizing VL guys until it is necessary
> >> > because
> >> > of actually starting to use FGL frames, in other words). This option
> is
> >> > more
> >> > work for the upgraded RBridges than option 1, and there still isn't
> good
> >> > coexistence with VL guys long-term (as option 3 will), but it does
> allow
> >> > upgrading RBridges one by one in a working campus without causing
> >> > disruption.
> >> >
> >> > Option 3: Two types of RBridges.  VL, of course, is the same as in
> >> > options 1
> >> > and 2.  This option makes FGL guys do more work, but allows maximal
> long
> >> > term coexistence of VL and FGL guys. In this option, FGL guys
> calculate
> >> > unicast paths to FGL edge guys that avoid any VL guys, and calculate
> at
> >> > least one FGL-friendly multicast tree that also avoids any VL guys. So
> >> > let's
> >> > say R1 is an FGL guy.  R1 discards all LSPs from VL guys (ones that
> >> > don't
> >> > advertise FGL capability in their LSP), when calculating paths to
> other
> >> > FGL
> >> > RBridges. Then R1 calculates paths to the VL guys using all LSPs.
> >> > Likewise,
> >> > when calculating an FGL-friendly tree, R1 calculates a tree through
> only
> >> > FGL
> >> > guys.   This option is more work for the upgraded RBridges (because
> they
> >> > have to calculate Dijkstra in two different ways, one for reaching VL
> >> > guys,
> >> > and one for reaching FGL guys).  However, it does allow having
> long-term
> >> > coexistence with VL guys.  For instance, you could forever keep some
> VL
> >> > edge
> >> > RBridges that communicate just fine through an FGL core.  They can
> stay
> >> > there forever, and still be able to communicate through the core to
> all
> >> > the
> >> > links attaching to their VLAN.
> >> >
> >> > _______________________________________________
> >> > trill mailing list
> >> > trill@ietf.org
> >> > https://www.ietf.org/mailman/listinfo/trill
> >> >
> >> _______________________________________________
> >> trill mailing list
> >> trill@ietf.org
> >> https://www.ietf.org/mailman/listinfo/trill
> >
> >
> >
> > _______________________________________________
> > trill mailing list
> > trill@ietf.org
> > https://www.ietf.org/mailman/listinfo/trill
> >
>

--20cf307f3aa0c89db704d473e408
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>Donald,<br><br></div>Yes, realize that would be way t=
o do it - not sure if these options are being captured in a document, if so=
 it would be great if this also added there. <br><br>Thanks,<br>Ayan<br><br=
>
</div><div class=3D"gmail_extra"><br><br><div class=3D"gmail_quote">On Tue,=
 Jan 29, 2013 at 12:49 PM, Donald Eastlake <span dir=3D"ltr">&lt;<a href=3D=
"mailto:d3e3e3@gmail.com" target=3D"_blank">d3e3e3@gmail.com</a>&gt;</span>=
 wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">Hi Ayan,<br>
<br>
On Tue, Jan 29, 2013 at 2:45 PM, Ayan Banerjee &lt;<a href=3D"mailto:ayaban=
er@gmail.com">ayabaner@gmail.com</a>&gt; wrote:<br>
&gt; I agree with Anoop here.<br>
&gt;<br>
&gt; I think that we are describing a network that is FGL-safe and ready an=
d then<br>
&gt; switch to FGL-mode. However, we need some statements to account for a<=
br>
&gt; VL-node coming-in/leaving-the network due to a flaky link.<br>
<br>
The general assumption is that you are migrating from VL to FGL. First<br>
you migrate to FGL-safe. Then you unleash the FGL traffic, by<br>
configuring FGL ports, which isolates any remaining (FGL-unsafe) VL<br>
switches while still being able to handle VL traffic between VL ports<br>
on FGL-safe switches.<br>
<br>
After you have unleashed FGL traffic, it would certainly be possible<br>
that a VL switch could be attached to the network or come and go due<br>
to a flakey link, but this would have almost no effect. With FGL<br>
traffic being handled, VL switches are isolated.<br>
<br>
Thanks,<br>
Donald<br>
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D<br>
=A0Donald E. Eastlake 3rd =A0 <a href=3D"tel:%2B1-508-333-2270" value=3D"+1=
5083332270">+1-508-333-2270</a> (cell)<br>
=A0155 Beaver Street, Milford, MA 01757 USA<br>
=A0<a href=3D"mailto:d3e3e3@gmail.com">d3e3e3@gmail.com</a><br>
<br>
&gt; Thanks,<br>
&gt; Ayan<br>
&gt;<br>
&gt;<br>
&gt;<br>
&gt; On Tue, Jan 29, 2013 at 6:42 AM, Anoop Ghanwani &lt;<a href=3D"mailto:=
anoop@alumni.duke.edu">anoop@alumni.duke.edu</a>&gt;<br>
&gt; wrote:<br>
&gt;&gt;<br>
&gt;&gt; At noted, option 1 looks dangerous so that should be ruled out.<br=
>
&gt;&gt;<br>
&gt;&gt; Option 2 seems reasonable. =A0It might be useful to clarify a coup=
le of<br>
&gt;&gt; things:<br>
&gt;&gt; - Can the FGL-safe campus have both VLAN &amp; FGL frames going ar=
ound?<br>
&gt;&gt; - Do all RBridges need to agree on when the campus is FGL-safe? =
=A0I would<br>
&gt;&gt; =A0 hope not. =A0But the problem I&#39;m thinking about is where R=
B1 thinks<br>
&gt;&gt; the campus is<br>
&gt;&gt; =A0 FGL-safe and starts announcing attachment to FGL links, but RB=
2 hadn&#39;t<br>
&gt;&gt; yet<br>
&gt;&gt; =A0 concluded that process and it see RB3 which happens to be VL.<=
br>
&gt;&gt;<br>
&gt;&gt; Option 3 seems to introduce more complexity to solve this problem =
than<br>
&gt;&gt; is necessary.<br>
&gt;&gt;<br>
&gt;&gt; Anoop<br>
&gt;&gt;<br>
&gt;&gt; On Fri, Jan 25, 2013 at 11:00 AM, Radia Perlman &lt;<a href=3D"mai=
lto:radiaperlman@gmail.com">radiaperlman@gmail.com</a>&gt;<br>
&gt;&gt; wrote:<br>
&gt;&gt; &gt; I&#39;m going to summarize three options for phasing-in FGL, =
and explain the<br>
&gt;&gt; &gt; tradeoffs, to help people have informed opinions about the im=
plications<br>
&gt;&gt; &gt; of<br>
&gt;&gt; &gt; the three approaches. =A0There are infinite variations, of co=
urse, but as<br>
&gt;&gt; &gt; I<br>
&gt;&gt; &gt; said, I&#39;ll describe three main ones.<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt; Option 1: =A0Original draft, no changes. =A0There are two typ=
es of RBridges.<br>
&gt;&gt; &gt; The<br>
&gt;&gt; &gt; first type is &quot;VL&quot; which only understands VLAN tags=
, and have unknown<br>
&gt;&gt; &gt; and<br>
&gt;&gt; &gt; possibly dangerous behavior when they receive an FGL-labeled =
packet.<br>
&gt;&gt; &gt; (e.g.,<br>
&gt;&gt; &gt; decapsulate a packet onto a link where it is not allowed, or =
mistakenly<br>
&gt;&gt; &gt; prune a multicast so it does not reach everywhere it should).=
 The second<br>
&gt;&gt; &gt; type is &quot;FGL&quot;. =A0FGL guys need to do two things: 1=
) understand FGL tags<br>
&gt;&gt; &gt; and<br>
&gt;&gt; &gt; do the right thing with them, and 2) ostrasize VL guys...mean=
ing that an<br>
&gt;&gt; &gt; FGL<br>
&gt;&gt; &gt; guy refuses to form an adjacency with a VL guy. =A0This optio=
n is very<br>
&gt;&gt; &gt; simple<br>
&gt;&gt; &gt; for implementers of FGL, but the implication is that you have=
 to upgrade<br>
&gt;&gt; &gt; your entire campus to FGL at once. =A0There is no coexistence=
.<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt; Option 2: Two types of RBridges, but the 2nd type is differen=
t from<br>
&gt;&gt; &gt; option<br>
&gt;&gt; &gt; 1. =A0VL guys of course are the same as in option 1...they ca=
nnot be<br>
&gt;&gt; &gt; trusted<br>
&gt;&gt; &gt; with FGL frames. The second type I will call &quot;FGL-safe&q=
uot;. An FGL-safe<br>
&gt;&gt; &gt; RBridge<br>
&gt;&gt; &gt; must advertise in its LSP that it is FGL-safe, and it must no=
t &quot;do<br>
&gt;&gt; &gt; anything<br>
&gt;&gt; &gt; bad&quot; with FGL frames, meaning that it is allowed to igno=
re pruning of<br>
&gt;&gt; &gt; FGL<br>
&gt;&gt; &gt; (or even VL frames) entirely...it just can&#39;t falsely drop=
 FGL frames.<br>
&gt;&gt; &gt; And<br>
&gt;&gt; &gt; it must not decapsulate an FGL frame onto a link for which th=
at FGL<br>
&gt;&gt; &gt; frame<br>
&gt;&gt; &gt; doesn&#39;t belong. In this option, all RBridges must be upgr=
aded to<br>
&gt;&gt; &gt; FGL-safe,<br>
&gt;&gt; &gt; but it need not happen instantaneously...it&#39;s fine to mix=
 FGL-safe<br>
&gt;&gt; &gt; RBridges<br>
&gt;&gt; &gt; with VL RBridges...it&#39;s just not safe to inject FGL frame=
s yet. =A0Once<br>
&gt;&gt; &gt; all<br>
&gt;&gt; &gt; the RBridges have been upgraded to be FGL-safe, then edge RBr=
idges can<br>
&gt;&gt; &gt; start<br>
&gt;&gt; &gt; announcing they are connected to an FGL link, and can start i=
njecting<br>
&gt;&gt; &gt; FGL<br>
&gt;&gt; &gt; packets. =A0It is considered a misconfiguration if you start =
injecting FGL<br>
&gt;&gt; &gt; frames before all the RBridges are upgraded to FGL-safe, so a=
n<br>
&gt;&gt; &gt; additional<br>
&gt;&gt; &gt; chore for an FGL-safe RBridge R1 is to examine LSPs, and if a=
ny RBridge<br>
&gt;&gt; &gt; claims to be attached to an FGL link, then R1 must ostrasize =
any VL<br>
&gt;&gt; &gt; neighbors. =A0(don&#39;t start ostrasizing VL guys until it i=
s necessary<br>
&gt;&gt; &gt; because<br>
&gt;&gt; &gt; of actually starting to use FGL frames, in other words). This=
 option is<br>
&gt;&gt; &gt; more<br>
&gt;&gt; &gt; work for the upgraded RBridges than option 1, and there still=
 isn&#39;t good<br>
&gt;&gt; &gt; coexistence with VL guys long-term (as option 3 will), but it=
 does allow<br>
&gt;&gt; &gt; upgrading RBridges one by one in a working campus without cau=
sing<br>
&gt;&gt; &gt; disruption.<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt; Option 3: Two types of RBridges. =A0VL, of course, is the sam=
e as in<br>
&gt;&gt; &gt; options 1<br>
&gt;&gt; &gt; and 2. =A0This option makes FGL guys do more work, but allows=
 maximal long<br>
&gt;&gt; &gt; term coexistence of VL and FGL guys. In this option, FGL guys=
 calculate<br>
&gt;&gt; &gt; unicast paths to FGL edge guys that avoid any VL guys, and ca=
lculate at<br>
&gt;&gt; &gt; least one FGL-friendly multicast tree that also avoids any VL=
 guys. So<br>
&gt;&gt; &gt; let&#39;s<br>
&gt;&gt; &gt; say R1 is an FGL guy. =A0R1 discards all LSPs from VL guys (o=
nes that<br>
&gt;&gt; &gt; don&#39;t<br>
&gt;&gt; &gt; advertise FGL capability in their LSP), when calculating path=
s to other<br>
&gt;&gt; &gt; FGL<br>
&gt;&gt; &gt; RBridges. Then R1 calculates paths to the VL guys using all L=
SPs.<br>
&gt;&gt; &gt; Likewise,<br>
&gt;&gt; &gt; when calculating an FGL-friendly tree, R1 calculates a tree t=
hrough only<br>
&gt;&gt; &gt; FGL<br>
&gt;&gt; &gt; guys. =A0 This option is more work for the upgraded RBridges =
(because they<br>
&gt;&gt; &gt; have to calculate Dijkstra in two different ways, one for rea=
ching VL<br>
&gt;&gt; &gt; guys,<br>
&gt;&gt; &gt; and one for reaching FGL guys). =A0However, it does allow hav=
ing long-term<br>
&gt;&gt; &gt; coexistence with VL guys. =A0For instance, you could forever =
keep some VL<br>
&gt;&gt; &gt; edge<br>
&gt;&gt; &gt; RBridges that communicate just fine through an FGL core. =A0T=
hey can stay<br>
&gt;&gt; &gt; there forever, and still be able to communicate through the c=
ore to all<br>
&gt;&gt; &gt; the<br>
&gt;&gt; &gt; links attaching to their VLAN.<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt; _______________________________________________<br>
&gt;&gt; &gt; trill mailing list<br>
&gt;&gt; &gt; <a href=3D"mailto:trill@ietf.org">trill@ietf.org</a><br>
&gt;&gt; &gt; <a href=3D"https://www.ietf.org/mailman/listinfo/trill" targe=
t=3D"_blank">https://www.ietf.org/mailman/listinfo/trill</a><br>
&gt;&gt; &gt;<br>
&gt;&gt; _______________________________________________<br>
&gt;&gt; trill mailing list<br>
&gt;&gt; <a href=3D"mailto:trill@ietf.org">trill@ietf.org</a><br>
&gt;&gt; <a href=3D"https://www.ietf.org/mailman/listinfo/trill" target=3D"=
_blank">https://www.ietf.org/mailman/listinfo/trill</a><br>
&gt;<br>
&gt;<br>
&gt;<br>
&gt; _______________________________________________<br>
&gt; trill mailing list<br>
&gt; <a href=3D"mailto:trill@ietf.org">trill@ietf.org</a><br>
&gt; <a href=3D"https://www.ietf.org/mailman/listinfo/trill" target=3D"_bla=
nk">https://www.ietf.org/mailman/listinfo/trill</a><br>
&gt;<br>
</blockquote></div><br></div>

--20cf307f3aa0c89db704d473e408--

From ghanwani@gmail.com  Tue Jan 29 16:03:19 2013
Return-Path: <ghanwani@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E957A21F886A for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 16:03:19 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.978
X-Spam-Level: 
X-Spam-Status: No, score=-1.978 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, NO_RELAYS=-0.001]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id AIuHcfdOFcpB for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 16:03:19 -0800 (PST)
Received: from mail-ia0-x229.google.com (ia-in-x0229.1e100.net [IPv6:2607:f8b0:4001:c02::229]) by ietfa.amsl.com (Postfix) with ESMTP id 8053321F8863 for <trill@ietf.org>; Tue, 29 Jan 2013 16:03:19 -0800 (PST)
Received: by mail-ia0-f169.google.com with SMTP id j5so1480331iaf.14 for <trill@ietf.org>; Tue, 29 Jan 2013 16:03:18 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:x-received:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type; bh=iLHRQkIf9b3yMKcMak8AbiphZle9mdEQC1G5bzNUxI0=; b=Fr7U4xmvJC/BkhuDypRe1UVOx32r0Wa8GX6YqZBHFJCOX652vgmi1PM/aBCYMCc3sG d7FGvaRnt9SPC5SaMj/tcpAU3jnp14KRBEvwmU141UgBLE6JGN4ee7tbx1W6FFgLrKOp CtmFVotDlX+kr+WXhL1lDsHzIlBSudvmA2nT+sxmvRfYW/lyi9Y0LPbQrsP8xR+umiC+ /1EAHtnNvW1aR333gLz1WGom7JJ30Po5n+ciUjily6vc/xoQ5E1BMvO1Jds6hvpDVwt8 2CHxkOJVZZarOym8RzyduRZSTuZMuhiA4w9bhqOF62CS6ujBIpLBBewZwNkuEFfhdAbK tRKQ==
MIME-Version: 1.0
X-Received: by 10.50.10.193 with SMTP id k1mr2259264igb.102.1359504198784; Tue, 29 Jan 2013 16:03:18 -0800 (PST)
Sender: ghanwani@gmail.com
Received: by 10.50.182.201 with HTTP; Tue, 29 Jan 2013 16:03:18 -0800 (PST)
In-Reply-To: <CAF4+nEG7YKGzGmObAMaOxackivjJsALymqHX=6PKBd5qU-Z+zQ@mail.gmail.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <CA+-tSzx0E3J4i5Dvfbj4egruCeGi7z1se1Jn5RXtx3RedQLPdg@mail.gmail.com> <CAHD03N89RNX4=XxRZpD+Y2PCktYf-FtVjyYJ5oFS3dui3SsXaA@mail.gmail.com> <CAF4+nEG7YKGzGmObAMaOxackivjJsALymqHX=6PKBd5qU-Z+zQ@mail.gmail.com>
Date: Tue, 29 Jan 2013 16:03:18 -0800
X-Google-Sender-Auth: JMUdA-6E9TNUzHyCmjndXMwBy3Q
Message-ID: <CA+-tSzz0=oh6iB1e-ESyeOx2sCvBFXeVpzuNbKmUVhRNLScDgQ@mail.gmail.com>
From: Anoop Ghanwani <anoop@alumni.duke.edu>
To: Donald Eastlake <d3e3e3@gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Cc: Radia Perlman <radiaperlman@gmail.com>, Ayan Banerjee <ayabaner@gmail.com>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 30 Jan 2013 00:03:20 -0000

On Tue, Jan 29, 2013 at 12:49 PM, Donald Eastlake <d3e3e3@gmail.com> wrote:

> The general assumption is that you are migrating from VL to FGL. First
> you migrate to FGL-safe. Then you unleash the FGL traffic, by
> configuring FGL ports, which isolates any remaining (FGL-unsafe) VL
> switches while still being able to handle VL traffic between VL ports
> on FGL-safe switches.

Based on what is written so far, it looks like as long as there is even
one VL RBridge in it's database, an RBridge would not be able to start
sourcing FGL traffic.  It may start doing so only when all RBridges in
its database are FGL-safe.  If we ever get into a condition where
an RBridge has started advertising FGL information, at that point,
does any other RBridge in the campus simply isolate any VL RBridge
(with the assumption that there was some kind of race condition?

The case I'm concerned about is we have 2 FGL RBridges, RB1 and RB2.
RB1 is connected to a campus of n RBs that are still VL only.  The VL link
Breaks, and RB1 starts advertising FGL information to RB2, but then
the link to the VL campus comes back up.  Does this mean RB1 and RB2
remain isolated from the campus (and operate as they own mini-campus)
until all the other RBridges are converted over to FGL-safe?

Anoop

From d3e3e3@gmail.com  Tue Jan 29 16:14:50 2013
Return-Path: <d3e3e3@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8E67921F870A for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 16:14:50 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.078
X-Spam-Level: 
X-Spam-Status: No, score=-103.078 tagged_above=-999 required=5 tests=[AWL=0.521, BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id x7xtlPZNb902 for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 16:14:49 -0800 (PST)
Received: from mail-ob0-f182.google.com (mail-ob0-f182.google.com [209.85.214.182]) by ietfa.amsl.com (Postfix) with ESMTP id 9835021F86F5 for <trill@ietf.org>; Tue, 29 Jan 2013 16:14:49 -0800 (PST)
Received: by mail-ob0-f182.google.com with SMTP id va7so1079996obc.13 for <trill@ietf.org>; Tue, 29 Jan 2013 16:14:49 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=x-received:mime-version:in-reply-to:references:from:date:message-id :subject:to:cc:content-type; bh=7FuHnn+yVDAXMMZyxlsJ/qhsBntRoSYgVy9CmeitYlI=; b=IP0eOXa6mEpyP1VlbIFTvN5GaiSmZfSWOAYWc9X9tSYyJg9sL5sLHQi/knKUq/+S9x zd9whZh2ltcHmfaPwDW/gt00tytbkXu2mmIsJKx9TGngNM4vh1nH+RAE6GeHSVpVgFr3 htkrwHbjAGO8FMo52k0+46K8qnsNk7CFyblqg4fpBsrflfqFAT7xGPyCL0L4WY4PO0D9 SCcsGO/awtJ0xFGJ7wX7TZbKsdPfqO6iyCOl7I873HOahoMoMl3+K4axLMa85CkLb9i1 /+Kld3YDnktOv93C0TTzRkvHviWDMNwJGVag/tATae3iPjgvBt3eOY9PflrjzQRCiWr8 iHRA==
X-Received: by 10.60.30.38 with SMTP id p6mr2320722oeh.2.1359504889203; Tue, 29 Jan 2013 16:14:49 -0800 (PST)
MIME-Version: 1.0
Received: by 10.76.98.168 with HTTP; Tue, 29 Jan 2013 16:14:28 -0800 (PST)
In-Reply-To: <CAHD03N8HMnMY9p2dm60qCnTfNXL-rawf0hkY0a=cKF_B+jknXA@mail.gmail.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <CA+-tSzx0E3J4i5Dvfbj4egruCeGi7z1se1Jn5RXtx3RedQLPdg@mail.gmail.com> <CAHD03N89RNX4=XxRZpD+Y2PCktYf-FtVjyYJ5oFS3dui3SsXaA@mail.gmail.com> <CAF4+nEG7YKGzGmObAMaOxackivjJsALymqHX=6PKBd5qU-Z+zQ@mail.gmail.com> <CAHD03N8HMnMY9p2dm60qCnTfNXL-rawf0hkY0a=cKF_B+jknXA@mail.gmail.com>
From: Donald Eastlake <d3e3e3@gmail.com>
Date: Tue, 29 Jan 2013 19:14:28 -0500
Message-ID: <CAF4+nEEL-FjzKxbxg93t6QTTJy_5uVtok1_RAS1=pR_0B_p1kQ@mail.gmail.com>
To: Ayan Banerjee <ayabaner@gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Cc: Anoop Ghanwani <anoop@alumni.duke.edu>, Radia Perlman <radiaperlman@gmail.com>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 30 Jan 2013 00:14:50 -0000

Ayan,

Yes, it would probably be good to have a brief informational section
describing the intended VL to FGL migration scenario.

Thanks,
Donald
=============================
 Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
 155 Beaver Street, Milford, MA 01757 USA
 d3e3e3@gmail.com


On Tue, Jan 29, 2013 at 4:15 PM, Ayan Banerjee <ayabaner@gmail.com> wrote:
> Donald,
>
> Yes, realize that would be way to do it - not sure if these options are
> being captured in a document, if so it would be great if this also added
> there.
>
> Thanks,
> Ayan
>
>
>
> On Tue, Jan 29, 2013 at 12:49 PM, Donald Eastlake <d3e3e3@gmail.com> wrote:
>>
>> Hi Ayan,
>>
>> On Tue, Jan 29, 2013 at 2:45 PM, Ayan Banerjee <ayabaner@gmail.com> wrote:
>> > I agree with Anoop here.
>> >
>> > I think that we are describing a network that is FGL-safe and ready and
>> > then
>> > switch to FGL-mode. However, we need some statements to account for a
>> > VL-node coming-in/leaving-the network due to a flaky link.
>>
>> The general assumption is that you are migrating from VL to FGL. First
>> you migrate to FGL-safe. Then you unleash the FGL traffic, by
>> configuring FGL ports, which isolates any remaining (FGL-unsafe) VL
>> switches while still being able to handle VL traffic between VL ports
>> on FGL-safe switches.
>>
>> After you have unleashed FGL traffic, it would certainly be possible
>> that a VL switch could be attached to the network or come and go due
>> to a flakey link, but this would have almost no effect. With FGL
>> traffic being handled, VL switches are isolated.
>>
>> Thanks,
>> Donald
>> =============================
>>  Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
>>  155 Beaver Street, Milford, MA 01757 USA
>>  d3e3e3@gmail.com
>>
>> > Thanks,
>> > Ayan
>> >
>> >
>> >
>> > On Tue, Jan 29, 2013 at 6:42 AM, Anoop Ghanwani <anoop@alumni.duke.edu>
>> > wrote:
>> >>
>> >> At noted, option 1 looks dangerous so that should be ruled out.
>> >>
>> >> Option 2 seems reasonable.  It might be useful to clarify a couple of
>> >> things:
>> >> - Can the FGL-safe campus have both VLAN & FGL frames going around?
>> >> - Do all RBridges need to agree on when the campus is FGL-safe?  I
>> >> would
>> >>   hope not.  But the problem I'm thinking about is where RB1 thinks
>> >> the campus is
>> >>   FGL-safe and starts announcing attachment to FGL links, but RB2
>> >> hadn't
>> >> yet
>> >>   concluded that process and it see RB3 which happens to be VL.
>> >>
>> >> Option 3 seems to introduce more complexity to solve this problem than
>> >> is necessary.
>> >>
>> >> Anoop
>> >>
>> >> On Fri, Jan 25, 2013 at 11:00 AM, Radia Perlman
>> >> <radiaperlman@gmail.com>
>> >> wrote:
>> >> > I'm going to summarize three options for phasing-in FGL, and explain
>> >> > the
>> >> > tradeoffs, to help people have informed opinions about the
>> >> > implications
>> >> > of
>> >> > the three approaches.  There are infinite variations, of course, but
>> >> > as
>> >> > I
>> >> > said, I'll describe three main ones.
>> >> >
>> >> > Option 1:  Original draft, no changes.  There are two types of
>> >> > RBridges.
>> >> > The
>> >> > first type is "VL" which only understands VLAN tags, and have unknown
>> >> > and
>> >> > possibly dangerous behavior when they receive an FGL-labeled packet.
>> >> > (e.g.,
>> >> > decapsulate a packet onto a link where it is not allowed, or
>> >> > mistakenly
>> >> > prune a multicast so it does not reach everywhere it should). The
>> >> > second
>> >> > type is "FGL".  FGL guys need to do two things: 1) understand FGL
>> >> > tags
>> >> > and
>> >> > do the right thing with them, and 2) ostrasize VL guys...meaning that
>> >> > an
>> >> > FGL
>> >> > guy refuses to form an adjacency with a VL guy.  This option is very
>> >> > simple
>> >> > for implementers of FGL, but the implication is that you have to
>> >> > upgrade
>> >> > your entire campus to FGL at once.  There is no coexistence.
>> >> >
>> >> > Option 2: Two types of RBridges, but the 2nd type is different from
>> >> > option
>> >> > 1.  VL guys of course are the same as in option 1...they cannot be
>> >> > trusted
>> >> > with FGL frames. The second type I will call "FGL-safe". An FGL-safe
>> >> > RBridge
>> >> > must advertise in its LSP that it is FGL-safe, and it must not "do
>> >> > anything
>> >> > bad" with FGL frames, meaning that it is allowed to ignore pruning of
>> >> > FGL
>> >> > (or even VL frames) entirely...it just can't falsely drop FGL frames.
>> >> > And
>> >> > it must not decapsulate an FGL frame onto a link for which that FGL
>> >> > frame
>> >> > doesn't belong. In this option, all RBridges must be upgraded to
>> >> > FGL-safe,
>> >> > but it need not happen instantaneously...it's fine to mix FGL-safe
>> >> > RBridges
>> >> > with VL RBridges...it's just not safe to inject FGL frames yet.  Once
>> >> > all
>> >> > the RBridges have been upgraded to be FGL-safe, then edge RBridges
>> >> > can
>> >> > start
>> >> > announcing they are connected to an FGL link, and can start injecting
>> >> > FGL
>> >> > packets.  It is considered a misconfiguration if you start injecting
>> >> > FGL
>> >> > frames before all the RBridges are upgraded to FGL-safe, so an
>> >> > additional
>> >> > chore for an FGL-safe RBridge R1 is to examine LSPs, and if any
>> >> > RBridge
>> >> > claims to be attached to an FGL link, then R1 must ostrasize any VL
>> >> > neighbors.  (don't start ostrasizing VL guys until it is necessary
>> >> > because
>> >> > of actually starting to use FGL frames, in other words). This option
>> >> > is
>> >> > more
>> >> > work for the upgraded RBridges than option 1, and there still isn't
>> >> > good
>> >> > coexistence with VL guys long-term (as option 3 will), but it does
>> >> > allow
>> >> > upgrading RBridges one by one in a working campus without causing
>> >> > disruption.
>> >> >
>> >> > Option 3: Two types of RBridges.  VL, of course, is the same as in
>> >> > options 1
>> >> > and 2.  This option makes FGL guys do more work, but allows maximal
>> >> > long
>> >> > term coexistence of VL and FGL guys. In this option, FGL guys
>> >> > calculate
>> >> > unicast paths to FGL edge guys that avoid any VL guys, and calculate
>> >> > at
>> >> > least one FGL-friendly multicast tree that also avoids any VL guys.
>> >> > So
>> >> > let's
>> >> > say R1 is an FGL guy.  R1 discards all LSPs from VL guys (ones that
>> >> > don't
>> >> > advertise FGL capability in their LSP), when calculating paths to
>> >> > other
>> >> > FGL
>> >> > RBridges. Then R1 calculates paths to the VL guys using all LSPs.
>> >> > Likewise,
>> >> > when calculating an FGL-friendly tree, R1 calculates a tree through
>> >> > only
>> >> > FGL
>> >> > guys.   This option is more work for the upgraded RBridges (because
>> >> > they
>> >> > have to calculate Dijkstra in two different ways, one for reaching VL
>> >> > guys,
>> >> > and one for reaching FGL guys).  However, it does allow having
>> >> > long-term
>> >> > coexistence with VL guys.  For instance, you could forever keep some
>> >> > VL
>> >> > edge
>> >> > RBridges that communicate just fine through an FGL core.  They can
>> >> > stay
>> >> > there forever, and still be able to communicate through the core to
>> >> > all
>> >> > the
>> >> > links attaching to their VLAN.
>> >> >
>> >> > _______________________________________________
>> >> > trill mailing list
>> >> > trill@ietf.org
>> >> > https://www.ietf.org/mailman/listinfo/trill
>> >> >
>> >> _______________________________________________
>> >> trill mailing list
>> >> trill@ietf.org
>> >> https://www.ietf.org/mailman/listinfo/trill
>> >
>> >
>> >
>> > _______________________________________________
>> > trill mailing list
>> > trill@ietf.org
>> > https://www.ietf.org/mailman/listinfo/trill
>> >
>
>

From ayabaner@gmail.com  Tue Jan 29 16:22:52 2013
Return-Path: <ayabaner@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8061B21F8626 for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 16:22:52 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.598
X-Spam-Level: 
X-Spam-Status: No, score=-3.598 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id PST-TuHsXZ2i for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 16:22:50 -0800 (PST)
Received: from mail-vc0-f171.google.com (mail-vc0-f171.google.com [209.85.220.171]) by ietfa.amsl.com (Postfix) with ESMTP id 72AF721F8718 for <trill@ietf.org>; Tue, 29 Jan 2013 16:22:50 -0800 (PST)
Received: by mail-vc0-f171.google.com with SMTP id p1so689992vcq.30 for <trill@ietf.org>; Tue, 29 Jan 2013 16:22:43 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:x-received:in-reply-to:references:date:message-id :subject:from:to:cc:content-type; bh=RM4X0ZU/FnBWTgXHd/YECTq06oQxGwIbsC1JJ5sFhr0=; b=PyJgGJyIzy8eTPz5t+NvNIx6s3LTs6qEo6+rFO4D2QyfhY3AziljGedNiXZTBvJ+iU cpNfEKsK+DOvFIAjitx1KnJosabGy107wQULZknCy1NOB9LZI5hfRs5N2wB3lRfnCS0U cDf9cpXiWduleEWQUG5MWoRiMcN9XmSspKyn4+wPumDhkIjuEy/eAOeppB5HECDjr7cX V/IHq7E+q5DkY+T9lnFuiDQmbV7xZ8Wr5H5PB4TvowDl119sI00K9G1s55Usoh7t8BoB LNEK1ppeqLJ4zIlLOas2+CWXaA498ec4GnfmUPACcodRIrcSVTR+MFWMvG8Zjlvtmn2M XSUQ==
MIME-Version: 1.0
X-Received: by 10.52.66.14 with SMTP id b14mr2897234vdt.0.1359505362874; Tue, 29 Jan 2013 16:22:42 -0800 (PST)
Received: by 10.220.144.78 with HTTP; Tue, 29 Jan 2013 16:22:42 -0800 (PST)
In-Reply-To: <CAF4+nEEL-FjzKxbxg93t6QTTJy_5uVtok1_RAS1=pR_0B_p1kQ@mail.gmail.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <CA+-tSzx0E3J4i5Dvfbj4egruCeGi7z1se1Jn5RXtx3RedQLPdg@mail.gmail.com> <CAHD03N89RNX4=XxRZpD+Y2PCktYf-FtVjyYJ5oFS3dui3SsXaA@mail.gmail.com> <CAF4+nEG7YKGzGmObAMaOxackivjJsALymqHX=6PKBd5qU-Z+zQ@mail.gmail.com> <CAHD03N8HMnMY9p2dm60qCnTfNXL-rawf0hkY0a=cKF_B+jknXA@mail.gmail.com> <CAF4+nEEL-FjzKxbxg93t6QTTJy_5uVtok1_RAS1=pR_0B_p1kQ@mail.gmail.com>
Date: Tue, 29 Jan 2013 16:22:42 -0800
Message-ID: <CAHD03N8T5rY4nFT46QX8EQ_KjarbugFNjFcjazk1ZiV4F4AymA@mail.gmail.com>
From: Ayan Banerjee <ayabaner@gmail.com>
To: Donald Eastlake <d3e3e3@gmail.com>
Content-Type: multipart/alternative; boundary=20cf3071c8aa396fdb04d47682b1
Cc: Anoop Ghanwani <anoop@alumni.duke.edu>, Radia Perlman <radiaperlman@gmail.com>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 30 Jan 2013 00:22:52 -0000

--20cf3071c8aa396fdb04d47682b1
Content-Type: text/plain; charset=ISO-8859-1

Thanks much Donald. I think that I would vote for approach 2.
Ayan



On Tue, Jan 29, 2013 at 4:14 PM, Donald Eastlake <d3e3e3@gmail.com> wrote:

> Ayan,
>
> Yes, it would probably be good to have a brief informational section
> describing the intended VL to FGL migration scenario.
>
> Thanks,
> Donald
> =============================
>  Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
>  155 Beaver Street, Milford, MA 01757 USA
>  d3e3e3@gmail.com
>
>
> On Tue, Jan 29, 2013 at 4:15 PM, Ayan Banerjee <ayabaner@gmail.com> wrote:
> > Donald,
> >
> > Yes, realize that would be way to do it - not sure if these options are
> > being captured in a document, if so it would be great if this also added
> > there.
> >
> > Thanks,
> > Ayan
> >
> >
> >
> > On Tue, Jan 29, 2013 at 12:49 PM, Donald Eastlake <d3e3e3@gmail.com>
> wrote:
> >>
> >> Hi Ayan,
> >>
> >> On Tue, Jan 29, 2013 at 2:45 PM, Ayan Banerjee <ayabaner@gmail.com>
> wrote:
> >> > I agree with Anoop here.
> >> >
> >> > I think that we are describing a network that is FGL-safe and ready
> and
> >> > then
> >> > switch to FGL-mode. However, we need some statements to account for a
> >> > VL-node coming-in/leaving-the network due to a flaky link.
> >>
> >> The general assumption is that you are migrating from VL to FGL. First
> >> you migrate to FGL-safe. Then you unleash the FGL traffic, by
> >> configuring FGL ports, which isolates any remaining (FGL-unsafe) VL
> >> switches while still being able to handle VL traffic between VL ports
> >> on FGL-safe switches.
> >>
> >> After you have unleashed FGL traffic, it would certainly be possible
> >> that a VL switch could be attached to the network or come and go due
> >> to a flakey link, but this would have almost no effect. With FGL
> >> traffic being handled, VL switches are isolated.
> >>
> >> Thanks,
> >> Donald
> >> =============================
> >>  Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
> >>  155 Beaver Street, Milford, MA 01757 USA
> >>  d3e3e3@gmail.com
> >>
> >> > Thanks,
> >> > Ayan
> >> >
> >> >
> >> >
> >> > On Tue, Jan 29, 2013 at 6:42 AM, Anoop Ghanwani <
> anoop@alumni.duke.edu>
> >> > wrote:
> >> >>
> >> >> At noted, option 1 looks dangerous so that should be ruled out.
> >> >>
> >> >> Option 2 seems reasonable.  It might be useful to clarify a couple of
> >> >> things:
> >> >> - Can the FGL-safe campus have both VLAN & FGL frames going around?
> >> >> - Do all RBridges need to agree on when the campus is FGL-safe?  I
> >> >> would
> >> >>   hope not.  But the problem I'm thinking about is where RB1 thinks
> >> >> the campus is
> >> >>   FGL-safe and starts announcing attachment to FGL links, but RB2
> >> >> hadn't
> >> >> yet
> >> >>   concluded that process and it see RB3 which happens to be VL.
> >> >>
> >> >> Option 3 seems to introduce more complexity to solve this problem
> than
> >> >> is necessary.
> >> >>
> >> >> Anoop
> >> >>
> >> >> On Fri, Jan 25, 2013 at 11:00 AM, Radia Perlman
> >> >> <radiaperlman@gmail.com>
> >> >> wrote:
> >> >> > I'm going to summarize three options for phasing-in FGL, and
> explain
> >> >> > the
> >> >> > tradeoffs, to help people have informed opinions about the
> >> >> > implications
> >> >> > of
> >> >> > the three approaches.  There are infinite variations, of course,
> but
> >> >> > as
> >> >> > I
> >> >> > said, I'll describe three main ones.
> >> >> >
> >> >> > Option 1:  Original draft, no changes.  There are two types of
> >> >> > RBridges.
> >> >> > The
> >> >> > first type is "VL" which only understands VLAN tags, and have
> unknown
> >> >> > and
> >> >> > possibly dangerous behavior when they receive an FGL-labeled
> packet.
> >> >> > (e.g.,
> >> >> > decapsulate a packet onto a link where it is not allowed, or
> >> >> > mistakenly
> >> >> > prune a multicast so it does not reach everywhere it should). The
> >> >> > second
> >> >> > type is "FGL".  FGL guys need to do two things: 1) understand FGL
> >> >> > tags
> >> >> > and
> >> >> > do the right thing with them, and 2) ostrasize VL guys...meaning
> that
> >> >> > an
> >> >> > FGL
> >> >> > guy refuses to form an adjacency with a VL guy.  This option is
> very
> >> >> > simple
> >> >> > for implementers of FGL, but the implication is that you have to
> >> >> > upgrade
> >> >> > your entire campus to FGL at once.  There is no coexistence.
> >> >> >
> >> >> > Option 2: Two types of RBridges, but the 2nd type is different from
> >> >> > option
> >> >> > 1.  VL guys of course are the same as in option 1...they cannot be
> >> >> > trusted
> >> >> > with FGL frames. The second type I will call "FGL-safe". An
> FGL-safe
> >> >> > RBridge
> >> >> > must advertise in its LSP that it is FGL-safe, and it must not "do
> >> >> > anything
> >> >> > bad" with FGL frames, meaning that it is allowed to ignore pruning
> of
> >> >> > FGL
> >> >> > (or even VL frames) entirely...it just can't falsely drop FGL
> frames.
> >> >> > And
> >> >> > it must not decapsulate an FGL frame onto a link for which that FGL
> >> >> > frame
> >> >> > doesn't belong. In this option, all RBridges must be upgraded to
> >> >> > FGL-safe,
> >> >> > but it need not happen instantaneously...it's fine to mix FGL-safe
> >> >> > RBridges
> >> >> > with VL RBridges...it's just not safe to inject FGL frames yet.
>  Once
> >> >> > all
> >> >> > the RBridges have been upgraded to be FGL-safe, then edge RBridges
> >> >> > can
> >> >> > start
> >> >> > announcing they are connected to an FGL link, and can start
> injecting
> >> >> > FGL
> >> >> > packets.  It is considered a misconfiguration if you start
> injecting
> >> >> > FGL
> >> >> > frames before all the RBridges are upgraded to FGL-safe, so an
> >> >> > additional
> >> >> > chore for an FGL-safe RBridge R1 is to examine LSPs, and if any
> >> >> > RBridge
> >> >> > claims to be attached to an FGL link, then R1 must ostrasize any VL
> >> >> > neighbors.  (don't start ostrasizing VL guys until it is necessary
> >> >> > because
> >> >> > of actually starting to use FGL frames, in other words). This
> option
> >> >> > is
> >> >> > more
> >> >> > work for the upgraded RBridges than option 1, and there still isn't
> >> >> > good
> >> >> > coexistence with VL guys long-term (as option 3 will), but it does
> >> >> > allow
> >> >> > upgrading RBridges one by one in a working campus without causing
> >> >> > disruption.
> >> >> >
> >> >> > Option 3: Two types of RBridges.  VL, of course, is the same as in
> >> >> > options 1
> >> >> > and 2.  This option makes FGL guys do more work, but allows maximal
> >> >> > long
> >> >> > term coexistence of VL and FGL guys. In this option, FGL guys
> >> >> > calculate
> >> >> > unicast paths to FGL edge guys that avoid any VL guys, and
> calculate
> >> >> > at
> >> >> > least one FGL-friendly multicast tree that also avoids any VL guys.
> >> >> > So
> >> >> > let's
> >> >> > say R1 is an FGL guy.  R1 discards all LSPs from VL guys (ones that
> >> >> > don't
> >> >> > advertise FGL capability in their LSP), when calculating paths to
> >> >> > other
> >> >> > FGL
> >> >> > RBridges. Then R1 calculates paths to the VL guys using all LSPs.
> >> >> > Likewise,
> >> >> > when calculating an FGL-friendly tree, R1 calculates a tree through
> >> >> > only
> >> >> > FGL
> >> >> > guys.   This option is more work for the upgraded RBridges (because
> >> >> > they
> >> >> > have to calculate Dijkstra in two different ways, one for reaching
> VL
> >> >> > guys,
> >> >> > and one for reaching FGL guys).  However, it does allow having
> >> >> > long-term
> >> >> > coexistence with VL guys.  For instance, you could forever keep
> some
> >> >> > VL
> >> >> > edge
> >> >> > RBridges that communicate just fine through an FGL core.  They can
> >> >> > stay
> >> >> > there forever, and still be able to communicate through the core to
> >> >> > all
> >> >> > the
> >> >> > links attaching to their VLAN.
> >> >> >
> >> >> > _______________________________________________
> >> >> > trill mailing list
> >> >> > trill@ietf.org
> >> >> > https://www.ietf.org/mailman/listinfo/trill
> >> >> >
> >> >> _______________________________________________
> >> >> trill mailing list
> >> >> trill@ietf.org
> >> >> https://www.ietf.org/mailman/listinfo/trill
> >> >
> >> >
> >> >
> >> > _______________________________________________
> >> > trill mailing list
> >> > trill@ietf.org
> >> > https://www.ietf.org/mailman/listinfo/trill
> >> >
> >
> >
>

--20cf3071c8aa396fdb04d47682b1
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>Thanks much Donald. I think that I would vote for app=
roach 2.<br></div>Ayan<br><br></div><div class=3D"gmail_extra"><br><br><div=
 class=3D"gmail_quote">On Tue, Jan 29, 2013 at 4:14 PM, Donald Eastlake <sp=
an dir=3D"ltr">&lt;<a href=3D"mailto:d3e3e3@gmail.com" target=3D"_blank">d3=
e3e3@gmail.com</a>&gt;</span> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">Ayan,<br>
<br>
Yes, it would probably be good to have a brief informational section<br>
describing the intended VL to FGL migration scenario.<br>
<div class=3D"im HOEnZb"><br>
Thanks,<br>
Donald<br>
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D<br>
=A0Donald E. Eastlake 3rd =A0 <a href=3D"tel:%2B1-508-333-2270" value=3D"+1=
5083332270">+1-508-333-2270</a> (cell)<br>
=A0155 Beaver Street, Milford, MA 01757 USA<br>
=A0<a href=3D"mailto:d3e3e3@gmail.com">d3e3e3@gmail.com</a><br>
<br>
<br>
</div><div class=3D"HOEnZb"><div class=3D"h5">On Tue, Jan 29, 2013 at 4:15 =
PM, Ayan Banerjee &lt;<a href=3D"mailto:ayabaner@gmail.com">ayabaner@gmail.=
com</a>&gt; wrote:<br>
&gt; Donald,<br>
&gt;<br>
&gt; Yes, realize that would be way to do it - not sure if these options ar=
e<br>
&gt; being captured in a document, if so it would be great if this also add=
ed<br>
&gt; there.<br>
&gt;<br>
&gt; Thanks,<br>
&gt; Ayan<br>
&gt;<br>
&gt;<br>
&gt;<br>
&gt; On Tue, Jan 29, 2013 at 12:49 PM, Donald Eastlake &lt;<a href=3D"mailt=
o:d3e3e3@gmail.com">d3e3e3@gmail.com</a>&gt; wrote:<br>
&gt;&gt;<br>
&gt;&gt; Hi Ayan,<br>
&gt;&gt;<br>
&gt;&gt; On Tue, Jan 29, 2013 at 2:45 PM, Ayan Banerjee &lt;<a href=3D"mail=
to:ayabaner@gmail.com">ayabaner@gmail.com</a>&gt; wrote:<br>
&gt;&gt; &gt; I agree with Anoop here.<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt; I think that we are describing a network that is FGL-safe and=
 ready and<br>
&gt;&gt; &gt; then<br>
&gt;&gt; &gt; switch to FGL-mode. However, we need some statements to accou=
nt for a<br>
&gt;&gt; &gt; VL-node coming-in/leaving-the network due to a flaky link.<br=
>
&gt;&gt;<br>
&gt;&gt; The general assumption is that you are migrating from VL to FGL. F=
irst<br>
&gt;&gt; you migrate to FGL-safe. Then you unleash the FGL traffic, by<br>
&gt;&gt; configuring FGL ports, which isolates any remaining (FGL-unsafe) V=
L<br>
&gt;&gt; switches while still being able to handle VL traffic between VL po=
rts<br>
&gt;&gt; on FGL-safe switches.<br>
&gt;&gt;<br>
&gt;&gt; After you have unleashed FGL traffic, it would certainly be possib=
le<br>
&gt;&gt; that a VL switch could be attached to the network or come and go d=
ue<br>
&gt;&gt; to a flakey link, but this would have almost no effect. With FGL<b=
r>
&gt;&gt; traffic being handled, VL switches are isolated.<br>
&gt;&gt;<br>
&gt;&gt; Thanks,<br>
&gt;&gt; Donald<br>
&gt;&gt; =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D<br>
&gt;&gt; =A0Donald E. Eastlake 3rd =A0 <a href=3D"tel:%2B1-508-333-2270" va=
lue=3D"+15083332270">+1-508-333-2270</a> (cell)<br>
&gt;&gt; =A0155 Beaver Street, Milford, MA 01757 USA<br>
&gt;&gt; =A0<a href=3D"mailto:d3e3e3@gmail.com">d3e3e3@gmail.com</a><br>
&gt;&gt;<br>
&gt;&gt; &gt; Thanks,<br>
&gt;&gt; &gt; Ayan<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt; On Tue, Jan 29, 2013 at 6:42 AM, Anoop Ghanwani &lt;<a href=
=3D"mailto:anoop@alumni.duke.edu">anoop@alumni.duke.edu</a>&gt;<br>
&gt;&gt; &gt; wrote:<br>
&gt;&gt; &gt;&gt;<br>
&gt;&gt; &gt;&gt; At noted, option 1 looks dangerous so that should be rule=
d out.<br>
&gt;&gt; &gt;&gt;<br>
&gt;&gt; &gt;&gt; Option 2 seems reasonable. =A0It might be useful to clari=
fy a couple of<br>
&gt;&gt; &gt;&gt; things:<br>
&gt;&gt; &gt;&gt; - Can the FGL-safe campus have both VLAN &amp; FGL frames=
 going around?<br>
&gt;&gt; &gt;&gt; - Do all RBridges need to agree on when the campus is FGL=
-safe? =A0I<br>
&gt;&gt; &gt;&gt; would<br>
&gt;&gt; &gt;&gt; =A0 hope not. =A0But the problem I&#39;m thinking about i=
s where RB1 thinks<br>
&gt;&gt; &gt;&gt; the campus is<br>
&gt;&gt; &gt;&gt; =A0 FGL-safe and starts announcing attachment to FGL link=
s, but RB2<br>
&gt;&gt; &gt;&gt; hadn&#39;t<br>
&gt;&gt; &gt;&gt; yet<br>
&gt;&gt; &gt;&gt; =A0 concluded that process and it see RB3 which happens t=
o be VL.<br>
&gt;&gt; &gt;&gt;<br>
&gt;&gt; &gt;&gt; Option 3 seems to introduce more complexity to solve this=
 problem than<br>
&gt;&gt; &gt;&gt; is necessary.<br>
&gt;&gt; &gt;&gt;<br>
&gt;&gt; &gt;&gt; Anoop<br>
&gt;&gt; &gt;&gt;<br>
&gt;&gt; &gt;&gt; On Fri, Jan 25, 2013 at 11:00 AM, Radia Perlman<br>
&gt;&gt; &gt;&gt; &lt;<a href=3D"mailto:radiaperlman@gmail.com">radiaperlma=
n@gmail.com</a>&gt;<br>
&gt;&gt; &gt;&gt; wrote:<br>
&gt;&gt; &gt;&gt; &gt; I&#39;m going to summarize three options for phasing=
-in FGL, and explain<br>
&gt;&gt; &gt;&gt; &gt; the<br>
&gt;&gt; &gt;&gt; &gt; tradeoffs, to help people have informed opinions abo=
ut the<br>
&gt;&gt; &gt;&gt; &gt; implications<br>
&gt;&gt; &gt;&gt; &gt; of<br>
&gt;&gt; &gt;&gt; &gt; the three approaches. =A0There are infinite variatio=
ns, of course, but<br>
&gt;&gt; &gt;&gt; &gt; as<br>
&gt;&gt; &gt;&gt; &gt; I<br>
&gt;&gt; &gt;&gt; &gt; said, I&#39;ll describe three main ones.<br>
&gt;&gt; &gt;&gt; &gt;<br>
&gt;&gt; &gt;&gt; &gt; Option 1: =A0Original draft, no changes. =A0There ar=
e two types of<br>
&gt;&gt; &gt;&gt; &gt; RBridges.<br>
&gt;&gt; &gt;&gt; &gt; The<br>
&gt;&gt; &gt;&gt; &gt; first type is &quot;VL&quot; which only understands =
VLAN tags, and have unknown<br>
&gt;&gt; &gt;&gt; &gt; and<br>
&gt;&gt; &gt;&gt; &gt; possibly dangerous behavior when they receive an FGL=
-labeled packet.<br>
&gt;&gt; &gt;&gt; &gt; (e.g.,<br>
&gt;&gt; &gt;&gt; &gt; decapsulate a packet onto a link where it is not all=
owed, or<br>
&gt;&gt; &gt;&gt; &gt; mistakenly<br>
&gt;&gt; &gt;&gt; &gt; prune a multicast so it does not reach everywhere it=
 should). The<br>
&gt;&gt; &gt;&gt; &gt; second<br>
&gt;&gt; &gt;&gt; &gt; type is &quot;FGL&quot;. =A0FGL guys need to do two =
things: 1) understand FGL<br>
&gt;&gt; &gt;&gt; &gt; tags<br>
&gt;&gt; &gt;&gt; &gt; and<br>
&gt;&gt; &gt;&gt; &gt; do the right thing with them, and 2) ostrasize VL gu=
ys...meaning that<br>
&gt;&gt; &gt;&gt; &gt; an<br>
&gt;&gt; &gt;&gt; &gt; FGL<br>
&gt;&gt; &gt;&gt; &gt; guy refuses to form an adjacency with a VL guy. =A0T=
his option is very<br>
&gt;&gt; &gt;&gt; &gt; simple<br>
&gt;&gt; &gt;&gt; &gt; for implementers of FGL, but the implication is that=
 you have to<br>
&gt;&gt; &gt;&gt; &gt; upgrade<br>
&gt;&gt; &gt;&gt; &gt; your entire campus to FGL at once. =A0There is no co=
existence.<br>
&gt;&gt; &gt;&gt; &gt;<br>
&gt;&gt; &gt;&gt; &gt; Option 2: Two types of RBridges, but the 2nd type is=
 different from<br>
&gt;&gt; &gt;&gt; &gt; option<br>
&gt;&gt; &gt;&gt; &gt; 1. =A0VL guys of course are the same as in option 1.=
..they cannot be<br>
&gt;&gt; &gt;&gt; &gt; trusted<br>
&gt;&gt; &gt;&gt; &gt; with FGL frames. The second type I will call &quot;F=
GL-safe&quot;. An FGL-safe<br>
&gt;&gt; &gt;&gt; &gt; RBridge<br>
&gt;&gt; &gt;&gt; &gt; must advertise in its LSP that it is FGL-safe, and i=
t must not &quot;do<br>
&gt;&gt; &gt;&gt; &gt; anything<br>
&gt;&gt; &gt;&gt; &gt; bad&quot; with FGL frames, meaning that it is allowe=
d to ignore pruning of<br>
&gt;&gt; &gt;&gt; &gt; FGL<br>
&gt;&gt; &gt;&gt; &gt; (or even VL frames) entirely...it just can&#39;t fal=
sely drop FGL frames.<br>
&gt;&gt; &gt;&gt; &gt; And<br>
&gt;&gt; &gt;&gt; &gt; it must not decapsulate an FGL frame onto a link for=
 which that FGL<br>
&gt;&gt; &gt;&gt; &gt; frame<br>
&gt;&gt; &gt;&gt; &gt; doesn&#39;t belong. In this option, all RBridges mus=
t be upgraded to<br>
&gt;&gt; &gt;&gt; &gt; FGL-safe,<br>
&gt;&gt; &gt;&gt; &gt; but it need not happen instantaneously...it&#39;s fi=
ne to mix FGL-safe<br>
&gt;&gt; &gt;&gt; &gt; RBridges<br>
&gt;&gt; &gt;&gt; &gt; with VL RBridges...it&#39;s just not safe to inject =
FGL frames yet. =A0Once<br>
&gt;&gt; &gt;&gt; &gt; all<br>
&gt;&gt; &gt;&gt; &gt; the RBridges have been upgraded to be FGL-safe, then=
 edge RBridges<br>
&gt;&gt; &gt;&gt; &gt; can<br>
&gt;&gt; &gt;&gt; &gt; start<br>
&gt;&gt; &gt;&gt; &gt; announcing they are connected to an FGL link, and ca=
n start injecting<br>
&gt;&gt; &gt;&gt; &gt; FGL<br>
&gt;&gt; &gt;&gt; &gt; packets. =A0It is considered a misconfiguration if y=
ou start injecting<br>
&gt;&gt; &gt;&gt; &gt; FGL<br>
&gt;&gt; &gt;&gt; &gt; frames before all the RBridges are upgraded to FGL-s=
afe, so an<br>
&gt;&gt; &gt;&gt; &gt; additional<br>
&gt;&gt; &gt;&gt; &gt; chore for an FGL-safe RBridge R1 is to examine LSPs,=
 and if any<br>
&gt;&gt; &gt;&gt; &gt; RBridge<br>
&gt;&gt; &gt;&gt; &gt; claims to be attached to an FGL link, then R1 must o=
strasize any VL<br>
&gt;&gt; &gt;&gt; &gt; neighbors. =A0(don&#39;t start ostrasizing VL guys u=
ntil it is necessary<br>
&gt;&gt; &gt;&gt; &gt; because<br>
&gt;&gt; &gt;&gt; &gt; of actually starting to use FGL frames, in other wor=
ds). This option<br>
&gt;&gt; &gt;&gt; &gt; is<br>
&gt;&gt; &gt;&gt; &gt; more<br>
&gt;&gt; &gt;&gt; &gt; work for the upgraded RBridges than option 1, and th=
ere still isn&#39;t<br>
&gt;&gt; &gt;&gt; &gt; good<br>
&gt;&gt; &gt;&gt; &gt; coexistence with VL guys long-term (as option 3 will=
), but it does<br>
&gt;&gt; &gt;&gt; &gt; allow<br>
&gt;&gt; &gt;&gt; &gt; upgrading RBridges one by one in a working campus wi=
thout causing<br>
&gt;&gt; &gt;&gt; &gt; disruption.<br>
&gt;&gt; &gt;&gt; &gt;<br>
&gt;&gt; &gt;&gt; &gt; Option 3: Two types of RBridges. =A0VL, of course, i=
s the same as in<br>
&gt;&gt; &gt;&gt; &gt; options 1<br>
&gt;&gt; &gt;&gt; &gt; and 2. =A0This option makes FGL guys do more work, b=
ut allows maximal<br>
&gt;&gt; &gt;&gt; &gt; long<br>
&gt;&gt; &gt;&gt; &gt; term coexistence of VL and FGL guys. In this option,=
 FGL guys<br>
&gt;&gt; &gt;&gt; &gt; calculate<br>
&gt;&gt; &gt;&gt; &gt; unicast paths to FGL edge guys that avoid any VL guy=
s, and calculate<br>
&gt;&gt; &gt;&gt; &gt; at<br>
&gt;&gt; &gt;&gt; &gt; least one FGL-friendly multicast tree that also avoi=
ds any VL guys.<br>
&gt;&gt; &gt;&gt; &gt; So<br>
&gt;&gt; &gt;&gt; &gt; let&#39;s<br>
&gt;&gt; &gt;&gt; &gt; say R1 is an FGL guy. =A0R1 discards all LSPs from V=
L guys (ones that<br>
&gt;&gt; &gt;&gt; &gt; don&#39;t<br>
&gt;&gt; &gt;&gt; &gt; advertise FGL capability in their LSP), when calcula=
ting paths to<br>
&gt;&gt; &gt;&gt; &gt; other<br>
&gt;&gt; &gt;&gt; &gt; FGL<br>
&gt;&gt; &gt;&gt; &gt; RBridges. Then R1 calculates paths to the VL guys us=
ing all LSPs.<br>
&gt;&gt; &gt;&gt; &gt; Likewise,<br>
&gt;&gt; &gt;&gt; &gt; when calculating an FGL-friendly tree, R1 calculates=
 a tree through<br>
&gt;&gt; &gt;&gt; &gt; only<br>
&gt;&gt; &gt;&gt; &gt; FGL<br>
&gt;&gt; &gt;&gt; &gt; guys. =A0 This option is more work for the upgraded =
RBridges (because<br>
&gt;&gt; &gt;&gt; &gt; they<br>
&gt;&gt; &gt;&gt; &gt; have to calculate Dijkstra in two different ways, on=
e for reaching VL<br>
&gt;&gt; &gt;&gt; &gt; guys,<br>
&gt;&gt; &gt;&gt; &gt; and one for reaching FGL guys). =A0However, it does =
allow having<br>
&gt;&gt; &gt;&gt; &gt; long-term<br>
&gt;&gt; &gt;&gt; &gt; coexistence with VL guys. =A0For instance, you could=
 forever keep some<br>
&gt;&gt; &gt;&gt; &gt; VL<br>
&gt;&gt; &gt;&gt; &gt; edge<br>
&gt;&gt; &gt;&gt; &gt; RBridges that communicate just fine through an FGL c=
ore. =A0They can<br>
&gt;&gt; &gt;&gt; &gt; stay<br>
&gt;&gt; &gt;&gt; &gt; there forever, and still be able to communicate thro=
ugh the core to<br>
&gt;&gt; &gt;&gt; &gt; all<br>
&gt;&gt; &gt;&gt; &gt; the<br>
&gt;&gt; &gt;&gt; &gt; links attaching to their VLAN.<br>
&gt;&gt; &gt;&gt; &gt;<br>
&gt;&gt; &gt;&gt; &gt; _______________________________________________<br>
&gt;&gt; &gt;&gt; &gt; trill mailing list<br>
&gt;&gt; &gt;&gt; &gt; <a href=3D"mailto:trill@ietf.org">trill@ietf.org</a>=
<br>
&gt;&gt; &gt;&gt; &gt; <a href=3D"https://www.ietf.org/mailman/listinfo/tri=
ll" target=3D"_blank">https://www.ietf.org/mailman/listinfo/trill</a><br>
&gt;&gt; &gt;&gt; &gt;<br>
&gt;&gt; &gt;&gt; _______________________________________________<br>
&gt;&gt; &gt;&gt; trill mailing list<br>
&gt;&gt; &gt;&gt; <a href=3D"mailto:trill@ietf.org">trill@ietf.org</a><br>
&gt;&gt; &gt;&gt; <a href=3D"https://www.ietf.org/mailman/listinfo/trill" t=
arget=3D"_blank">https://www.ietf.org/mailman/listinfo/trill</a><br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt; _______________________________________________<br>
&gt;&gt; &gt; trill mailing list<br>
&gt;&gt; &gt; <a href=3D"mailto:trill@ietf.org">trill@ietf.org</a><br>
&gt;&gt; &gt; <a href=3D"https://www.ietf.org/mailman/listinfo/trill" targe=
t=3D"_blank">https://www.ietf.org/mailman/listinfo/trill</a><br>
&gt;&gt; &gt;<br>
&gt;<br>
&gt;<br>
</div></div></blockquote></div><br></div>

--20cf3071c8aa396fdb04d47682b1--

From d3e3e3@gmail.com  Tue Jan 29 16:37:24 2013
Return-Path: <d3e3e3@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 18E7321F8684 for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 16:37:24 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.143
X-Spam-Level: 
X-Spam-Status: No, score=-103.143 tagged_above=-999 required=5 tests=[AWL=0.456, BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id hMvCZqVoYNxZ for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 16:37:23 -0800 (PST)
Received: from mail-oa0-f42.google.com (mail-oa0-f42.google.com [209.85.219.42]) by ietfa.amsl.com (Postfix) with ESMTP id 7D32721F83EF for <trill@ietf.org>; Tue, 29 Jan 2013 16:37:23 -0800 (PST)
Received: by mail-oa0-f42.google.com with SMTP id i18so833718oag.1 for <trill@ietf.org>; Tue, 29 Jan 2013 16:37:23 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=x-received:mime-version:in-reply-to:references:from:date:message-id :subject:to:cc:content-type; bh=h94FNRQay3F+veKoqD2of8XyeQKkGOY2G2DSJMdzyR4=; b=DcG1syGWRoxfNDJ+S5NtkDZVh3bx11sDzYdiot6dD9Kgdpn6hxhbpLdLMVWXv+HlFg u5FnZBu0zko1YW/BM0nSNjz+FIEG4MCNl65Q90HT4HnqmMwg0S7DU4j1RNn9yTAZ2MC3 ewXzDcqI9fHEg7OwZQymLurX1lBR/DJ+V6+Eij+19fVq19jfalMwj8d7ItFiMn0YKA8a pTG6AdFvDkZdrDn0CbxtzQ6yPkHsxr5jA3/pZns6rwqlZjZLj+Rt2EMvLapcsoQQGibl kD+ZWvvghQKOfM3UI4CpdTDAGsUuEv8yoT6yxSDdsvyP/5MpGjuG3kPbLsYQ14T4S+MG J0Tw==
X-Received: by 10.182.123.49 with SMTP id lx17mr2289499obb.63.1359506243073; Tue, 29 Jan 2013 16:37:23 -0800 (PST)
MIME-Version: 1.0
Received: by 10.76.98.168 with HTTP; Tue, 29 Jan 2013 16:37:02 -0800 (PST)
In-Reply-To: <CA+-tSzz0=oh6iB1e-ESyeOx2sCvBFXeVpzuNbKmUVhRNLScDgQ@mail.gmail.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <CA+-tSzx0E3J4i5Dvfbj4egruCeGi7z1se1Jn5RXtx3RedQLPdg@mail.gmail.com> <CAHD03N89RNX4=XxRZpD+Y2PCktYf-FtVjyYJ5oFS3dui3SsXaA@mail.gmail.com> <CAF4+nEG7YKGzGmObAMaOxackivjJsALymqHX=6PKBd5qU-Z+zQ@mail.gmail.com> <CA+-tSzz0=oh6iB1e-ESyeOx2sCvBFXeVpzuNbKmUVhRNLScDgQ@mail.gmail.com>
From: Donald Eastlake <d3e3e3@gmail.com>
Date: Tue, 29 Jan 2013 19:37:02 -0500
Message-ID: <CAF4+nEFLrJL=CgX9e6on-cgz89ugmu93P6u5kmODhUia3H7RLg@mail.gmail.com>
To: Anoop Ghanwani <anoop@alumni.duke.edu>
Content-Type: text/plain; charset=ISO-8859-1
Cc: Radia Perlman <radiaperlman@gmail.com>, Ayan Banerjee <ayabaner@gmail.com>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 30 Jan 2013 00:37:24 -0000

Hi Anoop,

On Tue, Jan 29, 2013 at 7:03 PM, Anoop Ghanwani <anoop@alumni.duke.edu> wrote:
> On Tue, Jan 29, 2013 at 12:49 PM, Donald Eastlake <d3e3e3@gmail.com> wrote:
>
>> The general assumption is that you are migrating from VL to FGL. First
>> you migrate to FGL-safe. Then you unleash the FGL traffic, by
>> configuring FGL ports, which isolates any remaining (FGL-unsafe) VL
>> switches while still being able to handle VL traffic between VL ports
>> on FGL-safe switches.
>
> Based on what is written so far, it looks like as long as there is even
> one VL RBridge in it's database, an RBridge would not be able to start
> sourcing FGL traffic.  It may start doing so only when all RBridges in

Well, it can start sourcing FGL traffic but their are consequences as
you note below.

> its database are FGL-safe.  If we ever get into a condition where
> an RBridge has started advertising FGL information, at that point,
> does any other RBridge in the campus simply isolate any VL RBridge
> (with the assumption that there was some kind of race condition?

Yes. (This could just be due to some misconfiguration.)

> The case I'm concerned about is we have 2 FGL RBridges, RB1 and RB2.
> RB1 is connected to a campus of n RBs that are still VL only.  The VL link
> Breaks, and RB1 starts advertising FGL information to RB2, but then
> the link to the VL campus comes back up.  Does this mean RB1 and RB2
> remain isolated from the campus (and operate as they own mini-campus)
> until all the other RBridges are converted over to FGL-safe?

Yes. But I don't think this as much of a problem given that I believe
existing TRILL switches can be software upgraded to be FGL-safe. I
view the configuration of FGL ports on RB1/RB2 and the consequent
routing of actual FGL traffic to be an active network management
decision, not something that might just happen because there appear to
be no more VL TRILL switches in the topology.

Thanks,
Donald
=============================
 Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
 155 Beaver Street, Milford, MA 01757 USA
 d3e3e3@gmail.com

> Anoop

From haoweiguo@huawei.com  Tue Jan 29 17:21:55 2013
Return-Path: <haoweiguo@huawei.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 528CF21F868E for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 17:21:55 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 2.599
X-Spam-Level: **
X-Spam-Status: No, score=2.599 tagged_above=-999 required=5 tests=[AWL=0.151,  BAYES_00=-2.599, CHARSET_FARAWAY_HEADER=3.2, MIME_8BIT_HEADER=0.3,  MIME_BASE64_TEXT=1.753, MIME_CHARSET_FARAWAY=2.45, RCVD_IN_DNSWL_MED=-4, SARE_SUB_ENC_GB2312=1.345]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id EnPL--+eTd5b for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 17:21:54 -0800 (PST)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) by ietfa.amsl.com (Postfix) with ESMTP id 3B8A221F88A1 for <trill@ietf.org>; Tue, 29 Jan 2013 17:21:53 -0800 (PST)
Received: from 172.18.7.190 (EHLO lhreml204-edg.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.5-GA FastPath queued) with ESMTP id APF66916; Wed, 30 Jan 2013 01:21:51 +0000 (GMT)
Received: from LHREML406-HUB.china.huawei.com (10.201.5.243) by lhreml204-edg.china.huawei.com (172.18.7.223) with Microsoft SMTP Server (TLS) id 14.1.323.7; Wed, 30 Jan 2013 01:21:19 +0000
Received: from NKGEML408-HUB.china.huawei.com (10.98.56.39) by lhreml406-hub.china.huawei.com (10.201.5.243) with Microsoft SMTP Server (TLS) id 14.1.323.7; Wed, 30 Jan 2013 01:21:51 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.101]) by nkgeml408-hub.china.huawei.com ([10.98.56.39]) with mapi id 14.01.0323.007; Wed, 30 Jan 2013 09:21:47 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: Donald Eastlake <d3e3e3@gmail.com>
Thread-Topic: =?gb2312?B?W3RyaWxsXSC08Li0OiC08Li0OiBFeHBsYWluaW5nIHRocmVlIG9wdGlvbnMg?= =?gb2312?Q?for_upgrading_to_FGL_(fine-grained-labeling)?=
Thread-Index: AQHN/j8cC6e1ZiC2qUuJTbAQVN5MC5hhEez4
Date: Wed, 30 Jan 2013 01:21:46 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F65150E@nkgeml501-mbs.china.huawei.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <DD5FC8DE455C3348B94340C0AB5517334F651415@nkgeml501-mbs.china.huawei.com> <CAFOuuo597npOuDLT1di-0V+Q1Fh7nYvKD3jHG3bbZPB81hc=Tw@mail.gmail.com> <DD5FC8DE455C3348B94340C0AB5517334F6514AD@nkgeml501-mbs.china.huawei.com>, <CAF4+nEHR9SadJoAFXGqD-s_=K-js4Z3E7ONsLWxBSHjPMN+hzQ@mail.gmail.com>
In-Reply-To: <CAF4+nEHR9SadJoAFXGqD-s_=K-js4Z3E7ONsLWxBSHjPMN+hzQ@mail.gmail.com>
Accept-Language: zh-CN, en-US
Content-Language: zh-CN
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.135.22.248]
Content-Type: text/plain; charset="gb2312"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Cc: Radia Perlman <radiaperlman@gmail.com>, "trill@ietf.org" <trill@ietf.org>
Subject: [trill] =?gb2312?b?tPC4tDogILTwuLQ6ILTwuLQ6IEV4cGxhaW5pbmcgdGhy?= =?gb2312?b?ZWUgb3B0aW9ucyBmb3IgdXBncmFkaW5nIHRvIEZHTCAoZmluZS1ncmFpbmVk?= =?gb2312?b?LWxhYmVsaW5nKQ==?=
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 30 Jan 2013 01:21:55 -0000
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From liyizhou@huawei.com  Tue Jan 29 20:25:47 2013
Return-Path: <liyizhou@huawei.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D324021F8967 for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 20:25:47 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.599
X-Spam-Level: 
X-Spam-Status: No, score=-6.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 6E-GG0t6SSAw for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 20:25:47 -0800 (PST)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) by ietfa.amsl.com (Postfix) with ESMTP id 4FF8521F856D for <trill@ietf.org>; Tue, 29 Jan 2013 20:25:45 -0800 (PST)
Received: from 172.18.7.190 (EHLO lhreml203-edg.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.5-GA FastPath queued) with ESMTP id ANZ44372; Wed, 30 Jan 2013 04:25:44 +0000 (GMT)
Received: from LHREML403-HUB.china.huawei.com (10.201.5.217) by lhreml203-edg.huawei.com (172.18.7.221) with Microsoft SMTP Server (TLS) id 14.1.323.7; Wed, 30 Jan 2013 04:25:07 +0000
Received: from SZXEML413-HUB.china.huawei.com (10.82.67.152) by lhreml403-hub.china.huawei.com (10.201.5.217) with Microsoft SMTP Server (TLS) id 14.1.323.7; Wed, 30 Jan 2013 04:25:43 +0000
Received: from l63746 (10.138.122.43) by smtpscn.huawei.com (10.82.67.152) with Microsoft SMTP Server (TLS) id 14.1.323.7; Wed, 30 Jan 2013 12:25:35 +0800
From: Yizhou Li <liyizhou@huawei.com>
To: "'Anoop Ghanwani'" <anoop@alumni.duke.edu>, "'Donald Eastlake'" <d3e3e3@gmail.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com>	<CA+-tSzx0E3J4i5Dvfbj4egruCeGi7z1se1Jn5RXtx3RedQLPdg@mail.gmail.com>	<CAHD03N89RNX4=XxRZpD+Y2PCktYf-FtVjyYJ5oFS3dui3SsXaA@mail.gmail.com>	<CAF4+nEG7YKGzGmObAMaOxackivjJsALymqHX=6PKBd5qU-Z+zQ@mail.gmail.com> <CA+-tSzz0=oh6iB1e-ESyeOx2sCvBFXeVpzuNbKmUVhRNLScDgQ@mail.gmail.com>
In-Reply-To: <CA+-tSzz0=oh6iB1e-ESyeOx2sCvBFXeVpzuNbKmUVhRNLScDgQ@mail.gmail.com>
Date: Wed, 30 Jan 2013 12:25:35 +0800
Message-ID: <049401cdfea1$d9ab3bc0$8d01b340$@com>
MIME-Version: 1.0
Content-Type: text/plain; charset="gb2312"
Content-Transfer-Encoding: quoted-printable
X-Mailer: Microsoft Office Outlook 12.0
Thread-Index: Ac3+fTnvU4O4WzT2SZO/1Z2uBQcFswAIdlxg
Content-Language: zh-cn
X-Originating-IP: [10.138.122.43]
X-CFilter-Loop: Reflected
Cc: 'Radia Perlman' <radiaperlman@gmail.com>, 'Ayan Banerjee' <ayabaner@gmail.com>, trill@ietf.org
Subject: Re: [trill] Explaining three options for upgrading to FGL	(fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 30 Jan 2013 04:25:47 -0000

With option 2, once RB starts to injecting FGL frame, isolated pure VL =
RB
will no longer able to exchange VL-tagged frame with FGL-safe RB and =
vice
versa. Such scenario may be caused by mis-configuration, flaky link, =
newly
connected cable, etc. Option 3 gives better handlings for such corner =
cases.
However comparing the complexity=A3=AC I would still vote for option 2. =
The most
possible smooth migration path would be gradually replacing VL RB with =
FGL
safe RB.=20
I think the administrator should roughly know the overall equipments
distribution and be responsible to decide when is the right time to =
actually
enable the injecting of FGL frame.=20

Yizhou


-----Original Message-----
From: trill-bounces@ietf.org [mailto:trill-bounces@ietf.org] On Behalf =
Of
Anoop Ghanwani
Sent: Wednesday, January 30, 2013 8:03 AM
To: Donald Eastlake
Cc: Radia Perlman; Ayan Banerjee; trill@ietf.org
Subject: Re: [trill] Explaining three options for upgrading to FGL
(fine-grained-labeling)

On Tue, Jan 29, 2013 at 12:49 PM, Donald Eastlake <d3e3e3@gmail.com> =
wrote:

> The general assumption is that you are migrating from VL to FGL. First
> you migrate to FGL-safe. Then you unleash the FGL traffic, by
> configuring FGL ports, which isolates any remaining (FGL-unsafe) VL
> switches while still being able to handle VL traffic between VL ports
> on FGL-safe switches.

Based on what is written so far, it looks like as long as there is even
one VL RBridge in it's database, an RBridge would not be able to start
sourcing FGL traffic.  It may start doing so only when all RBridges in
its database are FGL-safe.  If we ever get into a condition where
an RBridge has started advertising FGL information, at that point,
does any other RBridge in the campus simply isolate any VL RBridge
(with the assumption that there was some kind of race condition?

The case I'm concerned about is we have 2 FGL RBridges, RB1 and RB2.
RB1 is connected to a campus of n RBs that are still VL only.  The VL =
link
Breaks, and RB1 starts advertising FGL information to RB2, but then
the link to the VL campus comes back up.  Does this mean RB1 and RB2
remain isolated from the campus (and operate as they own mini-campus)
until all the other RBridges are converted over to FGL-safe?

Anoop
_______________________________________________
trill mailing list
trill@ietf.org
https://www.ietf.org/mailman/listinfo/trill


From talmi@marvell.com  Tue Jan 29 20:59:10 2013
Return-Path: <talmi@marvell.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4441121F8B0C for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 20:59:10 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.598
X-Spam-Level: 
X-Spam-Status: No, score=-2.598 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id brZyP26mREhS for <trill@ietfa.amsl.com>; Tue, 29 Jan 2013 20:59:09 -0800 (PST)
Received: from galiil.marvell.com (galiil.marvell.com [199.203.130.254]) by ietfa.amsl.com (Postfix) with ESMTP id 65B4121F86EA for <trill@ietf.org>; Tue, 29 Jan 2013 20:59:08 -0800 (PST)
From: Tal Mizrahi <talmi@marvell.com>
To: Radia Perlman <radiaperlman@gmail.com>, "trill@ietf.org" <trill@ietf.org>
Date: Wed, 30 Jan 2013 06:59:02 +0200
Thread-Topic: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
Thread-Index: Ac37LkSdwghuCroTRE6aoY+VHFikDADd9yMg
Message-ID: <74470498B659FA4687F0B0018C19A89C01A0F93685EF@IL-MB01.marvell.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com>
In-Reply-To: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US
Content-Type: multipart/alternative; boundary="_000_74470498B659FA4687F0B0018C19A89C01A0F93685EFILMB01marve_"
MIME-Version: 1.0
Subject: Re: [trill] Explaining three options for upgrading to FGL	(fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 30 Jan 2013 04:59:10 -0000

--_000_74470498B659FA4687F0B0018C19A89C01A0F93685EFILMB01marve_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

I would go with option 2, as it seems to offer a fairly smooth migration pr=
ocess, as well as a relatively simple software upgrade to turn VL nodes int=
o FGL-safe nodes.

Tal.

From: trill-bounces@ietf.org [mailto:trill-bounces@ietf.org] On Behalf Of R=
adia Perlman
Sent: Friday, January 25, 2013 9:00 PM
To: trill@ietf.org
Subject: [trill] Explaining three options for upgrading to FGL (fine-graine=
d-labeling)

I'm going to summarize three options for phasing-in FGL, and explain the tr=
adeoffs, to help people have informed opinions about the implications of th=
e three approaches.  There are infinite variations, of course, but as I sai=
d, I'll describe three main ones.

Option 1:  Original draft, no changes.  There are two types of RBridges. Th=
e first type is "VL" which only understands VLAN tags, and have unknown and=
 possibly dangerous behavior when they receive an FGL-labeled packet. (e.g.=
, decapsulate a packet onto a link where it is not allowed, or mistakenly p=
rune a multicast so it does not reach everywhere it should). The second typ=
e is "FGL".  FGL guys need to do two things: 1) understand FGL tags and do =
the right thing with them, and 2) ostrasize VL guys...meaning that an FGL g=
uy refuses to form an adjacency with a VL guy.  This option is very simple =
for implementers of FGL, but the implication is that you have to upgrade yo=
ur entire campus to FGL at once.  There is no coexistence.

Option 2: Two types of RBridges, but the 2nd type is different from option =
1.  VL guys of course are the same as in option 1...they cannot be trusted =
with FGL frames. The second type I will call "FGL-safe". An FGL-safe RBridg=
e must advertise in its LSP that it is FGL-safe, and it must not "do anythi=
ng bad" with FGL frames, meaning that it is allowed to ignore pruning of FG=
L (or even VL frames) entirely...it just can't falsely drop FGL frames.  An=
d it must not decapsulate an FGL frame onto a link for which that FGL frame=
 doesn't belong. In this option, all RBridges must be upgraded to FGL-safe,=
 but it need not happen instantaneously...it's fine to mix FGL-safe RBridge=
s with VL RBridges...it's just not safe to inject FGL frames yet.  Once all=
 the RBridges have been upgraded to be FGL-safe, then edge RBridges can sta=
rt announcing they are connected to an FGL link, and can start injecting FG=
L packets.  It is considered a misconfiguration if you start injecting FGL =
frames before all the RBridges are upgraded to FGL-safe, so an additional c=
hore for an FGL-safe RBridge R1 is to examine LSPs, and if any RBridge clai=
ms to be attached to an FGL link, then R1 must ostrasize any VL neighbors. =
 (don't start ostrasizing VL guys until it is necessary because of actually=
 starting to use FGL frames, in other words). This option is more work for =
the upgraded RBridges than option 1, and there still isn't good coexistence=
 with VL guys long-term (as option 3 will), but it does allow upgrading RBr=
idges one by one in a working campus without causing disruption.

Option 3: Two types of RBridges.  VL, of course, is the same as in options =
1 and 2.  This option makes FGL guys do more work, but allows maximal long =
term coexistence of VL and FGL guys. In this option, FGL guys calculate uni=
cast paths to FGL edge guys that avoid any VL guys, and calculate at least =
one FGL-friendly multicast tree that also avoids any VL guys. So let's say =
R1 is an FGL guy.  R1 discards all LSPs from VL guys (ones that don't adver=
tise FGL capability in their LSP), when calculating paths to other FGL RBri=
dges. Then R1 calculates paths to the VL guys using all LSPs.  Likewise, wh=
en calculating an FGL-friendly tree, R1 calculates a tree through only FGL =
guys.   This option is more work for the upgraded RBridges (because they ha=
ve to calculate Dijkstra in two different ways, one for reaching VL guys, a=
nd one for reaching FGL guys).  However, it does allow having long-term coe=
xistence with VL guys.  For instance, you could forever keep some VL edge R=
Bridges that communicate just fine through an FGL core.  They can stay ther=
e forever, and still be able to communicate through the core to all the lin=
ks attaching to their VLAN.

--_000_74470498B659FA4687F0B0018C19A89C01A0F93685EFILMB01marve_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micr=
osoft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:x=3D"urn:schemas-microsoft-com:office:excel" xmlns:m=3D"http://schema=
s.microsoft.com/office/2004/12/omml" xmlns=3D"http://www.w3.org/TR/REC-html=
40"><head><META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; charset=
=3Dus-ascii"><meta name=3DGenerator content=3D"Microsoft Word 14 (filtered =
medium)"><style><!--
/* Font Definitions */
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:Tahoma;
	panose-1:2 11 6 4 3 5 4 4 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman","serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
span.EmailStyle17
	{mso-style-type:personal-reply;
	font-family:"Calibri","sans-serif";
	color:#1F497D;}
.MsoChpDefault
	{mso-style-type:export-only;
	font-family:"Calibri","sans-serif";}
@page WordSection1
	{size:8.5in 11.0in;
	margin:1.0in 1.25in 1.0in 1.25in;}
div.WordSection1
	{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]--></head><body lang=3DEN-US link=3Dblue vli=
nk=3Dpurple><div class=3DWordSection1><p class=3DMsoNormal><span style=3D'f=
ont-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>I would g=
o with option 2, as it seems to offer a fairly smooth migration process, as=
 well as a relatively simple software upgrade to turn VL nodes into FGL-saf=
e nodes.<o:p></o:p></span></p><p class=3DMsoNormal><span style=3D'font-size=
:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'><o:p>&nbsp;</o:p>=
</span></p><p class=3DMsoNormal><span style=3D'font-size:10.0pt;font-family=
:"Arial","sans-serif";color:#1F497D'>Tal.<o:p></o:p></span></p><p class=3DM=
soNormal><span style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif"=
;color:#1F497D'><o:p>&nbsp;</o:p></span></p><p class=3DMsoNormal><b><span s=
tyle=3D'font-size:10.0pt;font-family:"Tahoma","sans-serif"'>From:</span></b=
><span style=3D'font-size:10.0pt;font-family:"Tahoma","sans-serif"'> trill-=
bounces@ietf.org [mailto:trill-bounces@ietf.org] <b>On Behalf Of </b>Radia =
Perlman<br><b>Sent:</b> Friday, January 25, 2013 9:00 PM<br><b>To:</b> tril=
l@ietf.org<br><b>Subject:</b> [trill] Explaining three options for upgradin=
g to FGL (fine-grained-labeling)<o:p></o:p></span></p><p class=3DMsoNormal>=
<o:p>&nbsp;</o:p></p><div><p class=3DMsoNormal style=3D'background:white'><=
span style=3D'font-size:10.0pt;font-family:"Arial","sans-serif";color:#2222=
22'>I'm going to summarize three options for phasing-in FGL, and explain th=
e tradeoffs, to help people have informed opinions about the implications o=
f the three approaches. &nbsp;There are infinite variations, of course, but=
 as I said, I'll describe three main ones.<o:p></o:p></span></p></div><div>=
<p class=3DMsoNormal style=3D'background:white'><span style=3D'font-size:10=
.0pt;font-family:"Arial","sans-serif";color:#222222'><o:p>&nbsp;</o:p></spa=
n></p></div><div><p class=3DMsoNormal style=3D'background:white'><span styl=
e=3D'font-size:10.0pt;font-family:"Arial","sans-serif";color:#222222'>Optio=
n 1: &nbsp;Original draft, no changes. &nbsp;There are two types of RBridge=
s. The first type is &quot;VL&quot; which only understands VLAN tags, and h=
ave unknown and possibly dangerous behavior when they receive an FGL-labele=
d packet. (e.g., decapsulate a packet onto a link where it is not allowed, =
or mistakenly prune a multicast so it does not reach everywhere it should).=
 The second type is &quot;FGL&quot;. &nbsp;FGL guys need to do two things: =
1) understand FGL tags and do the right thing with them, and 2) ostrasize V=
L guys...meaning that an FGL guy refuses to form an adjacency with a VL guy=
. &nbsp;This option is very simple for implementers of FGL, but the implica=
tion is that you have to upgrade your entire campus to FGL at once. &nbsp;T=
here is no coexistence.<o:p></o:p></span></p></div><div><p class=3DMsoNorma=
l style=3D'background:white'><span style=3D'font-size:10.0pt;font-family:"A=
rial","sans-serif";color:#222222'><o:p>&nbsp;</o:p></span></p></div><div><p=
 class=3DMsoNormal style=3D'background:white'><span style=3D'font-size:10.0=
pt;font-family:"Arial","sans-serif";color:#222222'>Option 2: Two types of R=
Bridges, but the 2nd type is different from option 1. &nbsp;VL guys of cour=
se are the same as in option 1...they cannot be trusted with FGL frames. Th=
e second type I will call &quot;FGL-safe&quot;. An FGL-safe RBridge must ad=
vertise in its LSP that it is FGL-safe, and it must not &quot;do anything b=
ad&quot; with FGL frames, meaning that it is allowed to ignore pruning of F=
GL (or even VL frames) entirely...it just can't falsely drop FGL frames. &n=
bsp;And it must not decapsulate an FGL frame onto a link for which that FGL=
 frame doesn't belong. In this option, all RBridges must be upgraded to FGL=
-safe, but it need not happen instantaneously...it's fine to mix FGL-safe R=
Bridges with VL RBridges...it's just not safe to inject FGL frames yet. &nb=
sp;Once all the RBridges have been upgraded to be FGL-safe, then edge RBrid=
ges can start announcing they are connected to an FGL link, and can start i=
njecting FGL packets. &nbsp;It is considered a misconfiguration if you star=
t injecting FGL frames before all the RBridges are upgraded to FGL-safe, so=
 an additional chore for an FGL-safe RBridge R1 is to examine LSPs, and if =
any RBridge claims to be attached to an FGL link, then R1 must ostrasize an=
y VL neighbors. &nbsp;(don't start ostrasizing VL guys until it is necessar=
y because of actually starting to use FGL frames, in other words). This opt=
ion is more work for the upgraded RBridges than option 1, and there still i=
sn't good coexistence with VL guys long-term (as option 3 will), but it doe=
s allow upgrading RBridges one by one in a working campus without causing d=
isruption.<o:p></o:p></span></p></div><div><p class=3DMsoNormal style=3D'ba=
ckground:white'><span style=3D'font-size:10.0pt;font-family:"Arial","sans-s=
erif";color:#222222'><o:p>&nbsp;</o:p></span></p></div><div><p class=3DMsoN=
ormal style=3D'background:white'><span style=3D'font-size:10.0pt;font-famil=
y:"Arial","sans-serif";color:#222222'>Option 3: Two types of RBridges. &nbs=
p;VL, of course, is the same as in options 1 and 2. &nbsp;This option makes=
 FGL guys do more work, but allows maximal long term coexistence of VL and =
FGL guys. In this option, FGL guys calculate unicast paths to FGL edge guys=
 that avoid any VL guys, and calculate at least one FGL-friendly multicast =
tree that also avoids any VL guys. So let's say R1 is an FGL guy. &nbsp;R1 =
discards all LSPs from VL guys (ones that don't advertise FGL capability in=
 their LSP), when calculating paths to other FGL RBridges. Then R1 calculat=
es paths to the VL guys using all LSPs. &nbsp;Likewise, when calculating an=
 FGL-friendly tree, R1 calculates a tree through only FGL guys. &nbsp; This=
 option is more work for the upgraded RBridges (because they have to calcul=
ate Dijkstra in two different ways, one for reaching VL guys, and one for r=
eaching FGL guys). &nbsp;However, it does allow having long-term coexistenc=
e with VL guys. &nbsp;For instance, you could forever keep some VL edge RBr=
idges that communicate just fine through an FGL core. &nbsp;They can stay t=
here forever, and still be able to communicate through the core to all the =
links attaching to their VLAN.<o:p></o:p></span></p></div></div></body></ht=
ml>=

--_000_74470498B659FA4687F0B0018C19A89C01A0F93685EFILMB01marve_--

From hu.fangwei@zte.com.cn  Tue Jan 29 22:27:29 2013
Return-Path: <hu.fangwei@zte.com.cn>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 64F9F21F84B6; Tue, 29 Jan 2013 22:27:29 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -98.395
X-Spam-Level: 
X-Spam-Status: No, score=-98.395 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, MIME_BASE64_TEXT=1.753, MIME_CHARSET_FARAWAY=2.45, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id waOlhZjP+urm; Tue, 29 Jan 2013 22:27:28 -0800 (PST)
Received: from zte.com.cn (mx5.zte.com.cn [63.217.80.70]) by ietfa.amsl.com (Postfix) with ESMTP id E1CCE21F84A1; Tue, 29 Jan 2013 22:27:27 -0800 (PST)
Received: from mse02.zte.com.cn (unknown [10.30.3.21]) by Websense Email Security Gateway with ESMTPS id 78F691245691; Wed, 30 Jan 2013 14:28:19 +0800 (CST)
Received: from notes_smtp.zte.com.cn ([10.30.1.239]) by mse02.zte.com.cn with ESMTP id r0U6R3Ag048006; Wed, 30 Jan 2013 14:27:03 +0800 (GMT-8) (envelope-from hu.fangwei@zte.com.cn)
In-Reply-To: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com>
To: Radia Perlman <radiaperlman@gmail.com>
MIME-Version: 1.0
X-Mailer: Lotus Notes Release 6.0.4 June 01, 2004
Message-ID: <OF7D4AED17.C9650586-ON48257B03.0022E885-48257B03.00237369@zte.com.cn>
From: hu.fangwei@zte.com.cn
Date: Wed, 30 Jan 2013 14:27:01 +0800
X-MIMETrack: Serialize by Router on notes_smtp/zte_ltd(Release 8.5.3FP1 HF212|May 23, 2012) at 2013-01-30 14:26:59, Serialize complete at 2013-01-30 14:26:59
Content-Type: multipart/alternative; boundary="=_alternative 0023736548257B03_="
X-MAIL: mse02.zte.com.cn r0U6R3Ag048006
Cc: trill-bounces@ietf.org, trill@ietf.org
Subject: Re: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 30 Jan 2013 06:27:29 -0000

This is a multipart message in MIME format.
--=_alternative 0023736548257B03_=
Content-Type: text/plain; charset="GB2312"
Content-Transfer-Encoding: base64
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--=_alternative 0023736548257B03_=
Content-Type: text/html; charset="GB2312"
Content-Transfer-Encoding: base64
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--=_alternative 0023736548257B03_=--

From zhai.hongjun@zte.com.cn  Wed Jan 30 00:12:04 2013
Return-Path: <zhai.hongjun@zte.com.cn>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0656821F8B35; Wed, 30 Jan 2013 00:12:04 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -96.999
X-Spam-Level: 
X-Spam-Status: No, score=-96.999 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, MIME_BASE64_TEXT=1.753, MIME_CHARSET_FARAWAY=2.45, USER_IN_WHITELIST=-100, WEIRD_QUOTING=1.396]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id WgDBvkDXQXEm; Wed, 30 Jan 2013 00:12:03 -0800 (PST)
Received: from zte.com.cn (mx5.zte.com.cn [63.217.80.70]) by ietfa.amsl.com (Postfix) with ESMTP id 8EA8E21F8B34; Wed, 30 Jan 2013 00:12:02 -0800 (PST)
Received: from zte.com.cn (unknown [192.168.168.119]) by Websense Email Security Gateway with ESMTP id 417C5125FC4D; Wed, 30 Jan 2013 16:12:50 +0800 (CST)
Received: from mse01.zte.com.cn (unknown [10.30.3.20]) by Websense Email Security Gateway with ESMTPS id 54B05193A682; Wed, 30 Jan 2013 16:13:58 +0800 (CST)
Received: from notes_smtp.zte.com.cn ([10.30.1.239]) by mse01.zte.com.cn with ESMTP id r0U8BaaB063208; Wed, 30 Jan 2013 16:11:36 +0800 (GMT-8) (envelope-from zhai.hongjun@zte.com.cn)
In-Reply-To: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com>
To: Radia Perlman <radiaperlman@gmail.com>
MIME-Version: 1.0
X-KeepSent: 84D5F30E:ED4AC6D3-48257B03:0026CB28; type=4; name=$KeepSent
X-Mailer: Lotus Notes Release 6.5.6 March 06, 2007
Message-ID: <OF84D5F30E.ED4AC6D3-ON48257B03.0026CB28-48257B03.002D59F3@zte.com.cn>
From: zhai.hongjun@zte.com.cn
Date: Wed, 30 Jan 2013 16:11:39 +0800
X-MIMETrack: Serialize by Router on notes_smtp/zte_ltd(Release 8.5.3FP1 HF212|May 23, 2012) at 2013-01-30 16:11:31, Serialize complete at 2013-01-30 16:11:31
Content-Type: multipart/alternative; boundary="=_alternative 002D59F048257B03_="
X-MAIL: mse01.zte.com.cn r0U8BaaB063208
Cc: trill-bounces@ietf.org, trill@ietf.org
Subject: Re: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 30 Jan 2013 08:12:04 -0000

This is a multipart message in MIME format.
--=_alternative 002D59F048257B03_=
Content-Type: text/plain; charset="GB2312"
Content-Transfer-Encoding: base64

SW4gdGhlIHByb2Nlc3Mgb2YgYW4gY3VycmVudCBUUklMTCBjYW1wdXMgbWlncmF0ZXMgc21vb3Ro
bHkgdG8gRkdMIGNhbXB1cywgDQppdCBpcyBwb3NzaWJsZSB0aGF0IHRoZSBjb3JlIFJCcmlnZXMg
YXJlIGZpcnN0IGNoYW5nZWQgdG8gRkdMLVJCcmlnZXMgKG9yIA0KRkdMLXNhZmUgUmJyaWRnZXMp
LCB0aGVuIHNvbWUgZWRnZSBSYnJpZGdlcyBhcmUgY2hhbmdlZCB0byBGR0wuIA0KDQpEdXJpbmcg
dGhlIHByb2Nlc3Mgb2YgdGhhdCBtaWdyYXRpb24sIFZMIHRyYWZmaWMgYW5kIEZHTCB0cmFmZmlj
IHdpbGwgZXhpdCANCmluIHRoZSBUUklMTCBjYW1wdXMgc2ltdWx0YW5lb3VzbHkuIE9wdGlvbiAz
IHByb3ZpZGVzIGEgd2F5IGZvciB0aGUgdHdvIA0KdHlwZXMgb2YgdHJhZmZpYyBiZWluZyBmb3J3
YXJkZWQgaW4gdGhhdCBUUklMTCBjYW1wdXMgc2FmZWx5LiANCg0KRnVydGhlcm1vcmUsIEkgYW0g
bm90IHN1cmUgdGhhdCBhIFZMLVJCcmlkZ2UgY2FuIGJlIGNoYW5nZWQgdG8gRkdMLXNhZmUgDQpq
dXN0IGJ5IHVwZGF0aW5nIGl0cyBzb2Z0d2FyZS4NCg0KU28gSSBwcmVmZXIgT3B0aW9uIDMsIGFs
dGhvdWdoIGl0IHNlZW1zIG1vcmUgY29tcGxleCB0aGFuIE9wdGlvbiAyLg0KDQoNCg0KQmVzdCBS
ZWdhcmRzLA0KWmhhaSBIb25nanVuDQoiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIi
IiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiDQogUHJvdG9jb2wgRGV2ZWxvcG1l
bnQgRGVwdC5WSSwgQ2VudHJhbCBSJkQgSW5zdGl0dXRlLCBaVEUgQ29ycG9yYXRpb24NCiBOby4g
NjgsIFppamluZ2h1YSBSb2FkLCBZdWh1YXRhaSBEaXN0cmljdCwgTmFuamluZywgUC5SLkNoaW5h
LCAyMTAwMTINCiANCiBaaGFpIEhvbmdqdW4NCiANCiBUZWw6ICs4Ni0yNS01Mjg3NzM0NQ0KIEVt
YWlsOiB6aGFpLmhvbmdqdW5AenRlLmNvbS5jbg0KIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIi
IiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiINCg0KDQoNCg0KDQpS
YWRpYSBQZXJsbWFuIDxyYWRpYXBlcmxtYW5AZ21haWwuY29tPiANCreivP7IyzogIHRyaWxsLWJv
dW5jZXNAaWV0Zi5vcmcNCjIwMTMtMDEtMjYgMDM6MDANCg0KytW8/sjLDQp0cmlsbEBpZXRmLm9y
Zw0Ks63LzQ0KDQrW98ziDQpbdHJpbGxdIEV4cGxhaW5pbmcgdGhyZWUgb3B0aW9ucyBmb3IgdXBn
cmFkaW5nIHRvIEZHTCANCihmaW5lLWdyYWluZWQtbGFiZWxpbmcpDQoNCg0KDQoNCg0KDQpJJ20g
Z29pbmcgdG8gc3VtbWFyaXplIHRocmVlIG9wdGlvbnMgZm9yIHBoYXNpbmctaW4gRkdMLCBhbmQg
ZXhwbGFpbiB0aGUgDQp0cmFkZW9mZnMsIHRvIGhlbHAgcGVvcGxlIGhhdmUgaW5mb3JtZWQgb3Bp
bmlvbnMgYWJvdXQgdGhlIGltcGxpY2F0aW9ucyBvZiANCnRoZSB0aHJlZSBhcHByb2FjaGVzLiAg
VGhlcmUgYXJlIGluZmluaXRlIHZhcmlhdGlvbnMsIG9mIGNvdXJzZSwgYnV0IGFzIEkgDQpzYWlk
LCBJJ2xsIGRlc2NyaWJlIHRocmVlIG1haW4gb25lcy4NCg0KT3B0aW9uIDE6ICBPcmlnaW5hbCBk
cmFmdCwgbm8gY2hhbmdlcy4gIFRoZXJlIGFyZSB0d28gdHlwZXMgb2YgUkJyaWRnZXMuIA0KVGhl
IGZpcnN0IHR5cGUgaXMgIlZMIiB3aGljaCBvbmx5IHVuZGVyc3RhbmRzIFZMQU4gdGFncywgYW5k
IGhhdmUgdW5rbm93biANCmFuZCBwb3NzaWJseSBkYW5nZXJvdXMgYmVoYXZpb3Igd2hlbiB0aGV5
IHJlY2VpdmUgYW4gRkdMLWxhYmVsZWQgcGFja2V0LiANCihlLmcuLCBkZWNhcHN1bGF0ZSBhIHBh
Y2tldCBvbnRvIGEgbGluayB3aGVyZSBpdCBpcyBub3QgYWxsb3dlZCwgb3IgDQptaXN0YWtlbmx5
IHBydW5lIGEgbXVsdGljYXN0IHNvIGl0IGRvZXMgbm90IHJlYWNoIGV2ZXJ5d2hlcmUgaXQgc2hv
dWxkKS4gDQpUaGUgc2Vjb25kIHR5cGUgaXMgIkZHTCIuICBGR0wgZ3V5cyBuZWVkIHRvIGRvIHR3
byB0aGluZ3M6IDEpIHVuZGVyc3RhbmQgDQpGR0wgdGFncyBhbmQgZG8gdGhlIHJpZ2h0IHRoaW5n
IHdpdGggdGhlbSwgYW5kIDIpIG9zdHJhc2l6ZSBWTCANCmd1eXMuLi5tZWFuaW5nIHRoYXQgYW4g
RkdMIGd1eSByZWZ1c2VzIHRvIGZvcm0gYW4gYWRqYWNlbmN5IHdpdGggYSBWTCBndXkuIA0KIFRo
aXMgb3B0aW9uIGlzIHZlcnkgc2ltcGxlIGZvciBpbXBsZW1lbnRlcnMgb2YgRkdMLCBidXQgdGhl
IGltcGxpY2F0aW9uIA0KaXMgdGhhdCB5b3UgaGF2ZSB0byB1cGdyYWRlIHlvdXIgZW50aXJlIGNh
bXB1cyB0byBGR0wgYXQgb25jZS4gIFRoZXJlIGlzIA0Kbm8gY29leGlzdGVuY2UuDQoNCk9wdGlv
biAyOiBUd28gdHlwZXMgb2YgUkJyaWRnZXMsIGJ1dCB0aGUgMm5kIHR5cGUgaXMgZGlmZmVyZW50
IGZyb20gb3B0aW9uIA0KMS4gIFZMIGd1eXMgb2YgY291cnNlIGFyZSB0aGUgc2FtZSBhcyBpbiBv
cHRpb24gMS4uLnRoZXkgY2Fubm90IGJlIHRydXN0ZWQgDQp3aXRoIEZHTCBmcmFtZXMuIFRoZSBz
ZWNvbmQgdHlwZSBJIHdpbGwgY2FsbCAiRkdMLXNhZmUiLiBBbiBGR0wtc2FmZSANClJCcmlkZ2Ug
bXVzdCBhZHZlcnRpc2UgaW4gaXRzIExTUCB0aGF0IGl0IGlzIEZHTC1zYWZlLCBhbmQgaXQgbXVz
dCBub3QgImRvIA0KYW55dGhpbmcgYmFkIiB3aXRoIEZHTCBmcmFtZXMsIG1lYW5pbmcgdGhhdCBp
dCBpcyBhbGxvd2VkIHRvIGlnbm9yZSANCnBydW5pbmcgb2YgRkdMIChvciBldmVuIFZMIGZyYW1l
cykgZW50aXJlbHkuLi5pdCBqdXN0IGNhbid0IGZhbHNlbHkgZHJvcCANCkZHTCBmcmFtZXMuICBB
bmQgaXQgbXVzdCBub3QgZGVjYXBzdWxhdGUgYW4gRkdMIGZyYW1lIG9udG8gYSBsaW5rIGZvciAN
CndoaWNoIHRoYXQgRkdMIGZyYW1lIGRvZXNuJ3QgYmVsb25nLiBJbiB0aGlzIG9wdGlvbiwgYWxs
IFJCcmlkZ2VzIG11c3QgYmUgDQp1cGdyYWRlZCB0byBGR0wtc2FmZSwgYnV0IGl0IG5lZWQgbm90
IGhhcHBlbiBpbnN0YW50YW5lb3VzbHkuLi5pdCdzIGZpbmUgDQp0byBtaXggRkdMLXNhZmUgUkJy
aWRnZXMgd2l0aCBWTCBSQnJpZGdlcy4uLml0J3MganVzdCBub3Qgc2FmZSB0byBpbmplY3QgDQpG
R0wgZnJhbWVzIHlldC4gIE9uY2UgYWxsIHRoZSBSQnJpZGdlcyBoYXZlIGJlZW4gdXBncmFkZWQg
dG8gYmUgRkdMLXNhZmUsIA0KdGhlbiBlZGdlIFJCcmlkZ2VzIGNhbiBzdGFydCBhbm5vdW5jaW5n
IHRoZXkgYXJlIGNvbm5lY3RlZCB0byBhbiBGR0wgbGluaywgDQphbmQgY2FuIHN0YXJ0IGluamVj
dGluZyBGR0wgcGFja2V0cy4gIEl0IGlzIGNvbnNpZGVyZWQgYSBtaXNjb25maWd1cmF0aW9uIA0K
aWYgeW91IHN0YXJ0IGluamVjdGluZyBGR0wgZnJhbWVzIGJlZm9yZSBhbGwgdGhlIFJCcmlkZ2Vz
IGFyZSB1cGdyYWRlZCB0byANCkZHTC1zYWZlLCBzbyBhbiBhZGRpdGlvbmFsIGNob3JlIGZvciBh
biBGR0wtc2FmZSBSQnJpZGdlIFIxIGlzIHRvIGV4YW1pbmUgDQpMU1BzLCBhbmQgaWYgYW55IFJC
cmlkZ2UgY2xhaW1zIHRvIGJlIGF0dGFjaGVkIHRvIGFuIEZHTCBsaW5rLCB0aGVuIFIxIA0KbXVz
dCBvc3RyYXNpemUgYW55IFZMIG5laWdoYm9ycy4gIChkb24ndCBzdGFydCBvc3RyYXNpemluZyBW
TCBndXlzIHVudGlsIA0KaXQgaXMgbmVjZXNzYXJ5IGJlY2F1c2Ugb2YgYWN0dWFsbHkgc3RhcnRp
bmcgdG8gdXNlIEZHTCBmcmFtZXMsIGluIG90aGVyIA0Kd29yZHMpLiBUaGlzIG9wdGlvbiBpcyBt
b3JlIHdvcmsgZm9yIHRoZSB1cGdyYWRlZCBSQnJpZGdlcyB0aGFuIG9wdGlvbiAxLCANCmFuZCB0
aGVyZSBzdGlsbCBpc24ndCBnb29kIGNvZXhpc3RlbmNlIHdpdGggVkwgZ3V5cyBsb25nLXRlcm0g
KGFzIG9wdGlvbiAzIA0Kd2lsbCksIGJ1dCBpdCBkb2VzIGFsbG93IHVwZ3JhZGluZyBSQnJpZGdl
cyBvbmUgYnkgb25lIGluIGEgd29ya2luZyBjYW1wdXMgDQp3aXRob3V0IGNhdXNpbmcgZGlzcnVw
dGlvbi4NCg0KT3B0aW9uIDM6IFR3byB0eXBlcyBvZiBSQnJpZGdlcy4gIFZMLCBvZiBjb3Vyc2Us
IGlzIHRoZSBzYW1lIGFzIGluIG9wdGlvbnMgDQoxIGFuZCAyLiAgVGhpcyBvcHRpb24gbWFrZXMg
RkdMIGd1eXMgZG8gbW9yZSB3b3JrLCBidXQgYWxsb3dzIG1heGltYWwgbG9uZyANCnRlcm0gY29l
eGlzdGVuY2Ugb2YgVkwgYW5kIEZHTCBndXlzLiBJbiB0aGlzIG9wdGlvbiwgRkdMIGd1eXMgY2Fs
Y3VsYXRlIA0KdW5pY2FzdCBwYXRocyB0byBGR0wgZWRnZSBndXlzIHRoYXQgYXZvaWQgYW55IFZM
IGd1eXMsIGFuZCBjYWxjdWxhdGUgYXQgDQpsZWFzdCBvbmUgRkdMLWZyaWVuZGx5IG11bHRpY2Fz
dCB0cmVlIHRoYXQgYWxzbyBhdm9pZHMgYW55IFZMIGd1eXMuIFNvIA0KbGV0J3Mgc2F5IFIxIGlz
IGFuIEZHTCBndXkuICBSMSBkaXNjYXJkcyBhbGwgTFNQcyBmcm9tIFZMIGd1eXMgKG9uZXMgdGhh
dCANCmRvbid0IGFkdmVydGlzZSBGR0wgY2FwYWJpbGl0eSBpbiB0aGVpciBMU1ApLCB3aGVuIGNh
bGN1bGF0aW5nIHBhdGhzIHRvIA0Kb3RoZXIgRkdMIFJCcmlkZ2VzLiBUaGVuIFIxIGNhbGN1bGF0
ZXMgcGF0aHMgdG8gdGhlIFZMIGd1eXMgdXNpbmcgYWxsIA0KTFNQcy4gIExpa2V3aXNlLCB3aGVu
IGNhbGN1bGF0aW5nIGFuIEZHTC1mcmllbmRseSB0cmVlLCBSMSBjYWxjdWxhdGVzIGEgDQp0cmVl
IHRocm91Z2ggb25seSBGR0wgZ3V5cy4gICBUaGlzIG9wdGlvbiBpcyBtb3JlIHdvcmsgZm9yIHRo
ZSB1cGdyYWRlZCANClJCcmlkZ2VzIChiZWNhdXNlIHRoZXkgaGF2ZSB0byBjYWxjdWxhdGUgRGlq
a3N0cmEgaW4gdHdvIGRpZmZlcmVudCB3YXlzLCANCm9uZSBmb3IgcmVhY2hpbmcgVkwgZ3V5cywg
YW5kIG9uZSBmb3IgcmVhY2hpbmcgRkdMIGd1eXMpLiAgSG93ZXZlciwgaXQgDQpkb2VzIGFsbG93
IGhhdmluZyBsb25nLXRlcm0gY29leGlzdGVuY2Ugd2l0aCBWTCBndXlzLiAgRm9yIGluc3RhbmNl
LCB5b3UgDQpjb3VsZCBmb3JldmVyIGtlZXAgc29tZSBWTCBlZGdlIFJCcmlkZ2VzIHRoYXQgY29t
bXVuaWNhdGUganVzdCBmaW5lIA0KdGhyb3VnaCBhbiBGR0wgY29yZS4gIFRoZXkgY2FuIHN0YXkg
dGhlcmUgZm9yZXZlciwgYW5kIHN0aWxsIGJlIGFibGUgdG8gDQpjb21tdW5pY2F0ZSB0aHJvdWdo
IHRoZSBjb3JlIHRvIGFsbCB0aGUgbGlua3MgYXR0YWNoaW5nIHRvIHRoZWlyIFZMQU4uDQpfX19f
X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fXw0KdHJpbGwgbWFpbGlu
ZyBsaXN0DQp0cmlsbEBpZXRmLm9yZw0KaHR0cHM6Ly93d3cuaWV0Zi5vcmcvbWFpbG1hbi9saXN0
aW5mby90cmlsbA0KDQoNCg==
--=_alternative 002D59F048257B03_=
Content-Type: text/html; charset="GB2312"
Content-Transfer-Encoding: base64
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--=_alternative 002D59F048257B03_=--

From aldrin.ietf@gmail.com  Wed Jan 30 11:55:37 2013
Return-Path: <aldrin.ietf@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3D3C821F88B6 for <trill@ietfa.amsl.com>; Wed, 30 Jan 2013 11:55:37 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.599
X-Spam-Level: 
X-Spam-Status: No, score=-3.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 5u+vKxXL+D1M for <trill@ietfa.amsl.com>; Wed, 30 Jan 2013 11:55:36 -0800 (PST)
Received: from mail-pb0-f51.google.com (mail-pb0-f51.google.com [209.85.160.51]) by ietfa.amsl.com (Postfix) with ESMTP id 3524821F88A9 for <trill@ietf.org>; Wed, 30 Jan 2013 11:55:36 -0800 (PST)
Received: by mail-pb0-f51.google.com with SMTP id un15so1153447pbc.24 for <trill@ietf.org>; Wed, 30 Jan 2013 11:55:35 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=x-received:content-type:mime-version:subject:from:in-reply-to:date :cc:content-transfer-encoding:message-id:references:to:x-mailer; bh=g/3F7QEGvrPqgPLpBdvfFddHZ8VBdnOcW08OWh6yuHU=; b=ZQ2AvZnHe4CEmfzRhu29UwC+Nc1pSZgv+qNk4LDWYLXuqSpaL4CHTX4uuzepsGP+Aj kmF3winsL+IKdnT97aJxd6dITnBliO0A7FowSTS/FDJMC7afix3kZmHB0idTow+XBqwk oOKa9VUcupvkgSzpl835NVB6iy2BzQPclTF6Oqu1vzBKJ6bkwQKNb+wgOE1AOCN7KALF eOttLpLRMwJkLiqscAh69KdALsaUEz4UAk/u3d2FqjI3w4mjVmGBqpAtsV0nN57KKOM1 bvLLKVwPK2ndx+gFqsAaPy0P/MesSawWRjyuYbcYl5z+xn5tKjfkYQ4qwe+cI9YM5+3J Geag==
X-Received: by 10.68.219.36 with SMTP id pl4mr15493703pbc.124.1359575735272; Wed, 30 Jan 2013 11:55:35 -0800 (PST)
Received: from [192.168.255.224] (107-1-141-74-ip-static.hfc.comcastbusiness.net. [107.1.141.74]) by mx.google.com with ESMTPS id rk6sm2466477pbc.20.2013.01.30.11.55.33 (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Wed, 30 Jan 2013 11:55:34 -0800 (PST)
Content-Type: text/plain; charset=us-ascii
Mime-Version: 1.0 (Mac OS X Mail 6.2 \(1499\))
From: Sam Aldrin <aldrin.ietf@gmail.com>
In-Reply-To: <CAF4+nEFLrJL=CgX9e6on-cgz89ugmu93P6u5kmODhUia3H7RLg@mail.gmail.com>
Date: Wed, 30 Jan 2013 11:55:35 -0800
Content-Transfer-Encoding: quoted-printable
Message-Id: <1F5A81AE-4B3C-45FD-B431-9BA0011EB085@gmail.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <CA+-tSzx0E3J4i5Dvfbj4egruCeGi7z1se1Jn5RXtx3RedQLPdg@mail.gmail.com> <CAHD03N89RNX4=XxRZpD+Y2PCktYf-FtVjyYJ5oFS3dui3SsXaA@mail.gmail.com> <CAF4+nEG7YKGzGmObAMaOxackivjJsALymqHX=6PKBd5qU-Z+zQ@mail.gmail.com> <CA+-tSzz0=oh6iB1e-ESyeOx2sCvBFXeVpzuNbKmUVhRNLScDgQ@mail.gmail.com> <CAF4+nEFLrJL=CgX9e6on-cgz89ugmu93P6u5kmODhUia3H7RLg@mail.gmail.com>
To: Donald Eastlake <d3e3e3@gmail.com>
X-Mailer: Apple Mail (2.1499)
Cc: Anoop Ghanwani <anoop@alumni.duke.edu>, Radia Perlman <radiaperlman@gmail.com>, Ayan Banerjee <ayabaner@gmail.com>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 30 Jan 2013 19:55:37 -0000

[still catching up on email, so the delay]

When everyone says, upgrade RBridges, what is the upgrade model they =
have in mind?
How can one do software upgrade when hardware/silicon cannot support?
Secondly, any actual users, not vendors, commented on how they plan to =
deploy FGL?

-sam

On Jan 29, 2013, at 4:37 PM, Donald Eastlake <d3e3e3@gmail.com> wrote:

> Hi Anoop,
>=20
> On Tue, Jan 29, 2013 at 7:03 PM, Anoop Ghanwani =
<anoop@alumni.duke.edu> wrote:
>> On Tue, Jan 29, 2013 at 12:49 PM, Donald Eastlake <d3e3e3@gmail.com> =
wrote:
>>=20
>>> The general assumption is that you are migrating from VL to FGL. =
First
>>> you migrate to FGL-safe. Then you unleash the FGL traffic, by
>>> configuring FGL ports, which isolates any remaining (FGL-unsafe) VL
>>> switches while still being able to handle VL traffic between VL =
ports
>>> on FGL-safe switches.
>>=20
>> Based on what is written so far, it looks like as long as there is =
even
>> one VL RBridge in it's database, an RBridge would not be able to =
start
>> sourcing FGL traffic.  It may start doing so only when all RBridges =
in
>=20
> Well, it can start sourcing FGL traffic but their are consequences as
> you note below.
>=20
>> its database are FGL-safe.  If we ever get into a condition where
>> an RBridge has started advertising FGL information, at that point,
>> does any other RBridge in the campus simply isolate any VL RBridge
>> (with the assumption that there was some kind of race condition?
>=20
> Yes. (This could just be due to some misconfiguration.)
>=20
>> The case I'm concerned about is we have 2 FGL RBridges, RB1 and RB2.
>> RB1 is connected to a campus of n RBs that are still VL only.  The VL =
link
>> Breaks, and RB1 starts advertising FGL information to RB2, but then
>> the link to the VL campus comes back up.  Does this mean RB1 and RB2
>> remain isolated from the campus (and operate as they own mini-campus)
>> until all the other RBridges are converted over to FGL-safe?
>=20
> Yes. But I don't think this as much of a problem given that I believe
> existing TRILL switches can be software upgraded to be FGL-safe. I
> view the configuration of FGL ports on RB1/RB2 and the consequent
> routing of actual FGL traffic to be an active network management
> decision, not something that might just happen because there appear to
> be no more VL TRILL switches in the topology.
>=20
> Thanks,
> Donald
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D
> Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
> 155 Beaver Street, Milford, MA 01757 USA
> d3e3e3@gmail.com
>=20
>> Anoop
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill


From linda.dunbar@huawei.com  Wed Jan 30 12:17:57 2013
Return-Path: <linda.dunbar@huawei.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D113921F87D7 for <trill@ietfa.amsl.com>; Wed, 30 Jan 2013 12:17:57 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.598
X-Spam-Level: 
X-Spam-Status: No, score=-6.598 tagged_above=-999 required=5 tests=[AWL=-0.000, BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ywlcV-gTAXeP for <trill@ietfa.amsl.com>; Wed, 30 Jan 2013 12:17:57 -0800 (PST)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) by ietfa.amsl.com (Postfix) with ESMTP id 070B121F87CA for <trill@ietf.org>; Wed, 30 Jan 2013 12:17:55 -0800 (PST)
Received: from 172.18.7.190 (EHLO lhreml203-edg.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.5-GA FastPath queued) with ESMTP id APG44481; Wed, 30 Jan 2013 20:17:54 +0000 (GMT)
Received: from LHREML402-HUB.china.huawei.com (10.201.5.241) by lhreml203-edg.huawei.com (172.18.7.221) with Microsoft SMTP Server (TLS) id 14.1.323.7; Wed, 30 Jan 2013 20:17:15 +0000
Received: from DFWEML405-HUB.china.huawei.com (10.193.5.102) by lhreml402-hub.china.huawei.com (10.201.5.241) with Microsoft SMTP Server (TLS) id 14.1.323.7; Wed, 30 Jan 2013 20:17:54 +0000
Received: from DFWEML505-MBX.china.huawei.com ([10.124.31.100]) by dfweml405-hub.china.huawei.com ([10.193.5.102]) with mapi id 14.01.0323.007; Wed, 30 Jan 2013 12:17:47 -0800
From: Linda Dunbar <linda.dunbar@huawei.com>
To: Tal Mizrahi <talmi@marvell.com>, Radia Perlman <radiaperlman@gmail.com>, "trill@ietf.org" <trill@ietf.org>
Thread-Topic: [trill] Explaining three options for upgrading to	FGL (fine-grained-labeling)
Thread-Index: AQHN/qaR4UWGgmXTIECI+Xc5tmBPVphiUFHQ
Date: Wed, 30 Jan 2013 20:17:47 +0000
Message-ID: <4A95BA014132FF49AE685FAB4B9F17F6451010BB@dfweml505-mbx>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <74470498B659FA4687F0B0018C19A89C01A0F93685EF@IL-MB01.marvell.com>
In-Reply-To: <74470498B659FA4687F0B0018C19A89C01A0F93685EF@IL-MB01.marvell.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.192.11.236]
Content-Type: multipart/alternative; boundary="_000_4A95BA014132FF49AE685FAB4B9F17F6451010BBdfweml505mbx_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Subject: Re: [trill] Explaining three options for upgrading to	FGL	(fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 30 Jan 2013 20:17:57 -0000

--_000_4A95BA014132FF49AE685FAB4B9F17F6451010BBdfweml505mbx_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

I think Option 2 is better.

Linda Dunbar

From: trill-bounces@ietf.org [mailto:trill-bounces@ietf.org] On Behalf Of R=
adia Perlman
Sent: Friday, January 25, 2013 9:00 PM
To: trill@ietf.org
Subject: [trill] Explaining three options for upgrading to FGL (fine-graine=
d-labeling)

I'm going to summarize three options for phasing-in FGL, and explain the tr=
adeoffs, to help people have informed opinions about the implications of th=
e three approaches.  There are infinite variations, of course, but as I sai=
d, I'll describe three main ones.

Option 1:  Original draft, no changes.  There are two types of RBridges. Th=
e first type is "VL" which only understands VLAN tags, and have unknown and=
 possibly dangerous behavior when they receive an FGL-labeled packet. (e.g.=
, decapsulate a packet onto a link where it is not allowed, or mistakenly p=
rune a multicast so it does not reach everywhere it should). The second typ=
e is "FGL".  FGL guys need to do two things: 1) understand FGL tags and do =
the right thing with them, and 2) ostrasize VL guys...meaning that an FGL g=
uy refuses to form an adjacency with a VL guy.  This option is very simple =
for implementers of FGL, but the implication is that you have to upgrade yo=
ur entire campus to FGL at once.  There is no coexistence.

Option 2: Two types of RBridges, but the 2nd type is different from option =
1.  VL guys of course are the same as in option 1...they cannot be trusted =
with FGL frames. The second type I will call "FGL-safe". An FGL-safe RBridg=
e must advertise in its LSP that it is FGL-safe, and it must not "do anythi=
ng bad" with FGL frames, meaning that it is allowed to ignore pruning of FG=
L (or even VL frames) entirely...it just can't falsely drop FGL frames.  An=
d it must not decapsulate an FGL frame onto a link for which that FGL frame=
 doesn't belong. In this option, all RBridges must be upgraded to FGL-safe,=
 but it need not happen instantaneously...it's fine to mix FGL-safe RBridge=
s with VL RBridges...it's just not safe to inject FGL frames yet.  Once all=
 the RBridges have been upgraded to be FGL-safe, then edge RBridges can sta=
rt announcing they are connected to an FGL link, and can start injecting FG=
L packets.  It is considered a misconfiguration if you start injecting FGL =
frames before all the RBridges are upgraded to FGL-safe, so an additional c=
hore for an FGL-safe RBridge R1 is to examine LSPs, and if any RBridge clai=
ms to be attached to an FGL link, then R1 must ostrasize any VL neighbors. =
 (don't start ostrasizing VL guys until it is necessary because of actually=
 starting to use FGL frames, in other words). This option is more work for =
the upgraded RBridges than option 1, and there still isn't good coexistence=
 with VL guys long-term (as option 3 will), but it does allow upgrading RBr=
idges one by one in a working campus without causing disruption.

Option 3: Two types of RBridges.  VL, of course, is the same as in options =
1 and 2.  This option makes FGL guys do more work, but allows maximal long =
term coexistence of VL and FGL guys. In this option, FGL guys calculate uni=
cast paths to FGL edge guys that avoid any VL guys, and calculate at least =
one FGL-friendly multicast tree that also avoids any VL guys. So let's say =
R1 is an FGL guy.  R1 discards all LSPs from VL guys (ones that don't adver=
tise FGL capability in their LSP), when calculating paths to other FGL RBri=
dges. Then R1 calculates paths to the VL guys using all LSPs.  Likewise, wh=
en calculating an FGL-friendly tree, R1 calculates a tree through only FGL =
guys.   This option is more work for the upgraded RBridges (because they ha=
ve to calculate Dijkstra in two different ways, one for reaching VL guys, a=
nd one for reaching FGL guys).  However, it does allow having long-term coe=
xistence with VL guys.  For instance, you could forever keep some VL edge R=
Bridges that communicate just fine through an FGL core.  They can stay ther=
e forever, and still be able to communicate through the core to all the lin=
ks attaching to their VLAN.

--_000_4A95BA014132FF49AE685FAB4B9F17F6451010BBdfweml505mbx_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micr=
osoft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:x=3D"urn:schemas-microsoft-com:office:excel" xmlns:m=3D"http://schema=
s.microsoft.com/office/2004/12/omml" xmlns=3D"http://www.w3.org/TR/REC-html=
40">
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Dus-ascii"=
>
<meta name=3D"Generator" content=3D"Microsoft Word 12 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
	{font-family:SimSun;
	panose-1:2 1 6 0 3 1 1 1 1 1;}
@font-face
	{font-family:"Cambria Math";
	panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:Tahoma;
	panose-1:2 11 6 4 3 5 4 4 2 4;}
@font-face
	{font-family:SimSun;
	panose-1:2 1 6 0 3 1 1 1 1 1;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman","serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
span.EmailStyle17
	{mso-style-type:personal;
	font-family:"Calibri","sans-serif";
	color:#1F497D;}
span.EmailStyle18
	{mso-style-type:personal-reply;
	font-family:"Calibri","sans-serif";
	color:#1F497D;}
.MsoChpDefault
	{mso-style-type:export-only;
	font-size:10.0pt;}
@page WordSection1
	{size:8.5in 11.0in;
	margin:1.0in 1.25in 1.0in 1.25in;}
div.WordSection1
	{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang=3D"EN-US" link=3D"blue" vlink=3D"purple">
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">I think Option 2 is bette=
r.
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D">Linda Dunbar<o:p></o:p></=
span></p>
<div style=3D"border:none;border-left:solid blue 1.5pt;padding:0in 0in 0in =
4.0pt">
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span><=
/p>
<p class=3D"MsoNormal"><b><span style=3D"font-size:10.0pt;font-family:&quot=
;Tahoma&quot;,&quot;sans-serif&quot;">From:</span></b><span style=3D"font-s=
ize:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;"> trill-bo=
unces@ietf.org [mailto:trill-bounces@ietf.org]
<b>On Behalf Of </b>Radia Perlman<br>
<b>Sent:</b> Friday, January 25, 2013 9:00 PM<br>
<b>To:</b> trill@ietf.org<br>
<b>Subject:</b> [trill] Explaining three options for upgrading to FGL (fine=
-grained-labeling)<o:p></o:p></span></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<div>
<p class=3D"MsoNormal" style=3D"background:white"><span style=3D"font-size:=
10.0pt;font-family:&quot;Arial&quot;,&quot;sans-serif&quot;;color:#222222">=
I'm going to summarize three options for phasing-in FGL, and explain the tr=
adeoffs, to help people have informed opinions about the implications
 of the three approaches. &nbsp;There are infinite variations, of course, b=
ut as I said, I'll describe three main ones.<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"background:white"><span style=3D"font-size:=
10.0pt;font-family:&quot;Arial&quot;,&quot;sans-serif&quot;;color:#222222">=
<o:p>&nbsp;</o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"background:white"><span style=3D"font-size:=
10.0pt;font-family:&quot;Arial&quot;,&quot;sans-serif&quot;;color:#222222">=
Option 1: &nbsp;Original draft, no changes. &nbsp;There are two types of RB=
ridges. The first type is &quot;VL&quot; which only understands VLAN tags, =
and
 have unknown and possibly dangerous behavior when they receive an FGL-labe=
led packet. (e.g., decapsulate a packet onto a link where it is not allowed=
, or mistakenly prune a multicast so it does not reach everywhere it should=
). The second type is &quot;FGL&quot;. &nbsp;FGL
 guys need to do two things: 1) understand FGL tags and do the right thing =
with them, and 2) ostrasize VL guys...meaning that an FGL guy refuses to fo=
rm an adjacency with a VL guy. &nbsp;This option is very simple for impleme=
nters of FGL, but the implication is
 that you have to upgrade your entire campus to FGL at once. &nbsp;There is=
 no coexistence.<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"background:white"><span style=3D"font-size:=
10.0pt;font-family:&quot;Arial&quot;,&quot;sans-serif&quot;;color:#222222">=
<o:p>&nbsp;</o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"background:white"><span style=3D"font-size:=
10.0pt;font-family:&quot;Arial&quot;,&quot;sans-serif&quot;;color:#222222">=
Option 2: Two types of RBridges, but the 2nd type is different from option =
1. &nbsp;VL guys of course are the same as in option 1...they cannot
 be trusted with FGL frames. The second type I will call &quot;FGL-safe&quo=
t;. An FGL-safe RBridge must advertise in its LSP that it is FGL-safe, and =
it must not &quot;do anything bad&quot; with FGL frames, meaning that it is=
 allowed to ignore pruning of FGL (or even VL frames)
 entirely...it just can't falsely drop FGL frames. &nbsp;And it must not de=
capsulate an FGL frame onto a link for which that FGL frame doesn't belong.=
 In this option, all RBridges must be upgraded to FGL-safe, but it need not=
 happen instantaneously...it's fine to
 mix FGL-safe RBridges with VL RBridges...it's just not safe to inject FGL =
frames yet. &nbsp;Once all the RBridges have been upgraded to be FGL-safe, =
then edge RBridges can start announcing they are connected to an FGL link, =
and can start injecting FGL packets.
 &nbsp;It is considered a misconfiguration if you start injecting FGL frame=
s before all the RBridges are upgraded to FGL-safe, so an additional chore =
for an FGL-safe RBridge R1 is to examine LSPs, and if any RBridge claims to=
 be attached to an FGL link, then R1
 must ostrasize any VL neighbors. &nbsp;(don't start ostrasizing VL guys un=
til it is necessary because of actually starting to use FGL frames, in othe=
r words). This option is more work for the upgraded RBridges than option 1,=
 and there still isn't good coexistence
 with VL guys long-term (as option 3 will), but it does allow upgrading RBr=
idges one by one in a working campus without causing disruption.<o:p></o:p>=
</span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"background:white"><span style=3D"font-size:=
10.0pt;font-family:&quot;Arial&quot;,&quot;sans-serif&quot;;color:#222222">=
<o:p>&nbsp;</o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"background:white"><span style=3D"font-size:=
10.0pt;font-family:&quot;Arial&quot;,&quot;sans-serif&quot;;color:#222222">=
Option 3: Two types of RBridges. &nbsp;VL, of course, is the same as in opt=
ions 1 and 2. &nbsp;This option makes FGL guys do more work, but allows
 maximal long term coexistence of VL and FGL guys. In this option, FGL guys=
 calculate unicast paths to FGL edge guys that avoid any VL guys, and calcu=
late at least one FGL-friendly multicast tree that also avoids any VL guys.=
 So let's say R1 is an FGL guy.
 &nbsp;R1 discards all LSPs from VL guys (ones that don't advertise FGL cap=
ability in their LSP), when calculating paths to other FGL RBridges. Then R=
1 calculates paths to the VL guys using all LSPs. &nbsp;Likewise, when calc=
ulating an FGL-friendly tree, R1 calculates
 a tree through only FGL guys. &nbsp; This option is more work for the upgr=
aded RBridges (because they have to calculate Dijkstra in two different way=
s, one for reaching VL guys, and one for reaching FGL guys). &nbsp;However,=
 it does allow having long-term coexistence
 with VL guys. &nbsp;For instance, you could forever keep some VL edge RBri=
dges that communicate just fine through an FGL core. &nbsp;They can stay th=
ere forever, and still be able to communicate through the core to all the l=
inks attaching to their VLAN.<o:p></o:p></span></p>
</div>
</div>
</div>
</body>
</html>

--_000_4A95BA014132FF49AE685FAB4B9F17F6451010BBdfweml505mbx_--

From jon.hudson@gmail.com  Wed Jan 30 12:41:00 2013
Return-Path: <jon.hudson@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B377221F88C7 for <trill@ietfa.amsl.com>; Wed, 30 Jan 2013 12:41:00 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.203
X-Spam-Level: 
X-Spam-Status: No, score=-2.203 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, MIME_QP_LONG_LINE=1.396, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id HX4QQnvSconL for <trill@ietfa.amsl.com>; Wed, 30 Jan 2013 12:40:59 -0800 (PST)
Received: from mail-gh0-f179.google.com (mail-gh0-f179.google.com [209.85.160.179]) by ietfa.amsl.com (Postfix) with ESMTP id 454B221F88B6 for <trill@ietf.org>; Wed, 30 Jan 2013 12:40:59 -0800 (PST)
Received: by mail-gh0-f179.google.com with SMTP id r14so355888ghr.24 for <trill@ietf.org>; Wed, 30 Jan 2013 12:40:58 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=x-received:references:mime-version:in-reply-to:content-type :content-transfer-encoding:message-id:cc:x-mailer:from:subject:date :to; bh=dAd1u+UhClckytY4DjnSl/yN3ZXdHUmmwyJtGtSvNoA=; b=NmMX0wlrfZb8yi9gR/KvKPs8NxxWU25y0Mwxa5pC3Kcio1XBPzWcNc0XVhSQLjFwIx U3dkRSvtYOmvIXUo+SiWJz7IUqYGq2RMqXg2V74Ei3ZFdRln9LeHxZk4KQ73o/2HUk7Q RZAe/vMp5A0Jhye6mRsFnyQf93duHonY0KIdw7mAndh3uBUwGl9Yu1aRxIItltJXJrZs 3JDBkkfrWFX7ReftkjvsIAHBmtcw7dBb5/P5Xe2Vg6brS5nJWD+/9Hz7BLh1/s4xmEn3 2fnyAcBTZ4fj40OFgJ4QES9D4fmJzrkTtnru9GFVNmKlt8gLDrz70GcPj/3P+vU+/Tgp W1Eg==
X-Received: by 10.236.131.9 with SMTP id l9mr7304279yhi.93.1359578458610; Wed, 30 Jan 2013 12:40:58 -0800 (PST)
Received: from [10.163.181.190] (mobile-166-147-115-015.mycingular.net. [166.147.115.15]) by mx.google.com with ESMTPS id s3sm4108359yhm.10.2013.01.30.12.40.52 (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Wed, 30 Jan 2013 12:40:57 -0800 (PST)
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <CA+-tSzx0E3J4i5Dvfbj4egruCeGi7z1se1Jn5RXtx3RedQLPdg@mail.gmail.com> <CAHD03N89RNX4=XxRZpD+Y2PCktYf-FtVjyYJ5oFS3dui3SsXaA@mail.gmail.com> <CAF4+nEG7YKGzGmObAMaOxackivjJsALymqHX=6PKBd5qU-Z+zQ@mail.gmail.com> <CA+-tSzz0=oh6iB1e-ESyeOx2sCvBFXeVpzuNbKmUVhRNLScDgQ@mail.gmail.com> <CAF4+nEFLrJL=CgX9e6on-cgz89ugmu93P6u5kmODhUia3H7RLg@mail.gmail.com> <1F5A81AE-4B3C-45FD-B431-9BA0011EB085@gmail.com>
Mime-Version: 1.0 (1.0)
In-Reply-To: <1F5A81AE-4B3C-45FD-B431-9BA0011EB085@gmail.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: quoted-printable
Message-Id: <DC8898F8-75C0-4535-BA61-C54279901332@gmail.com>
X-Mailer: iPhone Mail (10B143)
From: Jon Hudson <jon.hudson@gmail.com>
Date: Thu, 31 Jan 2013 04:40:43 +0800
To: Sam Aldrin <aldrin.ietf@gmail.com>
Cc: Donald Eastlake <d3e3e3@gmail.com>, Anoop Ghanwani <anoop@alumni.duke.edu>, Radia Perlman <radiaperlman@gmail.com>, Ayan Banerjee <ayabaner@gmail.com>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 30 Jan 2013 20:41:00 -0000

Can you clarify on what you mean by deploy?

I ask because I have many customers waiting for this functionality. Everyone=
 of the ~1000 installs we have, that also have multitenancy environments pla=
n to deploy this as soon as possible.=20

But with no one have I gone through and done a Visio of how their Rbridges a=
re wired up today and how FGL rbridges would be integrated gracefully into t=
he environment.

On Jan 31, 2013, at 3:55 AM, Sam Aldrin <aldrin.ietf@gmail.com> wrote:

> [still catching up on email, so the delay]
>=20
> When everyone says, upgrade RBridges, what is the upgrade model they have i=
n mind?
> How can one do software upgrade when hardware/silicon cannot support?
> Secondly, any actual users, not vendors, commented on how they plan to dep=
loy FGL?
>=20
> -sam
>=20
> On Jan 29, 2013, at 4:37 PM, Donald Eastlake <d3e3e3@gmail.com> wrote:
>=20
>> Hi Anoop,
>>=20
>> On Tue, Jan 29, 2013 at 7:03 PM, Anoop Ghanwani <anoop@alumni.duke.edu> w=
rote:
>>> On Tue, Jan 29, 2013 at 12:49 PM, Donald Eastlake <d3e3e3@gmail.com> wro=
te:
>>>=20
>>>> The general assumption is that you are migrating from VL to FGL. First
>>>> you migrate to FGL-safe. Then you unleash the FGL traffic, by
>>>> configuring FGL ports, which isolates any remaining (FGL-unsafe) VL
>>>> switches while still being able to handle VL traffic between VL ports
>>>> on FGL-safe switches.
>>>=20
>>> Based on what is written so far, it looks like as long as there is even
>>> one VL RBridge in it's database, an RBridge would not be able to start
>>> sourcing FGL traffic.  It may start doing so only when all RBridges in
>>=20
>> Well, it can start sourcing FGL traffic but their are consequences as
>> you note below.
>>=20
>>> its database are FGL-safe.  If we ever get into a condition where
>>> an RBridge has started advertising FGL information, at that point,
>>> does any other RBridge in the campus simply isolate any VL RBridge
>>> (with the assumption that there was some kind of race condition?
>>=20
>> Yes. (This could just be due to some misconfiguration.)
>>=20
>>> The case I'm concerned about is we have 2 FGL RBridges, RB1 and RB2.
>>> RB1 is connected to a campus of n RBs that are still VL only.  The VL li=
nk
>>> Breaks, and RB1 starts advertising FGL information to RB2, but then
>>> the link to the VL campus comes back up.  Does this mean RB1 and RB2
>>> remain isolated from the campus (and operate as they own mini-campus)
>>> until all the other RBridges are converted over to FGL-safe?
>>=20
>> Yes. But I don't think this as much of a problem given that I believe
>> existing TRILL switches can be software upgraded to be FGL-safe. I
>> view the configuration of FGL ports on RB1/RB2 and the consequent
>> routing of actual FGL traffic to be an active network management
>> decision, not something that might just happen because there appear to
>> be no more VL TRILL switches in the topology.
>>=20
>> Thanks,
>> Donald
>> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D
>> Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
>> 155 Beaver Street, Milford, MA 01757 USA
>> d3e3e3@gmail.com
>>=20
>>> Anoop
>> _______________________________________________
>> trill mailing list
>> trill@ietf.org
>> https://www.ietf.org/mailman/listinfo/trill
>=20
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill

From aldrin.ietf@gmail.com  Wed Jan 30 15:25:37 2013
Return-Path: <aldrin.ietf@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6DC5221F87B3 for <trill@ietfa.amsl.com>; Wed, 30 Jan 2013 15:25:37 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.901
X-Spam-Level: 
X-Spam-Status: No, score=-2.901 tagged_above=-999 required=5 tests=[AWL=-0.698, BAYES_00=-2.599, MIME_QP_LONG_LINE=1.396, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id eiG9yIhepbiE for <trill@ietfa.amsl.com>; Wed, 30 Jan 2013 15:25:36 -0800 (PST)
Received: from mail-pa0-f41.google.com (mail-pa0-f41.google.com [209.85.220.41]) by ietfa.amsl.com (Postfix) with ESMTP id 9193F21F8639 for <trill@ietf.org>; Wed, 30 Jan 2013 15:25:36 -0800 (PST)
Received: by mail-pa0-f41.google.com with SMTP id fa11so697629pad.28 for <trill@ietf.org>; Wed, 30 Jan 2013 15:25:35 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=x-received:content-type:mime-version:subject:from:in-reply-to:date :cc:content-transfer-encoding:message-id:references:to:x-mailer; bh=fRuhvatqSMRVdjpANCbWXGxzh6XQz3p5hwastQgedLI=; b=wj3IjmEMZb7u0ihNLb1UmkzDHlBfyBSo2/rAL0j5NljrVrcyvY89ICLvFTPvHjkSM5 yx03sfC6Fpjdh1HYu6uQSv5XvhhqdLXLDDp8olyx/33ZdQGu1gaMqfY3hAwYmEoGCSZp I3aHWINI0KbDcqck1qbfuUZUtAAqoCVFh9RWBWfQ4TlpiFscQjU9DleEZugkYzaMNzS1 UjP3uRfUO/K1RzSTsAmvYPf1Lt7Otf0BDW8ROwQWCnOxttPIP4N/DRrKcnf/gJ2BvUXU 3jxVUjE181mZQTQUdOn3EFbMY4tAZnp5J4CO92kzt+D+N0nsOONbz9Wb1jv3ajeGh4ud WnDw==
X-Received: by 10.66.89.199 with SMTP id bq7mr15388101pab.26.1359588335766; Wed, 30 Jan 2013 15:25:35 -0800 (PST)
Received: from [192.168.255.224] (107-1-141-74-ip-static.hfc.comcastbusiness.net. [107.1.141.74]) by mx.google.com with ESMTPS id mn4sm2913288pbc.12.2013.01.30.15.25.33 (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Wed, 30 Jan 2013 15:25:35 -0800 (PST)
Content-Type: text/plain; charset=us-ascii
Mime-Version: 1.0 (Mac OS X Mail 6.2 \(1499\))
From: Sam Aldrin <aldrin.ietf@gmail.com>
In-Reply-To: <DC8898F8-75C0-4535-BA61-C54279901332@gmail.com>
Date: Wed, 30 Jan 2013 15:25:32 -0800
Content-Transfer-Encoding: quoted-printable
Message-Id: <43DF00BE-B5F6-48B9-B313-8D179354C7B7@gmail.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <CA+-tSzx0E3J4i5Dvfbj4egruCeGi7z1se1Jn5RXtx3RedQLPdg@mail.gmail.com> <CAHD03N89RNX4=XxRZpD+Y2PCktYf-FtVjyYJ5oFS3dui3SsXaA@mail.gmail.com> <CAF4+nEG7YKGzGmObAMaOxackivjJsALymqHX=6PKBd5qU-Z+zQ@mail.gmail.com> <CA+-tSzz0=oh6iB1e-ESyeOx2sCvBFXeVpzuNbKmUVhRNLScDgQ@mail.gmail.com> <CAF4+nEFLrJL=CgX9e6on-cgz89ugmu93P6u5kmODhUia3H7RLg@mail.gmail.com> <1F5A81AE-4B3C-45FD-B431-9BA0011EB085@gmail.com> <DC8898F8-75C0-4535-BA61-C54279901332@gmail.com>
To: Jon Hudson <jon.hudson@gmail.com>
X-Mailer: Apple Mail (2.1499)
Cc: Donald Eastlake <d3e3e3@gmail.com>, Anoop Ghanwani <anoop@alumni.duke.edu>, Radia Perlman <radiaperlman@gmail.com>, Ayan Banerjee <ayabaner@gmail.com>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 30 Jan 2013 23:25:37 -0000

Here is what I meant.

Deployment scenarios could be  upgrading to FGL RBridges via 1) new =
device insertion OR 2) replacing older devices with FGL supported =
RBridges OR 3) the so called upgrade to existing hardware to be FGL =
safe.
In each of those cases, the way to do it could differ for the reasons =
highlighted in various emails.
At the end of the day, by moving towards FGL based network, we cannot =
afford to disrupt existing TRILL network or force to forklift upgrade of =
non-FGL aware hardware.

Lastly, it would be good to hear from the actual user to understand the =
pain points in doing one way or the other.

-sam

On Jan 30, 2013, at 12:40 PM, Jon Hudson <jon.hudson@gmail.com> wrote:

> Can you clarify on what you mean by deploy?
>=20
> I ask because I have many customers waiting for this functionality. =
Everyone of the ~1000 installs we have, that also have multitenancy =
environments plan to deploy this as soon as possible.=20
>=20
> But with no one have I gone through and done a Visio of how their =
Rbridges are wired up today and how FGL rbridges would be integrated =
gracefully into the environment.
>=20
> On Jan 31, 2013, at 3:55 AM, Sam Aldrin <aldrin.ietf@gmail.com> wrote:
>=20
>> [still catching up on email, so the delay]
>>=20
>> When everyone says, upgrade RBridges, what is the upgrade model they =
have in mind?
>> How can one do software upgrade when hardware/silicon cannot support?
>> Secondly, any actual users, not vendors, commented on how they plan =
to deploy FGL?
>>=20
>> -sam
>>=20
>> On Jan 29, 2013, at 4:37 PM, Donald Eastlake <d3e3e3@gmail.com> =
wrote:
>>=20
>>> Hi Anoop,
>>>=20
>>> On Tue, Jan 29, 2013 at 7:03 PM, Anoop Ghanwani =
<anoop@alumni.duke.edu> wrote:
>>>> On Tue, Jan 29, 2013 at 12:49 PM, Donald Eastlake =
<d3e3e3@gmail.com> wrote:
>>>>=20
>>>>> The general assumption is that you are migrating from VL to FGL. =
First
>>>>> you migrate to FGL-safe. Then you unleash the FGL traffic, by
>>>>> configuring FGL ports, which isolates any remaining (FGL-unsafe) =
VL
>>>>> switches while still being able to handle VL traffic between VL =
ports
>>>>> on FGL-safe switches.
>>>>=20
>>>> Based on what is written so far, it looks like as long as there is =
even
>>>> one VL RBridge in it's database, an RBridge would not be able to =
start
>>>> sourcing FGL traffic.  It may start doing so only when all RBridges =
in
>>>=20
>>> Well, it can start sourcing FGL traffic but their are consequences =
as
>>> you note below.
>>>=20
>>>> its database are FGL-safe.  If we ever get into a condition where
>>>> an RBridge has started advertising FGL information, at that point,
>>>> does any other RBridge in the campus simply isolate any VL RBridge
>>>> (with the assumption that there was some kind of race condition?
>>>=20
>>> Yes. (This could just be due to some misconfiguration.)
>>>=20
>>>> The case I'm concerned about is we have 2 FGL RBridges, RB1 and =
RB2.
>>>> RB1 is connected to a campus of n RBs that are still VL only.  The =
VL link
>>>> Breaks, and RB1 starts advertising FGL information to RB2, but then
>>>> the link to the VL campus comes back up.  Does this mean RB1 and =
RB2
>>>> remain isolated from the campus (and operate as they own =
mini-campus)
>>>> until all the other RBridges are converted over to FGL-safe?
>>>=20
>>> Yes. But I don't think this as much of a problem given that I =
believe
>>> existing TRILL switches can be software upgraded to be FGL-safe. I
>>> view the configuration of FGL ports on RB1/RB2 and the consequent
>>> routing of actual FGL traffic to be an active network management
>>> decision, not something that might just happen because there appear =
to
>>> be no more VL TRILL switches in the topology.
>>>=20
>>> Thanks,
>>> Donald
>>> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D
>>> Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
>>> 155 Beaver Street, Milford, MA 01757 USA
>>> d3e3e3@gmail.com
>>>=20
>>>> Anoop
>>> _______________________________________________
>>> trill mailing list
>>> trill@ietf.org
>>> https://www.ietf.org/mailman/listinfo/trill
>>=20
>> _______________________________________________
>> trill mailing list
>> trill@ietf.org
>> https://www.ietf.org/mailman/listinfo/trill


From jon.hudson@gmail.com  Wed Jan 30 23:16:17 2013
Return-Path: <jon.hudson@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3BA8921F86F4 for <trill@ietfa.amsl.com>; Wed, 30 Jan 2013 23:16:17 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.818
X-Spam-Level: 
X-Spam-Status: No, score=-0.818 tagged_above=-999 required=5 tests=[AWL=-1.385, BAYES_00=-2.599, FH_HOST_EQ_D_D_D_D=0.765, MIME_QP_LONG_LINE=1.396, RCVD_IN_PBL=0.905, RDNS_DYNAMIC=0.1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id tH+iTEXgu64b for <trill@ietfa.amsl.com>; Wed, 30 Jan 2013 23:16:16 -0800 (PST)
Received: from mail-yh0-x236.google.com (yh-in-x0236.1e100.net [IPv6:2607:f8b0:4002:c01::236]) by ietfa.amsl.com (Postfix) with ESMTP id A735221F86EB for <trill@ietf.org>; Wed, 30 Jan 2013 23:16:14 -0800 (PST)
Received: by mail-yh0-f54.google.com with SMTP id 29so2427yhr.13 for <trill@ietf.org>; Wed, 30 Jan 2013 23:16:13 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=x-received:references:mime-version:in-reply-to:content-type :content-transfer-encoding:message-id:cc:x-mailer:from:subject:date :to; bh=ITIV+EwkVWSAvtqt6ZK0aDb2Oo73XKCXUoJF+cD7dL8=; b=zujJORNR3Ky7zHjjBvfvwzjTAR01CJdETezAMF800Y3Uf9qsZ7S6FwzVGeS9yC93DK jD39+4gYBPT4/Xaa43tGtf4qfG6whaHz2hxwWYYOBPQokLbm9Sp1T1RCoQ/hLJS5rGuw P36tiq5d0kLQYbYIO90QRNnsdqRFCkaAa6wZ+mTUgooEvoq/Ip5TEnsOj1kH+0pqwNvG DOONbUWAGLdhBJX0jaFExZBzPooEs0m3GUJZHstJBiXrelO1ZmA1hg1vAwuqHXUIk/S5 cXkeE5OWiOj3WWlPN8gzDfkQF3NC3nTdSqUSUrYh4eKuD5nQrn5k4GhNbRWh/U1JzUPN 7Gxg==
X-Received: by 10.236.142.101 with SMTP id h65mr8862101yhj.95.1359616573625; Wed, 30 Jan 2013 23:16:13 -0800 (PST)
Received: from [10.10.184.9] (mobile-166-147-114-244.mycingular.net. [166.147.114.244]) by mx.google.com with ESMTPS id j1sm6597633yhn.3.2013.01.30.23.16.07 (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Wed, 30 Jan 2013 23:16:12 -0800 (PST)
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <CA+-tSzx0E3J4i5Dvfbj4egruCeGi7z1se1Jn5RXtx3RedQLPdg@mail.gmail.com> <CAHD03N89RNX4=XxRZpD+Y2PCktYf-FtVjyYJ5oFS3dui3SsXaA@mail.gmail.com> <CAF4+nEG7YKGzGmObAMaOxackivjJsALymqHX=6PKBd5qU-Z+zQ@mail.gmail.com> <CA+-tSzz0=oh6iB1e-ESyeOx2sCvBFXeVpzuNbKmUVhRNLScDgQ@mail.gmail.com> <CAF4+nEFLrJL=CgX9e6on-cgz89ugmu93P6u5kmODhUia3H7RLg@mail.gmail.com> <1F5A81AE-4B3C-45FD-B431-9BA0011EB085@gmail.com> <DC8898F8-75C0-4535-BA61-C54279901332@gmail.com> <43DF00BE-B5F6-48B9-B313-8D179354C7B7@gmail.com>
Mime-Version: 1.0 (1.0)
In-Reply-To: <43DF00BE-B5F6-48B9-B313-8D179354C7B7@gmail.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: quoted-printable
Message-Id: <A9B8F2D5-C3D7-4688-A76D-ED19EC47E59F@gmail.com>
X-Mailer: iPhone Mail (10B143)
From: Jon Hudson <jon.hudson@gmail.com>
Date: Thu, 31 Jan 2013 16:16:01 +0900
To: Sam Aldrin <aldrin.ietf@gmail.com>
Cc: Donald Eastlake <d3e3e3@gmail.com>, Anoop Ghanwani <anoop@alumni.duke.edu>, Radia Perlman <radiaperlman@gmail.com>, Ayan Banerjee <ayabaner@gmail.com>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 31 Jan 2013 07:16:17 -0000

Well most of my customers and bound by 3yr depreciation schedules. So any no=
n-FGL devices they have when Full FGl devices ship will not be available for=
 swap-pout. If they have been in service for 2hrs then maybe, but 2.5-3yrs i=
deal.

So we expect most scenarios to be the addition of net new full FGL rbridges.=
 Then (hopefully) a software update to the existing non FGLs to some quasi FG=
L mode ( just so you can take some advantage of the new functionality. Then w=
hen the older rbridges run out on the books, then they can be replaced with s=
hiny new full FGL rbridges and now have a full FGL compliant environment.

But that is just for the small number of customers already running non FGL c=
ompliant hardware. (~1000 for Brocade, and then probably more, but lets say 2=
,000for Cisco)

For a large majority it will be green field installs all net new.=20

We just don't want to punish those ~3000ish early adopters=20

I have had MUCH difficultly getting customers involved in mail list discussi=
ons. So getting someone who is today and operator (as apposed to a recoverin=
g operator like myself) isn't super easy.  Needless to say the volume of ema=
ils and tone of emails scares a lot of folks away. ( not just a TRILL WG thi=
ng, an IETF in general thing)


On Jan 31, 2013, at 8:25 AM, Sam Aldrin <aldrin.ietf@gmail.com> wrote:

> Here is what I meant.
>=20
> Deployment scenarios could be  upgrading to FGL RBridges via 1) new device=
 insertion OR 2) replacing older devices with FGL supported RBridges OR 3) t=
he so called upgrade to existing hardware to be FGL safe.
> In each of those cases, the way to do it could differ for the reasons high=
lighted in various emails.
> At the end of the day, by moving towards FGL based network, we cannot affo=
rd to disrupt existing TRILL network or force to forklift upgrade of non-FGL=
 aware hardware.
>=20
> Lastly, it would be good to hear from the actual user to understand the pa=
in points in doing one way or the other.
>=20
> -sam
>=20
> On Jan 30, 2013, at 12:40 PM, Jon Hudson <jon.hudson@gmail.com> wrote:
>=20
>> Can you clarify on what you mean by deploy?
>>=20
>> I ask because I have many customers waiting for this functionality. Every=
one of the ~1000 installs we have, that also have multitenancy environments p=
lan to deploy this as soon as possible.=20
>>=20
>> But with no one have I gone through and done a Visio of how their Rbridge=
s are wired up today and how FGL rbridges would be integrated gracefully int=
o the environment.
>>=20
>> On Jan 31, 2013, at 3:55 AM, Sam Aldrin <aldrin.ietf@gmail.com> wrote:
>>=20
>>> [still catching up on email, so the delay]
>>>=20
>>> When everyone says, upgrade RBridges, what is the upgrade model they hav=
e in mind?
>>> How can one do software upgrade when hardware/silicon cannot support?
>>> Secondly, any actual users, not vendors, commented on how they plan to d=
eploy FGL?
>>>=20
>>> -sam
>>>=20
>>> On Jan 29, 2013, at 4:37 PM, Donald Eastlake <d3e3e3@gmail.com> wrote:
>>>=20
>>>> Hi Anoop,
>>>>=20
>>>> On Tue, Jan 29, 2013 at 7:03 PM, Anoop Ghanwani <anoop@alumni.duke.edu>=
 wrote:
>>>>> On Tue, Jan 29, 2013 at 12:49 PM, Donald Eastlake <d3e3e3@gmail.com> w=
rote:
>>>>>=20
>>>>>> The general assumption is that you are migrating from VL to FGL. Firs=
t
>>>>>> you migrate to FGL-safe. Then you unleash the FGL traffic, by
>>>>>> configuring FGL ports, which isolates any remaining (FGL-unsafe) VL
>>>>>> switches while still being able to handle VL traffic between VL ports=

>>>>>> on FGL-safe switches.
>>>>>=20
>>>>> Based on what is written so far, it looks like as long as there is eve=
n
>>>>> one VL RBridge in it's database, an RBridge would not be able to start=

>>>>> sourcing FGL traffic.  It may start doing so only when all RBridges in=

>>>>=20
>>>> Well, it can start sourcing FGL traffic but their are consequences as
>>>> you note below.
>>>>=20
>>>>> its database are FGL-safe.  If we ever get into a condition where
>>>>> an RBridge has started advertising FGL information, at that point,
>>>>> does any other RBridge in the campus simply isolate any VL RBridge
>>>>> (with the assumption that there was some kind of race condition?
>>>>=20
>>>> Yes. (This could just be due to some misconfiguration.)
>>>>=20
>>>>> The case I'm concerned about is we have 2 FGL RBridges, RB1 and RB2.
>>>>> RB1 is connected to a campus of n RBs that are still VL only.  The VL l=
ink
>>>>> Breaks, and RB1 starts advertising FGL information to RB2, but then
>>>>> the link to the VL campus comes back up.  Does this mean RB1 and RB2
>>>>> remain isolated from the campus (and operate as they own mini-campus)
>>>>> until all the other RBridges are converted over to FGL-safe?
>>>>=20
>>>> Yes. But I don't think this as much of a problem given that I believe
>>>> existing TRILL switches can be software upgraded to be FGL-safe. I
>>>> view the configuration of FGL ports on RB1/RB2 and the consequent
>>>> routing of actual FGL traffic to be an active network management
>>>> decision, not something that might just happen because there appear to
>>>> be no more VL TRILL switches in the topology.
>>>>=20
>>>> Thanks,
>>>> Donald
>>>> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D
>>>> Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
>>>> 155 Beaver Street, Milford, MA 01757 USA
>>>> d3e3e3@gmail.com
>>>>=20
>>>>> Anoop
>>>> _______________________________________________
>>>> trill mailing list
>>>> trill@ietf.org
>>>> https://www.ietf.org/mailman/listinfo/trill
>>>=20
>>> _______________________________________________
>>> trill mailing list
>>> trill@ietf.org
>>> https://www.ietf.org/mailman/listinfo/trill
>=20

From aldrin.ietf@gmail.com  Wed Jan 30 23:31:46 2013
Return-Path: <aldrin.ietf@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A285821F86AE for <trill@ietfa.amsl.com>; Wed, 30 Jan 2013 23:31:46 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.203
X-Spam-Level: 
X-Spam-Status: No, score=-2.203 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, MIME_QP_LONG_LINE=1.396, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id iI0OmADP4rM7 for <trill@ietfa.amsl.com>; Wed, 30 Jan 2013 23:31:46 -0800 (PST)
Received: from mail-da0-f43.google.com (mail-da0-f43.google.com [209.85.210.43]) by ietfa.amsl.com (Postfix) with ESMTP id EAF9E21F869B for <trill@ietf.org>; Wed, 30 Jan 2013 23:31:45 -0800 (PST)
Received: by mail-da0-f43.google.com with SMTP id u36so1173499dak.2 for <trill@ietf.org>; Wed, 30 Jan 2013 23:31:45 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=x-received:references:mime-version:in-reply-to:content-type :content-transfer-encoding:message-id:cc:x-mailer:from:subject:date :to; bh=AiwPCSUL2zlX48FPLBYBjXKf/QAXuz6VuH+9NHN8Uhc=; b=VTIopOJdkeJJm3Qwa0XZq6BYkesXYc7xb94Uo9St+ioA3336tcUU/2UKdQiJ18KY9V Rsnt+AbSOGiPrB/rtSGpBnHUkLnSjzC3XsUCyJi2AVuD9/GvfSlaXeaCB3jts8D2GgPi MZx2S3nNJri3w9ElxL0JrbSR+0URleMjkSlaUWa0MBB/NX/Cj7oAWUUCiLDzk02c849g nqUEWwdFbZXK7ExHNvk5wLpydzgwbmZdJlhzC+oD+rJPvnPvuesHNrrVkbPNvJwH54Rw ImNAUJtuzN1PNI8AWVJEZARJCFxgyl5ylx9In2yynek11QMiSBG6zM/vts1cGC3QKMBk LUow==
X-Received: by 10.68.134.130 with SMTP id pk2mr19709002pbb.125.1359617505693;  Wed, 30 Jan 2013 23:31:45 -0800 (PST)
Received: from [192.168.1.3] (c-98-248-237-85.hsd1.ca.comcast.net. [98.248.237.85]) by mx.google.com with ESMTPS id wh4sm4197651pbc.18.2013.01.30.23.31.43 (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Wed, 30 Jan 2013 23:31:44 -0800 (PST)
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <CA+-tSzx0E3J4i5Dvfbj4egruCeGi7z1se1Jn5RXtx3RedQLPdg@mail.gmail.com> <CAHD03N89RNX4=XxRZpD+Y2PCktYf-FtVjyYJ5oFS3dui3SsXaA@mail.gmail.com> <CAF4+nEG7YKGzGmObAMaOxackivjJsALymqHX=6PKBd5qU-Z+zQ@mail.gmail.com> <CA+-tSzz0=oh6iB1e-ESyeOx2sCvBFXeVpzuNbKmUVhRNLScDgQ@mail.gmail.com> <CAF4+nEFLrJL=CgX9e6on-cgz89ugmu93P6u5kmODhUia3H7RLg@mail.gmail.com> <1F5A81AE-4B3C-45FD-B431-9BA0011EB085@gmail.com> <DC8898F8-75C0-4535-BA61-C54279901332@gmail.com> <43DF00BE-B5F6-48B9-B313-8D179354C7B7@gmail.com> <A9B8F2D5-C3D7-4688-A76D-ED19EC47E59F@gmail.com>
Mime-Version: 1.0 (1.0)
In-Reply-To: <A9B8F2D5-C3D7-4688-A76D-ED19EC47E59F@gmail.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: quoted-printable
Message-Id: <176D0FD3-C9F0-4396-91A5-A67DAD971A89@gmail.com>
X-Mailer: iPad Mail (10B141)
From: Sam Aldrin <aldrin.ietf@gmail.com>
Date: Wed, 30 Jan 2013 23:31:43 -0800
To: Jon Hudson <jon.hudson@gmail.com>
Cc: Donald Eastlake <d3e3e3@gmail.com>, Anoop Ghanwani <anoop@alumni.duke.edu>, Radia Perlman <radiaperlman@gmail.com>, Ayan Banerjee <ayabaner@gmail.com>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 31 Jan 2013 07:31:46 -0000

Jon,

There are only couple of vendors who have actual deployments, and that inclu=
des you. As long as you and other vendors who deployed give nod, it is fairl=
y ok to go ahead with model being chosen. For new deployments, which most ot=
her vendors likely to have, backward compatibility will not be an issue, w.r=
.t FGL.

True that it is difficult to get users to be on ietf lists. Atleast if those=
 concerns are echoed, indirectly through channels like you, would help immen=
sely.

Cheers
Sam

Sent from my iPad

On Jan 30, 2013, at 11:16 PM, Jon Hudson <jon.hudson@gmail.com> wrote:

> Well most of my customers and bound by 3yr depreciation schedules. So any n=
on-FGL devices they have when Full FGl devices ship will not be available fo=
r swap-pout. If they have been in service for 2hrs then maybe, but 2.5-3yrs i=
deal.
>=20
> So we expect most scenarios to be the addition of net new full FGL rbridge=
s. Then (hopefully) a software update to the existing non FGLs to some quasi=
 FGL mode ( just so you can take some advantage of the new functionality. Th=
en when the older rbridges run out on the books, then they can be replaced w=
ith shiny new full FGL rbridges and now have a full FGL compliant environmen=
t.
>=20
> But that is just for the small number of customers already running non FGL=
 compliant hardware. (~1000 for Brocade, and then probably more, but lets sa=
y 2,000for Cisco)
>=20
> For a large majority it will be green field installs all net new.=20
>=20
> We just don't want to punish those ~3000ish early adopters=20
>=20
> I have had MUCH difficultly getting customers involved in mail list discus=
sions. So getting someone who is today and operator (as apposed to a recover=
ing operator like myself) isn't super easy.  Needless to say the volume of e=
mails and tone of emails scares a lot of folks away. ( not just a TRILL WG t=
hing, an IETF in general thing)
>=20
>=20
> On Jan 31, 2013, at 8:25 AM, Sam Aldrin <aldrin.ietf@gmail.com> wrote:
>=20
>> Here is what I meant.
>>=20
>> Deployment scenarios could be  upgrading to FGL RBridges via 1) new devic=
e insertion OR 2) replacing older devices with FGL supported RBridges OR 3) t=
he so called upgrade to existing hardware to be FGL safe.
>> In each of those cases, the way to do it could differ for the reasons hig=
hlighted in various emails.
>> At the end of the day, by moving towards FGL based network, we cannot aff=
ord to disrupt existing TRILL network or force to forklift upgrade of non-FG=
L aware hardware.
>>=20
>> Lastly, it would be good to hear from the actual user to understand the p=
ain points in doing one way or the other.
>>=20
>> -sam
>>=20
>> On Jan 30, 2013, at 12:40 PM, Jon Hudson <jon.hudson@gmail.com> wrote:
>>=20
>>> Can you clarify on what you mean by deploy?
>>>=20
>>> I ask because I have many customers waiting for this functionality. Ever=
yone of the ~1000 installs we have, that also have multitenancy environments=
 plan to deploy this as soon as possible.=20
>>>=20
>>> But with no one have I gone through and done a Visio of how their Rbridg=
es are wired up today and how FGL rbridges would be integrated gracefully in=
to the environment.
>>>=20
>>> On Jan 31, 2013, at 3:55 AM, Sam Aldrin <aldrin.ietf@gmail.com> wrote:
>>>=20
>>>> [still catching up on email, so the delay]
>>>>=20
>>>> When everyone says, upgrade RBridges, what is the upgrade model they ha=
ve in mind?
>>>> How can one do software upgrade when hardware/silicon cannot support?
>>>> Secondly, any actual users, not vendors, commented on how they plan to d=
eploy FGL?
>>>>=20
>>>> -sam
>>>>=20
>>>> On Jan 29, 2013, at 4:37 PM, Donald Eastlake <d3e3e3@gmail.com> wrote:
>>>>=20
>>>>> Hi Anoop,
>>>>>=20
>>>>> On Tue, Jan 29, 2013 at 7:03 PM, Anoop Ghanwani <anoop@alumni.duke.edu=
> wrote:
>>>>>> On Tue, Jan 29, 2013 at 12:49 PM, Donald Eastlake <d3e3e3@gmail.com> w=
rote:
>>>>>>=20
>>>>>>> The general assumption is that you are migrating from VL to FGL. Fir=
st
>>>>>>> you migrate to FGL-safe. Then you unleash the FGL traffic, by
>>>>>>> configuring FGL ports, which isolates any remaining (FGL-unsafe) VL
>>>>>>> switches while still being able to handle VL traffic between VL port=
s
>>>>>>> on FGL-safe switches.
>>>>>>=20
>>>>>> Based on what is written so far, it looks like as long as there is ev=
en
>>>>>> one VL RBridge in it's database, an RBridge would not be able to star=
t
>>>>>> sourcing FGL traffic.  It may start doing so only when all RBridges i=
n
>>>>>=20
>>>>> Well, it can start sourcing FGL traffic but their are consequences as
>>>>> you note below.
>>>>>=20
>>>>>> its database are FGL-safe.  If we ever get into a condition where
>>>>>> an RBridge has started advertising FGL information, at that point,
>>>>>> does any other RBridge in the campus simply isolate any VL RBridge
>>>>>> (with the assumption that there was some kind of race condition?
>>>>>=20
>>>>> Yes. (This could just be due to some misconfiguration.)
>>>>>=20
>>>>>> The case I'm concerned about is we have 2 FGL RBridges, RB1 and RB2.
>>>>>> RB1 is connected to a campus of n RBs that are still VL only.  The VL=
 link
>>>>>> Breaks, and RB1 starts advertising FGL information to RB2, but then
>>>>>> the link to the VL campus comes back up.  Does this mean RB1 and RB2
>>>>>> remain isolated from the campus (and operate as they own mini-campus)=

>>>>>> until all the other RBridges are converted over to FGL-safe?
>>>>>=20
>>>>> Yes. But I don't think this as much of a problem given that I believe
>>>>> existing TRILL switches can be software upgraded to be FGL-safe. I
>>>>> view the configuration of FGL ports on RB1/RB2 and the consequent
>>>>> routing of actual FGL traffic to be an active network management
>>>>> decision, not something that might just happen because there appear to=

>>>>> be no more VL TRILL switches in the topology.
>>>>>=20
>>>>> Thanks,
>>>>> Donald
>>>>> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D
>>>>> Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
>>>>> 155 Beaver Street, Milford, MA 01757 USA
>>>>> d3e3e3@gmail.com
>>>>>=20
>>>>>> Anoop
>>>>> _______________________________________________
>>>>> trill mailing list
>>>>> trill@ietf.org
>>>>> https://www.ietf.org/mailman/listinfo/trill
>>>>=20
>>>> _______________________________________________
>>>> trill mailing list
>>>> trill@ietf.org
>>>> https://www.ietf.org/mailman/listinfo/trill
>>=20

From tsenevir@cisco.com  Thu Jan 31 07:18:53 2013
Return-Path: <tsenevir@cisco.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CF71421F85C6 for <trill@ietfa.amsl.com>; Thu, 31 Jan 2013 07:18:53 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -10.299
X-Spam-Level: 
X-Spam-Status: No, score=-10.299 tagged_above=-999 required=5 tests=[AWL=0.300, BAYES_00=-2.599, RCVD_IN_DNSWL_HI=-8]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id IIkNku4JDhL5 for <trill@ietfa.amsl.com>; Thu, 31 Jan 2013 07:18:52 -0800 (PST)
Received: from rcdn-iport-9.cisco.com (rcdn-iport-9.cisco.com [173.37.86.80]) by ietfa.amsl.com (Postfix) with ESMTP id 916D621F84F9 for <trill@ietf.org>; Thu, 31 Jan 2013 07:18:52 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=7106; q=dns/txt; s=iport; t=1359645532; x=1360855132; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-transfer-encoding:mime-version; bh=BE+KZR+6mDQ9ttnlKQl8rO4I5a9AE2D0Xs2awt1GILw=; b=CdPFGzU1/Ag7q37s6W/AUKddDOMxX0R3vXn0X303T+xKLePSijen/My2 4kxgstITYY5Rn+/Dzb3fEEra96pabuGynSPoFRF+cyMRJp1LB0PCVbrJB Axbqm3yZApbMPQQ0BNuxoYuxaySowz0RG0UX1uO56J8oEnx5W+YFSz57k I=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: AgEFAKGKClGtJV2a/2dsb2JhbABCA78dFnOCHgEBAQMBAQEBNzQLBQcEAgEIEQQBAQEKFAkHIQYLFAMGCAIEAQ0FCId3AwkGDLg9DYlSBIwUfQoGDTCCTWEDlDuNEYUSgneBZgIHFx4
X-IronPort-AV: E=Sophos;i="4.84,577,1355097600"; d="scan'208";a="168144406"
Received: from rcdn-core-3.cisco.com ([173.37.93.154]) by rcdn-iport-9.cisco.com with ESMTP; 31 Jan 2013 15:18:46 +0000
Received: from xhc-aln-x01.cisco.com (xhc-aln-x01.cisco.com [173.36.12.75]) by rcdn-core-3.cisco.com (8.14.5/8.14.5) with ESMTP id r0VFIjrj031925 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Thu, 31 Jan 2013 15:18:45 GMT
Received: from xmb-rcd-x08.cisco.com ([169.254.8.163]) by xhc-aln-x01.cisco.com ([173.36.12.75]) with mapi id 14.02.0318.004; Thu, 31 Jan 2013 09:18:45 -0600
From: "Tissa Senevirathne (tsenevir)" <tsenevir@cisco.com>
To: Jon Hudson <jon.hudson@gmail.com>, Sam Aldrin <aldrin.ietf@gmail.com>
Thread-Topic: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
Thread-Index: AQHN+y5DuS9tZw2pBEO8DLEnkCKGP5hgzAQAgABUdgCAABIkAIAANgUAgAAJbACAAUOygIAADJyAgAAuDQCAAIN0gIAAIEYQ
Date: Thu, 31 Jan 2013 15:18:44 +0000
Message-ID: <FBEA3E19AA24F847BA3AE74E2FE19356288F6978@xmb-rcd-x08.cisco.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <CA+-tSzx0E3J4i5Dvfbj4egruCeGi7z1se1Jn5RXtx3RedQLPdg@mail.gmail.com> <CAHD03N89RNX4=XxRZpD+Y2PCktYf-FtVjyYJ5oFS3dui3SsXaA@mail.gmail.com> <CAF4+nEG7YKGzGmObAMaOxackivjJsALymqHX=6PKBd5qU-Z+zQ@mail.gmail.com> <CA+-tSzz0=oh6iB1e-ESyeOx2sCvBFXeVpzuNbKmUVhRNLScDgQ@mail.gmail.com> <CAF4+nEFLrJL=CgX9e6on-cgz89ugmu93P6u5kmODhUia3H7RLg@mail.gmail.com> <1F5A81AE-4B3C-45FD-B431-9BA0011EB085@gmail.com> <DC8898F8-75C0-4535-BA61-C54279901332@gmail.com> <43DF00BE-B5F6-48B9-B313-8D179354C7B7@gmail.com> <A9B8F2D5-C3D7-4688-A76D-ED19EC47E59F@gmail.com>
In-Reply-To: <A9B8F2D5-C3D7-4688-A76D-ED19EC47E59F@gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.21.167.3]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Cc: Donald Eastlake <d3e3e3@gmail.com>, Anoop Ghanwani <anoop@alumni.duke.edu>, Radia Perlman <radiaperlman@gmail.com>, Ayan Banerjee <ayabaner@gmail.com>, "trill@ietf.org" <trill@ietf.org>
Subject: Re: [trill] Explaining three options for upgrading to FGL	(fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 31 Jan 2013 15:18:53 -0000

Jon

You brought up the most import element. Questions Sam asked are extremely v=
alid. What do you mean by "safe FGL", how can you make sure all of the so c=
alled "Safe FGL" have consistent behavior, what about some devices who are =
not "safe FGL" and so on, are we saying do not buy no safe FGL devices, if =
you do that you cannot get FGL interop ? We are not talking about control p=
lane we are talking about Hardware.

Based on your point and with the analogy (though not the exact numbers :) )=
, we are looking at a limited install base here. For that Do you agree if w=
e say, FGL will always be core of the network, and VL will be at the edge o=
f the network ?

Thanks
Tissa

-----Original Message-----
From: trill-bounces@ietf.org [mailto:trill-bounces@ietf.org] On Behalf Of J=
on Hudson
Sent: Wednesday, January 30, 2013 11:16 PM
To: Sam Aldrin
Cc: Donald Eastlake; Anoop Ghanwani; Radia Perlman; Ayan Banerjee; trill@ie=
tf.org
Subject: Re: [trill] Explaining three options for upgrading to FGL (fine-gr=
ained-labeling)

Well most of my customers and bound by 3yr depreciation schedules. So any n=
on-FGL devices they have when Full FGl devices ship will not be available f=
or swap-pout. If they have been in service for 2hrs then maybe, but 2.5-3yr=
s ideal.

So we expect most scenarios to be the addition of net new full FGL rbridges=
. Then (hopefully) a software update to the existing non FGLs to some quasi=
 FGL mode ( just so you can take some advantage of the new functionality. T=
hen when the older rbridges run out on the books, then they can be replaced=
 with shiny new full FGL rbridges and now have a full FGL compliant environ=
ment.

But that is just for the small number of customers already running non FGL =
compliant hardware. (~1000 for Brocade, and then probably more, but lets sa=
y 2,000for Cisco)

For a large majority it will be green field installs all net new.=20

We just don't want to punish those ~3000ish early adopters=20

I have had MUCH difficultly getting customers involved in mail list discuss=
ions. So getting someone who is today and operator (as apposed to a recover=
ing operator like myself) isn't super easy.  Needless to say the volume of =
emails and tone of emails scares a lot of folks away. ( not just a TRILL WG=
 thing, an IETF in general thing)


On Jan 31, 2013, at 8:25 AM, Sam Aldrin <aldrin.ietf@gmail.com> wrote:

> Here is what I meant.
>=20
> Deployment scenarios could be  upgrading to FGL RBridges via 1) new devic=
e insertion OR 2) replacing older devices with FGL supported RBridges OR 3)=
 the so called upgrade to existing hardware to be FGL safe.
> In each of those cases, the way to do it could differ for the reasons hig=
hlighted in various emails.
> At the end of the day, by moving towards FGL based network, we cannot aff=
ord to disrupt existing TRILL network or force to forklift upgrade of non-F=
GL aware hardware.
>=20
> Lastly, it would be good to hear from the actual user to understand the p=
ain points in doing one way or the other.
>=20
> -sam
>=20
> On Jan 30, 2013, at 12:40 PM, Jon Hudson <jon.hudson@gmail.com> wrote:
>=20
>> Can you clarify on what you mean by deploy?
>>=20
>> I ask because I have many customers waiting for this functionality. Ever=
yone of the ~1000 installs we have, that also have multitenancy environment=
s plan to deploy this as soon as possible.=20
>>=20
>> But with no one have I gone through and done a Visio of how their Rbridg=
es are wired up today and how FGL rbridges would be integrated gracefully i=
nto the environment.
>>=20
>> On Jan 31, 2013, at 3:55 AM, Sam Aldrin <aldrin.ietf@gmail.com> wrote:
>>=20
>>> [still catching up on email, so the delay]
>>>=20
>>> When everyone says, upgrade RBridges, what is the upgrade model they ha=
ve in mind?
>>> How can one do software upgrade when hardware/silicon cannot support?
>>> Secondly, any actual users, not vendors, commented on how they plan to =
deploy FGL?
>>>=20
>>> -sam
>>>=20
>>> On Jan 29, 2013, at 4:37 PM, Donald Eastlake <d3e3e3@gmail.com> wrote:
>>>=20
>>>> Hi Anoop,
>>>>=20
>>>> On Tue, Jan 29, 2013 at 7:03 PM, Anoop Ghanwani <anoop@alumni.duke.edu=
> wrote:
>>>>> On Tue, Jan 29, 2013 at 12:49 PM, Donald Eastlake <d3e3e3@gmail.com> =
wrote:
>>>>>=20
>>>>>> The general assumption is that you are migrating from VL to FGL.=20
>>>>>> First you migrate to FGL-safe. Then you unleash the FGL traffic,=20
>>>>>> by configuring FGL ports, which isolates any remaining=20
>>>>>> (FGL-unsafe) VL switches while still being able to handle VL=20
>>>>>> traffic between VL ports on FGL-safe switches.
>>>>>=20
>>>>> Based on what is written so far, it looks like as long as there is=20
>>>>> even one VL RBridge in it's database, an RBridge would not be able=20
>>>>> to start sourcing FGL traffic.  It may start doing so only when=20
>>>>> all RBridges in
>>>>=20
>>>> Well, it can start sourcing FGL traffic but their are consequences=20
>>>> as you note below.
>>>>=20
>>>>> its database are FGL-safe.  If we ever get into a condition where=20
>>>>> an RBridge has started advertising FGL information, at that point,=20
>>>>> does any other RBridge in the campus simply isolate any VL RBridge=20
>>>>> (with the assumption that there was some kind of race condition?
>>>>=20
>>>> Yes. (This could just be due to some misconfiguration.)
>>>>=20
>>>>> The case I'm concerned about is we have 2 FGL RBridges, RB1 and RB2.
>>>>> RB1 is connected to a campus of n RBs that are still VL only.  The=20
>>>>> VL link Breaks, and RB1 starts advertising FGL information to RB2,=20
>>>>> but then the link to the VL campus comes back up.  Does this mean=20
>>>>> RB1 and RB2 remain isolated from the campus (and operate as they=20
>>>>> own mini-campus) until all the other RBridges are converted over to F=
GL-safe?
>>>>=20
>>>> Yes. But I don't think this as much of a problem given that I=20
>>>> believe existing TRILL switches can be software upgraded to be=20
>>>> FGL-safe. I view the configuration of FGL ports on RB1/RB2 and the=20
>>>> consequent routing of actual FGL traffic to be an active network=20
>>>> management decision, not something that might just happen because=20
>>>> there appear to be no more VL TRILL switches in the topology.
>>>>=20
>>>> Thanks,
>>>> Donald
>>>> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D
>>>> Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
>>>> 155 Beaver Street, Milford, MA 01757 USA d3e3e3@gmail.com
>>>>=20
>>>>> Anoop
>>>> _______________________________________________
>>>> trill mailing list
>>>> trill@ietf.org
>>>> https://www.ietf.org/mailman/listinfo/trill
>>>=20
>>> _______________________________________________
>>> trill mailing list
>>> trill@ietf.org
>>> https://www.ietf.org/mailman/listinfo/trill
>=20
_______________________________________________
trill mailing list
trill@ietf.org
https://www.ietf.org/mailman/listinfo/trill

From radiaperlman@gmail.com  Thu Jan 31 12:24:23 2013
Return-Path: <radiaperlman@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2D6A721F86EF for <trill@ietfa.amsl.com>; Thu, 31 Jan 2013 12:24:23 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.999
X-Spam-Level: 
X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, J_CHICKENPOX_31=0.6, NO_RELAYS=-0.001]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id W5PdanWjltep for <trill@ietfa.amsl.com>; Thu, 31 Jan 2013 12:24:22 -0800 (PST)
Received: from mail-la0-x230.google.com (mail-la0-x230.google.com [IPv6:2a00:1450:4010:c03::230]) by ietfa.amsl.com (Postfix) with ESMTP id D952C21F86A9 for <trill@ietf.org>; Thu, 31 Jan 2013 12:24:21 -0800 (PST)
Received: by mail-la0-f48.google.com with SMTP id fq13so2231226lab.21 for <trill@ietf.org>; Thu, 31 Jan 2013 12:24:20 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:x-received:in-reply-to:references:date:message-id :subject:from:to:cc:content-type; bh=w8uMzQ05F2vCeSTyjtJvAymV8O1FC5lE+1foBZx/7OA=; b=s6N/OjCEkZ3kDTNffv+FzeHc4xogs9wbgwaEM6oKNXZv17K5jaL91nnZ5L94HP2SJB Lmv5b+qFlXv3RVwYkk4acYl0IsaFTuN/H38EnB3srRK0xw112FSc7L7Qp5g5jLSJjK2T GuUb1/Z7LxWLCqxp2fUrxyNKcFATqHNgnaC+xKz33yuHYPhGgKACri7MKVrYQIltfpGG ttRY56I9zuOQFxOctQcWtKWdwGlh7SzcozgSObuy7S5jp5YX4+gqk/U1UvS2TbBoaOjp O2AtTSwVIEujBudnh8azeipU+LBBEhiEdWaxEgSHtJ9vYrtiZPXRPDRMUMoCYu51xu7A oTnw==
MIME-Version: 1.0
X-Received: by 10.112.50.109 with SMTP id b13mr3847353lbo.8.1359663860409; Thu, 31 Jan 2013 12:24:20 -0800 (PST)
Received: by 10.112.64.17 with HTTP; Thu, 31 Jan 2013 12:24:20 -0800 (PST)
In-Reply-To: <FBEA3E19AA24F847BA3AE74E2FE19356288F6978@xmb-rcd-x08.cisco.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <CA+-tSzx0E3J4i5Dvfbj4egruCeGi7z1se1Jn5RXtx3RedQLPdg@mail.gmail.com> <CAHD03N89RNX4=XxRZpD+Y2PCktYf-FtVjyYJ5oFS3dui3SsXaA@mail.gmail.com> <CAF4+nEG7YKGzGmObAMaOxackivjJsALymqHX=6PKBd5qU-Z+zQ@mail.gmail.com> <CA+-tSzz0=oh6iB1e-ESyeOx2sCvBFXeVpzuNbKmUVhRNLScDgQ@mail.gmail.com> <CAF4+nEFLrJL=CgX9e6on-cgz89ugmu93P6u5kmODhUia3H7RLg@mail.gmail.com> <1F5A81AE-4B3C-45FD-B431-9BA0011EB085@gmail.com> <DC8898F8-75C0-4535-BA61-C54279901332@gmail.com> <43DF00BE-B5F6-48B9-B313-8D179354C7B7@gmail.com> <A9B8F2D5-C3D7-4688-A76D-ED19EC47E59F@gmail.com> <FBEA3E19AA24F847BA3AE74E2FE19356288F6978@xmb-rcd-x08.cisco.com>
Date: Thu, 31 Jan 2013 12:24:20 -0800
Message-ID: <CAFOuuo4jZMdvUU_2EzBj1MLp6k0e+AmW1+k1ai4Q_m4rA0ObMw@mail.gmail.com>
From: Radia Perlman <radiaperlman@gmail.com>
To: "Tissa Senevirathne (tsenevir)" <tsenevir@cisco.com>
Content-Type: multipart/alternative; boundary=f46d0401fbc969e57604d49b6943
Cc: Anoop Ghanwani <anoop@alumni.duke.edu>, Ayan Banerjee <ayabaner@gmail.com>, Sam Aldrin <aldrin.ietf@gmail.com>, "trill@ietf.org" <trill@ietf.org>, Donald Eastlake <d3e3e3@gmail.com>, Jon Hudson <jon.hudson@gmail.com>
Subject: Re: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 31 Jan 2013 20:24:23 -0000

--f46d0401fbc969e57604d49b6943
Content-Type: text/plain; charset=ISO-8859-1

Tissa said:
>> how can you make sure all of the so called "Safe FGL" have consistent
behavior

"Safe FGL" need not all have the *same* behavior as each other...they just
have to do "nothing bad" with an FGL frame.  It's worth summarizing (again)
what it means to be "FGL-safe", which allows you to coexist safely with FGL
frames.  To be FGL-safe you must do the following:

a) unicast: you MUST NOT drop a packet because of seeing the unrecognized
FGL tag.  You also MUST NOT get confused and do flow hashing improperly
(choosing different paths for packets for the same flow).

b) multicast pruning: it's OK not to do pruning at all, or to actually look
at all 24 bits of the FGL and prune as a proper FGL guy would do.  But an
example of what you canNOT do is look at the first 12 bits of FGL, pretend
it's the same as VL, and delete the multicast packet because nobody
downstream is listening to that 12 bit VL value

c) egress of either unicast or multicast:  you MUST NOT egress an FGL frame
onto a link that is not supposed to get that FGL value

d) you have to advertise that you are FGL-safe...otherwise other RBs will
assume you are VL ("FGL-unsafe")

---------
Also, TIssa said:
>>Do you agree if we say, FGL will always be core of the network, and VL
will be at the edge of the network ?

With option 2, a VL guy *cannot even be at the edge of a network with FGL
links* unless it is upgraded to FGL-safe AND advertises that it is
FGL-safe.  Option 2 says a VL guy is completely ostrasized once FGL-labeled
links are introduced.  There is NO coexistence of VL guys and FGL-links.
The only kinds of coexistence in option 2 are:

a) VL and FGL-safe RBridge, but no actual FGL links yet.
b) FGL-safe guys, even if some of them don't really understand FGL, but
just don't do "anything bad" with it, and FGL frames.

So...again...the migration phases in option 2 are
1) VL-only (today's deployment), then
2) A mixture of VL and FGL-safe, until
3) EVERYone upgraded to FGL-safe, then
4) FGL-links can be introduced.


----
Also, Tissa said:
>>what about some devices who are not "safe FGL" and so on, are we saying
do not buy no safe FGL devices, if you do that you cannot get FGL interop ?

*If* it's important to have VL guys as transit, and they cannot be upgraded
to FGL-safe because upgrading requires a hardware change (it involves
parsing data packets), then we are stuck with option 3.  In option 3, we
can forever have VL guys at the edge speaking through an FGL-safe or
full-FGL core.  In option 2, as I said, there is *no coexistence* of VL
guys, *even at the edg*e, once FGL links are introduced.  In option 2, a VL
guy, even at the edge, will be ostrasized completely once anyone advertises
connectivity to an FGL link.

I personally like option 2, because it's simple, but as Einstein said "make
things as simple as possible, but no simpler".  We have to make sure that
everyone understands the implications of option 2, and can live with them.
 To summarize again (since there seems to be some confusion about what the
implications of option 2 are),

If we can upgrade all existing VL implementations to FGL-safe through
software upgrade only, then option 2 provides smooth migration by changing
ALL RBridges (edge as well as transit) to FGL-safe, and only after ALL of
them are upgraded, then FGL-labeled links can be introduced.  If there are
any remaining VL guys once FGL-labeled links are introduced, the VL guys
are ostrasized, *even if they are just at the edge.*

If we canNOT upgrade all existing VL to FGL-safe, AND if it's important to
be able to mix VL guys into a network with FGL frames (say, having VL guys
at the edge), then option 3 allows this.

So hopefully everyone understands the implications of option 2 fully before
voting.  As I said, I don't know about whether there are VL chips that
would require a hardware upgrade to become FGL-safe.  Assuming there are
none, then I'm quite happy with option 2.   And actually, it looks like
people are converging on option 2 (though I hope it's based on fully
understanding option 2).  Now is the time for people to speak up, that know
of problems this would cause...for instance, needing to have non-upgradable
VL guys at the edge for a long time.

Radia

--f46d0401fbc969e57604d49b6943
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Tissa said:<div>&gt;&gt;=A0how can you make sure all of the so called &quot=
;Safe FGL&quot; have consistent behavior<br><div><br></div><div>&quot;Safe =
FGL&quot; need not all have the <i><b>same</b></i> behavior as each other..=
.they just have to do &quot;nothing bad&quot; with an FGL frame. =A0It&#39;=
s worth summarizing (again) what it means to be &quot;FGL-safe&quot;, which=
 allows you to coexist safely with FGL frames. =A0To be FGL-safe you must d=
o the following:</div>
<div><div><br></div><div>a) unicast: you MUST NOT drop a packet because of =
seeing the unrecognized FGL tag. =A0You also MUST NOT get confused and do f=
low hashing improperly (choosing different paths for packets for the same f=
low).</div>
<div><br></div><div>b) multicast pruning: it&#39;s OK not to do pruning at =
all, or to actually look at all 24 bits of the FGL and prune as a proper FG=
L guy would do. =A0But an example of what you canNOT do is look at the firs=
t 12 bits of FGL, pretend it&#39;s the same as VL, and delete the multicast=
 packet because nobody downstream is listening to that 12 bit VL value</div=
>
<div><br></div><div>c) egress of either unicast or multicast: =A0you MUST N=
OT egress an FGL frame onto a link that is not supposed to get that FGL val=
ue</div><div><br></div><div>d) you have to advertise that you are FGL-safe.=
..otherwise other RBs will assume you are VL (&quot;FGL-unsafe&quot;)</div>
<div><br></div><div>---------</div><div>Also, TIssa said:</div><div>&gt;&gt=
;Do you agree if we say, FGL will always be core of the network, and VL wil=
l be at the edge of the network ?</div><div><br></div><div>With option 2, a=
 VL guy <i><u>cannot even be at the edge of a network with FGL links</u></i=
> unless it is upgraded to FGL-safe AND advertises that it is FGL-safe. =A0=
Option 2 says a VL guy is completely ostrasized once FGL-labeled links are =
introduced. =A0There is NO coexistence of VL guys and FGL-links. The only k=
inds of coexistence in option 2 are:</div>
<div><br></div><div>a) VL and FGL-safe RBridge, but no actual FGL links yet=
.</div><div>b) FGL-safe guys, even if some of them don&#39;t really underst=
and FGL, but just don&#39;t do &quot;anything bad&quot; with it, and FGL fr=
ames.</div>
<div><br></div><div>So...again...the migration phases in option 2 are</div>=
<div>1) VL-only (today&#39;s deployment), then</div><div>2) A mixture of VL=
 and FGL-safe, until</div><div>3) EVERYone upgraded to FGL-safe, then</div>
<div>4) FGL-links can be introduced.</div><div><br></div><div><br></div><di=
v>----</div><div>Also, Tissa said:</div><div>&gt;&gt;what about some device=
s who are not &quot;safe FGL&quot; and so on, are we saying do not buy no s=
afe FGL devices, if you do that you cannot get FGL interop ?=A0</div>
<div><br></div><div><b><i><u>If</u></i></b> it&#39;s important to have VL g=
uys as transit, and they cannot be upgraded to FGL-safe because upgrading r=
equires a hardware change (it involves parsing data packets), then we are s=
tuck with option 3. =A0In option 3, we can forever have VL guys at the edge=
 speaking through an FGL-safe or full-FGL core. =A0In option 2, as I said, =
there is <i><u>no coexistence</u></i> of VL guys, <i><u>even at the edg</u>=
</i>e, once FGL links are introduced. =A0In option 2, a VL guy, even at the=
 edge, will be ostrasized completely once anyone advertises connectivity to=
 an FGL link.</div>
<div><br></div><div>I personally like option 2, because it&#39;s simple, bu=
t as Einstein said &quot;make things as simple as possible, but no simpler&=
quot;. =A0We have to make sure that everyone understands the implications o=
f option 2, and can live with them. =A0To summarize again (since there seem=
s to be some confusion about what the implications of option 2 are),</div>
<div><br></div><div>If we can upgrade all existing VL implementations to FG=
L-safe through software upgrade only, then option 2 provides smooth migrati=
on by changing ALL RBridges (edge as well as transit) to FGL-safe, and only=
 after ALL of them are upgraded, then FGL-labeled links can be introduced. =
=A0If there are any remaining VL guys once FGL-labeled links are introduced=
, the VL guys are ostrasized, <i><u>even if they are just at the edge.</u><=
/i></div>
<div><br></div><div>If we canNOT upgrade all existing VL to FGL-safe, AND i=
f it&#39;s important to be able to mix VL guys into a network with FGL fram=
es (say, having VL guys at the edge), then option 3 allows this.</div><div>
<br></div><div>So hopefully everyone understands the implications of option=
 2 fully before voting. =A0As I said, I don&#39;t know about whether there =
are VL chips that would require a hardware upgrade to become FGL-safe. =A0A=
ssuming there are none, then I&#39;m quite happy with option 2. =A0 And act=
ually, it looks like people are converging on option 2 (though I hope it&#3=
9;s based on fully understanding option 2). =A0Now is the time for people t=
o speak up, that know of problems this would cause...for instance, needing =
to have non-upgradable VL guys at the edge for a long time.</div>
<div><br></div><div>Radia</div><div><br></div><div><br></div></div></div>

--f46d0401fbc969e57604d49b6943--

From radiaperlman@gmail.com  Thu Jan 31 13:47:45 2013
Return-Path: <radiaperlman@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EBAC721F86AB for <trill@ietfa.amsl.com>; Thu, 31 Jan 2013 13:47:45 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.686
X-Spam-Level: 
X-Spam-Status: No, score=-2.686 tagged_above=-999 required=5 tests=[AWL=0.912,  BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id BoQitcItF2Mv for <trill@ietfa.amsl.com>; Thu, 31 Jan 2013 13:47:45 -0800 (PST)
Received: from mail-lb0-f177.google.com (mail-lb0-f177.google.com [209.85.217.177]) by ietfa.amsl.com (Postfix) with ESMTP id E8C5F21F867A for <trill@ietf.org>; Thu, 31 Jan 2013 13:47:44 -0800 (PST)
Received: by mail-lb0-f177.google.com with SMTP id go11so3814676lbb.8 for <trill@ietf.org>; Thu, 31 Jan 2013 13:47:43 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:x-received:date:message-id:subject:from:to:cc :content-type; bh=hKyqi/P9kS/ZBwOY9uOlZEPlUguapMk7DxCj57hcIxg=; b=fQHJPWWprm1c/xVOMW2FEFYRCvlmiTBEnpJYSlOQD3GIkKbc4rL5+eC8ZuKVMIEi5W kvOBQGiFlZCchpqZK31knQQOyI8DfVqUflyg1Pqai61UDT5tYEkQXq0GiOhPtsrODwzJ ZDHVKvYKqEVU7/8l1lMSNoH+Xu+vYCJYbfjjnq9DmSRfY/7c/eUH1brCrFSiJZwkUWNw n2VZQEjeQsvZBRne5Oy0CrqY6Ei7Gm+PRsqDxaezl7C10XyUUC5CCB8o5vJ8hxkQm8bf 2E/AX4rYLZ58wOq1NnuZ3xtO0W5r7FXLPljFQFo2+uIOm6iffkM8uVdAY/KV66L7Yh+x Zxfg==
MIME-Version: 1.0
X-Received: by 10.112.37.129 with SMTP id y1mr3924719lbj.17.1359668863843; Thu, 31 Jan 2013 13:47:43 -0800 (PST)
Received: by 10.112.64.17 with HTTP; Thu, 31 Jan 2013 13:47:43 -0800 (PST)
Date: Thu, 31 Jan 2013 13:47:43 -0800
Message-ID: <CAFOuuo5rHyAuQ1C6AWoYnkbG3T-if9cm8kS4MzWHdxS06Z6A1A@mail.gmail.com>
From: Radia Perlman <radiaperlman@gmail.com>
To: "Tissa Senevirathne (tsenevir)" <tsenevir@cisco.com>
Content-Type: multipart/alternative; boundary=e0cb4efe3306a43f5204d49c93bc
Cc: Anoop Ghanwani <anoop@alumni.duke.edu>, Ayan Banerjee <ayabaner@gmail.com>, Sam Aldrin <aldrin.ietf@gmail.com>, "trill@ietf.org" <trill@ietf.org>, Donald Eastlake <d3e3e3@gmail.com>, Jon Hudson <jon.hudson@gmail.com>
Subject: [trill] A tweak to option 2, if it's hard to upgrade VL guys to FGL-safe
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 31 Jan 2013 21:47:46 -0000

--e0cb4efe3306a43f5204d49c93bc
Content-Type: text/plain; charset=ISO-8859-1

 The other thread was getting a bit long, so I'm changing the subject line,
since here I'm proposing a tweak to option 2 that will make it less
restrictive.  To review, option 2 as originally proposed had this property:

>>With option 2, a VL guy cannot even be at the edge of a network with FGL
links unless it is upgraded to
>>FGL-safe AND advertises that it is FGL-safe.  Option 2 says a VL guy is
completely ostrasized once
>>FGL-labeled links are introduced.  There is NO coexistence of VL guys and
FGL-guys once FGL-links are
 >>advertised.

Here is the proposed tweak:

An FGL-safe guy has two choices of what to do with a VL neighbor (once FGL
links are advertised)

a) ostrasize the VL neighbor completely (as was in option 2)
b) advertise a very high cost to the VL neighbor AND ALSO discard any FGL
frames that would be forwarded to that VL neighbor.

If an FGL-safe guy is capable of differentiating between FGL and VL
frames...forwarding VL but discarding FGL to a VL neighbor, then this would
allow VL guys to be at the edge and successfully talk across an FGL core.

If an FGL-safe guy is not capable of differentiating between FGL and VL
frames, then the VL guy will be ostrasized.

As for transit...if a VL guy is in the core, its FGL neighbors will report
such high costs to it that if there is any FGL path, that will be the one
that is chosen, and if there isn't, then you're no worse off than if the VL
guy were completely ostrasized....it will look like there is a path, but
FGL packets will be lost on the path.

And of course, you shouldn't have the VL guy be the root of a multicast
tree, since link costs to it won't avoid it being on the path between FGL
guys.

Radia

--e0cb4efe3306a43f5204d49c93bc
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div class=3D"gmail_quote">
<div>The other thread was getting a bit long, so I&#39;m changing the subje=
ct line, since here I&#39;m proposing a tweak to option 2 that=A0will make =
it less restrictive.=A0 To review, option 2 as originally proposed had this=
 property:</div>

<div>=A0</div>
<div>&gt;&gt;With option 2, a VL guy cannot even be at the edge of a networ=
k with FGL links unless it is upgraded to </div>
<div>&gt;&gt;FGL-safe AND advertises that it is FGL-safe.=A0 Option 2 says =
a VL guy is completely ostrasized once </div>
<div>&gt;&gt;FGL-labeled links are introduced.=A0 There is NO coexistence o=
f VL guys and FGL-guys once=A0FGL-links are</div>
<div>=A0&gt;&gt;advertised. </div>
<div>=A0</div>
<div>Here is the proposed tweak:</div>
<div>=A0</div>
<div>An FGL-safe guy has two choices of what to do with a VL neighbor (once=
 FGL links are advertised)</div>
<div>=A0</div>
<div>a) ostrasize the VL neighbor completely (as was in option 2)</div>
<div>b) advertise a very high cost to the VL neighbor AND ALSO discard any =
FGL frames that would be forwarded to that VL neighbor.</div>
<div>=A0</div>
<div>If an FGL-safe guy is capable of differentiating between FGL and VL fr=
ames...forwarding VL but discarding FGL to a VL neighbor, then this would a=
llow VL guys to be at the edge and successfully talk across an FGL core.</d=
iv>

<div>=A0</div>
<div>If an FGL-safe guy is not capable of differentiating between FGL and V=
L frames, then the VL guy will be ostrasized.</div>
<div>=A0</div>
<div>As for transit...if a VL guy is in the core, its FGL neighbors will re=
port such high costs to it that if there is any FGL path, that will be the =
one that is chosen, and if there isn&#39;t, then you&#39;re no worse off th=
an if the VL guy were completely ostrasized....it will look like there is a=
 path, but FGL packets will be lost on the path.</div>

<div>=A0</div>
<div>And of course, you shouldn&#39;t have the VL guy be the root of a mult=
icast tree, since link costs to it won&#39;t avoid it being on the path bet=
ween FGL guys.</div>
<div>=A0</div>
<div>Radia</div>
<div>=A0</div>
<div>=A0</div></div>

--e0cb4efe3306a43f5204d49c93bc--

From ostokes@extremenetworks.com  Thu Jan 31 14:00:28 2013
Return-Path: <ostokes@extremenetworks.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3A12721F8505 for <trill@ietfa.amsl.com>; Thu, 31 Jan 2013 14:00:28 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.598
X-Spam-Level: 
X-Spam-Status: No, score=-2.598 tagged_above=-999 required=5 tests=[AWL=-0.000, BAYES_00=-2.599, HTML_MESSAGE=0.001]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id wPozOyJNqmIU for <trill@ietfa.amsl.com>; Thu, 31 Jan 2013 14:00:24 -0800 (PST)
Received: from ussc-casht-p1.extremenetworks.com (ussc-casht-p1.extremenetworks.com [207.179.9.62]) by ietfa.amsl.com (Postfix) with ESMTP id 80AB621F8440 for <trill@ietf.org>; Thu, 31 Jan 2013 14:00:24 -0800 (PST)
Received: from USEXCHANGE.corp.extremenetworks.com ([10.0.4.74]) by ussc-casht-p2.corp.extremenetworks.com ([10.255.181.88]) with mapi; Thu, 31 Jan 2013 14:00:24 -0800
From: Olen Stokes <ostokes@extremenetworks.com>
To: Radia Perlman <radiaperlman@gmail.com>, "Tissa Senevirathne (tsenevir)" <tsenevir@cisco.com>
Date: Thu, 31 Jan 2013 14:00:22 -0800
Thread-Topic: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
Thread-Index: Ac3/8RGYMXHAf2d5T9mBg4ykeRF8PgACGvPw
Message-ID: <A3C4E51A53661B4EBEE7C5F5E6FCDEB50276CA7B6A4D@USEXCHANGE.corp.extremenetworks.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <CA+-tSzx0E3J4i5Dvfbj4egruCeGi7z1se1Jn5RXtx3RedQLPdg@mail.gmail.com> <CAHD03N89RNX4=XxRZpD+Y2PCktYf-FtVjyYJ5oFS3dui3SsXaA@mail.gmail.com> <CAF4+nEG7YKGzGmObAMaOxackivjJsALymqHX=6PKBd5qU-Z+zQ@mail.gmail.com> <CA+-tSzz0=oh6iB1e-ESyeOx2sCvBFXeVpzuNbKmUVhRNLScDgQ@mail.gmail.com> <CAF4+nEFLrJL=CgX9e6on-cgz89ugmu93P6u5kmODhUia3H7RLg@mail.gmail.com> <1F5A81AE-4B3C-45FD-B431-9BA0011EB085@gmail.com> <DC8898F8-75C0-4535-BA61-C54279901332@gmail.com> <43DF00BE-B5F6-48B9-B313-8D179354C7B7@gmail.com> <A9B8F2D5-C3D7-4688-A76D-ED19EC47E59F@gmail.com> <FBEA3E19AA24F847BA3AE74E2FE19356288F6978@xmb-rcd-x08.cisco.com> <CAFOuuo4jZMdvUU_2EzBj1MLp6k0e+AmW1+k1ai4Q_m4rA0ObMw@mail.gmail.com>
In-Reply-To: <CAFOuuo4jZMdvUU_2EzBj1MLp6k0e+AmW1+k1ai4Q_m4rA0ObMw@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US
Content-Type: multipart/alternative; boundary="_000_A3C4E51A53661B4EBEE7C5F5E6FCDEB50276CA7B6A4DUSEXCHANGEc_"
MIME-Version: 1.0
Cc: Anoop Ghanwani <anoop@alumni.duke.edu>, Ayan Banerjee <ayabaner@gmail.com>, Sam Aldrin <aldrin.ietf@gmail.com>, "trill@ietf.org" <trill@ietf.org>, Donald Eastlake <d3e3e3@gmail.com>, Jon Hudson <jon.hudson@gmail.com>
Subject: Re: [trill] Explaining three options for upgrading to FGL	(fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 31 Jan 2013 22:00:28 -0000

--_000_A3C4E51A53661B4EBEE7C5F5E6FCDEB50276CA7B6A4DUSEXCHANGEc_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Thanks for consolidating these topics.  Please see two [OLS] comments in li=
ne...

From: trill-bounces@ietf.org [mailto:trill-bounces@ietf.org] On Behalf Of R=
adia Perlman
Sent: Thursday, January 31, 2013 3:24 PM
To: Tissa Senevirathne (tsenevir)
Cc: Anoop Ghanwani; Ayan Banerjee; Sam Aldrin; trill@ietf.org; Donald Eastl=
ake; Jon Hudson
Subject: Re: [trill] Explaining three options for upgrading to FGL (fine-gr=
ained-labeling)

Tissa said:
>> how can you make sure all of the so called "Safe FGL" have consistent be=
havior

"Safe FGL" need not all have the same behavior as each other...they just ha=
ve to do "nothing bad" with an FGL frame.  It's worth summarizing (again) w=
hat it means to be "FGL-safe", which allows you to coexist safely with FGL =
frames.  To be FGL-safe you must do the following:

a) unicast: you MUST NOT drop a packet because of seeing the unrecognized F=
GL tag.  You also MUST NOT get confused and do flow hashing improperly (cho=
osing different paths for packets for the same flow).

b) multicast pruning: it's OK not to do pruning at all, or to actually look=
 at all 24 bits of the FGL and prune as a proper FGL guy would do.  But an =
example of what you canNOT do is look at the first 12 bits of FGL, pretend =
it's the same as VL, and delete the multicast packet because nobody downstr=
eam is listening to that 12 bit VL value

[OLS]: I believe that there is still FGL-based pruning and multicast prunin=
g.  It should be OK to do FGL-based pruning and not do multicast pruning.  =
For FGL-based pruning, I agree that a FGL-safe node cannot look at the firs=
t 12 bits of FGL and just "pretend it's the same as VL".  However, it shoul=
d be possible for the control plane to determine all FGL's that have the sa=
me first 12 bits of FGL.  Such a control plane should be able to set up FGL=
-based pruning for each leg of a D-Tree based on the existence of listeners=
 downstream belonging to the collection of FGL's with the same first 12 bit=
s.
[/OLS]

c) egress of either unicast or multicast:  you MUST NOT egress an FGL frame=
 onto a link that is not supposed to get that FGL value

d) you have to advertise that you are FGL-safe...otherwise other RBs will a=
ssume you are VL ("FGL-unsafe")

---------
Also, TIssa said:
>>Do you agree if we say, FGL will always be core of the network, and VL wi=
ll be at the edge of the network ?

With option 2, a VL guy cannot even be at the edge of a network with FGL li=
nks unless it is upgraded to FGL-safe AND advertises that it is FGL-safe.  =
Option 2 says a VL guy is completely ostrasized once FGL-labeled links are =
introduced.  There is NO coexistence of VL guys and FGL-links. The only kin=
ds of coexistence in option 2 are:

a) VL and FGL-safe RBridge, but no actual FGL links yet.
b) FGL-safe guys, even if some of them don't really understand FGL, but jus=
t don't do "anything bad" with it, and FGL frames.

So...again...the migration phases in option 2 are
1) VL-only (today's deployment), then
2) A mixture of VL and FGL-safe, until
3) EVERYone upgraded to FGL-safe, then
4) FGL-links can be introduced.

[OLS]: Just to double check...  Following step #4 above, FGL-safe RBridges =
on the edge should still be able to send/receive VL frames.  This allows mi=
gration for devices on the edge that can be made FGL-safe but cannot ingres=
s/egress FGL frames.
[/OLS]

----
Also, Tissa said:
>>what about some devices who are not "safe FGL" and so on, are we saying d=
o not buy no safe FGL devices, if you do that you cannot get FGL interop ?

If it's important to have VL guys as transit, and they cannot be upgraded t=
o FGL-safe because upgrading requires a hardware change (it involves parsin=
g data packets), then we are stuck with option 3.  In option 3, we can fore=
ver have VL guys at the edge speaking through an FGL-safe or full-FGL core.=
  In option 2, as I said, there is no coexistence of VL guys, even at the e=
dge, once FGL links are introduced.  In option 2, a VL guy, even at the edg=
e, will be ostrasized completely once anyone advertises connectivity to an =
FGL link.

I personally like option 2, because it's simple, but as Einstein said "make=
 things as simple as possible, but no simpler".  We have to make sure that =
everyone understands the implications of option 2, and can live with them. =
 To summarize again (since there seems to be some confusion about what the =
implications of option 2 are),

If we can upgrade all existing VL implementations to FGL-safe through softw=
are upgrade only, then option 2 provides smooth migration by changing ALL R=
Bridges (edge as well as transit) to FGL-safe, and only after ALL of them a=
re upgraded, then FGL-labeled links can be introduced.  If there are any re=
maining VL guys once FGL-labeled links are introduced, the VL guys are ostr=
asized, even if they are just at the edge.

If we canNOT upgrade all existing VL to FGL-safe, AND if it's important to =
be able to mix VL guys into a network with FGL frames (say, having VL guys =
at the edge), then option 3 allows this.

So hopefully everyone understands the implications of option 2 fully before=
 voting.  As I said, I don't know about whether there are VL chips that wou=
ld require a hardware upgrade to become FGL-safe.  Assuming there are none,=
 then I'm quite happy with option 2.   And actually, it looks like people a=
re converging on option 2 (though I hope it's based on fully understanding =
option 2).  Now is the time for people to speak up, that know of problems t=
his would cause...for instance, needing to have non-upgradable VL guys at t=
he edge for a long time.

Radia



--_000_A3C4E51A53661B4EBEE7C5F5E6FCDEB50276CA7B6A4DUSEXCHANGEc_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micr=
osoft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" xmlns=3D"http:=
//www.w3.org/TR/REC-html40"><head><meta http-equiv=3DContent-Type content=
=3D"text/html; charset=3Dus-ascii"><meta name=3DGenerator content=3D"Micros=
oft Word 14 (filtered medium)"><style><!--
/* Font Definitions */
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:Tahoma;
	panose-1:2 11 6 4 3 5 4 4 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman","serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
span.EmailStyle17
	{mso-style-type:personal-reply;
	font-family:"Calibri","sans-serif";
	color:#1F497D;}
.MsoChpDefault
	{mso-style-type:export-only;
	font-family:"Calibri","sans-serif";}
@page WordSection1
	{size:8.5in 11.0in;
	margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
	{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]--></head><body lang=3DEN-US link=3Dblue vli=
nk=3Dpurple><div class=3DWordSection1><p class=3DMsoNormal><span style=3D'f=
ont-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>Thanks fo=
r consolidating these topics.&nbsp; Please see two [OLS] comments in line&#=
8230;<o:p></o:p></span></p><p class=3DMsoNormal><span style=3D'font-size:11=
.0pt;font-family:"Calibri","sans-serif";color:#1F497D'><o:p>&nbsp;</o:p></s=
pan></p><p class=3DMsoNormal><b><span style=3D'font-size:10.0pt;font-family=
:"Tahoma","sans-serif"'>From:</span></b><span style=3D'font-size:10.0pt;fon=
t-family:"Tahoma","sans-serif"'> trill-bounces@ietf.org [mailto:trill-bounc=
es@ietf.org] <b>On Behalf Of </b>Radia Perlman<br><b>Sent:</b> Thursday, Ja=
nuary 31, 2013 3:24 PM<br><b>To:</b> Tissa Senevirathne (tsenevir)<br><b>Cc=
:</b> Anoop Ghanwani; Ayan Banerjee; Sam Aldrin; trill@ietf.org; Donald Eas=
tlake; Jon Hudson<br><b>Subject:</b> Re: [trill] Explaining three options f=
or upgrading to FGL (fine-grained-labeling)<o:p></o:p></span></p><p class=
=3DMsoNormal><o:p>&nbsp;</o:p></p><p class=3DMsoNormal>Tissa said:<o:p></o:=
p></p><div><p class=3DMsoNormal>&gt;&gt;&nbsp;how can you make sure all of =
the so called &quot;Safe FGL&quot; have consistent behavior<o:p></o:p></p><=
div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p class=3DMsoNorm=
al>&quot;Safe FGL&quot; need not all have the <b><i>same</i></b> behavior a=
s each other...they just have to do &quot;nothing bad&quot; with an FGL fra=
me. &nbsp;It's worth summarizing (again) what it means to be &quot;FGL-safe=
&quot;, which allows you to coexist safely with FGL frames. &nbsp;To be FGL=
-safe you must do the following:<o:p></o:p></p></div><div><div><p class=3DM=
soNormal><o:p>&nbsp;</o:p></p></div><div><p class=3DMsoNormal>a) unicast: y=
ou MUST NOT drop a packet because of seeing the unrecognized FGL tag. &nbsp=
;You also MUST NOT get confused and do flow hashing improperly (choosing di=
fferent paths for packets for the same flow).<o:p></o:p></p></div><div><p c=
lass=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p class=3DMsoNormal>b) mu=
lticast pruning: it's OK not to do pruning at all, or to actually look at a=
ll 24 bits of the FGL and prune as a proper FGL guy would do. &nbsp;But an =
example of what you canNOT do is look at the first 12 bits of FGL, pretend =
it's the same as VL, and delete the multicast packet because nobody downstr=
eam is listening to that 12 bit VL value<o:p></o:p></p></div><div><p class=
=3DMsoNormal><span style=3D'color:#1F497D'><o:p>&nbsp;</o:p></span></p><p c=
lass=3DMsoNormal><span style=3D'font-size:11.0pt;font-family:"Calibri","san=
s-serif";color:#1F497D'>[OLS]: I believe that there is still FGL-based prun=
ing and multicast pruning.&nbsp; It should be OK to do FGL-based pruning an=
d not do multicast pruning.&nbsp; For FGL-based pruning, I agree that a FGL=
-safe node cannot look at the first 12 bits of FGL and just &#8220;pretend =
it&#8217;s the same as VL&#8221;.&nbsp; However, it should be possible for =
the control plane to determine all FGL&#8217;s that have the same first 12 =
bits of FGL.&nbsp; Such a control plane should be able to set up FGL-based =
pruning for each leg of a D-Tree based on the existence of listeners downst=
ream belonging to the collection of FGL&#8217;s with the same first 12 bits=
.&nbsp; </span><o:p></o:p></p><p class=3DMsoNormal><span style=3D'font-size=
:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>[/OLS]<o:p></o:p>=
</span></p><p class=3DMsoNormal><span style=3D'font-size:11.0pt;font-family=
:"Calibri","sans-serif";color:#1F497D'><o:p>&nbsp;</o:p></span></p></div><d=
iv><p class=3DMsoNormal>c) egress of either unicast or multicast: &nbsp;you=
 MUST NOT egress an FGL frame onto a link that is not supposed to get that =
FGL value<o:p></o:p></p></div><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></=
p></div><div><p class=3DMsoNormal>d) you have to advertise that you are FGL=
-safe...otherwise other RBs will assume you are VL (&quot;FGL-unsafe&quot;)=
<o:p></o:p></p></div><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><=
div><p class=3DMsoNormal>---------<o:p></o:p></p></div><div><p class=3DMsoN=
ormal>Also, TIssa said:<o:p></o:p></p></div><div><p class=3DMsoNormal>&gt;&=
gt;Do you agree if we say, FGL will always be core of the network, and VL w=
ill be at the edge of the network ?<o:p></o:p></p></div><div><p class=3DMso=
Normal><o:p>&nbsp;</o:p></p></div><div><p class=3DMsoNormal>With option 2, =
a VL guy <i><u>cannot even be at the edge of a network with FGL links</u></=
i> unless it is upgraded to FGL-safe AND advertises that it is FGL-safe. &n=
bsp;Option 2 says a VL guy is completely ostrasized once FGL-labeled links =
are introduced. &nbsp;There is NO coexistence of VL guys and FGL-links. The=
 only kinds of coexistence in option 2 are:<o:p></o:p></p></div><div><p cla=
ss=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p class=3DMsoNormal>a) VL a=
nd FGL-safe RBridge, but no actual FGL links yet.<o:p></o:p></p></div><div>=
<p class=3DMsoNormal>b) FGL-safe guys, even if some of them don't really un=
derstand FGL, but just don't do &quot;anything bad&quot; with it, and FGL f=
rames.<o:p></o:p></p></div><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p><=
/div><div><p class=3DMsoNormal>So...again...the migration phases in option =
2 are<o:p></o:p></p></div><div><p class=3DMsoNormal>1) VL-only (today's dep=
loyment), then<o:p></o:p></p></div><div><p class=3DMsoNormal>2) A mixture o=
f VL and FGL-safe, until<o:p></o:p></p></div><div><p class=3DMsoNormal>3) E=
VERYone upgraded to FGL-safe, then<o:p></o:p></p></div><div><p class=3DMsoN=
ormal>4) FGL-links can be introduced.<o:p></o:p></p></div><div><p class=3DM=
soNormal><span style=3D'color:#1F497D'><o:p>&nbsp;</o:p></span></p><p class=
=3DMsoNormal><span style=3D'font-size:11.0pt;font-family:"Calibri","sans-se=
rif";color:#1F497D'>[OLS]: Just to double check...&nbsp; Following step #4 =
above, FGL-safe RBridges on the edge should still be able to send/receive V=
L frames.&nbsp; This allows migration for devices on the edge that can be m=
ade FGL-safe but cannot ingress/egress FGL frames.&nbsp; <o:p></o:p></span>=
</p><p class=3DMsoNormal><span style=3D'font-size:11.0pt;font-family:"Calib=
ri","sans-serif";color:#1F497D'>[/OLS]<o:p></o:p></span></p></div><div><p c=
lass=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p class=3DMsoNormal>----<=
o:p></o:p></p></div><div><p class=3DMsoNormal>Also, Tissa said:<o:p></o:p><=
/p></div><div><p class=3DMsoNormal>&gt;&gt;what about some devices who are =
not &quot;safe FGL&quot; and so on, are we saying do not buy no safe FGL de=
vices, if you do that you cannot get FGL interop ?&nbsp;<o:p></o:p></p></di=
v><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p class=3DMsoN=
ormal><b><i><u>If</u></i></b> it's important to have VL guys as transit, an=
d they cannot be upgraded to FGL-safe because upgrading requires a hardware=
 change (it involves parsing data packets), then we are stuck with option 3=
. &nbsp;In option 3, we can forever have VL guys at the edge speaking throu=
gh an FGL-safe or full-FGL core. &nbsp;In option 2, as I said, there is <i>=
<u>no coexistence</u></i> of VL guys, <i><u>even at the edg</u></i>e, once =
FGL links are introduced. &nbsp;In option 2, a VL guy, even at the edge, wi=
ll be ostrasized completely once anyone advertises connectivity to an FGL l=
ink.<o:p></o:p></p></div><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></d=
iv><div><p class=3DMsoNormal>I personally like option 2, because it's simpl=
e, but as Einstein said &quot;make things as simple as possible, but no sim=
pler&quot;. &nbsp;We have to make sure that everyone understands the implic=
ations of option 2, and can live with them. &nbsp;To summarize again (since=
 there seems to be some confusion about what the implications of option 2 a=
re),<o:p></o:p></p></div><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></d=
iv><div><p class=3DMsoNormal>If we can upgrade all existing VL implementati=
ons to FGL-safe through software upgrade only, then option 2 provides smoot=
h migration by changing ALL RBridges (edge as well as transit) to FGL-safe,=
 and only after ALL of them are upgraded, then FGL-labeled links can be int=
roduced. &nbsp;If there are any remaining VL guys once FGL-labeled links ar=
e introduced, the VL guys are ostrasized, <i><u>even if they are just at th=
e edge.</u></i><o:p></o:p></p></div><div><p class=3DMsoNormal><o:p>&nbsp;</=
o:p></p></div><div><p class=3DMsoNormal>If we canNOT upgrade all existing V=
L to FGL-safe, AND if it's important to be able to mix VL guys into a netwo=
rk with FGL frames (say, having VL guys at the edge), then option 3 allows =
this.<o:p></o:p></p></div><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></=
div><div><p class=3DMsoNormal>So hopefully everyone understands the implica=
tions of option 2 fully before voting. &nbsp;As I said, I don't know about =
whether there are VL chips that would require a hardware upgrade to become =
FGL-safe. &nbsp;Assuming there are none, then I'm quite happy with option 2=
. &nbsp; And actually, it looks like people are converging on option 2 (tho=
ugh I hope it's based on fully understanding option 2). &nbsp;Now is the ti=
me for people to speak up, that know of problems this would cause...for ins=
tance, needing to have non-upgradable VL guys at the edge for a long time.<=
o:p></o:p></p></div><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><d=
iv><p class=3DMsoNormal>Radia<o:p></o:p></p></div><div><p class=3DMsoNormal=
><o:p>&nbsp;</o:p></p></div><div><p class=3DMsoNormal><o:p>&nbsp;</o:p></p>=
</div></div></div></div></body></html>=

--_000_A3C4E51A53661B4EBEE7C5F5E6FCDEB50276CA7B6A4DUSEXCHANGEc_--

From radiaperlman@gmail.com  Thu Jan 31 14:27:15 2013
Return-Path: <radiaperlman@gmail.com>
X-Original-To: trill@ietfa.amsl.com
Delivered-To: trill@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1F03021F845F for <trill@ietfa.amsl.com>; Thu, 31 Jan 2013 14:27:15 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.999
X-Spam-Level: 
X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, J_CHICKENPOX_31=0.6, NO_RELAYS=-0.001]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id urQ8VV+a277G for <trill@ietfa.amsl.com>; Thu, 31 Jan 2013 14:27:14 -0800 (PST)
Received: from mail-la0-x234.google.com (la-in-x0234.1e100.net [IPv6:2a00:1450:4010:c03::234]) by ietfa.amsl.com (Postfix) with ESMTP id 395B321F8467 for <trill@ietf.org>; Thu, 31 Jan 2013 14:27:12 -0800 (PST)
Received: by mail-la0-f52.google.com with SMTP id fs12so2302808lab.25 for <trill@ietf.org>; Thu, 31 Jan 2013 14:27:12 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:x-received:in-reply-to:references:date:message-id :subject:from:to:cc:content-type; bh=bxRAAuowylBVfWiBkIZYnmvSY5vma2o/IpTESTZTtmE=; b=TddmCvdWzlJ+IWdYxVkv/kPkFI/h+p4x7IzYzP/LkCXR+fNioBWrC4yRmoVYY5sqXo y95jCCZV1RgXuQis1qXOIUVXKw/bwBB/SCOAQi+0T9qPdSGZ5JzFzG6crtzpOd+4uo/d 7my1o3rTr27EZl8ZDOoWBDprjZ488pKGIAZgDUbsNrIFfNgHL+PlwFoPe+F8rMIbl11t LWhMfcZg7jy92fMT55nhaLZdmch8WGvZdeKoj/AG689612enahMaUSna9VIXEJ8j2eD/ f9VuV8gHqB73fmNnTWm9p33xB0tvlsfOmNv94WobWF2dE0zLO1kKaE8bJKD3f7QrUGak pSHA==
MIME-Version: 1.0
X-Received: by 10.152.121.212 with SMTP id lm20mr9204767lab.42.1359671232014;  Thu, 31 Jan 2013 14:27:12 -0800 (PST)
Received: by 10.112.64.17 with HTTP; Thu, 31 Jan 2013 14:27:11 -0800 (PST)
In-Reply-To: <A3C4E51A53661B4EBEE7C5F5E6FCDEB50276CA7B6A4D@USEXCHANGE.corp.extremenetworks.com>
References: <CAFOuuo67iBR2JkDtwPFCkrKHs3Fp4U_L1jkmNz5sfht6y05YMw@mail.gmail.com> <CA+-tSzx0E3J4i5Dvfbj4egruCeGi7z1se1Jn5RXtx3RedQLPdg@mail.gmail.com> <CAHD03N89RNX4=XxRZpD+Y2PCktYf-FtVjyYJ5oFS3dui3SsXaA@mail.gmail.com> <CAF4+nEG7YKGzGmObAMaOxackivjJsALymqHX=6PKBd5qU-Z+zQ@mail.gmail.com> <CA+-tSzz0=oh6iB1e-ESyeOx2sCvBFXeVpzuNbKmUVhRNLScDgQ@mail.gmail.com> <CAF4+nEFLrJL=CgX9e6on-cgz89ugmu93P6u5kmODhUia3H7RLg@mail.gmail.com> <1F5A81AE-4B3C-45FD-B431-9BA0011EB085@gmail.com> <DC8898F8-75C0-4535-BA61-C54279901332@gmail.com> <43DF00BE-B5F6-48B9-B313-8D179354C7B7@gmail.com> <A9B8F2D5-C3D7-4688-A76D-ED19EC47E59F@gmail.com> <FBEA3E19AA24F847BA3AE74E2FE19356288F6978@xmb-rcd-x08.cisco.com> <CAFOuuo4jZMdvUU_2EzBj1MLp6k0e+AmW1+k1ai4Q_m4rA0ObMw@mail.gmail.com> <A3C4E51A53661B4EBEE7C5F5E6FCDEB50276CA7B6A4D@USEXCHANGE.corp.extremenetworks.com>
Date: Thu, 31 Jan 2013 14:27:11 -0800
Message-ID: <CAFOuuo7=f14E-dGS09Ti=uOq9Ta0OPCJQTVbhdkcmREaxO1Uhw@mail.gmail.com>
From: Radia Perlman <radiaperlman@gmail.com>
To: Olen Stokes <ostokes@extremenetworks.com>
Content-Type: multipart/alternative; boundary=f46d04374ee1cbaa8b04d49d200a
Cc: Anoop Ghanwani <anoop@alumni.duke.edu>, Ayan Banerjee <ayabaner@gmail.com>, Sam Aldrin <aldrin.ietf@gmail.com>, Donald Eastlake <d3e3e3@gmail.com>, "trill@ietf.org" <trill@ietf.org>, "Tissa Senevirathne \(tsenevir\)" <tsenevir@cisco.com>, Jon Hudson <jon.hudson@gmail.com>
Subject: Re: [trill] Explaining three options for upgrading to FGL (fine-grained-labeling)
X-BeenThere: trill@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Developing a hybrid router/bridge." <trill.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/trill>, <mailto:trill-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/trill>
List-Post: <mailto:trill@ietf.org>
List-Help: <mailto:trill-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/trill>, <mailto:trill-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 31 Jan 2013 22:27:15 -0000

--f46d04374ee1cbaa8b04d49d200a
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

Hi Olen,

Re your first comment:  I was not intending to differentiate pruning based
on IP multicast vs FGL...I should have said "multidestination" instead of
"multicast".  So we are not disagreeing on anything here...you just
interpreted my word "multicast" differently than I intended.  Sorry for the
confusion.

Re your second comment:

[OLS]: Just to double check...  Following step #4 above, FGL-safe RBridges
on the edge should still be able to send/receive VL frames.  This allows
migration for devices on the edge that can be made FGL-safe but cannot
ingress/egress FGL frames.  ****

[/OLS]



Yes...that is true.  It is fine for an FGL-safe guy not to be quite smart
enough to ingress/egress FGL.  All it has to do is to be able to follow all
the rules for "FGL-safe".  But....see the new thread...there's a proposed
tweak to option 2 that even allows a VL-guy to coexist, so long as FGL guys
do a bit more work....see the description in the "tweak" thread.



Radia




On Thu, Jan 31, 2013 at 2:00 PM, Olen Stokes <ostokes@extremenetworks.com>w=
rote:

>  Thanks for consolidating these topics.  Please see two [OLS] comments in
> line=85****
>
> ** **
>
> *From:* trill-bounces@ietf.org [mailto:trill-bounces@ietf.org] *On Behalf
> Of *Radia Perlman
> *Sent:* Thursday, January 31, 2013 3:24 PM
> *To:* Tissa Senevirathne (tsenevir)
> *Cc:* Anoop Ghanwani; Ayan Banerjee; Sam Aldrin; trill@ietf.org; Donald
> Eastlake; Jon Hudson
>
> *Subject:* Re: [trill] Explaining three options for upgrading to FGL
> (fine-grained-labeling)****
>
> ** **
>
> Tissa said:****
>
> >> how can you make sure all of the so called "Safe FGL" have consistent
> behavior****
>
> ** **
>
> "Safe FGL" need not all have the *same* behavior as each other...they
> just have to do "nothing bad" with an FGL frame.  It's worth summarizing
> (again) what it means to be "FGL-safe", which allows you to coexist safel=
y
> with FGL frames.  To be FGL-safe you must do the following:****
>
> ** **
>
> a) unicast: you MUST NOT drop a packet because of seeing the unrecognized
> FGL tag.  You also MUST NOT get confused and do flow hashing improperly
> (choosing different paths for packets for the same flow).****
>
> ** **
>
> b) multicast pruning: it's OK not to do pruning at all, or to actually
> look at all 24 bits of the FGL and prune as a proper FGL guy would do.  B=
ut
> an example of what you canNOT do is look at the first 12 bits of FGL,
> pretend it's the same as VL, and delete the multicast packet because nobo=
dy
> downstream is listening to that 12 bit VL value****
>
> ** **
>
> [OLS]: I believe that there is still FGL-based pruning and multicast
> pruning.  It should be OK to do FGL-based pruning and not do multicast
> pruning.  For FGL-based pruning, I agree that a FGL-safe node cannot look
> at the first 12 bits of FGL and just =93pretend it=92s the same as VL=94.
> However, it should be possible for the control plane to determine all FGL=
=92s
> that have the same first 12 bits of FGL.  Such a control plane should be
> able to set up FGL-based pruning for each leg of a D-Tree based on the
> existence of listeners downstream belonging to the collection of FGL=92s =
with
> the same first 12 bits.  ****
>
> [/OLS]****
>
> ** **
>
> c) egress of either unicast or multicast:  you MUST NOT egress an FGL
> frame onto a link that is not supposed to get that FGL value****
>
> ** **
>
> d) you have to advertise that you are FGL-safe...otherwise other RBs will
> assume you are VL ("FGL-unsafe")****
>
> ** **
>
> ---------****
>
> Also, TIssa said:****
>
> >>Do you agree if we say, FGL will always be core of the network, and VL
> will be at the edge of the network ?****
>
> ** **
>
> With option 2, a VL guy *cannot even be at the edge of a network with FGL
> links* unless it is upgraded to FGL-safe AND advertises that it is
> FGL-safe.  Option 2 says a VL guy is completely ostrasized once FGL-label=
ed
> links are introduced.  There is NO coexistence of VL guys and FGL-links.
> The only kinds of coexistence in option 2 are:****
>
> ** **
>
> a) VL and FGL-safe RBridge, but no actual FGL links yet.****
>
> b) FGL-safe guys, even if some of them don't really understand FGL, but
> just don't do "anything bad" with it, and FGL frames.****
>
> ** **
>
> So...again...the migration phases in option 2 are****
>
> 1) VL-only (today's deployment), then****
>
> 2) A mixture of VL and FGL-safe, until****
>
> 3) EVERYone upgraded to FGL-safe, then****
>
> 4) FGL-links can be introduced.****
>
> ** **
>
> [OLS]: Just to double check...  Following step #4 above, FGL-safe RBridge=
s
> on the edge should still be able to send/receive VL frames.  This allows
> migration for devices on the edge that can be made FGL-safe but cannot
> ingress/egress FGL frames.  ****
>
> [/OLS]****
>
> ** **
>
> ----****
>
> Also, Tissa said:****
>
> >>what about some devices who are not "safe FGL" and so on, are we saying
> do not buy no safe FGL devices, if you do that you cannot get FGL interop=
 ?
> ****
>
> ** **
>
> *If* it's important to have VL guys as transit, and they cannot be
> upgraded to FGL-safe because upgrading requires a hardware change (it
> involves parsing data packets), then we are stuck with option 3.  In opti=
on
> 3, we can forever have VL guys at the edge speaking through an FGL-safe o=
r
> full-FGL core.  In option 2, as I said, there is *no coexistence* of VL
> guys, *even at the edg*e, once FGL links are introduced.  In option 2, a
> VL guy, even at the edge, will be ostrasized completely once anyone
> advertises connectivity to an FGL link.****
>
> ** **
>
> I personally like option 2, because it's simple, but as Einstein said
> "make things as simple as possible, but no simpler".  We have to make sur=
e
> that everyone understands the implications of option 2, and can live with
> them.  To summarize again (since there seems to be some confusion about
> what the implications of option 2 are),****
>
> ** **
>
> If we can upgrade all existing VL implementations to FGL-safe through
> software upgrade only, then option 2 provides smooth migration by changin=
g
> ALL RBridges (edge as well as transit) to FGL-safe, and only after ALL of
> them are upgraded, then FGL-labeled links can be introduced.  If there ar=
e
> any remaining VL guys once FGL-labeled links are introduced, the VL guys
> are ostrasized, *even if they are just at the edge.*****
>
> ** **
>
> If we canNOT upgrade all existing VL to FGL-safe, AND if it's important t=
o
> be able to mix VL guys into a network with FGL frames (say, having VL guy=
s
> at the edge), then option 3 allows this.****
>
> ** **
>
> So hopefully everyone understands the implications of option 2 fully
> before voting.  As I said, I don't know about whether there are VL chips
> that would require a hardware upgrade to become FGL-safe.  Assuming there
> are none, then I'm quite happy with option 2.   And actually, it looks li=
ke
> people are converging on option 2 (though I hope it's based on fully
> understanding option 2).  Now is the time for people to speak up, that kn=
ow
> of problems this would cause...for instance, needing to have non-upgradab=
le
> VL guys at the edge for a long time.****
>
> ** **
>
> Radia****
>
> ** **
>
> ** **
>

--f46d04374ee1cbaa8b04d49d200a
Content-Type: text/html; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

<div>Hi Olen,</div>
<div>=A0</div>
<div>Re your first comment:=A0 I was not intending to differentiate pruning=
 based on IP multicast vs FGL...I should have said &quot;multidestination&q=
uot; instead of &quot;multicast&quot;.=A0 So we are not disagreeing on anyt=
hing here...you just interpreted my word &quot;multicast&quot; differently =
than I intended.=A0 Sorry for the confusion.</div>

<div>=A0</div>
<div>Re your second comment:<br><br>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY:&#39;Calibri&#39;,&#39;sa=
ns-serif&#39;;COLOR:#1f497d;FONT-SIZE:11pt">[OLS]: Just to double check...=
=A0 Following step #4 above, FGL-safe RBridges on the edge should still be =
able to send/receive VL frames.=A0 This allows migration for devices on the=
 edge that can be made FGL-safe but cannot ingress/egress FGL frames.=A0 <u=
></u><u></u></span></p>

<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY:&#39;Calibri&#39;,&#39;sa=
ns-serif&#39;;COLOR:#1f497d;FONT-SIZE:11pt">[/OLS]</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY:&#39;Calibri&#39;,&#39;sa=
ns-serif&#39;;COLOR:#1f497d;FONT-SIZE:11pt"></span>=A0</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY:&#39;Calibri&#39;,&#39;sa=
ns-serif&#39;;COLOR:#1f497d;FONT-SIZE:11pt">Yes...that is true.=A0 It is fi=
ne for an FGL-safe guy not to be quite smart enough to ingress/egress FGL.=
=A0 All it has to do is to be able to follow all the rules for &quot;FGL-sa=
fe&quot;.=A0 But....see the new thread...there&#39;s a proposed tweak to op=
tion 2 that even allows a VL-guy to coexist, so long as FGL guys do a bit m=
ore work....see the description in the &quot;tweak&quot; thread.</span></p>

<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY:&#39;Calibri&#39;,&#39;sa=
ns-serif&#39;;COLOR:#1f497d;FONT-SIZE:11pt"></span>=A0</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY:&#39;Calibri&#39;,&#39;sa=
ns-serif&#39;;COLOR:#1f497d;FONT-SIZE:11pt">Radia</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY:&#39;Calibri&#39;,&#39;sa=
ns-serif&#39;;COLOR:#1f497d;FONT-SIZE:11pt"></span>=A0</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY:&#39;Calibri&#39;,&#39;sa=
ns-serif&#39;;COLOR:#1f497d;FONT-SIZE:11pt"></span>=A0</p></div>
<div class=3D"gmail_quote">On Thu, Jan 31, 2013 at 2:00 PM, Olen Stokes <sp=
an dir=3D"ltr">&lt;<a href=3D"mailto:ostokes@extremenetworks.com" target=3D=
"_blank">ostokes@extremenetworks.com</a>&gt;</span> wrote:<br>
<blockquote style=3D"BORDER-LEFT:#ccc 1px solid;MARGIN:0px 0px 0px 0.8ex;PA=
DDING-LEFT:1ex" class=3D"gmail_quote">
<div lang=3D"EN-US" vlink=3D"purple" link=3D"blue">
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY:&#39;Calibri&#39;,&#39;sa=
ns-serif&#39;;COLOR:#1f497d;FONT-SIZE:11pt">Thanks for consolidating these =
topics.=A0 Please see two [OLS] comments in line=85<u></u><u></u></span></p=
>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY:&#39;Calibri&#39;,&#39;sa=
ns-serif&#39;;COLOR:#1f497d;FONT-SIZE:11pt"><u></u>=A0<u></u></span></p>
<p class=3D"MsoNormal"><b><span style=3D"FONT-FAMILY:&#39;Tahoma&#39;,&#39;=
sans-serif&#39;;FONT-SIZE:10pt">From:</span></b><span style=3D"FONT-FAMILY:=
&#39;Tahoma&#39;,&#39;sans-serif&#39;;FONT-SIZE:10pt"> <a href=3D"mailto:tr=
ill-bounces@ietf.org" target=3D"_blank">trill-bounces@ietf.org</a> [mailto:=
<a href=3D"mailto:trill-bounces@ietf.org" target=3D"_blank">trill-bounces@i=
etf.org</a>] <b>On Behalf Of </b>Radia Perlman<br>
<b>Sent:</b> Thursday, January 31, 2013 3:24 PM<br><b>To:</b> Tissa Senevir=
athne (tsenevir)<br><b>Cc:</b> Anoop Ghanwani; Ayan Banerjee; Sam Aldrin; <=
a href=3D"mailto:trill@ietf.org" target=3D"_blank">trill@ietf.org</a>; Dona=
ld Eastlake; Jon Hudson=20
<div class=3D"im"><br><b>Subject:</b> Re: [trill] Explaining three options =
for upgrading to FGL (fine-grained-labeling)<u></u><u></u></div></span>
<p></p>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p>
<p class=3D"MsoNormal">Tissa said:<u></u><u></u></p>
<div>
<div class=3D"im">
<p class=3D"MsoNormal">&gt;&gt;=A0how can you make sure all of the so calle=
d &quot;Safe FGL&quot; have consistent behavior<u></u><u></u></p>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p></div>
<div>
<p class=3D"MsoNormal">&quot;Safe FGL&quot; need not all have the <b><i>sam=
e</i></b> behavior as each other...they just have to do &quot;nothing bad&q=
uot; with an FGL frame. =A0It&#39;s worth summarizing (again) what it means=
 to be &quot;FGL-safe&quot;, which allows you to coexist safely with FGL fr=
ames. =A0To be FGL-safe you must do the following:<u></u><u></u></p>
</div></div>
<div>
<div class=3D"im">
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p></div>
<div>
<p class=3D"MsoNormal">a) unicast: you MUST NOT drop a packet because of se=
eing the unrecognized FGL tag. =A0You also MUST NOT get confused and do flo=
w hashing improperly (choosing different paths for packets for the same flo=
w).<u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p></div>
<div>
<p class=3D"MsoNormal">b) multicast pruning: it&#39;s OK not to do pruning =
at all, or to actually look at all 24 bits of the FGL and prune as a proper=
 FGL guy would do. =A0But an example of what you canNOT do is look at the f=
irst 12 bits of FGL, pretend it&#39;s the same as VL, and delete the multic=
ast packet because nobody downstream is listening to that 12 bit VL value<u=
></u><u></u></p>
</div></div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR:#1f497d"><u></u>=A0<u></u></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY:&#39;Calibri&#39;,&#39;sa=
ns-serif&#39;;COLOR:#1f497d;FONT-SIZE:11pt">[OLS]: I believe that there is =
still FGL-based pruning and multicast pruning.=A0 It should be OK to do FGL=
-based pruning and not do multicast pruning.=A0 For FGL-based pruning, I ag=
ree that a FGL-safe node cannot look at the first 12 bits of FGL and just =
=93pretend it=92s the same as VL=94.=A0 However, it should be possible for =
the control plane to determine all FGL=92s that have the same first 12 bits=
 of FGL.=A0 Such a control plane should be able to set up FGL-based pruning=
 for each leg of a D-Tree based on the existence of listeners downstream be=
longing to the collection of FGL=92s with the same first 12 bits.=A0 </span=
><u></u><u></u></p>

<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY:&#39;Calibri&#39;,&#39;sa=
ns-serif&#39;;COLOR:#1f497d;FONT-SIZE:11pt">[/OLS]<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY:&#39;Calibri&#39;,&#39;sa=
ns-serif&#39;;COLOR:#1f497d;FONT-SIZE:11pt"><u></u>=A0<u></u></span></p></d=
iv>
<div class=3D"im">
<div>
<p class=3D"MsoNormal">c) egress of either unicast or multicast: =A0you MUS=
T NOT egress an FGL frame onto a link that is not supposed to get that FGL =
value<u></u><u></u></p></div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p></div>
<div>
<p class=3D"MsoNormal">d) you have to advertise that you are FGL-safe...oth=
erwise other RBs will assume you are VL (&quot;FGL-unsafe&quot;)<u></u><u><=
/u></p></div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p></div>
<div>
<p class=3D"MsoNormal">---------<u></u><u></u></p></div>
<div>
<p class=3D"MsoNormal">Also, TIssa said:<u></u><u></u></p></div>
<div>
<p class=3D"MsoNormal">&gt;&gt;Do you agree if we say, FGL will always be c=
ore of the network, and VL will be at the edge of the network ?<u></u><u></=
u></p></div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p></div>
<div>
<p class=3D"MsoNormal">With option 2, a VL guy <i><u>cannot even be at the =
edge of a network with FGL links</u></i> unless it is upgraded to FGL-safe =
AND advertises that it is FGL-safe. =A0Option 2 says a VL guy is completely=
 ostrasized once FGL-labeled links are introduced. =A0There is NO coexisten=
ce of VL guys and FGL-links. The only kinds of coexistence in option 2 are:=
<u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p></div>
<div>
<p class=3D"MsoNormal">a) VL and FGL-safe RBridge, but no actual FGL links =
yet.<u></u><u></u></p></div>
<div>
<p class=3D"MsoNormal">b) FGL-safe guys, even if some of them don&#39;t rea=
lly understand FGL, but just don&#39;t do &quot;anything bad&quot; with it,=
 and FGL frames.<u></u><u></u></p></div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p></div>
<div>
<p class=3D"MsoNormal">So...again...the migration phases in option 2 are<u>=
</u><u></u></p></div>
<div>
<p class=3D"MsoNormal">1) VL-only (today&#39;s deployment), then<u></u><u><=
/u></p></div>
<div>
<p class=3D"MsoNormal">2) A mixture of VL and FGL-safe, until<u></u><u></u>=
</p></div>
<div>
<p class=3D"MsoNormal">3) EVERYone upgraded to FGL-safe, then<u></u><u></u>=
</p></div>
<div>
<p class=3D"MsoNormal">4) FGL-links can be introduced.<u></u><u></u></p></d=
iv></div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR:#1f497d"><u></u>=A0<u></u></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY:&#39;Calibri&#39;,&#39;sa=
ns-serif&#39;;COLOR:#1f497d;FONT-SIZE:11pt">[OLS]: Just to double check...=
=A0 Following step #4 above, FGL-safe RBridges on the edge should still be =
able to send/receive VL frames.=A0 This allows migration for devices on the=
 edge that can be made FGL-safe but cannot ingress/egress FGL frames.=A0 <u=
></u><u></u></span></p>

<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY:&#39;Calibri&#39;,&#39;sa=
ns-serif&#39;;COLOR:#1f497d;FONT-SIZE:11pt">[/OLS]<u></u><u></u></span></p>=
</div>
<div class=3D"im">
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p></div>
<div>
<p class=3D"MsoNormal">----<u></u><u></u></p></div>
<div>
<p class=3D"MsoNormal">Also, Tissa said:<u></u><u></u></p></div>
<div>
<p class=3D"MsoNormal">&gt;&gt;what about some devices who are not &quot;sa=
fe FGL&quot; and so on, are we saying do not buy no safe FGL devices, if yo=
u do that you cannot get FGL interop ?=A0<u></u><u></u></p></div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p></div>
<div>
<p class=3D"MsoNormal"><b><i><u>If</u></i></b> it&#39;s important to have V=
L guys as transit, and they cannot be upgraded to FGL-safe because upgradin=
g requires a hardware change (it involves parsing data packets), then we ar=
e stuck with option 3. =A0In option 3, we can forever have VL guys at the e=
dge speaking through an FGL-safe or full-FGL core. =A0In option 2, as I sai=
d, there is <i><u>no coexistence</u></i> of VL guys, <i><u>even at the edg<=
/u></i>e, once FGL links are introduced. =A0In option 2, a VL guy, even at =
the edge, will be ostrasized completely once anyone advertises connectivity=
 to an FGL link.<u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p></div>
<div>
<p class=3D"MsoNormal">I personally like option 2, because it&#39;s simple,=
 but as Einstein said &quot;make things as simple as possible, but no simpl=
er&quot;. =A0We have to make sure that everyone understands the implication=
s of option 2, and can live with them. =A0To summarize again (since there s=
eems to be some confusion about what the implications of option 2 are),<u><=
/u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p></div>
<div>
<p class=3D"MsoNormal">If we can upgrade all existing VL implementations to=
 FGL-safe through software upgrade only, then option 2 provides smooth migr=
ation by changing ALL RBridges (edge as well as transit) to FGL-safe, and o=
nly after ALL of them are upgraded, then FGL-labeled links can be introduce=
d. =A0If there are any remaining VL guys once FGL-labeled links are introdu=
ced, the VL guys are ostrasized, <i><u>even if they are just at the edge.</=
u></i><u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p></div>
<div>
<p class=3D"MsoNormal">If we canNOT upgrade all existing VL to FGL-safe, AN=
D if it&#39;s important to be able to mix VL guys into a network with FGL f=
rames (say, having VL guys at the edge), then option 3 allows this.<u></u><=
u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p></div>
<div>
<p class=3D"MsoNormal">So hopefully everyone understands the implications o=
f option 2 fully before voting. =A0As I said, I don&#39;t know about whethe=
r there are VL chips that would require a hardware upgrade to become FGL-sa=
fe. =A0Assuming there are none, then I&#39;m quite happy with option 2. =A0=
 And actually, it looks like people are converging on option 2 (though I ho=
pe it&#39;s based on fully understanding option 2). =A0Now is the time for =
people to speak up, that know of problems this would cause...for instance, =
needing to have non-upgradable VL guys at the edge for a long time.<u></u><=
u></u></p>
</div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p></div>
<div>
<p class=3D"MsoNormal">Radia<u></u><u></u></p></div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p></div>
<div>
<p class=3D"MsoNormal"><u></u>=A0<u></u></p></div></div></div></div></p></d=
iv></div></blockquote></div><br>

--f46d04374ee1cbaa8b04d49d200a--
